Dynamic adhesion of immune cells to the vessel wall is a prerequisite for gut homing. Here, we present a protocol for a functional in vitro assay for the impact analysis of anti-integrin antibodies, chemokines or other factors on the dynamic cell adhesion of human cells using addressin-coated capillaries.
Gut homing of immune cells is important for the pathogenesis of inflammatory bowel diseases (IBD). Integrin-dependent cell adhesion to addressins is a crucial step in this process and therapeutic strategies interfering with adhesion have been successfully established. The anti-α4β7 integrin antibody, vedolizumab, is used for the clinical treatment of Crohn's disease (CD) and ulcerative colitis (UC) and further compounds are likely to follow.
The details of the adhesion procedure and the action mechanisms of anti-integrin antibodies are still unclear in many regards due to the limited available techniques for the functional research in this field.
Here, we present a dynamic adhesion assay for the functional analysis of human cell adhesion under flow conditions and the impact of anti-integrin therapies in the context of IBD. It is based on the perfusion of primary human cells through addressin-coated ultrathin glass capillaries with real-time microscopic analysis. The assay offers a variety of opportunities for refinements and modifications and holds potentials for mechanistic discoveries and translational applications.
Cell motion is a tightly regulated process indispensable for the development and function of multi-cellular organisms, but is also implicated in the pathogenesis of a multitude of diseases1. Recently, the homing process of immune cells from the blood stream to the peripheral tissues has gained increasing attention, since it contributes to replenishment and expansion of pathogenic cells in inflamed tissues in immunologically mediated diseases2,3. In particular, homing has been shown to have translational relevance in inflammatory bowel diseases (IBD). The therapeutic anti-α4β7 integrin antibody vedolizumab interfering with gut homing has shown efficacy in large clinical trials4,5 and has been successfully used in real-world clinical practice6,7,8. Further compounds are likely to follow9,10. Similarly, the therapeutic anti-α4 integrin antibody, natalizumab, is used for the treatment of multiple sclerosis (MS)11.
However, our functional understanding of the homing process in general and the mechanism of action of such therapeutic antibodies in particular is still limited. It is well established that homing consists of several steps including cell tethering and rolling with subsequent cell adhesion leading to firm arrest followed by trans endothelial migration12,13. The above-mentioned antibodies neutralize integrins on the cell surface preventing interaction with addressins on the endothelium of the vessel wall. This is thought to impede firm cell adhesion14,15. Yet, we are only beginning to understand the differential relevance of specific integrins for cell homing of distinct cell subsets. Moreover, the effects of anti-integrin antibodies on different cell subsets and dose-response associations are largely unknown leading to lots of open questions in the field of gut homing and anti-adhesion therapies in IBD.
Therefore, convenient tools to address such questions are desperately needed. The effect of anti-integrin antibodies on integrin-addressin interaction has so far predominantly been evaluated by assessing binding efficacy/binding inhibition with flow cytometry or through static adhesion assays16,17,18,19,20, thus with apparent simplification and deviation from the physiological situation. We recently established a dynamic adhesion assay to study integrin-dependent adhesion of human cells to addressins and the effects of anti-integrin antibodies under shear stress2. The principle of the technique has earlier been demonstrated with mouse cells21,22. Here, it was adapted and developed to address the above mentioned translational questions, opening novel avenues to better understand the mechanisms of therapy with anti-integrin antibodies in vivo.
The studies described in the following sections have been performed according to approval of the Ethics committee of the Friedrich-Alexander University Erlangen-Nuremberg.
1. Preparation of Capillaries
2. Cell Isolation, Treatment and Preparation for Microscopy
NOTE: These steps can be performed during the incubation periods necessary for the capillary preparation.
3. Live Cell Imaging of Dynamic Adhesion
4. Evaluation and Cell Counting
The method presented in this manuscript aims to simulate the in vivo process of human cell adhesion to the endothelial wall as closely as possible to functionally assess cell adhesion and the role of interfering antibodies. Therefore, ultrathin capillaries are coated with addressins and perfused with fluorescently labeled human cells of interest using a perfusion pump. Using live cell imaging the adhesion of human cells to the addressins can be observed in real time (Figure 1).
This method is particularly suitable to elucidate the mechanisms of anti-adhesion therapies in IBD. As shown in Figure 2A, perfusion of capillaries coated with ICAM-1, VCAM-1, MAdCAM-1 and Isotype Fc chimera with human PBMCs leads to marked adhesion, when one of the three addressins is present. In contrast, adhesion to isotype-coated capillaries is minimal. Inhibition of integrin-addressin interactions using neutralizing antibodies typically leads to a marked decrease of dynamic adhesion that is absent, when isotype control antibodies are added (representatively illustrated in Figure 2B, pooled quantitative data in Figure 2C).
