Electrotransformation and expression of bacterial genes encoding hygromycin phosphotransferase and beta-galactosidase in the pathogenic fungus Histoplasma capsulatum

Abstract

We developed an efficient electrotransformation system for the pathogenic fungus Histoplasma capsulatum and used it to examine the effects of features of the transforming DNA on transformation efficiency and fate of the transforming DNA and to demonstrate fungal expression of two recombinant Escherichia coli genes, hph and lacZ. Linearized DNA and plasmids containing Histoplasma telomeric sequences showed the greatest transformation efficiencies, while the plasmid vector had no significant effect, nor did the derivation of the selectable URA5 marker (native Histoplasma gene or a heterologous Podospora anserina gene). Electrotransformation resulted in more frequent multimerization, other modification, or possibly chromosomal integration of transforming telomeric plasmids when saturating amounts of DNA were used, but this effect was not observed with smaller amounts of transforming DNA. We developed another selection system using a hygromycin B resistance marker from plasmid pAN7-1, consisting of the E. coli hph gene flanked by Aspergillus nidulans promoter and terminator sequences. Much of the heterologous fungal sequences could be removed without compromising function in H. capsulatum, allowing construction of a substantially smaller effective marker fragment. Transformation efficiency increased when nonselective conditions were maintained for a time after electrotransformation before selection with the protein synthesis inhibitor hygromycin B was imposed. Finally, we constructed a readily detectable and quantifiable reporter gene by fusing Histoplasma URA5 with E. coli lacZ, resulting in expression of functional beta-galactosidase in H. capsulatum. Demonstration of expression of bacterial genes as effective selectable markers and reporters, together with a highly efficient electrotransformation system, provide valuable approaches for molecular genetic analysis and manipulation of H. capsulatum, which have proven useful for examination of targeted gene disruption, regulated gene expression, and potential virulence determinants in this fungus.

FIG. 1

Diagrams of selected plasmids used in this study. When a restriction enzyme is designated on a plasmid map, then all of the sites for that enzyme on that plasmid are indicated. “TEL” indicates an H. capsulatum telomeric sequence comprised of repeats of the hexamer GGGTTA. Other plasmid elements are described in the text.

FIG. 2

Southern blot of undigested genomic DNA from 10 strain G184ASura5-11 transformants (lanes 1 to 10) obtained after chemical transformation using HpaI-linearized pWU51, probed with ³²P-labeled vector pBR328 (4.9 kb). The leftmost lane contains linearized pWU51 (6.3 kb). Positions of molecular size markers are indicated to the left. Transformants 2 to 10 carry monomeric linear plasmids. Transformant 1 carries a larger linear plasmid. Transformants 3 and 8 show faint hybridization with the broad smear of chromosomal fragments in undigested H. capsulatum genomic DNA, indicating either very large linear plasmids or the occurrence of chromosomal integration of the transforming DNA at a barely detectable level. We did not attempt to distinguish between the two fates of chromosomal integration and very large linear plasmid formation in this study.

FIG. 3

Effect of amount of transforming DNA on the occurrence of monomeric linear plasmids after electroporation of strain G184ASura5-11 with HpaI-linearized pWU44. Each Southern blot was probed with the ³²P-labeled selectable marker PaURA5 (1.6-kb EcoRI fragment from pWU44) and contains in the leftmost lane linearized pWU44 (6.7 kb). Positions of molecular size markers are indicated on the left, as is the location of the broad smear of chromosomal fragments in undigested H. capsulatum genomic DNA on the precedent agarose gels. We did not attempt to distinguish between the two fates of chromosomal integration and very large linear plasmid formation in this study. (A) Undigested genomic DNA from 14 transformants (lanes 1 to 14) obtained by using 1 ng of transforming DNA. All transformants carry monomeric linear plasmids except for transformant 2, which shows evidence of either very large linear plasmids or chromosomal integration of the transforming DNA. (B) Undigested genomic DNA from 13 transformants (lanes 1 to 13) obtained by using 1 μg of transforming DNA. Nine of the transformants carry monomeric linear plasmids, alone or with modified linear plasmids; seven of the transformants carry modified linear plasmids, alone or with a monomeric linear plasmid; and one transformant (transformant 8) shows evidence of either very large linear plasmids or chromosomal integration of the transforming DNA. (C) Undigested genomic DNA prepared from pools of transformants. Lane 1, pool of 21 transformants obtained by using 1 ng of transforming DNA; lane 2, 265 transformants obtained by using 10 ng; lane 3, 123 transformants obtained by using 100 ng; lane 4, 780 transformants obtained by using 1 μg; lane 5, 1,512 transformants obtained by using 10 μg. Each pool shows the presence of monomeric linear plasmids, and additionally modified linear plasmids and either very large linear plasmids or chromosomal integration of the transforming DNA are shown in pools obtained by using larger amounts of transforming DNA.

FIG. 4

Colorimetric detection of HcURA5-lacZ fusion gene expression in colonies on solid medium. Strain G184ASura5-11 was electrotransformed with PacI-digested pWU77 carrying the fusion gene (top line, blue color) or with HpaI-linearized plasmid pWU44 lacking the fusion gene (bottom line, white color) and streaked in a pattern on HMM-agarose. Following growth, X-Gal solution was added to the streaks, leading to the appearance of blue color within minutes.