Epub ahead of print May 27, 2015 - doi:10.1189/jlb.1AB1014-488RR
Brief Conclusive Report
Fluorescent Ly6G antibodies determine macrophage phagocytosis of neutrophils and alter the retrieval of neutrophils in mice Kirsten Bucher,* Fee Schmitt,* Stella E. Autenrieth,† Inken Dillmann,* Bernd N¨ urnberg,* Katja Schenke-Layland,‡,§,{ and Sandra Beer-Hammer*,1 *Department of Pharmacology and Experimental Therapy, Institute of Experimental and Clinical Pharmacology and Toxicology, and ‡University Women’s Hospital, Eberhard-Karls-University T¨ubingen, Germany; †Department of Hematology, Oncology, Immunology, Rheumatology and Pulmology, University Hospital Tu¨ bingen, Germany; §Department of Cell and Tissue Engineering, Fraunhofer Institute for Interfacial Engineering and Biotechnology, Stuttgart, Germany; and {Department of Medicine/Cardiology, Cardiovascular Research Laboratories, David Geffen School of Medicine at University of California Los Angeles, California, USA RECEIVED OCTOBER 16, 2014; REVISED MAY 5, 2015; ACCEPTED MAY 5, 2015. DOI: 10.1189/jlb.1AB1014-488RR
ABSTRACT
Introduction
Fluorescently labeled Ly6G antibodies enable the tracking of neutrophils in mice, whereas purified antiLy6G rapidly depletes neutrophils from the circulation. The mechanisms underlying neutrophil depletion are still under debate. Here, we examined how identical Ly6G antibodies coupled to different fluorochromes affect neutrophil fate in vivo. BM cells stained with Ly6G antibodies were injected into mice. The number of retrieved anti-Ly6G-FITC+ cells was reduced significantly in comparison with anti-Ly6G-APC + or antiLy6G-PE+ cells. Flow cytometry and multispectral imaging flow cytometry analyses revealed that antiLy6G-FITC+ neutrophils were preferentially phagocytosed by BMMs in vitro and by splenic, hepatic, and BM macrophages in vivo. Direct antibody injection of antiLy6G-FITC but not anti-Ly6G-PE depleted neutrophils to the same degree as purified anti-Ly6G, indicating that the FITC-coupled antibody eliminates neutrophils by a similar mechanism as the uncoupled antibody. With the use of a protein G-binding assay, we demonstrated that APC and PE but not FITC coupling inhibited access to interaction sites on the anti-Ly6G antibody. We conclude the following: 1) that neutrophil phagocytosis by macrophages is a central mechanism in anti-Ly6G-induced neutrophil depletion and 2) that fluorochrome-coupling can affect functional properties of anti-Ly6G antibodies, thereby modifying macrophage uptake of Ly6G-labeled neutrophils and neutrophil retrieval following adoptive cell transfer or injection of fluorescent anti-Ly6G. J. Leukoc. Biol. 98: 000–000; 2015.
Ly6G is a glycosylphosphatidylinositol-linked cell-surface protein that is expressed predominantly on murine neutrophils [1]. The function of Ly6G has still not been fully clarified, although a recent study implicates Ly6G in neutrophil migration [2]. There are 2 mAb, i.e., RB6-8C5 and 1A8, which recognize Ly6G [3]. Whereas clone 1A8 specifically binds to Ly6G, clone RB6-8C5 additionally cross-reacts with Ly6C [3], which is also present on dendritic cells, subpopulations of macrophages, monocytes, and lymphocytes [4–7]. RB6-8C5 and 1A8 are important tools in neutrophil research, both in a fluorochrome-coupled and in an untagged form. Untagged antibodies are commonly referred to as purified antibodies. In vivo, administration of high doses of purified antiLy6G rapidly depletes neutrophils from the circulation. This is a standard method to investigate the role of neutrophils in a wide variety of disease models, including experimental encephalitis [8], as well as microbial and parasitic infections [9, 10]. The mechanisms leading to neutrophil depletion remain to be determined. Abbitt et al. [11] have shown that anti-Ly6G-mediated neutrophil depletion involves the cross-linking of Ly6G and is dependent on the Fc fragment of the antibody but does not require FcgR engagement. Moreover, the mechanism appears to involve opsonization of anti-Ly6G-coated cells with complement component C3b, as C32/2 mice but not mice treated with anti-C5 antibodies are protected against anti-Ly6G-induced neutrophil depletion [12]. However, studies analyzing the role of anti-Ly6G in the induction of apoptosis led to contradictory results [2, 13]. Whereas high doses of purified anti-Ly6G antibodies are used for neutrophil depletion, low doses of fluorochrome-coupled anti-Ly6G antibodies are used to analyze neutrophil trafficking by flow cytometry [14] or intravital microscopy [11, 15–19]. This might imply that the dose of the antibody determines whether anti-Ly6G affects neutrophil function and survival. However, it
Abbreviations: 7-AAD = 7-aminoactinomycin D, APC = allophycocyanin, BM = bone marrow, BMM = bone marrow-derived macrophage, SA = streptavidin The online version of this paper, found at www.jleukbio.org, includes supplemental information.
0741-5400/15/0098-0001 © Society for Leukocyte Biology
1. Correspondence: Institute of Experimental and Clinical Pharmacology and Toxicology, University Hospital Tuebingen, Wilhelmstrasse 56, 72074 Tuebingen, Germany. E-mail: [email protected]
Volume 98, September 2015
Journal of Leukocyte Biology 1
Copyright 2015 by The Society for Leukocyte Biology.
was recently shown that furthermore, low doses of purified antiLy6G impair neutrophil recruitment and attenuate experimental arthritis [2]. Interestingly, Yipp and Kubes [19] demonstrated that low doses of fluorochrome-coupled anti-Ly6G do not influence neutrophil migration. This suggests that the diverse effects induced by anti-Ly6G antibodies may not only be dose dependent but are also determined by coupled fluorochromes. Accordingly, Lee et al. [1] speculated that fluorochromes may alter the functional properties of the anti-Ly6G antibody. Here, we studied how fluorochrome-coupled anti-Ly6G antibodies affect neutrophil retrieval in mice. We demonstrate the following: 1) that identical antibodies conjugated with distinct fluorochromes differentially determine the fate of the labeled neutrophils and 2) that neutrophil phagocytosis by macrophages may be a central mechanism involved in anti-Ly6G-induced neutrophil depletion.
MATERIALS AND METHODS
Mice C57BL/6 mice (Charles River Laboratories, Sulzfeld, Germany) were kept under specific pathogen-free conditions. All animal work was performed according to the German animal care regulations and was approved by the local authorities.
Preparation of leukocyte suspensions BM cells and splenocytes were isolated as described previously [20]. Lungs were cut into pieces and incubated with 3 ml collagenase D solution (IMDM containing 5% FCS; both from PAA, Linz, Austria) and 0.2 mg/ml Collagenase D and 0.01 mg/ml DNase I; both from Roche, Mannheim, Germany) at 37°C for 30 min. Then, 2 ml IMDM containing 0.2 mg/ml Collagenase/Dispase (Roche) were added, and the tissue was incubated at 37°C for another 15 min. EDTA solution (0.5 M) was added to a final concentration of 5 mM EDTA and incubated for 5 min. Hepatocytes were isolated according to Schmitz et al. [21] with the following modifications: livers were perfused at 37°C, first with calcium-free buffer for 2 min, followed by calcium-containing collagenase buffer (15 mg/100 ml collagenase type IV; Sigma, St. Louis, MO, USA) for 10 min. Macrophages were enriched in the supernatant by centrifugation at 20 g for 2 min. Leukocytes from single-cell suspensions of BM, lung, heparinized peripheral blood, liver, and spleen were prepared by RBC lysis in ACK buffer (0.155 M NH4Cl, 0.01 M KHCO3, 0.1 mM EDTA).
Flow cytometry and ImageStream analysis For characterization of cells, the following antibodies or peptides were used: 7-AAD; Annexin V-V500; CD3e-Pacific Blue (500A2); CD11b-PE.Cy7 (M1/70); CD16/32 (2.4G2); CD19-V450 (1D3); CD45R-APC, -FITC, and -PE (RA3-6B2); Ly6G-APC, -FITC, and -PE (1A8); Ly6G-FITC and -PE (RB6-8C5); and SAPerCP-Cy5.5 (all BD Biosciences, Heidelberg, Germany). Also used were: F4/ 80 Pacific Blue (BM8; BioLegend, San Diego, CA, USA); Ly6G-APC (RB6-8C5; eBioscience, San Diego, CA, USA); Ly6G-APC, -FITC, and -PE (1A8; Miltenyi Biotec, Bergisch Gladbach, Germany); and protein G-biotin (Pierce Biotechnology, Rockford, IL, USA). CompBeads were from BD Biosciences. Flow cytometry measurements were performed on a FACSCanto II (BD Biosciences), and data were evaluated with FlowJo software. Where indicated, cells were additionally analyzed by use of ImageStreamX, and data were evaluated with Integrated Design and Engineering Analysis (IDEAS) software. Cells were gated on focus, followed by F4/80+ gating to identify BMM, which were visually counted for attached (1A8+ cells at the plasma membrane of F4/80+ BMM) or intracellular (1A8+ cells completely inside the cytoplasma of F4/80+ BMM), fluorescently labeled neutrophils. BMMs (1000 cells) were counted in each sample.
Adoptive transfer of anti-Ly6G-labeled BM leukocytes Equal fractions of freshly prepared BM leukocytes were separately stained with anti-Ly6G-APC, -FITC, or -PE (1A8) at a concentration of 0.25 mg/million cells
2
Journal of Leukocyte Biology
Volume 98, September 2015
in 100 ml RPMI plus 2% FCS on ice. After washing twice with PBS, the separately stained anti-Ly6G-APC+, -FITC+, and -PE+ BM leukocytes were pooled, and 2.5 3 107 cells in 150 ml PBS were injected into the tail vein of each recipient mouse. In each experiment, an aliquot of the injection suspension was retained to determine percentages of APC-, FITC-, and PElabeled neutrophils. After the indicated time, the recipients were euthanized, and leukocytes from organs were prepared as detailed above.
Generation of BMM and coculture of anti-Ly6G-labeled BM leukocytes with BMM BMMs were generated from murine BM cells in RPMI 1640 (with 10% FCS, 100 units/ml penicillin, 100 mg/ml streptomycin, 2 mM L-glutamine, and 0.05 mM b-ME (all from PAA) in the presence of 30 ng/ml mouse rM-CSF (R&D Systems, Wiesbaden, Germany). Medium was exchanged on days 2, 4, and 6, and BMMs were harvested on day 8. BM cells stained with 1A8-APC, -FITC, or -PE were prepared and pooled as described above. BMMs were seeded at 2 3 105 cells/well into 96-well plates. Following incubation at 5% CO2 and 37°C for 3 h, the medium was removed and the labeled BM leukocytes (1 3 106 cells in 200 ml) were added to each well and incubated for 10, 20, 30, 45, and 60 min. Wells containing only BM cells or only BMM served as gating controls. In addition, samples of BM cells and BMM were coincubated on ice and served as 4°C control. Cultured cells were harvested by use of ice-cold PBS-EDTA and subsequently analyzed by flow cytometry or ImageStreamx.
Statistical analysis Statistical analysis was performed as indicated in the figure legends. A value of P # 0.05 was considered statistically significant.