Similarly, the impact of chemokines on integrin-affinity modulation can be investigated in this system by pre-incubation with such peptides in vitro. As shown by representative data in Figure 3, treatment of CD4+ human T cells with either CCL2 or CXCL10 leads to increased adhesion to MAdCAM-1 compared to untreated human cells.
Figure 1: Schematic representation of the workflow. Preparation of addressin coated ultrathin capillaries and fluorescently labeled human cells (left) is followed by perfusion of cells through the capillary using a peristaltic pump (middle) and time-lapse microscopy (right). Modified with permission from reference23. Please click here to view a larger version of this figure.
Figure 2: Dynamic adhesion of human peripheral blood mononuclear cells (PBMCs) from control donors to different addressins. (A) Number of newly adhering PBMCs to capillaries coated either with specific addressins or with Fc Isotype control (n = 6). (B) Representative images of dynamic adhesion in ICAM-1 coated capillaries perfused with untreated (NT) PBMCs or cells incubated with 10 µg/mL anti-CD18 antibody or 10 µg/mL IgG isotype control. Beginning: Merged first three images of the 3 min sequence; End: Merged last three images of the 3 min sequence; adhering cells are depicted white. Their numbers as well as the difference (i.e., newly adhering cells) are indicated. Scale bar = 100 µm. (C) Impact of anti-integrin antibodies (each used at a concentration of 10 µg/mL) on dynamic adhesion. Left: Number of newly adhering cells to capillaries coated with ICAM-1 and perfused with untreated, anti-CD18-treated or IgG isotype control-treated cells; Middle: Number of newly adhering cells to capillaries coated with VCAM-1 and perfused with untreated, natalizumab-treated or IgG isotype control-treated cells; Right: Number of newly adhering cells to capillaries coated with MAdCAM-1 and perfused with untreated, vedolizumab-treated or IgG isotype control-treated cells (n = 5–6). Bars indicate means with standard error of the mean (SEM). Statistical testing was performed with one-way ANOVA followed by Newman-Keuls Multiple comparison test (* p <0.05; ** p <0.01). Modified with permission from reference23. Please click here to view a larger version of this figure.
Figure 3: Chemokine-mediated dynamic adhesion of human CD4+ T cells to MAdCAM-1. Images from a representative dynamic adhesion assay in MAdCAM-1-coated capillaries perfused with untreated (NT) CD4+ T cells or cells incubated with 10 ng/mL rhCXCL10 or with 100 ng/mL rhCCL2. Beginning: Merged first three images of the 3 min sequence; End: Merged last three images of the 3 min sequence; adhering cells are depicted white. Their numbers as well as the difference (i.e., newly adhering cells) are indicated. Scale bar = 100 µm. Preincubation with chemokines leads to increased dynamic adhesion, presumably due to integrin-affinity modulation. Please click here to view a larger version of this figure.
SupplementaryVideos: Movies from three minutes image sequences from MAdCAM-1 coated capillaries perfused with fluorescently labeled PBMCs. (A) Capillary perfused with untreated cells. (B) Capillary perfused with cells treated with 10 µg/mL vedolizumab. (C) Capillary perfused with cells treated with 10 µg/mL human IgG isotype control. Cell flow from bottom to top, newly adhering cells are indicated by white arrows. Please click here to download this file.
The above protocol describes a useful technique to study dynamic adhesion of human immune cells to endothelial ligands. Through variation of the coated ligands, the perfused cell types or subsets, incubation with additional stimuli or different neutralizing antibodies, it has almost unlimited potential applications. Therefore, such dynamic adhesion assays may be useful to answer both fundamental questions of basic research as well as translational queries that might help to develop and optimize clinical therapy with drugs interfering with the adhesion process.
The utility of the assay to investigate the effects of clinical anti-adhesion therapies has recently been demonstrated. Moreover, different cell populations expressing different types and amounts of integrins show consistent levels of dynamic adhesion to the respective addressins, supporting the validity of the assay23.
On the whole, the completion of the protocol needs about 6 hours of work and presents intermediate technical challenges. Once the coating of the capillary has started, a critical issue is to avoid dehydration. This requires accurate sealing during incubation steps and careful handling of the capillaries, when exchanging solutions. In addition, the initiation of the perfusion process necessitates some caution. Due to the dead volume of the tubing, it is not feasible to fill the whole tubing with the final perfusion flow rate. On the other hand, changing the flow rate after flushing the capillary with high speed entails the risk of flow interruption leading to the possibility of static adhesion limiting the validity in regard to dynamic adhesion. Therefore, a timely switch to the final flow rate is essential. The final flow rate should be chosen dependent on the pump, tubing and capillary size used. To mimic human blood flow in high endothelial venules as closely as possible, a volume flow of 0.1 to 5 mm/s is recommended24,25.
