Mast cell activation contributes to sickle cell pathobiology and pain in mice

Abstract

Sickle cell anemia (SCA) is an inherited disorder associated with severe lifelong pain and significant morbidity. The mechanisms of pain in SCA remain poorly understood. We show that mast cell activation/degranulation contributes to sickle pain pathophysiology by promoting neurogenic inflammation and nociceptor activation via the release of substance P in the skin and dorsal root ganglion. Mast cell inhibition with imatinib ameliorated cytokine release from skin biopsies and led to a correlative decrease in granulocyte-macrophage colony-stimulating factor and white blood cells in transgenic sickle mice. Targeting mast cells by genetic mutation or pharmacologic inhibition with imatinib ameliorates tonic hyperalgesia and prevents hypoxia/reoxygenation-induced hyperalgesia in sickle mice. Pretreatment with the mast cell stabilizer cromolyn sodium improved analgesia following low doses of morphine that were otherwise ineffective. Mast cell activation therefore underlies sickle pathophysiology leading to inflammation, vascular dysfunction, pain, and requirement for high doses of morphine. Pharmacological targeting of mast cells with imatinib may be a suitable approach to address pain and perhaps treat SCA.

Figure 1

Mast cell activation occurs in sickle cell anemia. (A) ir pixels for tryptase, SP, and CGRP in the dorsal skin. *P < .05, **P < .01 vs corresponding HbAA-BERK (ANOVA, with Bonferroni). (B) Representative confocal images of dorsal skin sections showing mast cell–specific costaining for c-kit/CD117 (red), FcεRI (green), and tryptase (blue). (C) Skin mast cells in culture stained for c-kit/CD117 (red), FcεRI (green), and tryptase (blue). (D-F) Gene expression of c-kit/CD117 (D), FcεRI (E), and TLR-4 (F) in mast cells derived from skin, measured by reverse-transcription quantitative polymerase chain reaction (RT-qPCR), normalized to GAPDH mRNA, and shown relative to HbAA-BERK, which was given an arbitrary value of 1. ***P < .001 (Student t test). (G) Tryptase levels in the supernatant of skin-derived mast cells after incubation with vehicle (Veh) or SP for indicated time. Black bars, HbAA-BERK; red bars, HbSS-BERK. *P < .05, **P < .01 vs corresponding HbAA-BERK; #P < .05 vs vehicle HbSS-BERK (ANOVA, with Bonferroni). (H-I) Skin-derived mast cells incubated with imatinib mesylate (Imat, 10 μM) and/or morphine sulfate (MS, 1 μM) for 24 hours. Culture medium showing tryptase (H) and SP (I) levels. *P < .05, **P < .01, ***P < .001 vs corresponding HbAA-BERK, #P < .05, ##P < .01 vs HbSS-BERK vehicle, $P < .05, $$P < .01 vs HbAA-BERK vehicle (Student t test). Each value is the mean ± SEM of 5 mice of each type, and each image represents images from the skin or skin-derived mast cells from 5 different mice.

Figure 2

Mast cell activation contributes to neuro-inflammation in SCA. (A-K) HbSS-BERK mice were treated with saline (Veh), CS, or imatinib mesylate (Imat) for 5 days followed by analysis as indicated with each figure. (A) Representative images of Toluidine blue–stained dorsal skin sections. n = 6; bar = 100 μm. (B) Ratio of degranulating/total mast cells. *P < .01 vs HbAA-BERK, #P < .05 vs HbSS-BERK Veh (ANOVA, with Bonferroni). (C-D) Levels of tryptase, β-hexosaminidase (β-hex), SAP, SP, and CGRP expressed as the percentage of HbAA-BERK values. Red bars, HbSS-veh; magenta bars, HbSS-CS; blue bars, HbSS-Imat. *P < .05, **P < .01 vs HbAA-BERK, #P < .05, ##P < .01 vs HbSS-BERK Veh (ANOVA, with Bonferroni). (E-F) Cytokine released following 24 hours of incubation of skin from mice treated with vehicle or imatinib for 5 days. Black bars, HbAA-BERK Veh; red bars, HbSS-BERK Veh; blue bars, HbSS-BERK imatinib. Values are expressed as a percent of HbAA (E) or HbSS veh (F). *P < .05, **P < .01 vs HbAA-BERK Veh (E) or HbSS-BERK Veh (F) (Student t test). (G) Linear regression analysis of the expression of plasma GM-CSF and total WBC counts. (H-K) Neuropeptide release from the skin (H-I) and DRG (J-K) following 24 hours in culture. *P < .05, **P < .01 vs HbAA-BERK (H,J). *P < .05 vs HbSS-BERK Veh, **P < .01 vs HbSS-BERK Veh (I,K) (ANOVA, with Bonferroni). Each value is the mean ± SEM of 6 mice per group (in B-K). (L) Percentage of degranulating mast cells in DRG of mice treated with vehicle or imatinib for 5 days. *P < .001 vs HbAA-BERK Veh, #P < .05 vs HbSS-BERK Veh (ANOVA, with Bonferroni). (M-N) Representative confocal images showing coexpression of ATF3 (green) and GFAP (red). Scale bar, 100 µm; n = 6.

