Drebrin 1 in dendritic cells regulates phagocytosis and cell surface receptor expression through recycling for efficient antigen presentation

Summary

Phagocytosis, macropinocytosis and antigen presentation by dendritic cells (DC) requires reorganization of the actin cytoskeleton. Drebrin (Dbn1) is an actin binding and stabilizing protein with roles in endocytosis, formation of dendrite spines in neurons and coordinating cell–cell synapses in immune cells. However, its role in DC phagocytosis and antigen presentation is unknown. These studies now report that silencing of Dbn1 in DC resulted in restrained cell surface display of receptors, most notably MHC class I and II and co‐stimulatory molecules. This, as expected, resulted in impaired antigen‐specific T‐cell activation and proliferation. Studies additionally revealed that knockdown of Dbn1 in DC impaired macropinocytosis and phagocytosis. However, there was a concomitant increase in fluid‐phase uptake, suggesting that Dbn1 is responsible for the differential control of macropinocytosis versus micropinocytosis activities. Taken together, these findings now reveal that Dbn1 plays a major role in coordinating the actin cytoskeletal activities responsible for antigen presentation in DC.

Abbreviations

DC

dendritic cell

Dbn1

Drebrin 1

F‐actin

filamentous actin

oligos

oligonucleotides

siControl

siRNA scrambled control oligonucleotides transfected into DC

siDbn1

siRNA targeted Dbn1 knockdown in DC

Introduction

Dendritic cells (DC) are antigen‐presenting cells capable of initiating and sustaining adaptive immune responses. This capacity is largely due to their innate ability to capture antigen by phagocytosis, receptor‐mediated endocytosis, and/or pinocytosis in peripheral tissues. Upon antigen uptake, DC process the protein into peptide fragments for loading onto MHC class molecules. The cells then extravasate out of tissues for migration into draining lymph nodes to present antigens to responder T cells. Efficient priming requires the responder T cell to recognize the correct peptide presented by DC in conjunction with co‐stimulatory and adhesion molecule engagement.1

Reorientation, bundling, and assembly of the actin cytoskeleton can support diverse cellular processes. These activities are required to employ mechanical work to drive cell motility, macropinocytosis, endocytic recycling, mechanosensation, and scaffold assembly of signaling molecules. The coordinated assembly and disassembly of actin are required for efficient phagocytosis. This process is regulated through signaling events and actin regulatory proteins such as the Rho‐family GTPases and calcium‐responsive actin‐bundling molecules.2, 3, 4, 5 Actin cytoskeletal remodeling is notably important for DC migration and engagement with cognate T cells (through formation of the immunological synapse) for efficient antigen presentation.6, 7

Developmentally regulated brain protein [Drebrin 1 (Dbn1)] is an F‐actin binding and remodeling protein that directly competes against fascin, α‐actinin, and tropomyosin for access to actin filaments.8 The protein plays a significant role in increasing filament stiffness through direct binding interactions.9 Dbn1 therefore has a critical role in neuronal dendrite cell spine morphogenesis, synaptogenesis, synaptic plasticity, cell migration, communication, and differentiation.10, 11, 12 In myeloid cells, inhibition of Dbn1 impairs actin cytoskeleton reorganization and Ca²⁺ influx, resulting in impaired FcεRI‐mediated degranulation and histamine release in mast cells.13 ^.

Although Dbn1 has been shown to have functional roles in lymphocytes and mast cells, the role of Dbn1 in DC antigen presentation and innate immunity is unknown. Here, studies now show that DC silenced for Dbn1 expression had decreased phagocytic and endocytosis capacities, but increased fluid‐phase uptake. In the context of antigen presentation, silencing of Dbn1 resulted in impaired cell surface expression of both MHC class I and II and co‐stimulatory molecules that led to abrogated T‐cell responses.

Materials and methods

Animals

Wild‐type (WT; C57BL/B6), ovalbumin (OVA) transgenic for MHC class I (OT‐I) and OVA transgenic for MHC class II (OT‐II), mice were purchased from Jackson Laboratories (Bar Harbor, ME) and bred in pathogen‐free facilities at Howard University (Washington, DC). All animal procedures were performed in accordance with Institutional Animal Care and Use Committees (IACUC).