RESULTS AND DISCUSSION
1A8-FITC labeling reduces retrieval of neutrophils following adoptive transfer Here, we studied whether and how anti-Ly6G-coupled fluorochromes affect neutrophil retrieval/tracking in vivo and compared the retrieval rates of 1A8-APC-, -FITC-, or -PE-labeled BM cells following adoptive transfer in mice. Equal fractions of BM cells were separately stained with 1A8-APC, -FITC, or -PE. Then, cells were pooled and intravenously injected into recipients. After 1 h, leukocyte suspensions from BM, lungs, blood, and spleens were analyzed by use of flow cytometry (Fig. 1A and B). The frequency of retrieved 1A8-APC+ and -PE+ neutrophils was comparable in all organs, resulting in a 1.1 ratio of 1A8-PE+:1A8-APC+ cells. In contrast, the frequency of retrieved 1A8-FITC+ cells was reduced by ;60%, resulting in a 0.47 ratio of 1A8-FITC+:1A8-APC+ cells and a ratio of 0.43 of 1A8-FITC+:1A8-PE+ cells (Fig. 1C and D). The decreased detectability of the 1A8-FITC+ neutrophils may result from a degradation of the FITC molecule as a result of its lower thermal stability in vivo. Alternatively, as fluorochrome conjugation can reduce the avidity of a given antibody [22, 23], the decreased retrieval of 1A8-FITC+ neutrophils might be explained by a reduced binding ability of 1A8-FITC to Ly6G. Therefore, we tested labeling of neutrophils with fluorescent 1A8 under physiologic conditions. To this end, 3 fractions of BM cells were separately stained with 1A8-APC, -FITC, or -PE. Thereafter, the cells were pooled, incubated at 37°C, and analyzed by flow cytometry. We found that percentages of 1A8FITC+ cells, as well as of 1A8-APC+ and -PE+ cells, remained stable under cell culture conditions (Fig. 1E). This showed that the decreased detectability of the 1A8-FITC+ neutrophils (Fig. 1C and D) was neither a result of a reduced binding ability of
www.jleukbio.org
Bucher et al. Fluorochromes alter functional properties of anti-Ly6G
1A8-FITC to neutrophils nor of lower thermal stability of the FITC molecule. Furthermore, by use of AnnexinV/7-AAD staining, we did not observe an elevated apoptosis rate or cell death of 1A8-FITC-labeled neutrophils compared with 1A8-APCor -PE-labeled cells (Supplemental Fig. 1A and B), suggesting that the decreased retrieval of 1A8-FITC-labeled neutrophils in mice was not a result of an increased induction of apoptosis. Moreover, although the expression of the activation marker CD11b decreased over time under cell culture conditions, it was similar among 1A8-APC-, -FITC-, and -PE-stained cells at all timepoints (Supplemental Fig. 1A and C). These results suggest that 1A8-FITC labeling did not differentially alter the activation state of neutrophils. Therefore, we hypothesized that the 1A8FITC-labeled neutrophils become rapidly cleared in the mouse.
1A8-FITC-labeled BM cells are preferentially phagocytosed by macrophages in vitro and in vivo
Figure 1. Fluorochromes coupled to anti-Ly6G (clone 1A8) differentially influence neutrophil retrieval in mice. (A) Experimental set-up to compare the effect of anti-Ly6G-APC, -FITC, and -PE (clone 1A8; BD Biosciences) labeling on the retrieval of adoptively transferred neutrophils. BM leukocytes from donor mice were separately stained with 1A8APC, -FITC, or -PE and coinjected into recipients. One hour later, leukocytes from BM, blood, spleens, and lungs from recipients were stained with antibodies against CD3e, CD11b, and CD19 and were analyzed by flow cytometry. (B) Myeloid cells were gated as CD11b+ CD3e2 CD192 singlet leukocytes. They were analyzed for the presence of 1A8-APC-, -FITC-, or -PE-labeled neutrophils. FSC-A, Forward scatter-area; FSC-W, FSC-width; SSC-A, side-scatter-area. (C) 1A8-APC-, -FITC-, or -PElabeled neutrophils retrieved in BM, blood, spleen, and lung of each recipient mouse. Data are expressed as percentages of 1A8-labeled cells of all CD11b+ cells and are shown as means 6 SD. Data were analyzed by ANOVA, followed by Bonferroni’s multiple comparison test; n = 3; *P , 0.05; **P , 0.01; ***P , 0.001. (D) For all organs of each mouse, ratios were calculated as percent retrieved 1A8-PE+ cells/percent retrieved 1A8-APC+ cells (d), percent retrieved 1A8-FITC+ cells/percent retrieved 1A8-APC+ cells (n), and percent retrieved A8-FITC+ cells/percent retrieved 1A8-PE+ cells (:). All calculated ratios are depicted. Bars indicate means 6 SD. (E) Coincubation of 1A8-APC-, -FITC-, and -PE-labeled BM cells on ice [4°C control (ctrl)] or at 37°C and 5% CO2 for 5, 30, and 60 min. Neutrophil labeling with 1A8-APC, -FITC, or -PE is stable at 37°C at all investigated time-points.
Phagocytic macrophages in the BM are known to mediate the clearance of neutrophils [24]. Thus, we tested whether macrophages preferentially take up 1A8-FITC-labeled neutrophils. Unstimulated BMMs were coincubated with 3 equal fractions of BM leukocytes, which had been separately stained with 1A8-APC, -FITC, or -PE, pooled, and analyzed by flow cytometry (Fig. 2A and B). Low percentages of 1A8-APC+ and -PE+ BMMs were detectable at all time-points following the onset of coincubation (Fig. 2B and C). Following coincubation on ice (4°C control) the percentage of 1A8-FITC+ BMM was similar to that of 1A8-APC+ or -PE+ BMM. However, percentages of 1A8-FITC+ BMM increased up to .10% and significantly exceeded those of 1A8-APC+ or -PE+ macrophages between 10 and 60 min of coincubation at 37°C (Fig. 2C). On the one hand, the high percentage of 1A8-FITC+ BMM may result from a preferential uptake of 1A8-FITC-labeled neutrophils. On the other hand, intercellular membrane exchanges by trogocytosis have been reported [25, 26]. These exchanges by themselves may lead to a transfer of the fluorescent antibodyantigen complex from 1 cell type to another. Thus, to examine the processes leading to a preferential increase in 1A8-FITC+ BMM, 1A8-labeled BM cells were coincubated with BMM, as described above. Cells were measured by multispectral imaging flow cytometry (ImageStreamX), a method that allows quantifying the location of fluorescent probes on, in, or between cells. We analyzed cell–cell interactions between macrophages and neutrophils, differentiating between attached and internalized neutrophils (Fig. 3). Ten minutes after coincubation, 1A8-FITC labeling of neutrophils resulted in an enhanced attachment to BMM (Fig. 3A and C), and a significantly higher number of 1A8FITC-labeled neutrophils was internalized by BMM compared with 1A8-APC (15-fold)- and -PE (30-fold)-stained cells (Fig. 3B and C). This confirmed that 1A8-FITC labeling resulted in an increased phagocytosis of neutrophils by macrophages. Percentages of 1A8-FITC+ BMM peaked between 10 and 20 min of coincubation and decreased thereafter (Fig. 2C), pointing to the possibility that internalized neutrophils were rapidly digested in the macrophages. This assumption would be in line with phagocytosis experiments that use human neutrophils, where macrophages engulfed and degraded neutrophils within 20–60 min [27].
www.jleukbio.org
Volume 98, September 2015
Journal of Leukocyte Biology 3
To test whether phagocytic macrophages preferentially eliminate 1A8-FITC+ neutrophils in vivo, we measured macrophage uptake of adoptively transferred 1A8-FITC- and -PE-labeled neutrophils by flow cytometry. As it is known that adoptively transferred neutrophils preferentially accumulate in the spleen, BM, and liver [24, 28], we analyzed macrophages in spleen, BM, and liver of recipient mice. Fifteen minutes following coinjection of 1A8-FITC- and -PE-labeled BM cells, the frequency of 1A8FITC+ F4/80+ macrophages was significantly higher than that of 1A8-PE+ macrophages in spleen and BM (Fig. 4A and B). Vice versa, the frequency of 1A8-FITC+ neutrophils was reduced significantly compared with 1A8-PE+ cells in both organs (Fig. 4C and D). Whereas leukocyte populations of spleen and BM could be isolated without delay, the preparation of liver cells required organ digestion by collagenase perfusion at 37°C [21]. Again, the frequency of 1A8-FITC+ F4/80+ macrophages was significantly higher than that of 1A8-PE+ macrophages in the liver (Fig. 4E). Taken together, our findings indicate that phagocytic macrophages preferentially clear 1A8-FITC-labeled neutrophils, corroborating our concept that neutrophil phagocytosis by
Figure 2. 1A8-FITC labeling of BM cells increases percentages of FITC+ BMM after coincubation of BM cells with BMM. Equal fractions of BM leukocytes were separately labeled with 1A8-APC, -FITC, or -PE (BD Biosciences). Fractions were pooled and coincubated with BMM at a ratio of 5:1. Cells were harvested at the indicated time-points, stained with anti-F4/80 and anti-CD11b, and analyzed by flow cytometry. (A) BMMs were gated as CD11b+ F4/80+ singlet leukocytes and analyzed for 1A8-APC, -FITC, or -PE staining. (B) Representative flow cytometry plots showing the frequencies of 1A8-APC+, -FITC+, or -PE+ cells within the BMM population after 20 or 60 min of coincubation at 37°C (BMM + BM cells). BMM and BM cells coincubated on ice served as a 4°C control (4°C ctrl). Pure BMM suspensions were analyzed as gating controls (BMM). (C) Percentages of 1A8-APC+, -FITC+, and -PE+ BMM between 0 and 60 min of coincubation. Between 10 and 60 min, percentages 1A8-FITC+ cells significantly exceeded those of 1A8-APC+ and -PE+ cells. Bars represent means 6 SD (n = 3; for 10 min, n = 13). Data were pooled from 6 independent experiments. Data were analyzed by ANOVA, followed by Bonferroni’s multiple comparison tests. ***P , 0.001.
4
Journal of Leukocyte Biology
Volume 98, September 2015
Figure 3. 1A8-FITC-labeled neutrophils are preferentially internalized by BMM. BM cells were labeled with 1A8-APC, -FITC, or -PE (BD Biosciences) and were coincubated with BMM, as described in Fig. 2. After harvest, cells were stained with anti-F4/80, and cell–cell interactions of F4/80+ BMM with 1A8-APC+, -FITC+, or -PE+ neutrophils were assessed by multispectral imaging flow cytometry (ImageStreamX). Representative bright field (BF) and fluorescence images of attached (A) and intracellular (B) 1A8-APC+, -FITC+, or -PE+ neutrophils are shown (defined as described in Materials and Methods). (C) Numbers of 1A8labeled attached or intracellular neutrophils/10,000 BMMs. Bars represent means 6 SD. Data were pooled from 2 independent experiments. Data were analyzed by one-way ANOVA, followed by Sidak’s multiple comparison test; n = 5; ****P , 0.0001.
www.jleukbio.org
Bucher et al. Fluorochromes alter functional properties of anti-Ly6G
Figure 4. Adoptive transfer of 1A8-labeled BM cells increases the frequency of FITC+ macrophages in the spleen, BM and liver of recipient mice. Equal fractions of 1A8-FITC- or -PE-labeled BM cells were pooled and coinjected into recipients. After 15 min, leukocytes from BM and spleen of recipient mice were isolated, stained with anti-F4/80 and antiCD11b and analyzed by flow cytometry. To analyze liver macrophages, recipient mice were terminally anesthetized 10 min following adoptive transfer, and livers were perfused at 37°C, first with calcium-free buffer for 2 min and then with collagenase buffer for 10 min. Liver cells were stained and analyzed as described above. (A) Macrophages from spleen, BM, and liver were gated as F4/80+ singlet leukocytes and analyzed for 1A8-FITC or -PE staining. (B) Percentages of 1A8-FITC+ and -PE+ macrophages of all gated splenic or BM macrophages. (C) To determine percentages of 1A8-FITC- or -PE-labeled neutrophils in spleen and BM, cells were gated as CD11b+ F4/802 singlet leukocytes and were analyzed for the presence of 1A8-FITC- or -PE-labeled neutrophils. (D) Percentages of 1A8-FITC- or -PE-labeled neutrophils of all gated CD11b+ F4/802 leukocytes 15 min after adoptive transfer. (E) Percentages of 1A8-FITC+ and -PE+ macrophages of all gated liver macrophages. Macrophages were gated as shown in A. Bars represent means 6 SD. Data were analyzed with unpaired t-test; n = 6; **P , 0.01; ***P , 0.001.
