Previous Article | Next Article ![]()
Eukaryotic Cell, October 2004, p. 1241-1248, Vol. 3, No. 5
1535-9778/04/$08.00+0 DOI: 10.1128/EC.3.5.1241-1248.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Department of Biology, Graduate School of Science, Osaka University, Toyonaka, Osaka,1 Institute of Biological Sciences, University of Tsukuba, Tsukuba, Ibaraki Japan,3 Cell and Developmental Biology, Division of Biological Sciences, University of California San Diego, La Jolla, California2
Received 19 May 2004/ Accepted 7 July 2004
|
|
|---|
|
|
|---|
The last step in the biosynthesis of differentiation-inducing factor 1 (DIF-1) involves a methyltransferase. Mutant strains lacking this enzyme, DMT1, have been isolated and shown to accumulate no measurable DIF-1 (22). Nevertheless, these dmtA mutant cells differentiate into prestalk and prespore cells that sort out to the anterior and posterior of slugs, just as they do in wild-type slugs, and the mutant cells form fairly normal fruiting bodies complete with spores and stalk cells. However, a genetic marker for a subset of prestalk cells defined as being at the posterior of the prestalk region, the PST-O cells, was not expressed in DIF-less slugs, and there was an increase in the number of prespore cells (22). Expression of the marker construct, ecmO::gal, could be induced by codeveloping dmtA mutant cells with wild-type cells or adding exogenous DIF-1. These results led to the suggestion that, while not essential for all prestalk differentiations, DIF-1 is required for differentiation of PST-O cells (22).
A genome-wide microarray study subsequently uncovered a large number of prestalk-specific genes, and in situ hybridization showed that 30 of these were preferentially expressed in PST-O cells (14). While 18 of these PST-O genes were significantly reduced in slugs of the dmtA mutant strain, the remaining 12 genes were strongly expressed in PST-O cells. Thus, it appears that some aspects of PST-O differentiation are DIF independent. Our results indicate that analysis of only one gene is insufficient for understanding the in vivo function of DIF-1.
Previously, 79 cell type-specific genes that are preferentially expressed in one or another of the three prestalk subdomains of a slug (PST-A, PST-AB, and PST-O) based on their position were described (14). We can now add 54 prespore genes recognized by microarray expression studies and confirmed by in situ hybridizations. We used these cell type-specific genes for in situ hybridizations to compare wild-type and dmtA mutant slugs. In the absence of DIF-1 the prespore region at the posterior expanded from 80 to 90% of the length, with a compensatory shrinkage of the anterior prestalk region to 10% of the length of the slugs. While each prestalk subdomain was proportionally reduced, their order along the axis was retained, suggesting that their proportions are determined by factors other than DIF-1. Furthermore, we found that microsurgically isolated prestalk regions from dmtA mutant slugs were able to regulate within 6 h such that a prespore gene was expressed in the majority of the cells at the posterior, suggesting that a field-wide inhibitor of prespore differentiation other than DIF-1 emanates from the prespore domain.
|
|
|---|
Whole-mount in situ hybridization. Whole-mount in situ hybridization was performed as previously described (14). Our in situ studies were carried out with whole slugs or prestalk isolates of strain AX2 and dmtA mutant strain HM1030, which is defective in DIF-1 production (21). After fixation, samples from each stage were mixed and hybridized in a single tube or a dish. Digoxigenin- or fluorescein-labeled riboprobes were generated for the entire length of each cDNA clone for hybridization to either sense or antisense RNA.
For double staining, digoxigenin- and fluorescein-labeled riboprobes were used for hybridization at 42°C overnight. After rinsing and then blocking, samples were incubated with Alexa 488-labeled anti-fluorescein antibody (diluted 1/500) and rhodamine-labeled anti-digoxigenin antibody (diluted 1/500) (Roche Diagnostics, Basel, Switzerland) at 4°C overnight. After washing and blocking, signals were detected with an Olympus Epifluorescent microscope BX50.
Sequences of all cDNAs used in this study can be drawn from the Dicty-cDNA database (http://www.csm.biol.tsukuba.ac.jp/cDNAproject.html).
