Mast cells and Th17 cells contribute to the lymphoma-associated pro-inflammatory microenvironment of angioimmunoblastic T-cell lymphoma

Abstract

Reports focusing on the immunological microenvironment of peripheral T-cell lymphomas (PTCL) are rare. Here we studied the reciprocal contribution of regulatory (Treg) and interleukin-17-producing (Th17) T-cells to the composition of the lymphoma-associated microenvironment of angioimmunoblastic T-cell lymphoma (AITL) and PTCL not otherwise specified on tissue microarrays from 30 PTCLs not otherwise specified and 37 AITLs. We found that Th17 but not Treg cells were differently represented in the two lymphomas and correlated with the amount of mast cells (MCs) and granulocytes, which preferentially occurred in the cellular milieu of AITL cases. We observed that MCs directly synthesized interleukin-6 and thus contribute to the establishment of a pro-inflammatory, Th17 permissive environment in AITL. We further hypothesized that the AITL clone itself could be responsible for the preferential accumulation of MCs at sites of infiltration through the synthesis of CXCL-13 and its interaction with the CXCR3 and CXCR5 receptors expressed on MCs. Consistent with this hypothesis, we observed MCs efficiently migrating in response to CXCL-13. On these bases, we conclude that MCs have a role in molding the immunological microenvironment of AITL toward the maintenance of pro-inflammatory conditions prone to Th17 generation and autoimmunity.

Figure 1

Frequency of Foxp3⁺ and IL-17⁺ cells in 30 PTCL/NOS and 37 AITL cases. The amount of Foxp3⁺ cells is comparable in the two lymphoma subgroups, whereas that of IL-17⁺ cells is significantly higher in AITL.

Figure 2

A and C: Immunohistochemical expression of Foxp3 in two representative cases of PTCL/NOS (A) and AITL (C) showing the scattered distribution of immunoreactive cells. B and D: By immunohistochemistry, IL-17-expressing cells display a scattered distribution and different frequency in PTCL/NOS (B) and AITL (D) cases (Strept-ABC method with DAB chromogenic substrate; original magnification, ×100).

Figure 3

Immunohistochemical expression of IL-6 (A and C), IL-21 (B and D), and IL-23 (inset) in representative AITL and PTCL/NOS cases. A and C: IL-6 is expressed by vascular endothelia (green arrows), small lymphoid cells (black arrows), and large cells showing intense staining (red arrows, inset). B and D: IL-21 is diffusely expressed in AITL cells, whereas its expression is confined to scattered lymphoid cells in PTCL/NOS (black arrows). Inset: IL-23 expression is mainly confined to the dendritic cell/macrophage meshwork (Strept-ABC method with DAB chromogen; original magnification, ×200; inset, ×400).

Figure 4

Double immunofluorescence confocal microscopy for IL-6 and mast cell tryptase showing that cells with bright IL-6 expression (A and D) are reactive to tryptase (B and E). Co-expression of the two markers is evidenced by signal overlap (C and F). Release of IL-6 and tryptase from a degranulating mast cells is shown in the inset (double indirect immunofluorescence using Alexa 488-conjugated [green signal] and Alexa 568-conjugated [red signal] secondary antibodies; original magnifications: ×200 [A–F], ×1000 [inset]). Microphotographs are relative to two representative AITL (A–C) and PTCL/NOS (D–F) cases.

Figure 5

Amount and distribution of infiltrating mast cells in AITL and PTCL/NOS cases. A significantly higher number of tryptase-expressing mast cells are observed infiltrating AITL cases (C and D) as compared with those infiltrating PTCL/NOS ones (A and B). Mast cells display a scattered distribution and a round-to spindle morphology (Strept-ABC method with aminoethylcarbazole chromogen; original magnifications: ×100 [A and C]; ×200 [B and D]).

Figure 6

Analysis of the associations among regulatory T-cells (Foxp3), Th17 (IL-17), mast cells (tryptase), granulocytes (CD15), and microvascular density (CD34). A strong linear correlation is found between Th17 and mast cells and between Th17 and granulocytes, whereas the association between mast cells and microvascular density is slight. Regulatory T-cells do not show significant associations with any of the other variables.

Figure 7

A: Tryptase⁺ mast cells display cell–cell contact with Foxp3-expressing regulatory T-cells (black arrows). B: Lymphocytes surrounding mast cells express OX40, suggesting that interactions between mast cells and neighboring T cells can be complemented by OX40 engagement. C: Few mast cells are found in close contact with IL-17-expressing cells (black arrow). D: A higher density of both IL-6-expressing and IL-17-expressing cells are observed in AITL (upper panel) than in PTCL/NOS (lower panel) cases, but spatial contact between these two cell types is rarely observed. Pictures are relative to two representative AITL and one PTCL/NOS cases. A–C: Double immunohistochemistry using alkaline phosphatase anti-alkaline phosphatase and Strept-ABC methods with Fast-red (purple) and DAB (orange/brown) chromogens, respectively. D: Double indirect immunofluorescence using Alexa 488-conjugated (green signal) and Alexa 568-conjugated (purple signal) secondary antibodies. Original magnifications: ×400 (A); ×200 (B–D).

Figure 8

Results of the MC/Th17 co-localization analysis performed on full sections from nine AITL (A) and nine PTCL/NOS (B) cases following double immunostaining for mast cell tryptase and IL-17. In each case, the number of MCs (light green and light blue bars) and Th17 cells (dark green and dark blue bars) is reported as assessed in four HPFs with easily countable MCs (indicated as “High MC”) and in four HPFs almost devoid of MCs (indicated as “Low MC”). In AITL cases (A) the spatial association between MCs and Th17 was evident as HPFs rich in MCs displayed higher Th17 counts than those almost devoid of MCs. Conversely, in PTCL/NOS cases (B), this picture was less evident owing to the paucity of MCs and Th17 cells.

Figure 9

CXCR3 and CXCR5 expression in the human LAD2 mast cell line. CXCR3 and CXCR5 mRNA levels in ADMEC, BJAB, and LAD2 cells were determined by quantitative real-time PCR. Normalized fold expression represents the concentration ratio between the specific chemokine and the housekeeping mRNA (GADPH). Data are the mean ± SD of two separate experiments; n.d., nondetected amplificate.

Figure 10

Dose and time response of BJAB and LAD2 cells to CXCL-13-induced chemotaxis. Chemotaxis of BJAB and LAD2 cells to 10 ng/ml (A) and 100 ng/ml (B) recombinant human CXCL13 at indicated time points is shown. Data represent mean ± SD of two separate experiments performed in triplicate.

Figure 11

A schematic representation of the dynamics taking place in the AITL microenvironment. AITL neoplastic cells and FDC secrete CXCL-13 contributing to the recruitment of MCs and to FDC hyperplasia. MC-derived IL-6, along with AITL-clone-derived IL-21 and FDC-synthesized IL-23 contribute to the preferential generation of Th17 cells, which in turn favor granulocyte accumulation and possibly trigger autoimmune phenomena.