The difficulty of the assay may markedly increase, when modifications are applied. For example, analysis of specific T cell subsets might require demanding polarization or purification strategies26. Also, including the homing steps of tethering, rolling and cell activation12 into the experiment will further increase the difficulty, since it might be necessary to coat a combination of ligands to the inside of the capillaries or (pre-)treat the perfused cell population with chemo- or cytokines to account for selectin-dependent rolling and chemokine-induced cell activation and integrin affinity modulation27. However, these might be particularly interesting questions for future research, especially as the presented technique allows to functionally address such a sophisticated interaction of different molecules with human cells and ligands. As mentioned above, competitively analyzing dynamic adhesion of several differently stained cell populations in the same capillary may add another level of complexity. Moreover, it has also been demonstrated that endothelial cells can be used in fluidic systems instead of recombinant ligands28.
Although it is a functional assay, the technique is limited by the reduction of complex in vivo networks to a selected set of involved molecules in vitro. This means that the effects of unknown or unexpected additional factors might be overlooked or not sufficiently considered. However, this is a frequent problem in human research, because ethical considerations often forbid mechanistic in vivo investigations29.
Taken together, this protocol presents a functional assay for the analysis of integrin-dependent dynamic adhesion and anti-integrin therapies in inflammatory bowel diseases. While the basic principle is quite straight-forward and not too difficult to perform, it can be embellished and modified in several regards and has the potential to answer translational questions regarding anti-adhesion therapy in IBD.
The authors have nothing to disclose.
The research of CN, IA, MFN and SZ was supported by the Interdisciplinary Center for Clinical Research (IZKF) and the ELAN program of the University Erlangen-Nuremberg, the Else Kröner-Fresenius-Stiftung, the Fritz-Bender-Stiftung, the German Crohn's and Colitis Foundation (DCCV), the Clinical Research Group CEDER of the German Research Council (DFG), the DFG topic program on Microbiota, the Emerging Field Initiative and the DFG Collaborative Research Centers 643, 796 and 1181.
48-Well plate | Sarstedt | 833,923 | |
Adhesion buffer: 150mM NaCl + 1mM HEPES + 1mM MgCl2 + 1mM CaCl2 | |||
Blocking solution: 1x PBS in ddH2O + 5 % BSA | |||
Bovine Serum albumin (BSA) | Applichem | A1391,0100 | |
CaCl2 | Merck | 2382 | |
Capillaries: Rectangle Boro Tubing 0,20×2.00 mm ID, 50 mm length | CM Scientific | 3520-050 | |
CCL-2, human | Immunotools | 11343384 | |
CD4-Microbeads, human | Miltenyi Biotec | 130-045-101 | |
CellTrace™ CFSE Cell Proliferation Kit | ThermoFischer Scientific | C34554 | |
Centrifuge (Rotixa 50 RS) | Hettrich | ||
Coating buffer: 150 mM NaCl + 1 mM HEPES | |||
Confocal Microscope (TCS SP8) | Leica | ||
CXCL-10, human | Immunotools | 11343884 | |
Dextran 500 | Roth | 9219.3 | |
EDTA KE/9 ml Monovette | Sarstedt | ||
Falcons (50 mL) | Sarstedt | 62,547,004 | |
Fc chimera isotype control | R&D Systems | 110-HG | |
Flow Rates Peristaltic Pump (LabV1) | Baoding Shenchen Precision Pump Company | ||
HEPES | VWR | J848-100ML | |
Human IgG Isotype Control | ThermoFischer Scientific | 31154 | |
Intercellular Adhesion Molecule 1 (ICAM-1) Fc chimera | R&D Systems | 720-IC-050 | |
LS-Columns | Miltenyi Biotec | 130-042-401 | |
MgCl2 | Roth | ||
MnCl2 | Roth | ||
mouse IgG isotype control | Miltenyi Biotec | 130-106-545 | |
Mucosal Vascular Addressin Cell Adhesion Molecule 1 (MAdCAM-1) Fc chimera | R&D Systems | 6056-MC | |
NaCl | Roth | 3957.3 | |
Natalizumab | Biogen | ||
Neubauer Counting chamber | Roth | T729.1 | |
Pancoll, human | PAN Biotech | P04-601000 | |
Phosphate Buffered Saline (PBS) | Biochrom | L 182-10 | w/o Mg and Ca |
Plastic paraffin film: Parafilm (PM-996) | VWR | 52858-000 | |
purified anti-human CD18 | Biolegend | 302102 | |
RPMI Medium 1640 | Gibco Life Technologies | 61870-010 | |
Rubber tubing: SC0059T 3-Stop LMT-55 Tubing, 1.02mm ID, 406.4 mm length | Ismatec | SC0059 | |
Serological Pipetts | Sarstedt | 861,254,025 | |
Trypan blue | Roth | CN76.1 | |
Vascular Cell Adhesion Molecule 1 (VCAM-1) Fc chimera | Biolegend | 553706 | |
Vedolizumab (Entyvio) | Takeda |