Role of the atypical chemoattractant receptor CRAM in regulating CCL19 induced CCR7 responses in B-cell chronic lymphocytic leukemia

In a previous publication we had shown CRAM to be highly expressed in pro- and pre-B primary cells as well as in corresponding cell lines [15]. The characterization of CRAM and CCR7 expression at the surface of cell lines and primary B cells from healthy donor controls was completed in the present study. MEC-1 (a B-CLL derived cell line) showed high expressions of both receptors whilst Nalm6 (pre-B Acute Lymphoblastic Leukemia derived cell line) and Reh (Acute Lymphoblastic Leukemia) cells had a similar level of CRAM but no CCR7 (Figure 1A). The expression of both CRAM and CCR7 at the surface of B cells from different healthy controls was shown to be consistent (Figure 1B and [18]). In parallel, the expression of CCR7 and CRAM was investigated at the surface of B cells from 23 CLL patients. Whilst CCR7 had a high but limited range of expression level (Figure 2A, data not shown and [3]), CRAM expression level was considerably more variable with regard to geometric mean fluorescence values, ranging from 2 to 949 (Figure 2A and data not shown).

The apparent inconsistency of CRAM expression in B-CLL led us to investigate the potential consequences for CCL19 induced migration. Two B-CLL primary cells samples paired for their similar levels of CCR7 yet variable levels of CRAM were selected (Figure 2A) and tested for their capacity to respond to CCL19 induced chemotaxis. A higher expression of CRAM was associated with a lower efficacy of chemotaxis (Figure 2A). To further investigate this observation, migrated cells from the lower wells of chemotaxis chambers, after migration towards CCL19 or CCL21, were stained for CRAM and CCR7 (Figure 2B). The CCR7 level of expression was not altered on the migrated cells. However, CCL19-induced migration was associated with a lowered expression of CRAM, not observed for CCL21. This observation is the result of cell selection rather than of receptor down-regulation as CCL19 does not induce CRAM internalization (CRAM constitutively recycles in a ligand independent manner [16]). Importantly, neither CCL19 nor CCL21 interfered with CRAM detection by the antibody used in the previous experiment (Figure 2B, lower right histogram). High levels of CRAM reduce the ability of cells to migrate toward CCL19, whereas it is not important for migration to CCL21 under the same conditions.

MEC-1 cells express both CRAM and the classical receptor for CCL19: CCR7 (also the classical receptor for CCL21) and were used to investigate MAPK phosphorylation after stimulation by CCL19 or CCL21 (Figure 3A). To evaluate the implications of CRAM expression on CCR7 activity, we blocked CRAM functions utilising different methods and analyzed the effects on CCL19 induced CCR7 activation. As phosphorylation of p44/42 has been shown to be involved in CCR7 induced B cell activation [19‐21], we examined phosphorylation in the absence or presence of CRAM blocking antibodies. Addition of CRAM specific antibody potentiated the CCL19 triggered response. Furthermore, CRAM-antibody induces a longer signalling profile suggesting that CRAM could be involved in limiting CCR7 activation (Figure 3A). To assess the specificity of this observation, a similar experiment was conducted using CCL21 instead of CCL19. The antibody-induced potentiation of MAPK phosphorylation was not observed with this chemokine (Figure 3A). Similar results were obtained in chemotaxis assays (Figure 3B). The addition of the isotype control antibody did not alter the chemotactic profile of CCL19 or CCL21. However addition of antibodies against CRAM enhanced chemotaxis toward CCL19 but not toward CCL21.