Microscopy

Spleens were harvested from WT mice and fixed in 3% paraformaldehyde (PFA) overnight at 4°. Tissues were then placed in 10% sucrose for 1 hr before generating 20‐μm sections using a Thermo Scientific NX70 cryosectioner (Thermo Fisher Scientific, Waltham, MA). Next, the sections were permeabilized with 0·3% Triton X‐100 for 10 min, blocked with 5% bovine serum albumin for 1 hr, and then stained with primary antibodies to Dbn1 and CD11c at a 1 : 100 dilution for 1 hr. Respective isotype antibodies were used as internal controls. After extensive washing, secondary antibodies at a 1 : 3000 dilution were added for an additional 1 hr. Nuclear staining was performed by labeling with DAPI at a 1 : 100 dilution. Coverslips were mounted onto slides with mowiol. Images were acquired using an Olympus FSX100 fluorescent microscope and analysed using imagej software (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, MD).

Generation of bone‐marrow‐derived dendritic cells

Dendritic cells were generated from C57BL/6 moue bone marrow. Briefly, bone marrow cells were flushed from tibia and femurs with phosphate‐buffered saline (PBS) through a 70‐μm cell strainer. Cells were then suspended in Iscove's modified Dulbecco's medium supplemented with 10% fetal buffered serum (FBS), 5% penicillin/streptomycin, 5% l‐glutamine and cultured in the presence of 20 ng/ml granulocyte–macrophage colony‐stimulating factor for 7 days. For maturation, day 6‐DC were treated with 250 ng/ml lipopolysaccharide for 24 hr.

Small interfering RNA knockdown

On day 5, in vitro generated DC were purified by MACS positive selection using CD11c microbeads (Miltenyi Biotec, Auburn, CA). Positive selection resulted in ~93% ± 2·7% purity as assessed by flow cytometric analyses. The CD11c⁺ DC were electroporated with 1 nmol of small interfering RNA (siRNA) using an ECM 830 square wave electroporator (BTX, Holliston, MA) at 300 V for one pulse at 10 ms, as previously described by Elizondo et al.14 Dbn1 targeting siRNA oligonucleotide (oligo) sequences used were: 5′‐GGTTAAAGGAGCAGTCTATCT‐3′ (siDbn1‐α) or 5′‐GGCTGTGCTAACCTTCTTAAT‐3′ (siDbn1‐β). Respective non‐targeting scrambled siRNA oligonucleotides were used for controls: 5′‐GTCGGAAGACGTTTATAGCTA‐3′ (siControl‐α) or 5′‐GACCGTCGTTATTATAGTCCT‐3′ (siControl‐β). All siRNA oligonucleotides purchased from Thermo Fisher Scientific. After confirmation of knockdown efficiency and little‐to‐no off‐target effects, siDbn1‐α was predominately used throughout the studies. After transfection, cells were incubated for an additional 48–72 hr in culture. Knockdown efficiency was evaluated by quantitative PCR and flow cytometric analyses.

Quantitative PCR

To evaluate gene expression, siControl or siDbn1 cells were harvested and resuspended in Trizol (Thermo Fisher Scientific) before total RNA extraction. Total RNA was reverse transcribed into single‐stranded cDNA using the High‐Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). For quantitative PCR, Gene Expression Master Mix and the following probes purchased from Thermo Fisher Scientific were used: Dbn1 (Mm00517314_m1), interleukin‐10 (IL‐10) (Mm01288386_m1), IL‐12p35 subunit (Mm00434169_m1), CD11c (Mm00498701_m1), IL‐6 (Mm00446190_m1), tumor necrosis factor‐α (TNF‐α) (Mm00443258_m1), CD86 (Mm00444543_m1), MHC class I (Mm04208017_mH; H2‐D1), MHC class IIα (Mm00439216_m1), and MHC class IIβ (Mm00439211_m1). Expression levels of the target transcripts were calculated by the comparative Ct method (2−ΔΔCt formula) after normalization with the housekeeping genes GAPDH (Mm99999915_g1) and β‐actin (Mm02619580_g1).