www.jleukbio.org
macrophages plays a central role in the elimination of 1A8-FITCbut not 1A8-PE-labeled neutrophils in vivo. To test whether the neutrophil-eliminating effect of 1A8-FITC was a general characteristic of FITC-coupled anti-Ly6G antibodies, we analyzed APC-, FITC-, and PE-coupled antibodies of another anti-Ly6G clone (RB6-8C5) and a 1A8 clone from another manufacturing company [1A8(M)]. To this end, again, we determined the retrieval of labeled neutrophils after adoptive transfer in vivo and macrophage uptake of labeled neutrophils following coincubation with BMM in vitro. In a first set of experiments, we examined the cell-labeling stability of fluorescent RB6-8C5 and 1A8(M) at 37°C (Supplemental Fig. 2A and B). The fluorescent RB6-8C5 antibodies caused a dye transfer from the labeled cells to other neutrophil populations, which were differently labeled. This effect was most pronounced by use of RB6-8C5-APC (Supplemental Fig. 2A) and rendered the data not evaluable. Therefore, further analyses of RB6-8C5 were restricted to RB6-8C5-FITC and -PE. In contrast, neutrophillabeling with 1A8(M)-APC, -FITC, or -PE was stable at 37°C at all investigated time-points (Supplemental Fig. 2B). Next, we determined the effect of neutrophil labeling with fluorescent RB6-8C5 after adoptive transfer in vivo. One hour after injection of equal fractions of RB6-8C5-FITC- and -PE-stained BM cells, the frequency of retrieved RB6-8C5-FITC+ cells was reduced by .90% compared with RB6-8C5-PE+ cells in BM, blood, spleen, and lung (Supplemental Fig. 2C), resulting in a 0.08 ratio of 1A8FITC+:1A8-PE+ cells (Supplemental Fig. 2D). This suggested that the eliminating effect of RB6-8C5-FITC was greater than that of 1A8-FITC (BD Biosciences; Fig. 1C and D). Consistently, coincubation experiments with BMM and RB6-8C5-stained BM cells showed that percentages of RB6-8C5-FITC+ BMM increased up to .20% and significantly exceeded those of RB6-8C5-PE+ macrophages after 10 min of coincubation (Supplemental Fig. 2E). Then, we determined the effect of neutrophil labeling with the 1A8(M) antibody in vivo and in vitro. Adoptive transfer of 1A8(M)-labeled BM cells led to comparable frequencies of retrieved 1A8-APC+ and -PE+ neutrophils in all organs, but frequencies of retrieved 1A8-FITC+ cells were reduced by ;40% (Supplemental Fig. 2F), resulting in a 0.6 ratio of 1A8-FITC+:1A8PE+ cells (Supplemental Fig. 2G). When we coincubated 1A8(M)stained BM cells with BMM, no differences among 1A8-APC+, -FITC+, or -PE+ BMM became apparent (Supplemental Fig. 2H). This points toward the eliminating effect of 1A8 from Miltenyi Biotec (Supplemental Fig. 2H) being smaller than that of 1A8 from BD Biosciences (Fig. 2C) and the in vitro model being too insensitive to observe any phagocytic effect. The different degrees of elimination appear to be dependent on the clone of the antibody and the manufacturing company. This might be explained on the 1 hand by the different binding specificities of the 2 clones and on the other hand by the fact that differences in the method of antibody production influence the fluorochrome: antibody ratio and alter functional properties of the conjugated antibodies [23]. Despite these considerations, our findings suggest that the eliminating effect of FITC-coupled anti-Ly6G antibodies is not restricted to the 1A8 clone by 1 specific company (BD Biosciences) but can also be induced by a different antibody clone (RB6-8C5) and by the 1A8 clone produced by another company (Miltenyi Biotec).
Volume 98, September 2015
Journal of Leukocyte Biology 5
Collectively, these results demonstrate that in vivo applications of fluorescent anti-Ly6G at 37°C require rigorous in vitro and in vivo testing of the respective antibody. This is especially important taking into consideration that the used fluorescent antibodies are designed for in vitro labeling at markedly lower temperatures.
Direct antibody injection of 1A8-FITC depletes neutrophils to the same degree as injection of purified 1A8 The above findings open up the possibility that the preferential elimination of FITC-labeled neutrophils results from an immunogenic property of the antibody-conjugated FITC molecule. FITC is a well-known sensitizer in murine models of contact dermatitis [29, 30] and acquires its immunogenic capacity after binding to proteins [29, 30]. Upon dermal application, the fluorochrome preferentially accumulates in antigen-presenting cells and induces pronounced immune responses [31, 32]. Thus, if the observed effects were a result of the immunogenic properties of the antibody-coupled FITC molecule, then coincubation of BMM with neutrophils that were stained with 1A8-FITC should lead to
a similar phagocytosis rate as coincubation with beads labeled with identical antibody. To test this, CompBeads, which couple to the k light chain of IgG antibodies, were separately stained with 1A8APC, -FITC, or -PE. However, coincubation with BMM did not result in increased percentages of 1A8-FITC+ BMM (Supplemental Fig. 3). This implied that the internalization of the FITC-labeled neutrophils by macrophages was not a result of a general effect of 1A8-mediated FITC labeling or the immunogenic properties of FITC per se but specifically, required the binding of the 1A8-FITC antibody to the Ly6G protein on the neutrophils. Interestingly, the binding of purified anti-Ly6G antibodies to Ly6G proteins depletes neutrophils in vivo [33]. Taken together, this suggested that 1A8FITC, similar to purified anti-Ly6G antibodies, can exert a neutrophildepleting effect if used under in vivo conditions. To test this, 30 mg purified or FITC- or PE-coupled 1A8 was injected into the tail vein of mice. We found that 15 min after antibody injection, neutrophil numbers were significantly lower in blood and spleen of 1A8-purified and -FITC-treated mice compared with 1A8-PEtreated mice (Fig. 5A and B). The depleting effect of 1A8-FITC was similar to that of the purified antibody (Fig. 5B), which may suggest that 1A8-FITC depletes neutrophils by the same mechanism as purified 1A8.
Figure 5. Antibody injection of FITC-coupled 1A8 depletes neutrophils to the same degree as uncoupled 1A8. Mice were injected intravenously with 30 mg uncoupled or FITC- or PE-coupled 1A8. After 15 min, splenic and blood leukocytes from recipients were counterstained with antibodies against CD11b, F4/80, and 1A8-APC and were analyzed by flow cytometry. (A) To determine percentages of neutrophils in spleen and blood, cells were gated as CD11b+ F4/802 singlet leukocytes and were analyzed for the presence of SSCint 1A8-APC+ neutrophils. The circular gates in the far-right graphs contain 1A8-FITC/1A8-APCcolabeled neutrophils (far-right middle) and 1A8PE/1A8-APC-colabeled neutrophils (far-right bottom) of all gated CD11b+ F4/802 leukocytes. Note that the 1A8-FITC+ 1A8-APC+ population is less detectable than the 1A8-PE+ 1A8-APC+ population. (B and C) Percentages of SSCint 1A8-APC+ neutrophils of all gated CD11b+ F4/802 leukocytes in spleen (B) and blood (C). Bars represent means 6 SD. Data were pooled from 2 independent experiments. Data were analyzed by ANOVA, followed by Bonferroni’s multiple comparison test; n = 5; *P , 0.05; **P , 0.01.
6
Journal of Leukocyte Biology
Volume 98, September 2015
www.jleukbio.org
Bucher et al. Fluorochromes alter functional properties of anti-Ly6G
APC, FITC, and PE differentially modify access to binding sites on the 1A8 antibody It has been demonstrated that neutrophil depletion by anti-Ly6G involves cross-linking of Ly6G and requires the Fc fragment of the antibody [11]. Importantly, the Fc fragment of IgG antibodies harbors specific binding sites for FcgR and complement protein C1q. FcgRs are present on many cell types, including macrophages, and IgG binding to FcgR stimulates uptake of IgGcoated cells by macrophages [34]. C1q is produced by cells of the monocyte-macrophage lineage [35], and IgG coating can induce cell depletion via the activation of the complement system [36]. Interestingly, the used, 1A8-coupling fluorochromes APC, FITC, and PE show great differences in their chemical structure and molecular mass (APC, 150 kDa; FITC, 0.39 kDa; and PE, 240 kDa). Moreover, as antibody glycosylation is important for complement activation [37] and FcR binding [38, 39], the method of antibody production or the chemistry of fluorochrome conjugation may affect the glycosylation sites on the antibody, thereby enhancing or preventing binding to FcR or complement. Therefore, we hypothesized that fluorochrome coupling may alter the functional properties of 1A8 (;150 kDa) by interfering with functionally relevant interaction sites of the antibody. The degree of this effect might be determined by the molecular mass and/or the biochemical characteristics of the fluorochromes, their binding position on the antibody, as well as their potential influence on antibody glycosylation. Protein G is known to bind to Fab and Fc fragments of IgG [40]. Hence, to test whether APC, FITC, and PE can, in general, modify access to interaction sites on the anti-Ly6G antibody, we determined protein G-binding to the 3 fluorescent 1A8 antibodies. To this end, BM cells were separately stained with 1A8APC, -FITC, or -PE; pooled; and incubated with biotinylated protein G, followed by SA-PerCp-Cy5.5 staining. We found that protein G binding was significantly lower in 1A8-APC+ or -PE+ cells compared with 1A8-FITC+ cells (Fig. 6), indicating that APC and PE impaired protein G binding to 1A8 antibodies. These results demonstrate that coupling fluorochromes can, in principle, alter the accessibility to binding sites on 1A8. If these differences in the accessibility also extend to depletion-relevant binding sites, then one could assume that APC and PE modify the functional properties of 1A8 by impeding interaction sites required for the depleting function. In contrast, in FITC-coupled antibodies, these interaction sites would be accessible, enabling neutrophil depletion by 1A8. Alternately, fluorochrome conjugation may determine the depleting capacity of anti-Ly6G by other mechanisms, for instance, by affecting glycosylation sites on the antibody. In summary, we demonstrate that fluorochromes coupled to anti-Ly6G can differentially influence the functional properties of the Ly6G antibody. Fluorochromes control ligand binding to neutrophil-bound anti-Ly6G, determine macrophage uptake of anti-Ly6G-labeled neutrophils, and alter their retrieval after adoptive cell transfer or intravenous injection of anti-Ly6G antibodies into mice. Moreover, our results identify neutrophil phagocytosis by macrophages as a so-far-unrecognized mechanism involved in anti-Ly6G-induced neutrophil depletion. Taken together, this knowledge is essential for a valid interpretation and
www.jleukbio.org
Figure 6. APC and PE but not FITC coupling inhibits protein G binding to the 1A8 antibody. To determine protein G binding to fluorochromecoupled 1A8, equal fractions of 1A8-APC-, -FITC-, or -PE-labeled BM cells were pooled and incubated with or without biotinylated protein G. Thereafter, cells were washed and stained with SA-PerCP-Cy5.5 and with antibodies against CD3e, CD11b, and CD19. Then, cells were analyzed by flow cytometry. 1A8-APC-, -FITC-, or -PE-labeled neutrophils were gated as shown in Fig. 1, and protein G-binding to 1A8-labeled cell populations was determined by the geometric mean fluorescence intensity (GMFI) of SA-PerCP-Cy5.5. Bars represent means 6 SD. Data were pooled from 2 independent experiments and were analyzed by ANOVA, followed by Bonferroni’s multiple comparison test; n = 6; ***P , 0.001.
improved experimental design of in vivo experiments that use anti-Ly6G antibodies.