To support these observations, experiments were conducted without blocking antibodies and in a different pathway of activation. CRAM expression was down-regulated by siRNA transfection in MEC-1 cells, and chemotaxis assays were performed using CCL19 or CCL21. siRNA achieved a CRAM down-regulation of approximately 30% (Figure 1A), low albeit sufficient to effect chemotaxis (Figure 4A). CRAM down-regulation resulted in enhanced CCL19 mediated chemotaxis but did not affect CCL21 induced chemotaxis of MEC-1 cells. This suggests that CRAM regulates CCL19 and thereby CCR7 activation, probably by competition either of ligand binding or by modifying/influencing activation. Finally, we obtained concurring results in an additional MEC-1 (CRAM⁺/CCR7⁺) chemotaxis assay. In this experiment, CRAM was blocked with its alternative ligands: chemerin and CCL5 (Figure 4B). Addition of CCL5 to the upper part of the chemotaxis well significantly increased the migration toward CCL19, yet this did not modify migration toward CCL21. This observation suggests that addition of CCL5 was not responsible per se for this increased migration but that there was a cooperative effect with CCL19. CCL5 addition was tested at two concentrations (CCL5-High: 50 ng/ml and CCL5-Low: 5 ng/ml) which showed significant decreases of chemotaxis (p = 0.005 and 0.048 respectively). Chemerin addition also resulted in significantly enhanced migration toward CCL19, but only the highest dose tested (Chem-High: 500 ng/ml) was statistically significant (p = 0.032). As low levels of RNA from classical receptors for chemerin and CCL25 (respectively ChemR23 and CCR9) and proteins from CCR5 and CCX-CKR were detected in MEC-1 cells (Additional file 1, Figure S1), we proceeded to conduct additional chemotaxis controls. CCL5 (for CCR5 expression), chemerin (for ChemR23) and CCL25 (for CCR9 and CCX-CKR) influences on the MEC-1 migration capacity were defined by assessing chemotaxis (Figure 5A) or chemokinesis (Figure 5B). None of these assays showed any conclusive involvement of CCL5, CCL25 or chemerin in MEC-1 migratory behaviour. No significant alteration of CCL19-induced migration was observed with CCL25, indicating that there was no involvement of CCX-CKR in our observations (Figure 4B). In addition, we controlled CRAM versus CCX-CKR involvement by reproducing our chemotaxis regulatory experiment with CCL21. CCL21 induced migration was not enhanced by the addition of other CRAM ligands (Figure 4B). Overall these experiments provide some insight into the nature of CRAM involvement in CCR7 regulation, implying that CRAM expression influences CCL19 activity and that this system is affected by other factors, including other cytokines (CCL5, chemerin).

Finally, we examined CRAM modulation of CCL19 availability over time and performed a calcium-release assay. We stimulated MEC-1 (CRAM⁺/CCR7⁺) with Nalm6 (CRAM⁺/CCR7^-) that had been pre-incubated with chemokines; Nalm6 cells do not express CCX-CKR, the previously described scavenger receptor for CCL19 (Additional file 1, Figure S1) hence CCL19 regulation observed cannot be attributed to it. Incubation of the Nalm6 cells (CRAM⁺/CCR7^-) was done at 37°C to allow the internalisation of chemokines by CRAM and they were subsequently washed thoroughly with PBS buffer, before a further incubation (30 min) at 37°C or 4°C to allow or prevent respectively re-expression of potentially internalized chemokines. Chemokine incubated Nalm6 (CRAM⁺/CCR7^-) were then added to MEC-1 cells (CRAM⁺/CCR7⁺) to investigate whether calcium release could be induced from chemokines presented at the cell surface of the Nalm6 cells (Figure 6A). Nalm6 cells that were incubated at 4°C post wash induced a smaller calcium release in MEC-1 cells, upon coincubation at 37°C, relative to control, unwashwed Nalm6 cells. This indicates that all the chemokines used in the first incubation step were not available for stimulation of MEC-1 cells (CRAM⁺/CCR7⁺) and therefore were not presented at the cell surface. MEC-1 cells co-incubated with Nalm6 cells, previously incubated at 37°C, had induced calcium release; almost as intense as before washing and significantly higher than stimulation by 4°C incubated Nalm6 cells. This implies that a proportion of CCL19, that was internalized in Nalm6 cells during the first incubation period, was available for MEC-1 stimulation during co-incubation in this experimental setting (Figure 6D). In comparison with CCL21, no difference was seen when Nalm6 cells were incubated at 37°C versus 4°C, supporting the involvement of a CCL19 specific receptor. Additionally, we monitored the CCL19-Nalm6 induced calcium release in MEC-1 cells after the two temperature conditions of initial exposure of the Nalm6 cells to the chemokine (4°C or 37°C) followed by a washing step and variable time and temperature conditions for the second incubation step (Figure 6B). This setting allows the monitoring of the internalization of CCL19 by Nalm6 (CRAM⁺/CCR7^-) over time. When cells are initially incubated at 4°C, only surface bound CCL19 is present at the beginning of the experiment. Further incubation at 37°C did not change this observation, implying that the amount of chemokine associated with Nalm6 (CRAM⁺/CCR7^-) and available for stimulation of MEC-1 (CRAM⁺/CCR7⁺) is sustained (blue trace). Nalm6 cells were incubated at 37°C prior to the washing step, to allow CCL19 to be internalized. This created two different pools of chemokine: one present at the cell surface and one intracellular. The latter would only be available to stimulate MEC-1 (CRAM⁺/CCR7⁺) when re-expressed at the cell surface and if CRAM is efficiently recycling rather than degrading CCL19; a 37°C dependent process. A correlation between incubation time and calcium release intensity was observed with enhanced responses when the second incubation period is done at 37°C. This resulted in CCL19 cell surface re-expression (purple and red traces). Quantification of CCL19 by ELISA was used to ascertain the location of the chemokines in suspension with Nalm6 cells after the longest incubation time (90 min). Three fractions were separated: supernatant, cell surface associated or internalized (Figure 6C). CCL19 associated to cells incubated first at 4°C and then at 37°C was found at lower concentrations than in the two other incubation conditions, with a very low proportion of CCL19 internalized and/or associated with the cells surfaces. CCL19 associated to the Nalm6 cells in the two other incubation conditions (preliminary incubation at 37°C followed by a washing step and further incubation at 4°C or 37°C) and presented higher concentrations of CCL19, with a similar quantity of CCL19 either associated to the cell surface or released in the supernatant (sum of blue and pink bars). However, when both primary and secondary incubations are at 37°C, the proportion of CCL19 in the cells or associated to the cell surface is increased, probably supporting signalling over time, an observation already seen in Figure 3. These data provide support to our hypothesis of CRAM involvement in active regulation of circulating CCL19.