Antibodies, flow cytometry, and live/dead cell staining

Cells were washed in PBS supplemented with 2·5% FBS and 1 mm EDTA (FACS buffer). Surface staining was performed with fluorochrome‐tagged antibodies for 20 min at 4°. For DC immunophenotyping, the following antibodies were used: CD11c, CD80, CD86, CCR7, MHC class I and MHC class II. All antibodies were purchased from BioLegend (San Diego, CA). For intracellular analyses, cells were treated with permeabilization buffer (0·2% saponin in PBS supplemented with 2·5% FBS and 0·2 μm EDTA) for 1 hr before staining with a 1 : 200 dilution of Dbn1 antibody (clones: M2F6 and AB60933; Abcam, Cambridge, MA). For primary unconjugated antibodies, secondary‐tagged fluorochrome‐labeled antibodies were added to cells at a 1 : 2000 dilution before incubation for 2 hr at 4° followed by extensive washing. Staining with respective isotype control antibodies was performed to establish gating strategies. To study the viability of cells between control and Dbn1 knockdown, cells were stained with LIVE/DEAD™ Fixable Green (Thermo Fisher Scientific) for 30 min on ice. Cells were fixed with 3% PFA for 1 hr at 4° before flow cytometric analysis. Acquisition was performed using a BD Accuri C6 flow cytometric analyzer (BD Bioscience, San Diego, CA) and analyzed using flowjo v10 (TreeStar, Ashland, OR).

Enzyme‐linked immunosorbent assays and cytometric bead arrays

Supernatant from cell cultures was harvested at indicated time‐points. For DC phenotyping, cytometric bead arrays (CBA; BD Biosciences) were used to detect IIL‐6, IL‐10, monocyte chemoattractant protein‐1 (MCP‐1), interferon‐γ (IFN‐γ), TNF‐α and IL‐12p70, following manufacturer instructions. For evaluating OVA‐specific CD4⁺ T‐cell responses, CBA was used to measure IL‐2, IL‐4, IL‐6, IFN‐γ, TNF, IL‐17A, and IL‐10 cytokine levels. Additionally, IL‐2, IL‐4, IFN‐γ, IL‐12p70 and IL‐10 protein levels were measured using ELISA kits purchased from BioLegend to corroborate the results from CBA.

Fluid‐phase and phagocytosis uptake assays

At 4°, control or Dbn1 knockdown DC were plated with 250 ng/ml of zymosan‐FITC (Thermo Fisher Scientific; ~3 μm diameter size) or 400 ng/ml of dextran‐FITC (Thermo Fisher Scientific; 10 000 MW size), following the manufacturer‐recommended protocol. To initiate uptake, cells were transferred to 37° for varying time‐points up to 1 hr. Cells were then washed with ice‐cold PBS to stop uptake and remove excess unbound particles before extracellular co‐staining with CD11c antibodies (BioLegend). Cells were then fixed in 3% PFA and immediately acquired on a flow cytometric analyzer. Additionally, total fluorescence was measured using a multi‐mode microplate reader (Synergy HT; BioTek, Winooski, VT).

Presentation of MHC class I molecule Kb (H‐2Kb) bound to the peptide SIINFEKL

OVA_SIINFEKL peptide or OVA protein was adsorbed onto 3·0 or 0·1 μm latex beads for 30 min at 37°. The labeled beads were then cultured with control or Dbn1 knockdown DC for 4 hr. DC were then assessed for extracellular and intracellular expression of SIINFEKL peptide bound to MHC class I molecules using specific antibodies (Kb‐SIINFEKL; clone 25‐D1.16; BioLegend).

Flow cytometric quantification of recycling assays

Recycling assays were performed following a modified protocol by Blagojevic et al.15 Quantification was determined using flow cytometric approaches to evaluate recycling of MHC class I and CD11c. Briefly, cells were acid washed (0·2 m acetic acid and 0·5 m NaCl at pH 2·5) for 1 min to remove cell surface receptors at 4°. Staining immediately after acid wash stripping (time 0) was used to determine background noise or the non‐recycled proportion. For the recycled portion, after acid stripping, cells were placed at 37°. Recycling was stopped by placing cells on ice after 3, 10 and 30 min time‐points. Surface fluorescence signal was then measured by extracellular staining for MHC class I. Cells were labeled with fluorochrome‐conjugated MHC class I antibodies for 15 min at 4° before immediate fixation with 3% PFA. For measuring internalized pool, surface MHC class I was pre‐labeled and cells were allowed to incubate at 37° for 60 min. Cells were then washed and acid‐washed to strip uninternalized cell‐surface‐bound MHC class I. The amount of internalized MHC class I was then determined by flow cytometry. Percentage of MHC class I that recycled to the surface was calculated as: (∆MFI_recycled−∆MFI_{non‐recycled})/∆MFI_internalized × 100.15