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Prespore genes
|
![]() View larger version (33K): [in a new window] |
FIG. 1. Expression patterns of prespore genes. RNAs prepared at 2-h intervals throughout development on filters were analyzed on microarrays. Values were normalized to 1 at the time of initiation of development, and the fold changes are presented for the 54 prespore genes analyzed in this study. Values are the averages of four independent determinations. Genes showing the greatest changes are indicated by name.
|
![]() View larger version (108K): [in a new window] |
FIG. 2. In situ hybridization with prespore genes. The pattern of staining of posterior cells by SLB126 (cotC) for mRNA (a to d) or antisense RNA (a' to d') is representative of almost all of the prespore genes. Patterns for each of the 54 prespore genes can be inspected at http://www.csm.biol.tsukuba.ac.jp/ tools/bin/ISH/index.html. Only SSH707 and two other cDNAs derived from the same gene, which encodes a protein with no significant homologs, gave a significantly different pattern (e to h'). Up to the culmination stage its mRNA accumulated in posterior cells just like that of other prespore genes; however, its antisense RNA was seen in anterior cells (f'). During culmination, SSH704 mRNA disappeared from the prespore cells and both mRNA and antisense RNA accumulated in cells at the back of the prestalk domain.
|
![]() View larger version (84K): [in a new window] |
FIG. 3. Prespore regions in slugs of wild-type and DIF-less slugs. Expression of three prespore genes was analyzed by in situ hybridization on more than 20 slugs prepared from wild-type Ax2 and dmtA mutant strains. Proportional lengths were measured, and the ratio and standard deviations were calculated for each gene by NIH image.
|
|
View this table: [in a new window] |
TABLE 2. Expression ranges of cell type-specific genes in slugs
|
![]() View larger version (45K): [in a new window] |
FIG. 4. Double-staining in situ hybridization. (A) Slugs formed from wild-type Ax2 and dmtA mutant strains were hybridized with prestalk probes to generate a green signal and were hybridized simultaneously with prespore probe SSJ770 (pspA) to generate a red signal. (a to d') SLE474 (ecmA); (e to h') SSA854; (i to l') SSM184. (a to l) wild-type Ax2; (a' to l') dmtA mutant strain. (a, a', e, e', i, and i') Prestalk gene expression revealed by the Alexa 488 signal. (b, b', f, f', j, j') Expression of SSJ770 (pspA) revealed by the rhodamine signal. (c, c', g, g', k, k') Merged pictures. (d, d', h, h', l, l') Enlarged slug anteriors. The stained regions did not overlap and were separated by unstained cells in the case of SSA854. (B) Simultaneous analysis of spatial expression patterns of PST-AO and PST-O genes. Slugs formed from wild-type Ax2 (a to c) and dmtA mutant strain (a' to c'). Expression of SSM184 (a PST-O gene) was detected with rhodamine (a), while expression of SLE474 (ecmA, a PST-AO gene) in Ax2 slugs was detected with Alexa 488 (b). (c) Merged picture of the images shown in panels a and b. In dmtA mutant slugs, SSM184 was detected with Alexa 488 (a'), while SLE474 was detected with rhodamine (b'). (c') Merged picture of the images shown in panels a' and b'. Regions with cells expressing both the PST-O and the PST-AO genes appear yellow in the merged pictures.
|
To quantitate the change in proportions of prestalk cell types resulting from the lack of ability to synthesize DIF-1, we measured the relative lengths of regions stained with three PST-AO- and two PST-O-specific genes in 20 wild-type and 20 dmtA mutant slugs (Fig. 5). In each case the stained regions were reduced about twofold in the DIF-less slugs relative to the length of the region in wild-type slugs (Table 2). Staining of structures at later stages in development showed that these genes continued to be expressed in these prestalk domains through culmination of both wild-type and dmtA mutant strains. Likewise, PST-A specific genes (SLF308 and SSK861) stained the most anterior of slugs and the top of culminants, while PST-AB genes such as SSK348 and SLA128 stained cells in the anterior funnel of slugs and the stalks in both wild-type and dmtA mutant strains (data not shown). While the lack of DIF-1 reduced the proportion of PST-A and PST-O cells, it did not result in the mislocalization or absence of any of the prestalk cell types (Table 2).
![]() View larger version (53K): [in a new window] |
FIG. 5. Prestalk regions in wild-type and DIF-less slugs. Expression of three PST-O genes and two PST-O genes was analyzed by in situ hybridization of 20 slugs prepared from wild-type Ax2 and dmtA mutant strains. Proportional lengths were measured, and the ratio and standard deviations were calculated for each gene. The PST-O genes (SSH630 and SSM184) are not expressed in the most anterior cells. The anterior as well as the posterior borders were measured for these genes.
|
![]() View larger version (74K): [in a new window] |
FIG. 6. Cell type conversion in isolated prestalk fragments. The anterior 20% of wild-type Ax2 slugs and the anterior 10% of dmtA mutant slugs were dissected and incubated in isolation. At the indicated times (in hours) after removal of the posterior tissue, the prespore marker SSJ770 (pspA) was used for in situ hybridization. Cell type conversion was apparent in the back of the fragments within a few hours.
