Sequence elements necessary for transcriptional activation of BAD1 in the yeast phase of Blastomyces dermatitidis

Abstract

Blastomyces dermatitidis is a dimorphic fungal pathogen that converts from mycelia or conidia to a host-adapted yeast morphotype upon infection. Conversion to the yeast form is accompanied by the production of the virulence factor BAD1. Yeast-phase-specific expression of BAD1 is transcriptionally regulated, and its promoter shares homology with that of the yeast-phase-specific gene YPS3 of Histoplasma capsulatum. Serial truncations of the BAD1 upstream region were fused to the lacZ reporter to define functional areas in the promoter. Examination of PBAD1-lacZ fusions in B. dermatitidis indicated that BAD1 transcription is upregulated in the yeast phase. The 63-nucleotide box A region conserved in the YPS3 upstream region was shown to be an essential component of the minimal BAD1 promoter. A matched PYPS3-lacZ construct indicated that this same region was needed for minimal YPS3 promoter activity in B. dermatitidis transformants. Reporter activity in H. capsulatum transformants similarly showed a requirement for box A in the minimal BAD1 promoter. Several putative transcription factor binding sites were identified within box A of BAD1. Replacement of two of these predicted sites within box A--a cAMP responsive element and a Myb binding site--sharply reduced transcriptional activity, indicating that these regions are critical in dictating the yeast-phase-specific expression of this crucial virulence determinant of B. dermatitidis.

FIG. 1.

Serial truncations of the upstream regions of BAD1 and YPS3. Shown are schematic representations of amplified fragments that were fused to lacZ to assess the contributions of promoter regions to B. dermatitidis (Bd) BAD1 and H. capsulatum (Hc) YPS3 transcription (A and B, respectively). Transcriptional start sites are indicated by arrows. Numbering of the 5′ and 3′ (at the translation-initiating ATG) ends of the promoter fragments are indicated relative to the transcriptional start sites (+1). At the right are promoter fragment designations, based on either the distance upstream of the transcriptional start site or homology between the BAD1 and YPS3 promoters at boxes B and A, shown as shaded boxes.

FIG. 2.

β-Galactosidase activity in B. dermatitidis P_BAD1-lacZ, P_YPS3-lacZ, and control P_URA5-lacZ transformants. B. dermatitidis 26199 transformed with lacZ fusion vectors was analyzed in both the yeast and mycelial phase for β-galactosidase activity using an o-nitrophenyl β-d-galactopyranoside assay. Activity levels for yeast are shown as black bars, and the levels for mold are shown as shaded bars. Test promoters driving lacZ expression are indicated on the x axis according to the designations indicated in Table 1 and/or Fig. 2. Eight independent transformants of each vector were analyzed in three experiments, and the results were averaged. Values are reported in milliunits of β-galactosidase activity per milligram of protein and represent the mean ± standard error. (A) β-Galactosidase activity of transformants of BAD1 promoter truncations fused to lacZ. (B) β-Galactosidase activity of transformants of YPS3 promoter truncations fused to lacZ.

FIG. 3.

β-Galactosidase activity in H. capsulatum P_BAD1-lacZ and P_YPS3-lacZ transformants. Selected yeast-phase H. capsulatum strain G217B transformants in which the lacZ fusion vectors were episomal were analyzed for β-galactosidase activity using an o-nitrophenyl β-d-galactopyranoside assay. Independent transformants analyzed are indicated along the x axis according to the promoter fragment driving lacZ expression (see Fig. 2). Transformants were analyzed in three experiments, and the results were averaged. Values, scored in milliunits of β-galactosidase activity per milligram of protein, were normalized to the relative vector copy number per genome (see Materials and Methods) and are reported as relative milliunits of galactosidase activity per milligram of protein, representing the mean ± standard error. (A) β-Galactosidase activities of transformants of BAD1 promoter truncations fused to lacZ. (B) β-Galactosidase activities of transformants of YPS3 promoter truncations fused to lacZ.

FIG. 4.

β-Galactosidase activity in B. dermatitidis transformants of reporter fusions containing modified BAD1 or URA5 promoters. B. dermatitidis 26199 transformed with lacZ fusion vectors was analyzed in the yeast phase for β-galactosidase activity using an o-nitrophenyl β-d-galactopyranoside assay. Test promoters driving lacZ expression are indicated on the x axis according to the designations indicated in the Results section. Eight independent transformants of each vector were analyzed in three experiments, and the results were averaged. Values are reported as milliunits of β-galactosidase activity per milligram of protein and represent the mean ± standard error. (A) β-Galactosidase activities of transformants of constructs in which the BAD1 promoter is fused to lacZ, comparing activities from box A-containing vectors versus those in which box A was replaced with sequence from the actin gene (ACT1) of S. cerevisiae. (B) β-Galactosidase activities of transformants harboring the URA5 promoter fused to lacZ, with the upstream addition of either the box A region of the BAD1 promoter or a similarly sized fragment of the ACT1 coding region. The activity of the BAD1 box A-lacZ construct is included for comparison.

FIG. 5.

Possible transcription factor binding sites within the box A region and their effect on transcription from the BAD1 promoter. (A) Transcription factor binding sites in which the four core bases of the consensus were found at 100% identity and the entire consensus was 90% or greater by MatInspector matrix similarity. Sites are named for the proteins recruited, shown in shaded boxes over the box A sequence. (B and C) β-Galactosidase activities of yeast-phase B. dermatitidis transformants of BAD1 wild-type and modified box A-lacZ vectors. On the x axes, modified box A promoters, in which a 10-bp region is replaced, are indicated by a delta symbol, followed by the predicted transcription factor binding site (CRE), or the protein (Sox-5, Myb, NF-1, and AREB6) that may bind the site found within the altered region. Eight independent transformants of each vector were analyzed in three experiments, and the results were averaged. Values are reported as milliunits of β-galactosidase activity per milligram of protein and represent the mean ± standard error. Panels B and C represent analyses of transformant sets generated in separate experiments.

FIG. 6.

Locations of putative transcription factor binding sites of the BAD1 promoter outside of box A. Sites are labeled for the type of protein or proteins recruited (Myb, MADS box, GATA, Sox, AbaA, and StuA) or the sequence element itself (STRE and HSE) and are shown in their relative positions on the BAD1 promoter map. Binding sites listed share 100% identity with the four core bases of the transcription factor consensus binding site, and the entire consensus was 90% or greater by MatInspector matrix similarity. Note that the MADS box transcription factor binding sites were identified as matching that of the mammalian factor Mef2a (32). The BAD1 transcriptional start site (+1) is shown as an arrow, and boxes B and A are shown as shaded areas. Actual locations of the binding sites are as follows: STRE, −488 to −484; Myb, −463 to −458; CRE, −391 to −384; MADS box, −411 to −402 and +69 to +78; GATA, −352 to −347, −310 to −305, and −54 to −49; HSE, −277 to −264; Sox, −146 to −140; AbaA, +99 to +104; StuA, +114 to +121.