AUTHORSHIP K.B. designed and carried out the experiments, analyzed the data, and wrote the manuscript. F.S. and I.D. performed experiments. S.E.A. performed experiments and analyzed data. B.N. and K.S-L. provided conceptual advice and edited the manuscript. S.B.-H. designed experiments, provided experimental and conceptual advice, and wrote the manuscript.
ACKNOWLEDGMENTS The study was supported by FOR729 from the Deutsche Forschungsgemeinschaft and Interfaculty Centre for Pharmacogenomics and Pharma Research, Mouse Clinic, Eberhard-KarlsUniversity (Tubingen, ¨ Germany). The authors acknowledge Prof. Dr. Christian Harteneck and Dr. Veronika Leiss for helpful discussion, Dr. Claudia de Oliveira Franz for performing intravenous injections, Silvia Vetter for carrying out liver cell isolation, and Benedikt Mothes for experimental help. DISCLOSURES
The authors declare no conflict of interest.
Volume 98, September 2015
Journal of Leukocyte Biology 7
REFERENCES
1. Lee, P. Y., Wang, J. X., Parisini, E., Dascher, C. C., Nigrovic, P. A. (2013) Ly6 family proteins in neutrophil biology. J. Leukoc. Biol. 94, 585–594. 2. Wang, J. X., Bair, A. M., King, S. L., Shnayder, R., Huang, Y. F., Shieh, C. C., Soberman, R. J., Fuhlbrigge, R. C., Nigrovic, P. A. (2012) Ly6G ligation blocks recruitment of neutrophils via a b2-integrin-dependent mechanism. Blood 120, 1489–1498. 3. Fleming, T. J., Fleming, M. L., Malek, T. R. (1993) Selective expression of Ly-6G on myeloid lineage cells in mouse bone marrow. RB6-8C5 mAb to granulocyte-differentiation antigen (Gr-1) detects members of the Ly-6 family. J. Immunol. 151, 2399–2408. 4. Jutila, M. A., Kroese, F. G., Jutila, K. L., Stall, A. M., Fiering, S., Herzenberg, L. A., Berg, E. L., Butcher, E. C. (1988) Ly-6C is a monocyte/macrophage and endothelial cell differentiation antigen regulated by interferon-gamma. Eur. J. Immunol. 18, 1819–1826. 5. Kung, J. T., Castillo, M., Heard, P., Kerbacher, K., Thomas III, C. A. (1991) Subpopulations of CD8+ cytotoxic T cell precursors collaborate in the absence of conventional CD4+ helper T cells. J. Immunol. 146, 1783–1790. 6. Jutila, D. B., Kurk, S., Jutila, M. A. (1994) Differences in the expression of Ly-6C on neutrophils and monocytes following PI-PLC hydrolysis and cellular activation. Immunol. Lett. 41, 49–57. 7. Shortman, K., Naik, S. H. (2007) Steady-state and inflammatory dendritic-cell development. Nat. Rev. Immunol. 7, 19–30. 8. Zhou, J., Stohlman, S. A., Hinton, D. R., Marten, N. W. (2003) Neutrophils promote mononuclear cell infiltration during viral-induced encephalitis. J. Immunol. 170, 3331–3336. 9. Bliss, S. K., Gavrilescu, L. C., Alcaraz, A., Denkers, E. Y. (2001) Neutrophil depletion during Toxoplasma gondii infection leads to impaired immunity and lethal systemic pathology. Infect. Immun. 69, 4898–4905. 10. Tateda, K., Moore, T. A., Deng, J. C., Newstead, M. W., Zeng, X., Matsukawa, A., Swanson, M. S., Yamaguchi, K., Standiford, T. J. (2001) Early recruitment of neutrophils determines subsequent T1/T2 host responses in a murine model of Legionella pneumophila pneumonia. J. Immunol. 166, 3355–3361. 11. Abbitt, K. B., Cotter, M. J., Ridger, V. C., Crossman, D. C., Hellewell, P. G., Norman, K. E. (2009) Antibody ligation of murine Ly-6G induces neutropenia, blood flow cessation, and death via complement-dependent and independent mechanisms. J. Leukoc. Biol. 85, 55–63. 12. Morton, J., Coles, B., Wright, K., Gallimore, A., Morrow, J. D., Terry, E. S., Anning, P. B., Morgan, B. P., Dioszeghy, V., K¨uhn, H., Chaitidis, P., Hobbs, A. J., Jones, S. A., O’Donnell, V. B. (2008) Circulating neutrophils maintain physiological blood pressure by suppressing bacteria and IFNgamma-dependent iNOS expression in the vasculature of healthy mice. Blood 111, 5187–5194. 13. Ribechini, E., Leenen, P. J., Lutz, M. B. (2009) Gr-1 antibody induces STAT signaling, macrophage marker expression and abrogation of myeloid-derived suppressor cell activity in BM cells. Eur. J. Immunol. 39, 3538–3551. 14. Martin, C., Burdon, P. C., Bridger, G., Gutierrez-Ramos, J. C., Williams, T. J., Rankin, S. M. (2003) Chemokines acting via CXCR2 and CXCR4 control the release of neutrophils from the bone marrow and their return following senescence. Immunity 19, 583–593. 15. Phillipson, M., Heit, B., Colarusso, P., Liu, L., Ballantyne, C. M., Kubes, P. (2006) Intraluminal crawling of neutrophils to emigration sites: a molecularly distinct process from adhesion in the recruitment cascade. J. Exp. Med. 203, 2569–2575. 16. Hut´as, G., Bajnok, E., G´al, I., Finnegan, A., Glant, T. T., Mikecz, K. (2008) CD44-specific antibody treatment and CD44 deficiency exert distinct effects on leukocyte recruitment in experimental arthritis. Blood 112, 4999–5006. 17. McDonald, B., Pittman, K., Menezes, G. B., Hirota, S. A., Slaba, I., Waterhouse, C. C., Beck, P. L., Muruve, D. A., Kubes, P. (2010) Intravascular danger signals guide neutrophils to sites of sterile inflammation. Science 330, 362–366. 18. Devi, S., Li, A., Westhorpe, C. L., Lo, C. Y., Abeynaike, L. D., Snelgrove, S. L., Hall, P., Ooi, J. D., Sobey, C. G., Kitching, A. R., Hickey, M. J. (2013) Multiphoton imaging reveals a new leukocyte recruitment paradigm in the glomerulus. Nat. Med. 19, 107–112. 19. Yipp, B. G., Kubes, P. (2013) Antibodies against neutrophil LY6G do not inhibit leukocyte recruitment in mice in vivo. Blood 121, 241–242. 20. Beer-Hammer, S., Zebedin, E., von Holleben, M., Alferink, J., Reis, B., Dresing, P., Degrandi, D., Scheu, S., Hirsch, E., Sexl, V., Pfeffer, K., N¨urnberg, B., Piekorz, R. P. (2010) The catalytic PI3K isoforms p110gamma and p110delta contribute to B cell development and
8
Journal of Leukocyte Biology
Volume 98, September 2015
21.
22. 23. 24. 25.
26. 27.
28.
29. 30.
31.
32. 33. 34. 35. 36. 37.
38. 39. 40.
maintenance, transformation, and proliferation. J. Leukoc. Biol. 87, 1083–1095. Schmitz, H. J., Hagenmaier, A., Hagenmaier, H. P., Bock, K. W., Schrenk, D. (1995) Potency of mixtures of polychlorinated biphenyls as inducers of dioxin receptor-regulated CYP1A activity in rat hepatocytes and H4IIE cells. Toxicology 99, 47–54. McCormack, T., O’Keeffe, G., Mac Craith, B., O’Kennedy, R. (1996) Assessment of the effect of increased fluorophore labelling on the binding ability of an antibody. Anal. Lett. 29, 953–968. Vira, S., Mekhedov, E., Humphrey, G., Blank, P. S. (2010) Fluorescentlabeled antibodies: balancing functionality and degree of labeling. Anal. Biochem. 402, 146–150. Furze, R. C., Rankin, S. M. (2008) The role of the bone marrow in neutrophil clearance under homeostatic conditions in the mouse. FASEB J. 22, 3111–3119. Daubeuf, S., Puaux, A. L., Joly, E., Hudrisier, D. (2006) A simple trogocytosis-based method to detect, quantify, characterize and purify antigen-specific live lymphocytes by flow cytometry, via their capture of membrane fragments from antigen-presenting cells. Nat. Protoc. 1, 2536–2542. Gillette, J. M., Larochelle, A., Dunbar, C. E., Lippincott-Schwartz, J. (2009) Intercellular transfer to signalling endosomes regulates an ex vivo bone marrow niche. Nat. Cell Biol. 11, 303–311. Savill, J. S., Wyllie, A. H., Henson, J. E., Walport, M. J., Henson, P. M., Haslett, C. (1989) Macrophage phagocytosis of aging neutrophils in inflammation. Programmed cell death in the neutrophil leads to its recognition by macrophages. J. Clin. Invest. 83, 865–875. Suratt, B. T., Young, S. K., Lieber, J., Nick, J. A., Henson, P. M., Worthen, G. S. (2001) Neutrophil maturation and activation determine anatomic site of clearance from circulation. Am. J. Physiol. Lung Cell. Mol. Physiol. 281, L913–L921. Brinkley, M. (1992) A brief survey of methods for preparing protein conjugates with dyes, haptens, and cross-linking reagents. Bioconjug. Chem. 3, 2–13. Samuelsson, K., Simonsson, C., Jonsson, C. A., Westman, G., Ericson, M. B., Karlberg, A. T. (2009) Accumulation of FITC near stratum corneum-visualizing epidermal distribution of a strong sensitizer using two-photon microscopy. Contact Dermat. 61, 91–100. Macatonia, S. E., Knight, S. C., Edwards, A. J., Griffiths, S., Fryer, P. (1987) Localization of antigen on lymph node dendritic cells after exposure to the contact sensitizer fluorescein isothiocyanate. Functional and morphological studies. J. Exp. Med. 166, 1654–1667. Pior, J., Vogl, T., Sorg, C., MacHer, E. (1999) Free hapten molecules are dispersed by way of the bloodstream during contact sensitization to fluorescein isothiocyanate. J. Invest. Dermatol. 113, 888–893. Daley, J. M., Thomay, A. A., Connolly, M. D., Reichner, J. S., Albina, J. E. (2008) Use of Ly6G-specific monoclonal antibody to deplete neutrophils in mice. J. Leukoc. Biol. 83, 64–70. Chang, N. S., Leu, R. W., Rummage, J. A., Anderson, J. K., Mole, J. E. (1992) Regulation of macrophage Fc receptor expression and phagocytosis by histidine-rich glycoprotein. Immunology 77, 532–538. Petry, F., Botto, M., Holtappels, R., Walport, M. J., Loos, M. (2001) Reconstitution of the complement function in C1q-deficient (C1qa-/-) mice with wild-type bone marrow cells. J. Immunol. 167, 4033–4037. Duncan, A. R., Winter, G. (1988) The binding site for C1q on IgG. Nature 332, 738–740. Tao, M. H., Morrison, S. L. (1989) Studies of aglycosylated chimeric mouse-human IgG. Role of carbohydrate in the structure and effector functions mediated by the human IgG constant region. J. Immunol. 143, 2595–2601. Nose, M., Wigzell, H. (1983) Biological significance of carbohydrate chains on monoclonal antibodies. Proc. Natl. Acad. Sci. USA 80, 6632–6636. Wright, A., Morrison, S. L. (1997) Effect of glycosylation on antibody function: implications for genetic engineering. Trends Biotechnol. 15, 26–32. Aybay, C. (2003) Differential binding characteristics of protein G and protein A for Fc fragments of papain-digested mouse IgG. Immunol. Lett. 85, 231–235.