All cell line media were supplemented with 10% Fœtal Calf Serum (FCS) and 1% penicillin/streptomycin. Nalm6 and Reh cells were maintained in RPMI and MEC-1 cells in IMDM in a humidified atmosphere (5% CO₂) at 37°C (all from Invitrogen, Carlsbad, CA).

The murine monoclonal antibody against CRAM (anti-HCR/CRAM-A/B, Clone 152254) was produced by R&D Systems (Minneapolis, MN). The murine PE-labelled monoclonal antibodies against CCR7 (MAB197), CCR1 (FAB145P), CCR3 (FAB155P) and CCR5 (FAB182P) were purchased from R&D Systems. CCX-CKR (sc-46836) unlabelled antibody was purchased from Santa Cruz (Heidelberg, Germany). RPE-labelled rabbit anti-mouse IgG was from DakoCytomation (Glostrup, Denmark), Alexa-647 rabbit anti-goat (A21446) was from Invitrogen (Darmstadt, Germany) and chemokines were from R&D systems (Minneapolis, MN). PBMCs from Buffy coats of healthy donors were isolated by Ficoll density-gradient centrifugation. For isolation of chronic lymphocytic leukemia (CLL) cells, blood samples were collected from patients after informed consent. Peripheral blood mononuclear cells were isolated by density gradient centrifugation over Ficoll-Hypaque (Pharmacia, Uppsala, Sweden), and contained more than 90% CLL B cells [15].

5 × 10⁶ cells were incubated with the appropriate antibodies 30 min at 4°C in PBS, 0.5% BSA, and then washed in the same buffer prior to incubation with the secondary antibodies in the same conditions. After final washes, cell fluorescence was assessed using a FACScalibur counting 2 × 10⁴ cells. Flow cytometry data were analysed with Flowjo.

Cells were starved in FCS free medium for 2 h and stimulated with chemokines for the indicated times at 37°C. Protein lysates were prepared with 100 μl lysis buffer (20 mM Tris/HCl pH 8.0, 150 mM KCl, 1 mM EDTA, 0.2 mM Na₃VO₄, 1% Triton X-100, 0.5 mM PMSF) with protein inhibitor cocktail (complete^®, Roche Applied Science, Basel, Switzerland). Equal amounts of protein were separated by 10% SDS polyacrylamide gel electrophoresis and transferred onto PVDF membranes. Western blot analysis was performed using the appropriate antibodies recognizing the phosphorylated form of p44/p42 or pan-protein antibodies (#9101 and #9102, respectively) and were from Cell Signalling (Beverly, MA). Immunoreactive bands were visualized using horseradish peroxidase-conjugated secondary antibody and the enhanced chemiluminescence system (GE Amersham, Fairfield, CT). Densitometric quantification was done using Adobe Photoshop (Adobe Photoshop 9.0, Adobe, San Jose, CA, 2005). Data are presented as mean arbitrary density units normalized to the ratio between phosphorylated proteins to non-phosphorylated forms at time 0 in presence of the control isotype + SD. This statistical analysis as well as all the following ones were performed using Graphpad prism version 5.02.