T‐cell isolation and in vitro priming

CD4⁺ T cells were isolated from the spleen of OT‐II mice by negative depletion approaches. Briefly, antibodies to CD8 (clone 53‐6.7) and MHC class II (clone M5/114.15.2) were incubated with the harvested splenocytes before labeling with anti‐rat IgG magnetic microbeads (Qiagen, Hilden, Germany). Cells were then passed over a magnetic column to deplete CD8⁺ and MHC class II⁺ subsets. Results yielded > 94% purity. For CD8⁺ T‐cell isolation, cells were derived from OT‐I mice by negative depletion using antibodies to CD4 (clone RM4‐5) and MHC class II followed by labeling with anti‐rat IgG magnetic microbeads. Results yielded > 96% purity. For antigen presentation assays, siControl or siDbn1 DC were pulsed with OVA_323‐339 or OVA_SIINFEKL peptides for 4 hr at 37° before extensive washing. The purified CD4⁺ or CD8⁺ T cells were then added to DC a 10 : 1 ratio, respectively. To evaluate early T‐cell activation, cells were harvested after 24 hr and antibody‐stained for CD69, CD25 and CD62L; all antibodies purchased from BioLegend. To evaluate proliferation capacity, CD4⁺ T cells were labeled with 2·5 μm of CFSE (Thermo Fisher Scientific) before the culture with OVA‐pulsed mature siControl or siDbn1 DC. After 4 days of co‐culture, cells were assessed for proliferation. In addition, culture supernatant was collected for measurements of IL‐2, IFN‐γ, TNF‐α, IL‐4, IL‐6, IL‐10, and IL‐17A cytokines.

Statistical analysis

Statistical analysis was carried out using graphpad prism version 7 (GraphPad, La Jolla, CA). The unpaired two‐tailed Student's t‐test was used for the comparison of two groups. Data are representative of replicate independent experiments and expressed as mean ± SD. P values less than 0.05 were considered statistically significant; * = < 0·05, ** = < 0·01, and ns = non‐significant. Error bars for all figures indicate standard errors.

Results

Drebrin 1 is expressed in dendritic cells

Expression of Drebrin 1 (Dbn1) was assessed in CD11c⁺ DC. Previous studies have documented presence in mast cells, but no study has shown expression in other myeloid subsets. Co‐localization of CD11c, a predominant marker of DC, and Dbn1 was observed in spleen sections (Fig. 1a). Flow cytometric studies corroborated co‐expression of Drebrin 1 in CD11c⁺ MHC class II⁺ and CD8α ⁺ subsets (Fig. 1b). Studies reproducibly found that approximately 50–70% of CD11c⁺ MHC class II⁺ subsets expressed Dbn1. Lastly, in vitro generated bone marrow‐derived CD11c⁺ DC were additionally found to express Dbn1 (Fig. 1c).

Figure 1.

Drebrin 1 (Dbn1) is expressed in splenic and bone marrow‐derived CD11c⁺ dendritic cell (DC) subsets. (a) Cryosections from spleens of wild‐type (WT) mice were stained with Dbn1 (green), CD11c (red) and DAPI (blue). Fluorescence microscopy image was captured at magnifications of 4× and 20×. Images are representative of triplicate independent experiments. (b) Total splenocytes were harvested and stained for detection of CD11c, MHC class II, CD8, and Dbn1. Gating on CD11c⁺ subsets was used to evaluate co‐expression of MHC class II and Dbn1. Additionally, Dbn1 was assessed for expression with CD8 in the CD11c⁺ MHC class II ⁺ gated population. (c) In vitro generated bone‐marrow‐derived DC were evaluated for co‐expression of CD11c and Dbn1. All flow cytometric analyses used isotype controls to establish gating strategies. Data are representative of four independent experiments.