KEY WORDS: cell tracking APC FITC PE granulocytes •
•
•
•
www.jleukbio.org
Brief Conclusive Report
Fluorescent Ly6G antibodies determine macrophage phagocytosis of neutrophils and alter the retrieval of neutrophils in mice Kirsten Bucher,* Fee Schmitt,* Stella E. Autenrieth,† Inken Dillmann,* Bernd N¨ urnberg,* Katja Schenke-Layland,‡,§,{ and Sandra Beer-Hammer*,1 *Department of Pharmacology and Experimental Therapy, Institute of Experimental and Clinical Pharmacology and Toxicology, and ‡University Women’s Hospital, Eberhard-Karls-University T¨ubingen, Germany; †Department of Hematology, Oncology, Immunology, Rheumatology and Pulmology, University Hospital Tu¨ bingen, Germany; §Department of Cell and Tissue Engineering, Fraunhofer Institute for Interfacial Engineering and Biotechnology, Stuttgart, Germany; and {Department of Medicine/Cardiology, Cardiovascular Research Laboratories, David Geffen School of Medicine at University of California Los Angeles, California, USA RECEIVED OCTOBER 16, 2014; REVISED MAY 5, 2015; ACCEPTED MAY 5, 2015. DOI: 10.1189/jlb.1AB1014-488RR
ABSTRACT
Introduction
Fluorescently labeled Ly6G antibodies enable the tracking of neutrophils in mice, whereas purified antiLy6G rapidly depletes neutrophils from the circulation. The mechanisms underlying neutrophil depletion are still under debate. Here, we examined how identical Ly6G antibodies coupled to different fluorochromes affect neutrophil fate in vivo. BM cells stained with Ly6G antibodies were injected into mice. The number of retrieved anti-Ly6G-FITC+ cells was reduced significantly in comparison with anti-Ly6G-APC + or antiLy6G-PE+ cells. Flow cytometry and multispectral imaging flow cytometry analyses revealed that antiLy6G-FITC+ neutrophils were preferentially phagocytosed by BMMs in vitro and by splenic, hepatic, and BM macrophages in vivo. Direct antibody injection of antiLy6G-FITC but not anti-Ly6G-PE depleted neutrophils to the same degree as purified anti-Ly6G, indicating that the FITC-coupled antibody eliminates neutrophils by a similar mechanism as the uncoupled antibody. With the use of a protein G-binding assay, we demonstrated that APC and PE but not FITC coupling inhibited access to interaction sites on the anti-Ly6G antibody. We conclude the following: 1) that neutrophil phagocytosis by macrophages is a central mechanism in anti-Ly6G-induced neutrophil depletion and 2) that fluorochrome-coupling can affect functional properties of anti-Ly6G antibodies, thereby modifying macrophage uptake of Ly6G-labeled neutrophils and neutrophil retrieval following adoptive cell transfer or injection of fluorescent anti-Ly6G. J. Leukoc. Biol. 98: 000–000; 2015.
Ly6G is a glycosylphosphatidylinositol-linked cell-surface protein that is expressed predominantly on murine neutrophils [1]. The function of Ly6G has still not been fully clarified, although a recent study implicates Ly6G in neutrophil migration [2]. There are 2 mAb, i.e., RB6-8C5 and 1A8, which recognize Ly6G [3]. Whereas clone 1A8 specifically binds to Ly6G, clone RB6-8C5 additionally cross-reacts with Ly6C [3], which is also present on dendritic cells, subpopulations of macrophages, monocytes, and lymphocytes [4–7]. RB6-8C5 and 1A8 are important tools in neutrophil research, both in a fluorochrome-coupled and in an untagged form. Untagged antibodies are commonly referred to as purified antibodies. In vivo, administration of high doses of purified antiLy6G rapidly depletes neutrophils from the circulation. This is a standard method to investigate the role of neutrophils in a wide variety of disease models, including experimental encephalitis [8], as well as microbial and parasitic infections [9, 10]. The mechanisms leading to neutrophil depletion remain to be determined. Abbitt et al. [11] have shown that anti-Ly6G-mediated neutrophil depletion involves the cross-linking of Ly6G and is dependent on the Fc fragment of the antibody but does not require FcgR engagement. Moreover, the mechanism appears to involve opsonization of anti-Ly6G-coated cells with complement component C3b, as C32/2 mice but not mice treated with anti-C5 antibodies are protected against anti-Ly6G-induced neutrophil depletion [12]. However, studies analyzing the role of anti-Ly6G in the induction of apoptosis led to contradictory results [2, 13]. Whereas high doses of purified anti-Ly6G antibodies are used for neutrophil depletion, low doses of fluorochrome-coupled anti-Ly6G antibodies are used to analyze neutrophil trafficking by flow cytometry [14] or intravital microscopy [11, 15–19]. This might imply that the dose of the antibody determines whether anti-Ly6G affects neutrophil function and survival. However, it
Abbreviations: 7-AAD = 7-aminoactinomycin D, APC = allophycocyanin, BM = bone marrow, BMM = bone marrow-derived macrophage, SA = streptavidin The online version of this paper, found at www.jleukbio.org, includes supplemental information.
0741-5400/15/0098-0001 © Society for Leukocyte Biology
1. Correspondence: Institute of Experimental and Clinical Pharmacology and Toxicology, University Hospital Tuebingen, Wilhelmstrasse 56, 72074 Tuebingen, Germany. E-mail: [email protected]
Volume 98, September 2015
Journal of Leukocyte Biology 1
Copyright 2015 by The Society for Leukocyte Biology.
was recently shown that furthermore, low doses of purified antiLy6G impair neutrophil recruitment and attenuate experimental arthritis [2]. Interestingly, Yipp and Kubes [19] demonstrated that low doses of fluorochrome-coupled anti-Ly6G do not influence neutrophil migration. This suggests that the diverse effects induced by anti-Ly6G antibodies may not only be dose dependent but are also determined by coupled fluorochromes. Accordingly, Lee et al. [1] speculated that fluorochromes may alter the functional properties of the anti-Ly6G antibody. Here, we studied how fluorochrome-coupled anti-Ly6G antibodies affect neutrophil retrieval in mice. We demonstrate the following: 1) that identical antibodies conjugated with distinct fluorochromes differentially determine the fate of the labeled neutrophils and 2) that neutrophil phagocytosis by macrophages may be a central mechanism involved in anti-Ly6G-induced neutrophil depletion.
MATERIALS AND METHODS
Mice C57BL/6 mice (Charles River Laboratories, Sulzfeld, Germany) were kept under specific pathogen-free conditions. All animal work was performed according to the German animal care regulations and was approved by the local authorities.
Preparation of leukocyte suspensions BM cells and splenocytes were isolated as described previously [20]. Lungs were cut into pieces and incubated with 3 ml collagenase D solution (IMDM containing 5% FCS; both from PAA, Linz, Austria) and 0.2 mg/ml Collagenase D and 0.01 mg/ml DNase I; both from Roche, Mannheim, Germany) at 37°C for 30 min. Then, 2 ml IMDM containing 0.2 mg/ml Collagenase/Dispase (Roche) were added, and the tissue was incubated at 37°C for another 15 min. EDTA solution (0.5 M) was added to a final concentration of 5 mM EDTA and incubated for 5 min. Hepatocytes were isolated according to Schmitz et al. [21] with the following modifications: livers were perfused at 37°C, first with calcium-free buffer for 2 min, followed by calcium-containing collagenase buffer (15 mg/100 ml collagenase type IV; Sigma, St. Louis, MO, USA) for 10 min. Macrophages were enriched in the supernatant by centrifugation at 20 g for 2 min. Leukocytes from single-cell suspensions of BM, lung, heparinized peripheral blood, liver, and spleen were prepared by RBC lysis in ACK buffer (0.155 M NH4Cl, 0.01 M KHCO3, 0.1 mM EDTA).
Flow cytometry and ImageStream analysis For characterization of cells, the following antibodies or peptides were used: 7-AAD; Annexin V-V500; CD3e-Pacific Blue (500A2); CD11b-PE.Cy7 (M1/70); CD16/32 (2.4G2); CD19-V450 (1D3); CD45R-APC, -FITC, and -PE (RA3-6B2); Ly6G-APC, -FITC, and -PE (1A8); Ly6G-FITC and -PE (RB6-8C5); and SAPerCP-Cy5.5 (all BD Biosciences, Heidelberg, Germany). Also used were: F4/ 80 Pacific Blue (BM8; BioLegend, San Diego, CA, USA); Ly6G-APC (RB6-8C5; eBioscience, San Diego, CA, USA); Ly6G-APC, -FITC, and -PE (1A8; Miltenyi Biotec, Bergisch Gladbach, Germany); and protein G-biotin (Pierce Biotechnology, Rockford, IL, USA). CompBeads were from BD Biosciences. Flow cytometry measurements were performed on a FACSCanto II (BD Biosciences), and data were evaluated with FlowJo software. Where indicated, cells were additionally analyzed by use of ImageStreamX, and data were evaluated with Integrated Design and Engineering Analysis (IDEAS) software. Cells were gated on focus, followed by F4/80+ gating to identify BMM, which were visually counted for attached (1A8+ cells at the plasma membrane of F4/80+ BMM) or intracellular (1A8+ cells completely inside the cytoplasma of F4/80+ BMM), fluorescently labeled neutrophils. BMMs (1000 cells) were counted in each sample.
Adoptive transfer of anti-Ly6G-labeled BM leukocytes Equal fractions of freshly prepared BM leukocytes were separately stained with anti-Ly6G-APC, -FITC, or -PE (1A8) at a concentration of 0.25 mg/million cells
2
Journal of Leukocyte Biology
Volume 98, September 2015
in 100 ml RPMI plus 2% FCS on ice. After washing twice with PBS, the separately stained anti-Ly6G-APC+, -FITC+, and -PE+ BM leukocytes were pooled, and 2.5 3 107 cells in 150 ml PBS were injected into the tail vein of each recipient mouse. In each experiment, an aliquot of the injection suspension was retained to determine percentages of APC-, FITC-, and PElabeled neutrophils. After the indicated time, the recipients were euthanized, and leukocytes from organs were prepared as detailed above.
Generation of BMM and coculture of anti-Ly6G-labeled BM leukocytes with BMM BMMs were generated from murine BM cells in RPMI 1640 (with 10% FCS, 100 units/ml penicillin, 100 mg/ml streptomycin, 2 mM L-glutamine, and 0.05 mM b-ME (all from PAA) in the presence of 30 ng/ml mouse rM-CSF (R&D Systems, Wiesbaden, Germany). Medium was exchanged on days 2, 4, and 6, and BMMs were harvested on day 8. BM cells stained with 1A8-APC, -FITC, or -PE were prepared and pooled as described above. BMMs were seeded at 2 3 105 cells/well into 96-well plates. Following incubation at 5% CO2 and 37°C for 3 h, the medium was removed and the labeled BM leukocytes (1 3 106 cells in 200 ml) were added to each well and incubated for 10, 20, 30, 45, and 60 min. Wells containing only BM cells or only BMM served as gating controls. In addition, samples of BM cells and BMM were coincubated on ice and served as 4°C control. Cultured cells were harvested by use of ice-cold PBS-EDTA and subsequently analyzed by flow cytometry or ImageStreamx.
Statistical analysis Statistical analysis was performed as indicated in the figure legends. A value of P # 0.05 was considered statistically significant.