Chemotaxis experiments were done using NeuroProbe 96 well plate devices with 5 μm pore filter (Receptor Technologies, Warwickshire, UK). 50 μl of a 2 × 10⁶ cell/ml suspension in HBSS, 0.1% BSA were loaded on the upper part of the filter after filling up the lower part with buffer (HBSS, 0.1% BSA) or chemokines in suspension in the same buffer. When stated, cellular suspension was completed with the indicated concentration of chemokines or chemerin. Migration was done over 1.5 hours and cell number evaluated by resuspending the migrated cells from the lower part of the well in 200 μl buffer and counting them by flow cytometry for a fixed period of time. The numbers of replicates and details of statistical analyses are described in the figure legends.

10⁷/ml MEC-1 cells were incubated for 45 min at 37°C in PBS buffer containing Fluo-3/AM probe at a final concentration of 2 μM (Invitrogen Molecular Probes), then washed and resuspended in calcium buffer (RPMI, 0.11 μM CaCl₂, 0.1% BSA) to a final concentration of 2 × 10⁶cell/ml and incubated at 37°C for 5 min before proceeding to the calcium release assay at 37°C. The calcium release assay was performed by inducing MEC-1 with the addition of Nalm6 cells previously incubated with CCL19 or CCL21 (100 ng/ml) at 37°C (Figure 6A) or as indicated (Figure 6B). Nalm6 cells were then washed 3 times at 4°C in 5 ml PBS, 0.1%BSA, and incubated for a further 90 min in calcium buffer at 37°C (Figure 6A) or as indicated (Figure 6B) before being used to induce calcium release in MEC-1 cells. Results were monitored using a FACScalibur, FSC/SSC scattering allowing the differentiation between MEC-1 and Nalm6 cells according to their physical characteristics (Additional file 2, Figure S2).

Nalm6 cells were incubated in 500 μl of binding buffer (RPMI, 20 mM HEPES, 0.1%BSA) with or without 100 ng/ml CCL19 for 30 minutes at 37°C or 4°C as indicated. After this preliminary incubation step they were washed 3 times with 5 ml of PBS/0.1%BSA, and further incubated at 37°C or 4°C, as indicated, for 90 minutes in 500 μl of binding buffer. Three different sample collections were done: supernatant (1), cell associated fraction after a brief incubation of the cells with 50 μl of 10xPBS, to disrupt interactions between chemokines and their receptors (as described in [14]), and resuspension in a final volume of 500 μl PBS 1× for collection (2), finally the cells were lysed by 200 μl lysis buffer (20 mM Tris-HCl (pH 8.0), 150 mM KCl, 1 mM EDTA, 1 mM Na₃VO₄, 1 mM PMSF, 1% Triton X-100 and protease inhibitor mix) and resupended in a final volume of 500 μl with PBS (3). ELISAs were performed according to the manufacturer's instructions. Each fraction was normalised to its equivalent fractions collected in absence of CCL19.

As described in [15], total RNA was isolated using the RNeasy RNA isolation kit (Qiagen, Dusseldorf, Germany) according to the manufacturer's instructions. Residual DNA was removed by DNase I Digestion (Ambion, Austin, TX). The cDNA synthesis was performed using 1 μg RNA as a template for oligo-dT (12-18 mer) primers and 50 Units SuperScript II reverse transcriptase (Super Script first-strand synthesis system for reverse transcription (RT-) PCR; Invitrogen). The cDNA was amplified using Taq polymerase (Qiagen). The following primer pairs were used for PCR: CRAM-B: 5'-ATGGCCAATTACACGCTGGCACCAGAG-3' and the corresponding antisense primer 5'-CACTTCGGTGGAATGGTCAGGTTCTTCCCTC-3', GAPDH: the sense primer 5'-GGAGTCCACTGGCGTCTTCACC-3' and the antisense primer 5'-ATTGCTGATGATCTTGAGGCTGTTGTC-3' CCR9: 5'-GCCCAGGCCATGAGA-3' and the corresponding antisense primer 5'-AACCCACTGGGCCTGGCTGA-3' and ChemR23: the sense primer 5'-CCGGGACACAGCCAACCTGC-3' and the antisense primer 5'-CAGTGGCCAGGGGCAAACCC-3'. Annealing was done at 60°C.