Figure 3

Neurogenic inflammation occurs in SCA. (A-B) Evans blue leakage evoked by injection of saline, capsaicin, or SP in the hind paws (A) and dorsal skin (B) of mice treated with vehicle (saline), morphine sulfate (MS, 10 mg/kg), CS (100 mg/kg), or imatinib (Imat, 100 mg/kg). (C-D) Evans blue leakage. *P < .05, **P < .01 vs HbSS-BERK Veh, #P < .05, ##P < .01 vs HbAA-BERK Veh (ANOVA with Bonferroni). (E) Measures of cutaneous blood flow. **P < .01 (Student t test). (F) Temperature of the dorsal skin of mice with representative thermogram. *P < .05 (Student t test). (G) Calibration bar used.

Figure 4

Mast cells contribute to hyperalgesia in SCA. (A-D) HbSS-BERK mice were treated with vehicle (Veh), CS, or imatinib (Imat). On day 5, measures of pain were recorded 30 min after the injection of morphine (MS) or phosphate-buffered saline (PBS) as shown in (D). Measures of deep pain (A), mechanical hyperalgesia (B), and thermal sensitivity (C) are shown. BL, baseline obtained before the treatments. *P < .05, **P < .01 versus BL (ANOVA, with Bonferroni). #P < .05, ##P < .01 vs PBS for each corresponding treatment (Student t test). Each value is the mean ± SEM from 8 mice with 3 observations per mouse. (D) Treatment protocol. HbAA-BERK and HbSS-BERK mice were treated with saline, CS, or imatinib sulfate for 5 days. At day 5, a single injection of morphine sulfate or PBS was given and pain behaviors were followed for 240 min. i.p., intraperitoneal; s.c., subcutaneous; red arrow, pain testing. (E-J) Pain-related behaviors of age-matched HbAA BERK, HbSS-BERK, Kit^W/Wv (W/Wv), and HbSS-Kit^W/Wv (HbSS-W/Wv) mice. Each value is the mean ± SEM from 4-6 mice with 3 observations per mouse. *P < .05, **P < .01, ***P < .001 vs HbSS-BERK (ANOVA, with Bonferroni).

Figure 5

Mast cell activation contributes to hypoxia/reoxygenation-induced pain in SCA. (A-C) HbAA- and HbSS-BERK mice were treated with saline (Veh) or imatinib (imat) for 5 days. All mice were then treated with 3 hours hypoxia and 1 hour reoxygenation (H/R). Pain measures were obtained before starting the drug treatments on day 0 (D₀), at the conclusion of drug treatments, D₅ before inciting H/R, immediately after H/R, and 24 hours after H/R as indicated by red arrows in the schema in (G). Measures of deep pain (A), mechanical hyperalgesia (B), and thermal sensitivity (C) are shown. ¶P < .05, ¶¶P < .001 versus D₀ of matched treatment; #P < .05, ##P < .001 vs D₅ pre-H/R of matched treatment; *P < .05, **P < .01 versus HbSS Veh (ANOVA, with Bonferroni). Each value is the mean ± SEM from 6 mice with 3 observations per mouse. (D-F) Pain-related behaviors from age-matched HbSS-BERK and HbSS-KitW/Wv (HbSS-W/Wv) mice before (baseline, BL), immediately after (Post H/R), and 1 day post-H/R. *P < .05, **P < .01, ***P < .001 vs HbSS-BERK of matched pain-testing time point; #P < .05 vs HbSS-BERK BL; ¶P < .05 vs HbSS-W/Wv BL (Student t test). Each value is the mean ± SEM of 5 mice.

Figure 6

Activated mast cells contribute to a feed-forward cycle of neuropeptide release in the skin of sickle mice. (1) Tryptase from mast cells activates PAR2 on the peripheral nerve endings. (2) Activation of PAR2 sensitizes transient receptor potential vanilloid 1 (TRPV1). (3) Excited nociceptors stimulate the release of CGRP and SP from the sensory nerve endings. (4) CGRP interacts with the type 1 CGRP receptor on arterioles to induce dilatation. (5) Substance P activates plasma extravasation via neurokinin 1 (NK1) receptors. (6) SP released from nerve endings as well as from mast cells also acts on the mast cells themselves, thus promoting a vicious cycle of mast cell activation.