PD-1 Regulates Passive Anaphylaxis: A Possible Role of the Mast Cell Intracellular Inhibitory Signal

Dear Editor,

Antigen engagement on immunoreceptors initiates a series of intracellular phosphorylation/dephosphorylation events and regulates immune effector functions located downstream. Src-family protein tyrosine kinases phosphorylate the immunoreceptor tyrosine-based activation motif (ITAM), while protein tyrosine phosphatase non-receptor type 11 (SHP-2) dephosphorylates the immunoreceptor tyrosine-based inhibitory motif (ITIM) or the immunoreceptor tyrosine-based switch motif (ITSM). ITIM/ITSM in the cytoplasmic tail of programmed cell death-1 (PD-1)¹ dampens immune effector functions following the engagement of T-cell receptor (TCR)² or B-cell receptor (BCR).³

The mast cell (MC) can serve as a critical effector of adaptive immune responses; analogous to TCR/BCR signaling pathways, the high-affinity immunoglobulin E (IgE) receptor (FcεRI) signaling pathway involves ITAM-or ITIM/ITSM-mediated immune signaling.⁴ Although the MC reportedly does not express PD-1 on its surface,⁵ ligation of this receptor does lead to phosphorylation of SHP-1/SHP-2 in a mast cell line,⁵ suggesting that the presence of receptor cytoplasmic tail-mediated inhibitory signaling is conserved across the cell lineage.²,³ Given that PD-1 interacts with SHP-2 even without the engagement of its own receptor in T cells in vitro,² we hypothesized that FcεRI signaling is inhibited through the cytoplasmic tail. To this end, we employed an IgE-mediated passive anaphylaxis model to assess MC effector function in vivo.

In contrast to the results of MC-specific SHP-2 deletion,⁶ the number of the cutaneous or peritoneal MC populations of PD-1 knockout (PD-1KO) mice were not significantly altered (Supplementary Fig. S1A and B, Supplementary Data S1). This discrepancy suggests that PD-1 is dispensable for the KIT proto-oncogene/SHP-2 axis-mediated development of the tissue MC lineage.⁶ PD-1KO mice exhibited a significantly greater decline in temperature than wild-type (WT) mice (Figure A). Likewise, in passive cutaneous anaphylaxis, PD-1KO mice exhibited a significantly greater degree of ear swelling than WT mice (Figure B). However, monoclonal antibody (mAb)-mediated PD-1 blockade did not produce such an exacerbated anaphylactic response (Figure C-E). Conversely, during active anaphylaxis that solely depends on adaptive immunity, PD-1KO mice exhibited a decline in temperature comparable to WT mice (data not shown). These results suggest that exacerbated passive anaphylaxis is not necessarily attributable to broad immunologic abnormalities in PD-1KO mice.

Figure. Enhanced passive anaphylaxis in PD-1KO, but not in mAb-mediated PD-1 blockade.

(A) Both WT and PD-1KO mice were injected intraperitoneally (20 μg of anti-DNP IgE) and intravenously (1 mg of DNP-human serum albumin). Rectal temperature was monitored for 90 minutes following the challenge (n = 4, two-way ANOVA). (B) Both WT and PD-1KO mice were injected intravenously (2 μg of anti-DNP IgE), and their ear pinnae were painted (20 μL of 0.3% 1-fluoro-2,4-dinitrobenzene). Ear thickness was measured 24 hours after the challenge (n = 6, Welch's t test). (C) Administration of anti-PD-1mAb to WT mice preceded the induction of passive systemic anaphylaxis. Rectal temperature was monitored for 90 minutes following the challenge (n = 4, two-way ANOVA). (D) Administration of anti-PD-1mAb to WT mice preceded the induction of passive cutaneous anaphylaxis. Ear thickness was measured 24 hours after the challenge (n = 6, Welch's t test). (E) A possible role for intracytoplasmic PD-1 signaling in mast cells. Multivalent antigen-binding causes aggregation of mast cell surface FcεRI. Subsequently, cytoplasmic tyrosine kinases phosphorylate ITAMs in the immunoreceptor tail, resulting in the rapid release of soluble mediators from preformed granules.

Wild-type/Untreated (WT/UT): ITSM-containing PD-1 cytoplasmic tail recruits SHP-2 to oppose phosphorylation events.

PD-1 knockout (PD-1KO): Complete deletion of PD-1 abrogates interaction with SHP-2 (and thus opposition to phosphorylation events).

PD-1 blockade (PD-1mAb): Because the PD-1 cytoplasmic tail remains intact, FcεRI activation overcomes receptor engagement-dependent opposition to phosphorylation events.

PD-1KO, programmed cell death-1 knockout; mAb, monoclonal antibody; PD-1, programmed cell death-1; DNP, dinitrophenyl; IgE, immunoglobulin E; ANOVA, analysis of variance; WT, wild-type; FcεRI, high-affinity IgE receptor; ITAM, immunoreceptor tyrosine-based activation motif; ITIM, immunoreceptor tyrosine-based inhibitory motif; ITSM, immunoreceptor tyrosine-based switch motif; PBS, phosphate-buffered saline; Ag, antigen; APC, antigen presentation cell; CM, cell membrane; PD-L1, programmed death-ligand 1; SHP-2, protein tyrosine phosphatase non-receptor type 11; TK, tyrosine-protein kinase.

^*P < 0.05.

Immune checkpoint inhibitors can evoke immune-related adverse events (irAEs).⁷ Analogous to TCR signaling,² our results suggest that an augmented IgE humoral immune response requires complete deletion of PD-1, while PD-1 blockade is dispensable. Furthermore, our results indicate that PD-1 has co-evolved with other inhibitory immunoreceptors to regulate the adaptive immune response at the effector phase,⁷ and may additionally correlate with clinical observations; although urticarial eruptions are a common form of cutaneous irAE, its systemic counterpart–anaphylaxis–is rarely reported in this context. Our results may be an important guide for medical oncology practice in that the PD-1 blockade would hardly aggravate IgE-mediated anaphylaxis. Further studies are needed to elucidate PD-1 functions on MC.

SUPPLEMENTARY MATERIALS

Supplementary Data S1

Materials and methods

Supplementary Fig. S1

Unaltered mast cell homeostasis in the absence of PD-1.