RESULTS AND DISCUSSION
1A8-FITC labeling reduces retrieval of neutrophils following adoptive transfer Here, we studied whether and how anti-Ly6G-coupled fluorochromes affect neutrophil retrieval/tracking in vivo and compared the retrieval rates of 1A8-APC-, -FITC-, or -PE-labeled BM cells following adoptive transfer in mice. Equal fractions of BM cells were separately stained with 1A8-APC, -FITC, or -PE. Then, cells were pooled and intravenously injected into recipients. After 1 h, leukocyte suspensions from BM, lungs, blood, and spleens were analyzed by use of flow cytometry (Fig. 1A and B). The frequency of retrieved 1A8-APC+ and -PE+ neutrophils was comparable in all organs, resulting in a 1.1 ratio of 1A8-PE+:1A8-APC+ cells. In contrast, the frequency of retrieved 1A8-FITC+ cells was reduced by ;60%, resulting in a 0.47 ratio of 1A8-FITC+:1A8-APC+ cells and a ratio of 0.43 of 1A8-FITC+:1A8-PE+ cells (Fig. 1C and D). The decreased detectability of the 1A8-FITC+ neutrophils may result from a degradation of the FITC molecule as a result of its lower thermal stability in vivo. Alternatively, as fluorochrome conjugation can reduce the avidity of a given antibody [22, 23], the decreased retrieval of 1A8-FITC+ neutrophils might be explained by a reduced binding ability of 1A8-FITC to Ly6G. Therefore, we tested labeling of neutrophils with fluorescent 1A8 under physiologic conditions. To this end, 3 fractions of BM cells were separately stained with 1A8-APC, -FITC, or -PE. Thereafter, the cells were pooled, incubated at 37°C, and analyzed by flow cytometry. We found that percentages of 1A8FITC+ cells, as well as of 1A8-APC+ and -PE+ cells, remained stable under cell culture conditions (Fig. 1E). This showed that the decreased detectability of the 1A8-FITC+ neutrophils (Fig. 1C and D) was neither a result of a reduced binding ability of
www.jleukbio.org
Bucher et al. Fluorochromes alter functional properties of anti-Ly6G
1A8-FITC to neutrophils nor of lower thermal stability of the FITC molecule. Furthermore, by use of AnnexinV/7-AAD staining, we did not observe an elevated apoptosis rate or cell death of 1A8-FITC-labeled neutrophils compared with 1A8-APCor -PE-labeled cells (Supplemental Fig. 1A and B), suggesting that the decreased retrieval of 1A8-FITC-labeled neutrophils in mice was not a result of an increased induction of apoptosis. Moreover, although the expression of the activation marker CD11b decreased over time under cell culture conditions, it was similar among 1A8-APC-, -FITC-, and -PE-stained cells at all timepoints (Supplemental Fig. 1A and C). These results suggest that 1A8-FITC labeling did not differentially alter the activation state of neutrophils. Therefore, we hypothesized that the 1A8FITC-labeled neutrophils become rapidly cleared in the mouse.
1A8-FITC-labeled BM cells are preferentially phagocytosed by macrophages in vitro and in vivo
Figure 1. Fluorochromes coupled to anti-Ly6G (clone 1A8) differentially influence neutrophil retrieval in mice. (A) Experimental set-up to compare the effect of anti-Ly6G-APC, -FITC, and -PE (clone 1A8; BD Biosciences) labeling on the retrieval of adoptively transferred neutrophils. BM leukocytes from donor mice were separately stained with 1A8APC, -FITC, or -PE and coinjected into recipients. One hour later, leukocytes from BM, blood, spleens, and lungs from recipients were stained with antibodies against CD3e, CD11b, and CD19 and were analyzed by flow cytometry. (B) Myeloid cells were gated as CD11b+ CD3e2 CD192 singlet leukocytes. They were analyzed for the presence of 1A8-APC-, -FITC-, or -PE-labeled neutrophils. FSC-A, Forward scatter-area; FSC-W, FSC-width; SSC-A, side-scatter-area. (C) 1A8-APC-, -FITC-, or -PElabeled neutrophils retrieved in BM, blood, spleen, and lung of each recipient mouse. Data are expressed as percentages of 1A8-labeled cells of all CD11b+ cells and are shown as means 6 SD. Data were analyzed by ANOVA, followed by Bonferroni’s multiple comparison test; n = 3; *P , 0.05; **P , 0.01; ***P , 0.001. (D) For all organs of each mouse, ratios were calculated as percent retrieved 1A8-PE+ cells/percent retrieved 1A8-APC+ cells (d), percent retrieved 1A8-FITC+ cells/percent retrieved 1A8-APC+ cells (n), and percent retrieved A8-FITC+ cells/percent retrieved 1A8-PE+ cells (:). All calculated ratios are depicted. Bars indicate means 6 SD. (E) Coincubation of 1A8-APC-, -FITC-, and -PE-labeled BM cells on ice [4°C control (ctrl)] or at 37°C and 5% CO2 for 5, 30, and 60 min. Neutrophil labeling with 1A8-APC, -FITC, or -PE is stable at 37°C at all investigated time-points.
Phagocytic macrophages in the BM are known to mediate the clearance of neutrophils [24]. Thus, we tested whether macrophages preferentially take up 1A8-FITC-labeled neutrophils. Unstimulated BMMs were coincubated with 3 equal fractions of BM leukocytes, which had been separately stained with 1A8-APC, -FITC, or -PE, pooled, and analyzed by flow cytometry (Fig. 2A and B). Low percentages of 1A8-APC+ and -PE+ BMMs were detectable at all time-points following the onset of coincubation (Fig. 2B and C). Following coincubation on ice (4°C control) the percentage of 1A8-FITC+ BMM was similar to that of 1A8-APC+ or -PE+ BMM. However, percentages of 1A8-FITC+ BMM increased up to .10% and significantly exceeded those of 1A8-APC+ or -PE+ macrophages between 10 and 60 min of coincubation at 37°C (Fig. 2C). On the one hand, the high percentage of 1A8-FITC+ BMM may result from a preferential uptake of 1A8-FITC-labeled neutrophils. On the other hand, intercellular membrane exchanges by trogocytosis have been reported [25, 26]. These exchanges by themselves may lead to a transfer of the fluorescent antibodyantigen complex from 1 cell type to another. Thus, to examine the processes leading to a preferential increase in 1A8-FITC+ BMM, 1A8-labeled BM cells were coincubated with BMM, as described above. Cells were measured by multispectral imaging flow cytometry (ImageStreamX), a method that allows quantifying the location of fluorescent probes on, in, or between cells. We analyzed cell–cell interactions between macrophages and neutrophils, differentiating between attached and internalized neutrophils (Fig. 3). Ten minutes after coincubation, 1A8-FITC labeling of neutrophils resulted in an enhanced attachment to BMM (Fig. 3A and C), and a significantly higher number of 1A8FITC-labeled neutrophils was internalized by BMM compared with 1A8-APC (15-fold)- and -PE (30-fold)-stained cells (Fig. 3B and C). This confirmed that 1A8-FITC labeling resulted in an increased phagocytosis of neutrophils by macrophages. Percentages of 1A8-FITC+ BMM peaked between 10 and 20 min of coincubation and decreased thereafter (Fig. 2C), pointing to the possibility that internalized neutrophils were rapidly digested in the macrophages. This assumption would be in line with phagocytosis experiments that use human neutrophils, where macrophages engulfed and degraded neutrophils within 20–60 min [27].
www.jleukbio.org
Volume 98, September 2015
Journal of Leukocyte Biology 3
To test whether phagocytic macrophages preferentially eliminate 1A8-FITC+ neutrophils in vivo, we measured macrophage uptake of adoptively transferred 1A8-FITC- and -PE-labeled neutrophils by flow cytometry. As it is known that adoptively transferred neutrophils preferentially accumulate in the spleen, BM, and liver [24, 28], we analyzed macrophages in spleen, BM, and liver of recipient mice. Fifteen minutes following coinjection of 1A8-FITC- and -PE-labeled BM cells, the frequency of 1A8FITC+ F4/80+ macrophages was significantly higher than that of 1A8-PE+ macrophages in spleen and BM (Fig. 4A and B). Vice versa, the frequency of 1A8-FITC+ neutrophils was reduced significantly compared with 1A8-PE+ cells in both organs (Fig. 4C and D). Whereas leukocyte populations of spleen and BM could be isolated without delay, the preparation of liver cells required organ digestion by collagenase perfusion at 37°C [21]. Again, the frequency of 1A8-FITC+ F4/80+ macrophages was significantly higher than that of 1A8-PE+ macrophages in the liver (Fig. 4E). Taken together, our findings indicate that phagocytic macrophages preferentially clear 1A8-FITC-labeled neutrophils, corroborating our concept that neutrophil phagocytosis by
Figure 2. 1A8-FITC labeling of BM cells increases percentages of FITC+ BMM after coincubation of BM cells with BMM. Equal fractions of BM leukocytes were separately labeled with 1A8-APC, -FITC, or -PE (BD Biosciences). Fractions were pooled and coincubated with BMM at a ratio of 5:1. Cells were harvested at the indicated time-points, stained with anti-F4/80 and anti-CD11b, and analyzed by flow cytometry. (A) BMMs were gated as CD11b+ F4/80+ singlet leukocytes and analyzed for 1A8-APC, -FITC, or -PE staining. (B) Representative flow cytometry plots showing the frequencies of 1A8-APC+, -FITC+, or -PE+ cells within the BMM population after 20 or 60 min of coincubation at 37°C (BMM + BM cells). BMM and BM cells coincubated on ice served as a 4°C control (4°C ctrl). Pure BMM suspensions were analyzed as gating controls (BMM). (C) Percentages of 1A8-APC+, -FITC+, and -PE+ BMM between 0 and 60 min of coincubation. Between 10 and 60 min, percentages 1A8-FITC+ cells significantly exceeded those of 1A8-APC+ and -PE+ cells. Bars represent means 6 SD (n = 3; for 10 min, n = 13). Data were pooled from 6 independent experiments. Data were analyzed by ANOVA, followed by Bonferroni’s multiple comparison tests. ***P , 0.001.
4
Journal of Leukocyte Biology
Volume 98, September 2015
Figure 3. 1A8-FITC-labeled neutrophils are preferentially internalized by BMM. BM cells were labeled with 1A8-APC, -FITC, or -PE (BD Biosciences) and were coincubated with BMM, as described in Fig. 2. After harvest, cells were stained with anti-F4/80, and cell–cell interactions of F4/80+ BMM with 1A8-APC+, -FITC+, or -PE+ neutrophils were assessed by multispectral imaging flow cytometry (ImageStreamX). Representative bright field (BF) and fluorescence images of attached (A) and intracellular (B) 1A8-APC+, -FITC+, or -PE+ neutrophils are shown (defined as described in Materials and Methods). (C) Numbers of 1A8labeled attached or intracellular neutrophils/10,000 BMMs. Bars represent means 6 SD. Data were pooled from 2 independent experiments. Data were analyzed by one-way ANOVA, followed by Sidak’s multiple comparison test; n = 5; ****P , 0.0001.
www.jleukbio.org
Bucher et al. Fluorochromes alter functional properties of anti-Ly6G
Figure 4. Adoptive transfer of 1A8-labeled BM cells increases the frequency of FITC+ macrophages in the spleen, BM and liver of recipient mice. Equal fractions of 1A8-FITC- or -PE-labeled BM cells were pooled and coinjected into recipients. After 15 min, leukocytes from BM and spleen of recipient mice were isolated, stained with anti-F4/80 and antiCD11b and analyzed by flow cytometry. To analyze liver macrophages, recipient mice were terminally anesthetized 10 min following adoptive transfer, and livers were perfused at 37°C, first with calcium-free buffer for 2 min and then with collagenase buffer for 10 min. Liver cells were stained and analyzed as described above. (A) Macrophages from spleen, BM, and liver were gated as F4/80+ singlet leukocytes and analyzed for 1A8-FITC or -PE staining. (B) Percentages of 1A8-FITC+ and -PE+ macrophages of all gated splenic or BM macrophages. (C) To determine percentages of 1A8-FITC- or -PE-labeled neutrophils in spleen and BM, cells were gated as CD11b+ F4/802 singlet leukocytes and were analyzed for the presence of 1A8-FITC- or -PE-labeled neutrophils. (D) Percentages of 1A8-FITC- or -PE-labeled neutrophils of all gated CD11b+ F4/802 leukocytes 15 min after adoptive transfer. (E) Percentages of 1A8-FITC+ and -PE+ macrophages of all gated liver macrophages. Macrophages were gated as shown in A. Bars represent means 6 SD. Data were analyzed with unpaired t-test; n = 6; **P , 0.01; ***P , 0.001.
