Effect of a large deletion of the basic domain of mitranscription factor on differentiation of mast cells

The mi locus of mice encodes a member of the basic-helix-loop-helix-leucine zipper protein family of transcription factors (hereafter called MITF).1 MITF plays an important role in the development of mast cells.2-9Mast cells of mi/mi genotype express normal amounts of abnormal MITF (mi-MITF), whereas mast cells oftg/tg genotype do not express any MITFs.10-12Because tg is considered to be a null mutant allele, thetg/tg mice are useful as a standard for evaluating the function of other mutant MITFs.13-15 We previously compared the phenotype of mast cells of mi/mi mice with that of tg/tg mice.13 The transcription of mouse mast cell protease (mMCP)–4, mMCP-5, and mMCP-6 genes in cultured mast cells (CMCs) derived from mi/mi mice is reduced to the level comparable to that of tg/tg CMCs.13 The transcription of c-kit, granzyme B (Gr B), and tryptophan hydroxylase (TPH) genes is significantly reduced in mi/miCMCs, but the reduction of transactivation of these genes is moderate in tg/tg CMCs.13 This shows that themi-MITF possesses an inhibitory effect on the transcription of c-kit, Gr B, and TPH genes.13 Themi-MITF deletes one arginine in the basic domain, whereas MITF encoded by the mi^ew mutant allele (ew-MITF) deletes 16 of 21 amino acids of the basic domain.10 11 In the present study, we compared the mast cell phenotypes of mi^ew/mi^ew mice with those of tg/tg and mi/mi mice to examine the effect of a large deletion of the basic domain.

The C57BL/6-tg/tg and C57BL/6-mi/mi mice (hereafter tg/tg and mi/mi mice) were described previously.13 The C57BL/6-mi^ew/+mice16 were sent from the laboratory of M. L. Lamoreux, and C57BL/6-mi^ew/mi^ew mice were raised in our laboratory. C57BL/6-+/+ mice were used as a control. CMCs and P815 cells have been described.13 

Total RNA prepared from CMCs was electrophoresed and blotted. The intensity of the signal for each messenger RNA (mRNA) was quantified by a densitometer (Molecular Dynamics, Sunnyvale, CA).

The concentration of serotonin was measured using high-performance liquid chromatography as previously described.17 

Pieces of dorsal skin were removed from mice aged 20 days and embedded in paraffin. A section was stained with alcian blue and nuclear fast red, and an adjacent section was stained with berberine sulfate.18 

The reporter plasmid that contained Gr B promoter between nucleotide −910 and +42 (pGr B-910, +1 shows a transcription initiation site) and the reporter plasmid lacking Gr B promoter (pGr B-42) have been described.13 The luciferase activity was normalized as previously described.13 

The phenotype of mi^ew/mi^ew mast cells was compared with that of +/+, tg/tg, andmi/mi mast cells. First, CMCs were obtained from spleens of mice of each genotype.13 The expression of genes that were known to be affected by MITF19-23 was compared with Northern blot. The amounts of c-kit, Gr B, and TPH mRNAs inmi^ew/mi^ew CMCs were comparable to those of tg/tg CMCs, and the amounts were intermediate between those of +/+ and mi/mi CMCs (Figure1A). Although the amounts of mMCP-4, mMCP-5, and mMCP-6 mRNAs were very low in all kinds of CMCs examined here, their amounts in mi^ew/mi^ewCMCs were comparable to those of tg/tg and mi/miCMCs (Figure 1A).

We quantified the amounts of mRNAs using densitometry and calculated the ratio of each mRNA amount of +/+, mi/mi, ormi^ew/mi^ew CMCs to that oftg/tg CMCs (Figure 1B). We defined the ratio as relative mRNA amount. The relative mRNA amounts of all examined genes were more than 1.0 in +/+ CMCs, indicating that the +-MITF had positive effect on transcription. In mi/mi CMCs, the relative mRNA amounts of mMCP-4, mMCP-5, and mMCP-6 genes were about 1.0, but those of c-kit, Gr B, and TPH genes were less than 1.0. This suggested that the mi-MITF had negative effect on the transcription of c-kit, Gr B, and TPH genes. The relative mRNA amounts of all examined genes were about 1.0 inmi^ew/mi^ew CMCs, indicating that theew-MITF had neither positive nor negative effects on transcription.

Because TPH is the rate-limiting enzyme of the serotonin synthesis,24 the serotonin contents were compared among various CMCs. The serotonin content ofmi^ew/mi^ew CMCs was comparable to that of tg/tg CMCs. Both values were significantly smaller than that of +/+ CMCs and significantly greater than that ofmi/mi CMCs (Figure 1C, P < .05 by ttest). Serotonin contents of +/+, tg/tg, mi/mi,and mi^ew/mi^ew CMCs were well correlated with the mRNA expression levels of TPH gene in CMCs of each genotype (+/+ > mi^ew/mi^ew ∼tg/tg > mi/mi).

Second, the number of mast cells and the proportion of berberine sulfate–positive mast cells were compared among skin tissues of +/+,tg/tg, mi/mi, andmi^ew/mi^ew mice. The number of skin mast cells of mi^ew/mi^ew mice decreased to one third that of the +/+ mice and was comparable to that of tg/tg and mi/mi mice (Figure 1D). Most skin mast cells were berberine sulfate–positive in +/+ mice.25The proportion of berberine sulfate–positive to alcian blue–positive mast cells in the skin of mi^ew/mi^ewmice was comparable to that of +/+ mice and to that of tg/tgmice (Figure 1D). In contrast, the proportion of berberine sulfate–positive mast cells was only 3% in the skin of mi/mimice as reported previously.7 The abnormalities ofmi^ew/mi^ew skin mast cells were comparable to those of tg/tg skin mast cells rather than to those of mi/mi skin mast cells.

We examined the effect of ew-MITF on the transactivation of Gr B promoter using the transient cotransfection assay. The luciferase construct containing the Gr B promoter was cotransfected into P815 cells with the expression plasmid containing no insert, +-MITF,mi-MITF, or ew-MITF complementary DNA. The coexpression of +-MITF significantly increased the luciferase activity, whereas that of mi-MITF significantly reduced it as reported previously (Figure 2A).13The expression of ew-MITF showed luciferase activity comparable to the value obtained by the expression vector alone (Figure2A). The ew-MITF had no effect on the activity of Gr B promoter. This was consistent with the fact that the abnormalities of CMCs and skin mast cells ofmi^ew/mi^ew mice were similar to those of tg/tg mice. The part of the basic domain that was deleted in ew-MITF (Figure 2B) appeared necessary for both positive and negative functions.