|
|
|
|---|
Many years ago it was shown that the proportion of spores and stalk cells was invariant in fruiting bodies of widely different sizes (1). The process which results in this surprising result appears to occur much earlier during the slug stage. In situ hybridization of slugs that varied more than 10-fold in size with the prespore marker cotC showed that the proportions of prespore cells was essentially the same in large and small slugs (13). A mechanism that could account for size invariance has been proposed in which prespore cells produce an inhibitor to which they themselves are insensitive but which blocks prestalk cells from differentiating into prespore cells (20). If the inhibitor is broken down or inactivated by prestalk cells, this mechanism can robustly establish a constant proportion of the cell types over a wide range in the total cell number. DIF-1 was initially a candidate for the role of such an inhibitor, but the phenotype of strains unable to synthesize DIF-1 due to disruption of the enzyme DMT, which catalyzes the last step in DIF-1 biosynthesis, appeared to rule it out, because they expressed some prestalk genes in the anterior of slugs and were able to make fairly normal stalks (22). These dmtA mutant cells were shown to accumulate less than 5% as much DIF-1 as wild-type cells, below the limit of detection, but it is possible that there is some residual DIF-1 which functions in prestalk differentiation. The fact that slugs of a dmtA mutant strain do not express a PST-O marker construct and have an increased number of prespore cells indicates that DIF-1 plays an essential role in differentiation of PST-O cells (22). We have confirmed that certain PST-O genes, such as SSD764 (Table 2), are not expressed in the absence of DIF-1. However, other genes expressed in cells just anterior to prespore cells (PST-O cells) are expressed normally in DIF-less slugs of the dmtA mutant strain (Fig. 4 and 5). By quantitating the proportion of PST-A, PST-O, and prespore regions in wild-type and dmtA mutant slugs, we found that DIF-1 plays a significant role is regulating the proportion of the cell types, because the prestalk domain in wild-type slugs is double that in dmtA mutant cells (Table 2). However, DIF-1 does not appear to be the only signal involved in proportioning the cell types, because the anterior 10% of dmtA mutant slugs still express prestalk genes and do not express prespore genes. Moreover, the order of the prestalk subtypes recognized by patterns of expression of different prestalk genes is not affected by the absence of DIF-1. By quantitating the expression domains of a large number of prestalk genes, we could recognize at least five classes on the basis of their posterior boundaries (Table 2 and unpublished data). Subdivision of cells of the prestalk domain into PST-A and PST-O cells was based solely on expression patterns seen with portions of the ecmA regulatory region (9) and may be an oversimplification. A DIF-independent mechanism appears to pattern the prestalk domain into a series of distinct regions that can be distinguished by their unique transcriptional profiles.
When culmination occurs in anterior fragments soon after they are separated from the cells at the back, long stalks are made that carry almost no spores; however, if the isolated anterior fragments migrate for more than 6 h before culminating, the resulting fruiting bodies carry a considerable number of spores (16). These results indicated that prestalk cells had to regulate to be able to undergo terminal differentiation into spores. We were able to directly observe regulation in isolated prestalk fragments by in situ hybridization to a prespore gene (pspA). Initially, very few cells carried this mRNA, but the number increased within the first 3 h following microsurgical isolation. By 9 h most of the cells in the posterior 80% of the newly made small slugs had expressed the prespore gene. Assuming there was little or no axial mixing of the cells, it is of interest that the first cells to regulate into prespore cells appeared at the posterior, suggesting that cell type determination in these cells is more labile than that in ones closer to the front. The model that can account for size-invariant proportioning can also account for regulation following perturbation of the cell type proportions (20). When prespore cells are excised, the source of the inhibitor of prespore differentiation is removed while breakdown of the inhibitor can proceed. As cells at the back of the isolated prestalk fragments regulate to become prespore cells they start to produce the inhibitor, which increases until a threshold is reached that precludes the regulation of further prestalk cells. The nature of the proposed inhibitor is not known, but it does not appear to be DIF-1, because regulation was also observed in dmtA mutant anterior fragments. However, like DIF-1, it appears to be produced by prespore cells, because regulation among prestalk cells occurs only after the prespore cells are removed.
A DIF-insensitive mutant was recently characterized and shown to carry a disruption of the dimA gene, which encodes a DNA binding protein of the bZIP family (21). This mutant strain exhibits all of the phenotypes of dmtA mutant null cells, except that it produces DIF-1 and is not responsive to the addition of DIF-1. DimA appears to be responsible for optimal expression of a prestalk gene marker (ecmA) and repression of a prespore gene marker (pspA) in PST-O cells which may partially account for the mutually exclusive patterns of gene expression in prespore and prestalk cells. However, we have shown that a considerable number of prestalk-specific genes are expressed in a DIF-independent manner, indicating that other processes are involved in the initial divergence of cell types (14, 18).
This study was supported by grants from Research for the Future of the Japan Society for the Promotion of Science (JSPS-RFTF96L00105) and grants in aid for scientific research in priority areas "Genome Biology" from MEXT of Japan (12206001) to M.M. and to Y.T. and a grant from the National Institutes of Health (GM62350) to W.F.L.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»