Intestinal mast cell progenitors require CD49dbeta7 (alpha4beta7 integrin) for tissue-specific homing

Abstract

Mast cells (MCs) are centrally important in allergic inflammation of the airways, as well as in the intestinal immune response to helminth infection. A single lineage of bone marrow (BM)-derived progenitors emigrates from the circulation and matures into phenotypically distinct MCs in different tissues. Because the mechanisms of MC progenitor (MCp) homing to peripheral tissues have not been evaluated, we used limiting dilution analysis to measure the concentration of MCp in various tissues of mice deficient for candidate homing molecules. MCp were almost completely absent in the small intestine but were present in the lung, spleen, BM, and large intestine of beta7 integrin-deficient mice (on the C57BL/6 background), indicating that a beta7 integrin is critical for homing of these cells to the small intestine. MCp concentrations were not altered in the tissues of mice deficient in the alphaE integrin (CD103), the beta2 integrin (CD18), or the recombination activating gene (RAG)-2 gene either alone or in combination with the interleukin (IL)-receptor common gamma chain. Therefore, it is the alpha4beta7 integrin and not the alphaEbeta7 integrin that is critical, and lymphocytes and natural killer cells play no role in directing MCp migration under basal conditions. When MCp in BALB/c mice were eliminated with sublethal doses of gamma-radiation and then reconstituted with syngeneic BM, the administration of anti-alpha4beta7 integrin, anti-alpha4 integrin, anti-beta7 integrin, or anti-MAdCAM-1 monoclonal antibodies (mAbs) blocked the recovery of MCp in the small intestine. The blocking mAbs could be administered as late as 4 d after BM reconstitution with optimal inhibition, implying that the MCp must arise first in the BM, circulate in the vasculature, and then translocate into the intestine. Inasmuch as MCp are preserved in the lungs of beta7 integrin-deficient and anti-alpha4beta7 integrin-treated mice but not in the small intestine, alpha4beta7 integrin is critical for tissue specific extravasation for localization of MCp in the small intestine, but not the lungs.

Figure 1.

Concentrations of MCp in the small intestine and lung of C57BL/6 (white bars) and BALB/c (black bars) mice. MNCs were isolated from the intestine or the lung of normal C57BL/6J or BALB/c mice by enzymatic digestion. The MCp concentration is the number of MCp per 10⁶ MNCs as determined by limiting dilution analysis with the cytokine combinations of SCF plus IL-3, IL-3 alone, or SCF plus IL-6 plus IL-10 (Triad). The values are the means ± SEM for five experiments. ND, not done.

Figure 2.

Concentrations of MCp in the small intestine, lung, spleen, and BM of wild-type (C57BL/6, +/+, white bars) control and β7 integrin-deficient mice (−/−, black bars) evaluated in parallel. The data are from three separate experiments and the values are the mean ± SEM. SCF plus IL-3 was used to determine MCp concentration, expressed as the number of MCp/10⁶ MNCs.

Figure 3.

Concentrations of MCp in the small intestine, lung, and BM of wild-type (C57BL/6, +/+, white bars) controls and RAG-2/IL-receptor common γ-chain (γc) double-deficient mice (−/−, black bars). The data are the mean ± 1/2 range of MCp/10⁶ MNCs from two experiments using SCF plus IL-3 in which concentrations in wild-type and deficient mice were determined in parallel.

Figure 4.

Recovery of MCp in the small intestine of BALB/c mice after sublethal irradiation with (filled circles) or without (open circles) BM reconstitution (BMR). Values for MCp concentrations without BM reconstitution are the mean ± 1/2 range of two experiments. Values for animals receiving BM reconstitution are the mean ± SEM of the following number of experiments: day 7, n = 12; day 8, n = 9; day 9, n = 6; day 11, n = 5; day 14, n = 4. MCp concentrations were determined with SCF plus IL-3. The shaded area represents the mean ± SEM concentration of MCp in normal BALB/c small intestine (from Fig. 1, SCF plus IL-3).

Figure 5.

Effect of the blocking anti-integrin mAb, PS/2 (anti-α4, rat IgG2b, open circles) and DATK32 (anti-α4β7, rat IgG2a, filled circles), on the recovery of MCp in the small intestine of BALB/c mice. MCp concentrations were determined 9 d after sublethal irradiation and BM reconstitution and are expressed as the percent inhibition of MCp concentrations in the intestine relative to mice treated with a control mAb (rat anti–mouse IgD, IgG2a) or HBSS. mAb was diluted in HBSS and administered intraperitoneally every other day starting 4 h after BM reconstitution. The results are the average inhibition for two or three experiments with the following n values: For PS/2, 100 μg, n = 2; 30 μg, n = 3, 10 μg and 1 μg, n = 1; for DATK32, 30 μg, n = 3, for 10 μg, 1 μg, and 0.1 μg, n = 1. The average concentration of intestinal MCp for mice given the control mAb or HBSS was 903 ± 311 MCp/10⁶ MNCs (mean ± SEM, n = 5).

Figure 6.

Effect of delaying the time of initial administration of the blocking mAb DATK32 (anti-α4β7 integrin) on the recovery of MCp in the small intestine of BALB/c mice. The MCp concentrations were determined 7 d after sublethal irradiation and BM reconstitution and are expressed as the percent inhibition of MCp concentrations in the intestine relative to the mice treated with the isotype-matched rat anti–mouse IgD mAb. The control mice had an average MCp concentration of 632 ± 206 MCp/10⁶ MNCs (mean ± SEM, n = 6). Experimental values are the mean ± SEM of four (day 0 and day 2) or two (day 4 and day 6) experiments.

Figure 7.

Comparison of the inhibition by mAb to α4β7, β7, αE, or β1 integrins of the recovery of MCp in the small intestine of BALB/c mice. Animals received 30 μg of the indicated mAb, DATK32, FIB27, M290, or 9EG7, directed against α4β7, β7, αE, or β1 integrins, respectively, starting 0 or 2 d after sublethal irradiation and BM reconstitution. Values represent the mean ± SEM percent inhibition relative to control animals treated in parallel. MCp concentrations were analyzed 7 or 8 d after sublethal irradiation and BM reconstitution. Means are from seven (DATK32), three (FIB27, M290), or four (9EG7) experiments. The mean MCp concentration for the control mice was 622 ± 65 MCp/10⁶ MNCs (mean ± SEM). Asterisks indicate statistically significant inhibition relative to mice injected with anti-β1 integrin mAb (P < 0.03).

Figure 8.

Inhibition by mAb to α4β7 integrin of the recovery of MCp in the small intestine of BALB/c mice is accompanied by increased MCp in the lung. Animals received 30 μg of the mAb, DATK32, directed against the α4β7 integrin, or a control injection of HBSS or isotype matched rat mAb, starting 2 d after sublethal irradiation and BM reconstitution. Values represent the mean ± SEM (n = 4) of the MCp concentration in the indicated tissues expressed as a percentage of the value obtained from control animals treated in parallel. MCp concentrations were analyzed 7 d after sublethal irradiation and BM reconstitution. The MCp concentrations for the control mice were 1,068 ± 390 and 252 ± 51 MCp/10⁶ MNCs (mean ± SEM) for the intestine and lung, respectively.