www.jleukbio.org
macrophages plays a central role in the elimination of 1A8-FITCbut not 1A8-PE-labeled neutrophils in vivo. To test whether the neutrophil-eliminating effect of 1A8-FITC was a general characteristic of FITC-coupled anti-Ly6G antibodies, we analyzed APC-, FITC-, and PE-coupled antibodies of another anti-Ly6G clone (RB6-8C5) and a 1A8 clone from another manufacturing company [1A8(M)]. To this end, again, we determined the retrieval of labeled neutrophils after adoptive transfer in vivo and macrophage uptake of labeled neutrophils following coincubation with BMM in vitro. In a first set of experiments, we examined the cell-labeling stability of fluorescent RB6-8C5 and 1A8(M) at 37°C (Supplemental Fig. 2A and B). The fluorescent RB6-8C5 antibodies caused a dye transfer from the labeled cells to other neutrophil populations, which were differently labeled. This effect was most pronounced by use of RB6-8C5-APC (Supplemental Fig. 2A) and rendered the data not evaluable. Therefore, further analyses of RB6-8C5 were restricted to RB6-8C5-FITC and -PE. In contrast, neutrophillabeling with 1A8(M)-APC, -FITC, or -PE was stable at 37°C at all investigated time-points (Supplemental Fig. 2B). Next, we determined the effect of neutrophil labeling with fluorescent RB6-8C5 after adoptive transfer in vivo. One hour after injection of equal fractions of RB6-8C5-FITC- and -PE-stained BM cells, the frequency of retrieved RB6-8C5-FITC+ cells was reduced by .90% compared with RB6-8C5-PE+ cells in BM, blood, spleen, and lung (Supplemental Fig. 2C), resulting in a 0.08 ratio of 1A8FITC+:1A8-PE+ cells (Supplemental Fig. 2D). This suggested that the eliminating effect of RB6-8C5-FITC was greater than that of 1A8-FITC (BD Biosciences; Fig. 1C and D). Consistently, coincubation experiments with BMM and RB6-8C5-stained BM cells showed that percentages of RB6-8C5-FITC+ BMM increased up to .20% and significantly exceeded those of RB6-8C5-PE+ macrophages after 10 min of coincubation (Supplemental Fig. 2E). Then, we determined the effect of neutrophil labeling with the 1A8(M) antibody in vivo and in vitro. Adoptive transfer of 1A8(M)-labeled BM cells led to comparable frequencies of retrieved 1A8-APC+ and -PE+ neutrophils in all organs, but frequencies of retrieved 1A8-FITC+ cells were reduced by ;40% (Supplemental Fig. 2F), resulting in a 0.6 ratio of 1A8-FITC+:1A8PE+ cells (Supplemental Fig. 2G). When we coincubated 1A8(M)stained BM cells with BMM, no differences among 1A8-APC+, -FITC+, or -PE+ BMM became apparent (Supplemental Fig. 2H). This points toward the eliminating effect of 1A8 from Miltenyi Biotec (Supplemental Fig. 2H) being smaller than that of 1A8 from BD Biosciences (Fig. 2C) and the in vitro model being too insensitive to observe any phagocytic effect. The different degrees of elimination appear to be dependent on the clone of the antibody and the manufacturing company. This might be explained on the 1 hand by the different binding specificities of the 2 clones and on the other hand by the fact that differences in the method of antibody production influence the fluorochrome: antibody ratio and alter functional properties of the conjugated antibodies [23]. Despite these considerations, our findings suggest that the eliminating effect of FITC-coupled anti-Ly6G antibodies is not restricted to the 1A8 clone by 1 specific company (BD Biosciences) but can also be induced by a different antibody clone (RB6-8C5) and by the 1A8 clone produced by another company (Miltenyi Biotec).
Volume 98, September 2015
Journal of Leukocyte Biology 5
Collectively, these results demonstrate that in vivo applications of fluorescent anti-Ly6G at 37°C require rigorous in vitro and in vivo testing of the respective antibody. This is especially important taking into consideration that the used fluorescent antibodies are designed for in vitro labeling at markedly lower temperatures.
Direct antibody injection of 1A8-FITC depletes neutrophils to the same degree as injection of purified 1A8 The above findings open up the possibility that the preferential elimination of FITC-labeled neutrophils results from an immunogenic property of the antibody-conjugated FITC molecule. FITC is a well-known sensitizer in murine models of contact dermatitis [29, 30] and acquires its immunogenic capacity after binding to proteins [29, 30]. Upon dermal application, the fluorochrome preferentially accumulates in antigen-presenting cells and induces pronounced immune responses [31, 32]. Thus, if the observed effects were a result of the immunogenic properties of the antibody-coupled FITC molecule, then coincubation of BMM with neutrophils that were stained with 1A8-FITC should lead to
a similar phagocytosis rate as coincubation with beads labeled with identical antibody. To test this, CompBeads, which couple to the k light chain of IgG antibodies, were separately stained with 1A8APC, -FITC, or -PE. However, coincubation with BMM did not result in increased percentages of 1A8-FITC+ BMM (Supplemental Fig. 3). This implied that the internalization of the FITC-labeled neutrophils by macrophages was not a result of a general effect of 1A8-mediated FITC labeling or the immunogenic properties of FITC per se but specifically, required the binding of the 1A8-FITC antibody to the Ly6G protein on the neutrophils. Interestingly, the binding of purified anti-Ly6G antibodies to Ly6G proteins depletes neutrophils in vivo [33]. Taken together, this suggested that 1A8FITC, similar to purified anti-Ly6G antibodies, can exert a neutrophildepleting effect if used under in vivo conditions. To test this, 30 mg purified or FITC- or PE-coupled 1A8 was injected into the tail vein of mice. We found that 15 min after antibody injection, neutrophil numbers were significantly lower in blood and spleen of 1A8-purified and -FITC-treated mice compared with 1A8-PEtreated mice (Fig. 5A and B). The depleting effect of 1A8-FITC was similar to that of the purified antibody (Fig. 5B), which may suggest that 1A8-FITC depletes neutrophils by the same mechanism as purified 1A8.
Figure 5. Antibody injection of FITC-coupled 1A8 depletes neutrophils to the same degree as uncoupled 1A8. Mice were injected intravenously with 30 mg uncoupled or FITC- or PE-coupled 1A8. After 15 min, splenic and blood leukocytes from recipients were counterstained with antibodies against CD11b, F4/80, and 1A8-APC and were analyzed by flow cytometry. (A) To determine percentages of neutrophils in spleen and blood, cells were gated as CD11b+ F4/802 singlet leukocytes and were analyzed for the presence of SSCint 1A8-APC+ neutrophils. The circular gates in the far-right graphs contain 1A8-FITC/1A8-APCcolabeled neutrophils (far-right middle) and 1A8PE/1A8-APC-colabeled neutrophils (far-right bottom) of all gated CD11b+ F4/802 leukocytes. Note that the 1A8-FITC+ 1A8-APC+ population is less detectable than the 1A8-PE+ 1A8-APC+ population. (B and C) Percentages of SSCint 1A8-APC+ neutrophils of all gated CD11b+ F4/802 leukocytes in spleen (B) and blood (C). Bars represent means 6 SD. Data were pooled from 2 independent experiments. Data were analyzed by ANOVA, followed by Bonferroni’s multiple comparison test; n = 5; *P , 0.05; **P , 0.01.
6
Journal of Leukocyte Biology
Volume 98, September 2015
www.jleukbio.org
Bucher et al. Fluorochromes alter functional properties of anti-Ly6G
APC, FITC, and PE differentially modify access to binding sites on the 1A8 antibody It has been demonstrated that neutrophil depletion by anti-Ly6G involves cross-linking of Ly6G and requires the Fc fragment of the antibody [11]. Importantly, the Fc fragment of IgG antibodies harbors specific binding sites for FcgR and complement protein C1q. FcgRs are present on many cell types, including macrophages, and IgG binding to FcgR stimulates uptake of IgGcoated cells by macrophages [34]. C1q is produced by cells of the monocyte-macrophage lineage [35], and IgG coating can induce cell depletion via the activation of the complement system [36]. Interestingly, the used, 1A8-coupling fluorochromes APC, FITC, and PE show great differences in their chemical structure and molecular mass (APC, 150 kDa; FITC, 0.39 kDa; and PE, 240 kDa). Moreover, as antibody glycosylation is important for complement activation [37] and FcR binding [38, 39], the method of antibody production or the chemistry of fluorochrome conjugation may affect the glycosylation sites on the antibody, thereby enhancing or preventing binding to FcR or complement. Therefore, we hypothesized that fluorochrome coupling may alter the functional properties of 1A8 (;150 kDa) by interfering with functionally relevant interaction sites of the antibody. The degree of this effect might be determined by the molecular mass and/or the biochemical characteristics of the fluorochromes, their binding position on the antibody, as well as their potential influence on antibody glycosylation. Protein G is known to bind to Fab and Fc fragments of IgG [40]. Hence, to test whether APC, FITC, and PE can, in general, modify access to interaction sites on the anti-Ly6G antibody, we determined protein G-binding to the 3 fluorescent 1A8 antibodies. To this end, BM cells were separately stained with 1A8APC, -FITC, or -PE; pooled; and incubated with biotinylated protein G, followed by SA-PerCp-Cy5.5 staining. We found that protein G binding was significantly lower in 1A8-APC+ or -PE+ cells compared with 1A8-FITC+ cells (Fig. 6), indicating that APC and PE impaired protein G binding to 1A8 antibodies. These results demonstrate that coupling fluorochromes can, in principle, alter the accessibility to binding sites on 1A8. If these differences in the accessibility also extend to depletion-relevant binding sites, then one could assume that APC and PE modify the functional properties of 1A8 by impeding interaction sites required for the depleting function. In contrast, in FITC-coupled antibodies, these interaction sites would be accessible, enabling neutrophil depletion by 1A8. Alternately, fluorochrome conjugation may determine the depleting capacity of anti-Ly6G by other mechanisms, for instance, by affecting glycosylation sites on the antibody. In summary, we demonstrate that fluorochromes coupled to anti-Ly6G can differentially influence the functional properties of the Ly6G antibody. Fluorochromes control ligand binding to neutrophil-bound anti-Ly6G, determine macrophage uptake of anti-Ly6G-labeled neutrophils, and alter their retrieval after adoptive cell transfer or intravenous injection of anti-Ly6G antibodies into mice. Moreover, our results identify neutrophil phagocytosis by macrophages as a so-far-unrecognized mechanism involved in anti-Ly6G-induced neutrophil depletion. Taken together, this knowledge is essential for a valid interpretation and
www.jleukbio.org
Figure 6. APC and PE but not FITC coupling inhibits protein G binding to the 1A8 antibody. To determine protein G binding to fluorochromecoupled 1A8, equal fractions of 1A8-APC-, -FITC-, or -PE-labeled BM cells were pooled and incubated with or without biotinylated protein G. Thereafter, cells were washed and stained with SA-PerCP-Cy5.5 and with antibodies against CD3e, CD11b, and CD19. Then, cells were analyzed by flow cytometry. 1A8-APC-, -FITC-, or -PE-labeled neutrophils were gated as shown in Fig. 1, and protein G-binding to 1A8-labeled cell populations was determined by the geometric mean fluorescence intensity (GMFI) of SA-PerCP-Cy5.5. Bars represent means 6 SD. Data were pooled from 2 independent experiments and were analyzed by ANOVA, followed by Bonferroni’s multiple comparison test; n = 6; ***P , 0.001.
improved experimental design of in vivo experiments that use anti-Ly6G antibodies.