Silencing of Drebrin 1 in DC leads to impaired T‐cell responses

As DC are the principal initiators of antigen‐specific T‐cell responses, presentation and stimulatory capacity upon silencing of Dbn1 expression were assessed. Dbn1 was silenced in DC using siRNA before culturing with antigen‐specific CD4⁺ (OT‐II) T cells. Results revealed that siRNA knockdown of Dbn1 (siDbn1) OVA_323‐339 peptide‐pulsed DC led to decreased levels of responder T‐cell activation, as measured by lower levels of CD69 and CD25 (IL‐2Rα) and retained levels of CD62L on the surface, in comparison to control (siControl) DC (Fig. 2a). Concomitantly, the CD4⁺ T cells cultured with siDbn1 DC had reduced proliferation capacity, as measured at varying concentrations of OVA_323‐339 peptide pulsing (Fig. 2b). Evaluation of soluble released cytokines revealed no distinct skewing in polarization/differentiation states of the T helper cell subsets. However, there was an overall depression in expression of the hallmark T helper cytokines IL‐2, IFN‐γ, TNF‐α, IL‐4, and IL‐6 (Fig. 2c).

Figure 2.

RNAi‐mediated silencing of Dbn1 in dendritic cells (DC) impaired antigen‐specific CD4⁺ T‐cell responses. OVA _323‐339 peptide‐pulsed control (siControl) or Dbn1 knockdown (siDbn1) DC were used to stimulate CD4⁺ T cells at a 1 : 10 ratio, respectively, in vitro. (a) T cells were collected after 24 hr of stimulation and stained for CD69, CD25, and CD62L to measure early T‐cell activation. (b) CD4⁺ T cells were pre‐labeled with CFSE before stimulation with siControl and siDbn1 DC at varying concentrations of OVA _323‐339. Cells were harvested and assessed for proliferation on day 4. Data presented as histograms. Unfilled dashed lines represent unstimulated T cells as the control. (c) CD4⁺ T cells primed by DC were isolated and re‐stimulated in vitro on day 7. Supernatant was collected and assessed for IL‐2, IFN‐γ, TNF‐α, IL‐4, IL‐6, IL‐10, and IL‐17A cytokines by ELISA or Cytometric Bead Array (CBA). All data representative of three independent experiments. Bar graphs expressed as mean ± SD. P values less than 0.05 were considered statistically significant; * = < 0·05, ** = < 0·01, and ns = non‐significant. Error bars for all figures indicate standard errors.

Drebrin 1 knockdown DC have impaired display of cell surface receptors

Given impaired T‐cell responses, the next series of studies aimed to elucidate the mechanism that restrains DC from efficiently priming adaptive immune responses. Dbn1 knockdown in DC successfully reduced total gene transcript level by 61·43 ± 7·1% as assessed by quantitative PCR (qPCR; Fig. 3a). Live/dead cell staining reaffirmed that the siDbn1 DC were as viable as control siRNA‐transfected DC (Fig. 3b). Interestingly, in evaluating both MHC class molecules and co‐stimulatory molecule expression, qPCR assays revealed no significant impairment in transcript levels of MHC class I, MHC class IIα/β, CD11c, CD80, CD83, or CD86 upon silencing of Dbn1 in DC compared with controls (Fig. 3c). However, studies repeatedly found that total cell surface expression of MHC class I and II, as well as co‐stimulatory molecules CD80, CD83, and CD86, in siDbn1 DC were markedly lower than that of siControl DC (Fig. 3d). Both percentage and mean fluorescence intensity (MFI) were significantly reduced in MHC and co‐stimulatory molecules upon silencing of Dbn1 in DC. Lastly, studies also revealed impaired production of pro‐inflammatory cytokines IL‐6, MCP‐1, TNF‐α, and IL‐12p70 in siDbn1 DC upon lipopolysaccharide Toll‐like receptor‐agonist stimulation (Fig. 3e). However, there was no significant difference in transcript levels of IL‐6, expression of IL‐6, IL‐10, TNF‐α, or IL‐12p70 between upon Dbn1 knockdown in DC (Fig. 3f).

Figure 3.

