The spoIIE Homolog of Epulopiscium sp. Type B Is Expressed Early in Intracellular Offspring Development

Putative homologs of spoIIE, spoIIAA, spoIIAB, and sigF genes have been identified in the Epulopiscium sp. type B genome, a finding which suggests that Epulopiscium has an early offspring-specific sigma factor activated by the cascade of molecular interactions described for B. subtilis.

SpoIIE is a large protein with three distinct domains. The N terminus targets SpoIIE to the cell membrane (7, 9). Domain II facilitates interaction with FtsZ and SpoIIE oligomerization (34). Conserved amino acid residues in the C-terminal phosphatase domain place SpoIIE within the PP2C protein phosphatase family (1, 44). All strains, plasmids, and primers used in the present study are provided in Tables 1 and 2. The sequence of the spoIIE gene from Epulopiscium plus 269 bases upstream of the predicted start codon was determined. The gene is 2,352 bp long, and the predicted protein shares 22% amino acid identity and 50% similarity with B. subtilis SpoIIE (Fig. 2). A hydrophobicity plot of Epulopiscium SpoIIE reveals a potential N terminus transmembrane domain to approximately residue 304, similar to the 324-amino-acid hydrophobic domain I of B. subtilis SpoIIE (data not shown) (7, 9). Domain II is not highly conserved between known SpoIIE homologs, and the B. subtilis and Epulopiscium proteins are only 16% identical. The phosphatase domain (III) is highly conserved, with 35% identity and 64% similarity to B. subtilis SpoIIE. Two essential active-site aspartic acid residues and two downstream amino acid motifs (Fig. 2B) found in all PP2C type phosphatases (1) are conserved in Epulopiscium SpoIIE. The σ^A promoters of both B. subtilis and Epulopiscium spoIIE genes have an unusual spacing of 21 bp (instead of the typical 17 bp) between the −10 and −35 sequences (52).

The spoIIAA, spoIIAB, and sigF genes are found in an operon in the Epulopiscium genome. Alignments of the predicted protein sequences of SpoIIAA, SpoIIAB, and SigF from Epulopiscium with B. subtilis homologs indicate 27%, 41%, and 48% identity and 66%, 67%, and 72% similarity, respectively.

Naso tonganus, which feeds diurnally on plant material but retains food in the alimentary tract overnight (13), was collected by spearfishing on outer reefs in the vicinity of Lizard Island, Great Barrier Reef, Australia, between 1000 and 1400 h. Ethanol-fixed intestinal contents were stained with 2 μg/ml 4′,6-diamidino-2-phenylindole (DAPI) (50), and randomly selected Epulopiscium cells were classified by stage of development (Fig. 3). Since morphological transitions in Epulopiscium are similar to those in a sporulating cell (4), we used the classically described stages of endospore formation (42) to represent similar events in Epulopiscium offspring development.

Intestinal contents from N. tonganus were also fixed with RNAprotect (Qiagen). For each sample, total RNA isolated from a 1-ml aliquot was converted to cDNA and used in quantitative PCR (qPCR) assays using Power SYBR Green master mix (Applied Biosystems). All qPCR primers were designed using Primer Express software, and qPCR was performed using an ABI 7300 real-time PCR system with default cycling conditions. To control for differences in Epulopiscium cell density in these samples, the rpoB housekeeping gene was used to normalize the data. This approach has been used in other studies of population-specific gene expression (14, 19, 41, 47) and seemed appropriate here. Based on previously published microarray studies, rpoB gene expression appears constant during the transition to stationary phase and throughout sporulation in B. subtilis (12, 23, 49). In only one study, a modest, 3-fold drop in the rpoB gene transcript level at t₁ was observed (11). We found that for both Epulopiscium and B. subtilis qPCR results (described below), accounting for this drop in the rpoB gene transcript number at t₁ and beyond does not affect the trends observed or the interpretation of the spoIIE gene transcription analyses. Specificity of all qPCR products generated from intestinal samples was confirmed by sequence analysis. Samples 1 and 2 had the highest spoIIE-to-rpoB gene transcript ratios (0.2256 and 0.1452), indicating that the spoIIE gene is highly expressed prior to polar division (Fig. 4 A). Sample 3 had a lower transcript ratio (0.0560), which correlates well with the distribution of the cells that had advanced beyond polar septum formation. Samples 4, 5, and 6 had low transcript ratios (0.0021, 0.0035, and 0.0020, respectively), which are consistent with these postengulfment populations.

To the best of our knowledge, a reverse transcription (RT)-qPCR assay for spoIIE gene expression has never been reported for sporulating B. subtilis; thus, we wanted to validate this method. Samples of B. subtilis were taken every half hour from the onset of sporulation (t₀), induced by resuspension (37), and assayed (Fig. 4B). No-template controls for the spoIIE gene yielded values that were three orders of magnitude less than that for any sample. Prior to t₁, a low spoIIE-to-rpoB gene transcript ratio was observed (0.0067 for t₀, 0.0052 for t_0.5), and transcript ratios then increased for t₁ and t_1.5 (0.0219, 0.0214, respectively). Samples from t₂ and later had greater ratios (0.2466, 0.1828, 0.3143, and 0.1425). Microscopic examination of cells stained with FM4-64, MitoTracker green, and DAPI (45) was used to determine the proportion of cells at specific morphological stages of sporulation (no polar septum, straight polar septum, curved septum, completely engulfed forespore) at each time point (data not shown). Large numbers of cells with a polar septum begin to appear at t_1.5 (8.2% of the population), and this stage becomes more prominent at t₂ (25.6%). By t₃, cells with a fully engulfed forespore are more abundant than cells with a polar septum. The qPCR results are comparable to published spoIIE gene transcription profiles using β-galactosidase fusions (27), in which spoIIE expression increases considerably in populations with cells that are just beginning to divide asymmetrically and persists even when the population has progressed and engulfed cells are abundant. Compared with that for Epulopiscium, the timing of spoIIE gene expression in B. subtilis populations appears slightly later in the developmental progression, and expression persists for a longer developmental interval. This may reflect response heterogeneity of sporulating B. subtilis cultures (15, 48). Sporulation is a last resort, and there is a clear advantage to the population for some cells to delay the commitment to sporulate as long as possible. Conversely, intracellular offspring formation is essential for Epulopiscium reproduction, and synchronized development may presage recurrent fluctuations in nutrients (50); thus, there would be little advantage for some cells in the population to delay development.

Taken together, the RT-qPCR results from Epulopiscium and B. subtilis indicate that the spoIIE homolog in Epulopiscium is expressed in a manner that would support its predicted role in offspring development. Expression peaks just prior to polar division, when SpoIIE would be needed for polar septum formation and subsequent activation of σ^F. The use of cell-specific sigma factors could drive the alternative cell fates of the offspring and the mother cell in Epulopiscium, allowing for the growth of the offspring while the mother cell initially supports offspring growth but eventually progresses toward apoptosis (50). Further investigation of these potential transcription programs may indicate where the line is drawn between intracellular offspring formation and sporulation.

The sequence of the spoIIE gene from Epulopiscium plus 269 bases upstream of the predicted start codon was submitted to GenBank under accession number HQ149097.