AUTHORSHIP K.B. designed and carried out the experiments, analyzed the data, and wrote the manuscript. F.S. and I.D. performed experiments. S.E.A. performed experiments and analyzed data. B.N. and K.S-L. provided conceptual advice and edited the manuscript. S.B.-H. designed experiments, provided experimental and conceptual advice, and wrote the manuscript.
ACKNOWLEDGMENTS The study was supported by FOR729 from the Deutsche Forschungsgemeinschaft and Interfaculty Centre for Pharmacogenomics and Pharma Research, Mouse Clinic, Eberhard-KarlsUniversity (Tubingen, ¨ Germany). The authors acknowledge Prof. Dr. Christian Harteneck and Dr. Veronika Leiss for helpful discussion, Dr. Claudia de Oliveira Franz for performing intravenous injections, Silvia Vetter for carrying out liver cell isolation, and Benedikt Mothes for experimental help. DISCLOSURES
The authors declare no conflict of interest.
Volume 98, September 2015
Journal of Leukocyte Biology 7
REFERENCES
1. Lee, P. Y., Wang, J. X., Parisini, E., Dascher, C. C., Nigrovic, P. A. (2013) Ly6 family proteins in neutrophil biology. J. Leukoc. Biol. 94, 585–594. 2. Wang, J. X., Bair, A. M., King, S. L., Shnayder, R., Huang, Y. F., Shieh, C. C., Soberman, R. J., Fuhlbrigge, R. C., Nigrovic, P. A. (2012) Ly6G ligation blocks recruitment of neutrophils via a b2-integrin-dependent mechanism. Blood 120, 1489–1498. 3. Fleming, T. J., Fleming, M. L., Malek, T. R. (1993) Selective expression of Ly-6G on myeloid lineage cells in mouse bone marrow. RB6-8C5 mAb to granulocyte-differentiation antigen (Gr-1) detects members of the Ly-6 family. J. Immunol. 151, 2399–2408. 4. Jutila, M. A., Kroese, F. G., Jutila, K. L., Stall, A. M., Fiering, S., Herzenberg, L. A., Berg, E. L., Butcher, E. C. (1988) Ly-6C is a monocyte/macrophage and endothelial cell differentiation antigen regulated by interferon-gamma. Eur. J. Immunol. 18, 1819–1826. 5. Kung, J. T., Castillo, M., Heard, P., Kerbacher, K., Thomas III, C. A. (1991) Subpopulations of CD8+ cytotoxic T cell precursors collaborate in the absence of conventional CD4+ helper T cells. J. Immunol. 146, 1783–1790. 6. Jutila, D. B., Kurk, S., Jutila, M. A. (1994) Differences in the expression of Ly-6C on neutrophils and monocytes following PI-PLC hydrolysis and cellular activation. Immunol. Lett. 41, 49–57. 7. Shortman, K., Naik, S. H. (2007) Steady-state and inflammatory dendritic-cell development. Nat. Rev. Immunol. 7, 19–30. 8. Zhou, J., Stohlman, S. A., Hinton, D. R., Marten, N. W. (2003) Neutrophils promote mononuclear cell infiltration during viral-induced encephalitis. J. Immunol. 170, 3331–3336. 9. Bliss, S. K., Gavrilescu, L. C., Alcaraz, A., Denkers, E. Y. (2001) Neutrophil depletion during Toxoplasma gondii infection leads to impaired immunity and lethal systemic pathology. Infect. Immun. 69, 4898–4905. 10. Tateda, K., Moore, T. A., Deng, J. C., Newstead, M. W., Zeng, X., Matsukawa, A., Swanson, M. S., Yamaguchi, K., Standiford, T. J. (2001) Early recruitment of neutrophils determines subsequent T1/T2 host responses in a murine model of Legionella pneumophila pneumonia. J. Immunol. 166, 3355–3361. 11. Abbitt, K. B., Cotter, M. J., Ridger, V. C., Crossman, D. C., Hellewell, P. G., Norman, K. E. (2009) Antibody ligation of murine Ly-6G induces neutropenia, blood flow cessation, and death via complement-dependent and independent mechanisms. J. Leukoc. Biol. 85, 55–63. 12. Morton, J., Coles, B., Wright, K., Gallimore, A., Morrow, J. D., Terry, E. S., Anning, P. B., Morgan, B. P., Dioszeghy, V., K¨uhn, H., Chaitidis, P., Hobbs, A. J., Jones, S. A., O’Donnell, V. B. (2008) Circulating neutrophils maintain physiological blood pressure by suppressing bacteria and IFNgamma-dependent iNOS expression in the vasculature of healthy mice. Blood 111, 5187–5194. 13. Ribechini, E., Leenen, P. J., Lutz, M. B. (2009) Gr-1 antibody induces STAT signaling, macrophage marker expression and abrogation of myeloid-derived suppressor cell activity in BM cells. Eur. J. Immunol. 39, 3538–3551. 14. Martin, C., Burdon, P. C., Bridger, G., Gutierrez-Ramos, J. C., Williams, T. J., Rankin, S. M. (2003) Chemokines acting via CXCR2 and CXCR4 control the release of neutrophils from the bone marrow and their return following senescence. Immunity 19, 583–593. 15. Phillipson, M., Heit, B., Colarusso, P., Liu, L., Ballantyne, C. M., Kubes, P. (2006) Intraluminal crawling of neutrophils to emigration sites: a molecularly distinct process from adhesion in the recruitment cascade. J. Exp. Med. 203, 2569–2575. 16. Hut´as, G., Bajnok, E., G´al, I., Finnegan, A., Glant, T. T., Mikecz, K. (2008) CD44-specific antibody treatment and CD44 deficiency exert distinct effects on leukocyte recruitment in experimental arthritis. Blood 112, 4999–5006. 17. McDonald, B., Pittman, K., Menezes, G. B., Hirota, S. A., Slaba, I., Waterhouse, C. C., Beck, P. L., Muruve, D. A., Kubes, P. (2010) Intravascular danger signals guide neutrophils to sites of sterile inflammation. Science 330, 362–366. 18. Devi, S., Li, A., Westhorpe, C. L., Lo, C. Y., Abeynaike, L. D., Snelgrove, S. L., Hall, P., Ooi, J. D., Sobey, C. G., Kitching, A. R., Hickey, M. J. (2013) Multiphoton imaging reveals a new leukocyte recruitment paradigm in the glomerulus. Nat. Med. 19, 107–112. 19. Yipp, B. G., Kubes, P. (2013) Antibodies against neutrophil LY6G do not inhibit leukocyte recruitment in mice in vivo. Blood 121, 241–242. 20. Beer-Hammer, S., Zebedin, E., von Holleben, M., Alferink, J., Reis, B., Dresing, P., Degrandi, D., Scheu, S., Hirsch, E., Sexl, V., Pfeffer, K., N¨urnberg, B., Piekorz, R. P. (2010) The catalytic PI3K isoforms p110gamma and p110delta contribute to B cell development and
8
Journal of Leukocyte Biology
Volume 98, September 2015
21.
22. 23. 24. 25.
26. 27.
28.
29. 30.
31.
32. 33. 34. 35. 36. 37.
38. 39. 40.
maintenance, transformation, and proliferation. J. Leukoc. Biol. 87, 1083–1095. Schmitz, H. J., Hagenmaier, A., Hagenmaier, H. P., Bock, K. W., Schrenk, D. (1995) Potency of mixtures of polychlorinated biphenyls as inducers of dioxin receptor-regulated CYP1A activity in rat hepatocytes and H4IIE cells. Toxicology 99, 47–54. McCormack, T., O’Keeffe, G., Mac Craith, B., O’Kennedy, R. (1996) Assessment of the effect of increased fluorophore labelling on the binding ability of an antibody. Anal. Lett. 29, 953–968. Vira, S., Mekhedov, E., Humphrey, G., Blank, P. S. (2010) Fluorescentlabeled antibodies: balancing functionality and degree of labeling. Anal. Biochem. 402, 146–150. Furze, R. C., Rankin, S. M. (2008) The role of the bone marrow in neutrophil clearance under homeostatic conditions in the mouse. FASEB J. 22, 3111–3119. Daubeuf, S., Puaux, A. L., Joly, E., Hudrisier, D. (2006) A simple trogocytosis-based method to detect, quantify, characterize and purify antigen-specific live lymphocytes by flow cytometry, via their capture of membrane fragments from antigen-presenting cells. Nat. Protoc. 1, 2536–2542. Gillette, J. M., Larochelle, A., Dunbar, C. E., Lippincott-Schwartz, J. (2009) Intercellular transfer to signalling endosomes regulates an ex vivo bone marrow niche. Nat. Cell Biol. 11, 303–311. Savill, J. S., Wyllie, A. H., Henson, J. E., Walport, M. J., Henson, P. M., Haslett, C. (1989) Macrophage phagocytosis of aging neutrophils in inflammation. Programmed cell death in the neutrophil leads to its recognition by macrophages. J. Clin. Invest. 83, 865–875. Suratt, B. T., Young, S. K., Lieber, J., Nick, J. A., Henson, P. M., Worthen, G. S. (2001) Neutrophil maturation and activation determine anatomic site of clearance from circulation. Am. J. Physiol. Lung Cell. Mol. Physiol. 281, L913–L921. Brinkley, M. (1992) A brief survey of methods for preparing protein conjugates with dyes, haptens, and cross-linking reagents. Bioconjug. Chem. 3, 2–13. Samuelsson, K., Simonsson, C., Jonsson, C. A., Westman, G., Ericson, M. B., Karlberg, A. T. (2009) Accumulation of FITC near stratum corneum-visualizing epidermal distribution of a strong sensitizer using two-photon microscopy. Contact Dermat. 61, 91–100. Macatonia, S. E., Knight, S. C., Edwards, A. J., Griffiths, S., Fryer, P. (1987) Localization of antigen on lymph node dendritic cells after exposure to the contact sensitizer fluorescein isothiocyanate. Functional and morphological studies. J. Exp. Med. 166, 1654–1667. Pior, J., Vogl, T., Sorg, C., MacHer, E. (1999) Free hapten molecules are dispersed by way of the bloodstream during contact sensitization to fluorescein isothiocyanate. J. Invest. Dermatol. 113, 888–893. Daley, J. M., Thomay, A. A., Connolly, M. D., Reichner, J. S., Albina, J. E. (2008) Use of Ly6G-specific monoclonal antibody to deplete neutrophils in mice. J. Leukoc. Biol. 83, 64–70. Chang, N. S., Leu, R. W., Rummage, J. A., Anderson, J. K., Mole, J. E. (1992) Regulation of macrophage Fc receptor expression and phagocytosis by histidine-rich glycoprotein. Immunology 77, 532–538. Petry, F., Botto, M., Holtappels, R., Walport, M. J., Loos, M. (2001) Reconstitution of the complement function in C1q-deficient (C1qa-/-) mice with wild-type bone marrow cells. J. Immunol. 167, 4033–4037. Duncan, A. R., Winter, G. (1988) The binding site for C1q on IgG. Nature 332, 738–740. Tao, M. H., Morrison, S. L. (1989) Studies of aglycosylated chimeric mouse-human IgG. Role of carbohydrate in the structure and effector functions mediated by the human IgG constant region. J. Immunol. 143, 2595–2601. Nose, M., Wigzell, H. (1983) Biological significance of carbohydrate chains on monoclonal antibodies. Proc. Natl. Acad. Sci. USA 80, 6632–6636. Wright, A., Morrison, S. L. (1997) Effect of glycosylation on antibody function: implications for genetic engineering. Trends Biotechnol. 15, 26–32. Aybay, C. (2003) Differential binding characteristics of protein G and protein A for Fc fragments of papain-digested mouse IgG. Immunol. Lett. 85, 231–235.
KEY WORDS: cell tracking APC FITC PE granulocytes •
•
•
•
www.jleukbio.org