Dendritic cells (DC) silenced for Dbn1 reveals impaired cell surface receptor expression and deficient cytokine release. (a) Real‐time PCR analysis of Dbn1 mRNA expression after Dbn1 knockdown in DC compared with control. (b) Viability staining of control or Dbn1 knockdown DC. Solid line filled histogram represents siControl and dashed line unfilled is siDbn1 DC. Percentage of dead cells is shown in top right corner of histogram plot. (c) Relative mRNA expression of CD11c, MHC class I (H2‐D1), MHC class II, CD80, CD83 and CD86 measured in siControl versus siDbn1 DC by quantitative PCR. (d) Flow cytometric analysis for extracellular expression of the cell surface receptors CD11c, MHC class I, MHC class II, CD80, CD83, and CD86 in siControl versus siDbn1 DC was performed. Gating strategies were established using isotype controls. Percentages are displayed in top right corner. Mean fluorescence intensity (MFI) is displayed below the percentages. (e) Supernatant from non‐stimulated or lipopolysaccharide (LPS) ‐stimulated siControl and siDbn1 DC were collected 24 hr after post‐stimulation and assessed for interleukin‐6 (IL‐6), IL‐10, monocyte chemoattractant protein 1 (MCP‐1), interferon‐γ (IFN‐γ), tumor necrosis factor‐α (TNF‐α), and IL‐12p70 by CBA. (f) Relative mRNA expression of IL‐6, IL‐10, IFN‐ γ, TNF‐α, and IL‐12p70 measured in siControl versus siDbn1 mature DC by quantitative PCR. Data are representative of at least three independent experiments. Bar graphs expressed as mean ± SD.

Restrained phagocytosis and increased fluid‐phase uptake in DC silenced for Drebrin 1

To evaluate endocytic changes associated with Dbn1, alterations in fluid‐phase uptake versus phagocytosis were assessed. First, zymosan particle uptake was assessed to determine phagocytosis capacity. Dbn1 knockdown DC had reduced phagocytic uptake in comparison to control DC (Fig. 4a). This was marked by a lower overall expression of CD11c in the Dbn1 knockdown DC. Gating on CD11c⁺ subsets further revealed that the Dbn1 knockdown cohort had reduced ability to phagocytose zymosan particles. Total fluorescence intensity of DC cultures incubated with zymosan particles corroboratively revealed a significant reduction in phagocytic capacity by the siDbn1 cohort (Fig. 4b). Interestingly, there was a reciprocal increase in percentage and MFI uptake of dextran molecules in the Dbn1 knockdown DC compared with controls (Fig. 4c,d). The increase in dextran uptake in siDbn1 DC was largely in the CD11c‐negative population; there was a corresponding increase in dextran uptake with reduction in cell surface CD11c expression in the Dbn1 knockdown cohort.

Figure 4.

Loss of Dbn1 impedes phagocytosis, but increases fluid‐phase uptake of dendritic cells (DC). (a) Control or Dbn1‐silenced DC were incubated with 3·0‐μm zymosan particles to measure phagocytic ability. After allowing particle uptake, cells were co‐stained with CD11c antibody before flow cytometric analysis. Dot plots represent CD11c versus zymosan expression in siControl versus siDbn1 DC. Histogram plots show CD11c⁺ subsets for expression of zymosan. (b) Total fluorescence intensity of zymosan uptake of the treated DC cultures was measured using a multi‐mode microplate reader and presented as a bar graph. (c) To measure fluid‐phase uptake, dextran (10 000 MW) was fed to control or Dbn1 knockdown DC before co‐staining with CD11c antibody. Dot plot shows CD11c versus dextran in siControl versus siDbn1 DC. Histogram plot is gated from live total cells. Top number value in the histogram plot is percentage. Lower number value is the mean fluorescence intensity (MFI). (d) Total fluorescence intensity of dextran uptake of the treated DC cultures was measured using a multi‐mode microplate reader of the treated DC cultures and presented as a bar graph. All presented data are representative of at least four independent experiments. Bar graphs expressed as mean ± SD.

Silencing of Drebrin 1 in DC impairs receptor recycling activities

To further evaluate uptake and recycling activities associated with antigen presentation activities, specific presentation of MHC class I‐H‐2K^b bound to OVA_SIINFEKL peptide was assessed. DC were pulsed with OVA_SIINFEKL peptide. After allowing for antigen capture and processing, siControl or siDbn1 DC were measured for cell surface expression of MHC class I bearing OVA_SIINFEKL peptide. Figure 5(a) shows relative expression of total cell surface MHC class I (H‐2D/K^b) with that of the proportion of H‐2K^b specifically bound to OVA_SIINFEKL peptide. Results reveal an overall total reduction in expression of surface‐bound MHC class I. In a follow‐up study, 3‐μm latex beads labeled with OVA protein were fed to control or Dbn1 knockdown DC to assess macropinocytosis uptake and antigen presentation. Similarly, after allowing for antigen capture, siDbn1 DC were found to have impaired cell surface expression of MHC class I bearing the processed OVA_SIINFEKL peptide compared with siControl DC (Fig. 5b). As expected, this led to restrained antigen‐specific IFN‐γ production responses by responder CD8⁺ OT‐I T cells upon priming by the siDbn1 DC pulsed with OVA_SIINFEKL peptide or fed OVA protein‐labeled beads (Fig. 5c). In assessing total levels of actin, staining revealed a lower MFI in siDbn1 DC in comparison to control DC (Fig. 5d). Lastly, recycling studies were performed to assess how Dbn1 affected cell surface expression after acid wash. Studies reproducibly found defective recycling of MHC class I molecules to the surface (Fig. 5e). Time–course studies revealed deficient recycling over a 60‐min time course, as was assessed by comparing recycled versus the non‐recycled MHC class I pools (Fig. 5f). Taken together, results showed that Dbn1 expression in DC is important for antigen uptake and display of cell surface receptors through recycling events.

Figure 5.

Dbn1 expression is important for antigen presentation through endocytic recycling. (a) Control or Dbn1 silenced dendritic cells (DC) were cultured with OVA_SIINFEKL peptide. Flow cytometric dot plots show extracellular co‐expression of total MHC class I with MHC class I (H‐2K^b) bound to the OVA_SIINFEKL peptide. (b) Control or Dbn1‐silenced DC were fed ovalbumin (OVA) protein coated 3‐μm latex beads. Cells were then stained with antibodies targeting CD11c and OVA_SIINFEKL‐bound‐to‐MHC class I before flow cytometric analysis. (c) CD8⁺ OT‐I T cells were cultured with DC pulsed with OVA_SIINFEKL peptide‐ or fed OVA protein coated on 3‐μm latex beads at a 20 : 1 ratio, respectively. After 72 hr, supernatant was harvested and IFN‐γ production was measured by ELISA. (d) DC were stained with phalloidin to measure F‐actin. Histogram plots show control DC as dashed and Dbn1‐silenced DC as the solid line. Number values are the MFI. (e) DC were evaluated for ability to recycle MHC class I. Cells were acid‐washed for 1 min to strip surface‐bound receptors before incubation at 37° to allow recycling events to occur. Top panels are siControl DC and bottom are siDbn1 DC. Left dot plots show populations displayed as SSC versus FSC immediately after acid washing of DC. Middle dot plot shows extracellular staining of MHC class I (x‐axis) immediately after acid wash. Right dot plot shows extracellular staining of MHC class I (x‐axis) 20 min at 37° to allow recycling after initial acid wash. (f) Graph shows kinetics of recycling at 0, 6, 20, and 60 min after acid washing of siControl versus siDbn1 DC for cell surface expression of MHC class I. To calculate the fraction recycled, the change in mean fluorescence intensity (MFI) of the 6‐, 20‐, or 60‐min groups was subtracted from the change in MFI of the 0‐min group. This value was then divided by the total amount of MHC class I internalized. This was then displayed as a percentage and plotted on a graph. Data are representative of five independent experiments. Bar graphs expressed as mean ± SD. P values less than 0.05 were considered statistically significant; * = < 0·05, ** = < 0·01, and ns = non‐significant. Error bars for all figures indicate standard errors.

Discussion

Drebrin 1 was found to be expressed within CD11c⁺ splenocyte subsets. There was a large proportion of CD11c⁻ cells that expressed Dbn1, which can account for the reported presence in T cells and mast cells.13, 16, 17 The studies reproducibly found higher levels of Dbn1 within the CD8⁺ CD11c⁺ MHC class II⁺ DC splenic subsets, which have been shown to be important for cross‐presentation at steady state, more so than in the CD8⁻ DC populations.18 As noted in previous reports, Dbn1 is important in modulating the actin cytoskeletal network important for scaffold assembly, signaling cascades, endocytosis and phagocytosis.2, 8, 9 In this work, studies demonstrated that Dbn1 plays a prominent role in antigen uptake and presentation in DC. Results revealed that silencing of Dbn1 in DC impaired uptake of large particles, associated with macropinocytosis and phagocytosis, and recycling of cell surface receptors. However, in both fluid‐phase and macropinocytosis assays, DC silenced for Dbn1 had impaired display of MHC class I, suggesting that the major observed defect of Dbn1 silenced DC was the recycling of cell surface receptors required for efficient antigen presentation.

Dbn1 has been shown to be a potent regulator of F‐actin assembly and structural stability by increasing the overall stiffness of the polymerized state.9 Published reports have shown that down‐regulation of Dbn1 results in decreased polymerization states, which has been associated with enhanced viral‐mediated cellular entry.16 In this respect, Dbn1 negatively regulates HIV‐1 and rotavirus infections by controlling the reorganization of the actin into stiffened cytoskeletal networks to repress viral entry;16, 19 the rigid state of the actin network at the plasma membrane counters the force applied by the virus to mediate entry. Results in this report support the premise that Dbn1 plays dominant roles in actin cytoskeletal dynamics associated with internalization of large versus small particles in DC. Collectively, data suggest that Dbn1 exerts work on the actin cytoskeletal network in DC to exert force against the plasma membrane for coordinated steps involved in large particle uptake and endocytic recycling.

Work by August et al. shows that Dbn1 expression in mast cells effectively remodels actin cytoskeletal activities and drives Ca²⁺ responses.13 The presence of Dbn1 was further important for FcεR1‐triggered degranulation, release of histamine, and secretion of IL‐2 and granulocyte–macrophage colony‐stimulating factor. The results of this study support their findings, whereby DC silenced for Dbn1 had decreased levels of soluble cytokine release.13 Secretory release is largely dependent on actin cytoskeletal reorganization, particularly through the recycling endosomal pathway. However, what was unclear was how actin levels change. August et al. found increased actin levels in mast cells from Dbn1^−/− mice,13 whereas another group found decreased levels utilizing smooth muscle cells (SMC) from Dbn1^−/+.20 Specifically, Stiber et al. fractionated filamentous from globular actin in Dbn1^−/+ SMC, which expressed roughly 50% of Dbn1 compared with WT SMC, and found less total F‐actin.20 This report now additionally shows decreased levels of F‐actin upon knockdown of Dbn1 in DC. Given that both this study and the work by Stiber et al. used a depressed expression system, as opposed to a complete abolishment of Dbn1, it could explain the differential reports of total F‐actin levels reported in the literature. Nevertheless, as stabilization of actin is important for endocytic events, it collectively stands to reason that Dbn1 controls cytoskeletal dynamics for differential movement of large versus small sized particles. The work in this report shows that silencing of Dbn1 in DC impaired uptake of large particles associated with macropinocytosis, but enhanced that of small molecule movement facilitated by fluid‐phase endocytosis.

Bin et al. showed that Dbn1 suppresses dynamin‐mediated endocytosis events.19 Given that dynamin‐2 plays such pivotal roles in fluid‐phase uptake,21 it does suggest that defects of Dbn1 activity would lead to differential changes in DC macropinocytosis and endocytic recycling. Micropinocytosis involves internalization of particles less than 0·2 μm, whereas macropinocyotsis involves particles between 0·2 and 5·0 μm. As both phagocytosis and macropinocytosis are heavily dependent on the reorganization of the actin cytoskeleton, the absence of Dbn1 would suggest that the destabilized F‐actin would allow greater fluidity of the membrane for uptake of small particles (as opposed to the work required for deformation and invagination associated with large particle uptake). Therefore, these studies support reports that down‐regulation of Dbn1 allows dynamin‐dependent pathway internationalization of small cargo to occur, thereby leading to increased fluid‐phase uptake. However, for endocytic recycling and macropinocytosis activities, extensive actin remodeling is required at the cell surface, to which the presence of Dbn1 plays a critical element in DC. Future studies additionally warrant evaluating potential cross‐presentation defects, given that the CD8⁺ CD11c⁺ MHC class II⁺ DC splenic subsets were found to have a preferentially higher expression of Dbn1.

Taken together, these studies now propose that Dbn1 plays a principal role in endocytic events associated with macropinocytosis, phagocytosis, and recycling of cell surface receptors in DC. These collective activities are important for sustaining successful immune responses through uptake of large antigens and display of peptides for antigen presentation to T cells, as well as secretory release of inflammatory cytokines. In the absence of Dbn1, DC are deficient in capture of large particles and endocytic recycling activities required for adequate display of both MHC class I and II, as well as co‐stimulatory molecules, on the cell surface.

Disclosures

The authors declare no conflict of interests.