lines and bowl affect the specification of cyst stem cells and niche cells in the Drosophila testis - PubMed (original) (raw)
lines and bowl affect the specification of cyst stem cells and niche cells in the Drosophila testis
Stephen Dinardo et al. Development. 2011 May.
Abstract
To function properly, tissue-specific stem cells must reside in a niche. The Drosophila testis niche is one of few niches studied in vivo. Here, a single niche, comprising ten hub cells, maintains both germline stem cells (GSC) and somatic stem cells (CySC). Here, we show that lines is an essential CySC factor. Surprisingly, lines-depleted CySCs adopted several characteristics of hub cells, including the recruitment of new CySCs. This led us to examine the developmental relationship between CySCs and hub cells. In contrast to a previous report, we did not observe significant conversion of steady-state CySC progeny to hub fate. However, we found that these two cell types derive from a common precursor pool during gonadogenesis. Furthermore, lines mutant embryos exhibited gonads containing excess hub cells, indicating that lines represses hub cell fate during gonadogenesis. In many tissues, lines acts antagonistically to bowl, and we found that this is true for hub specification, establishing bowl as a positively acting factor in the development of the testis niche.
Figures
Fig. 1.
lines mutant CySCs initially proliferate and form aggregates. (A-A″) A control CySC clone, 2 days post-induction. The clone is marked by GFP expression (A′, arrow) among the first tier of somatic cells (Tj; A″) adjacent to the hub (arrowhead, A″). One of several GFP+ progeny of this CySC is indicated (arrowhead, A′). (B-B″) A lines g2 mutant clone, 2 days post-induction. A GFP+ cell (arrow, B′) is adjacent to the hub (arrowhead, B″). This mutant CySC has generated about eight progeny since inception (GFP+ cells, bracket, B′); mutant cells remain associated, and they tend to express lowered levels of Tj (B″, bracket). (C-C″) A lines g2 mutant clone, 6 days post-induction. The aggregate is larger (bracket, C′), and the repression of Tj more complete (C″). The arrowhead in C″ marks the hub. This testis also contained a phenotypically normal germline clone, a portion of which is indicated (asterisks, C′). Scale bar: 25 μm.
Fig. 2.
lines mutant aggregates express markers of hub fate. (A,A′) lines13B mutant cells (arrows) do not accumulate Eya (A′; a clone at day 5). Wild-type somatic cells directly adjacent to the aggregates also do not accumulate Eya (arrowheads; see Fig. 6 and text). (B-C′) linesg2 mutant aggregates (arrows) express Hh-lacZ (B′) and Cactus (C,C′; clones at days 4-6). (B) An example where both the endogenous hub (asterisk) and the aggregates were in the same focal plane (shown in inset). Each of these examples contains two separate aggregates; we do not know whether that reflects the initial induction of two lines mutant CySCs. Scale bar: 25 μm.
Fig. 3.
lines mutant aggregates do not recruit new GSCs. Control (A-A‴) and lines g2 (B-B‴) clones analyzed 5-7 days after induction. (A,A′) STAT accumulates among first tier germline (arrow, GSC) and somatic (arrowhead, CySCs) cells. (A″) GSCs contain dot fusomes (arrow). Arrowhead indicates a CySC. (A‴) There are several marked CySCs (e.g. arrowhead) and their descendants. (B,B′) Testes with _lines_13B or linesg2 (shown) clones exhibit STAT enrichment (arrow) around the endogenous hub. Arrow marks the position of a GSC in B-B″. We also observe a haze of accumulation (B′, bracket) across regions of the testis containing lines mutant aggregates (arrowheads in B,B′,B‴). However, in wild type there is variable accumulation of STAT protein in somatic cells away from the hub (data not shown) as well as within differentiating germ cells (A; note spermatocyte nucleoli, along right-hand side). Whether this is cross-reaction or uncharacterized STAT activation is unclear, but it precludes us from interpreting the haze of STAT accumulation in testes bearing lines− clones. (B″) Although normal GSCs exhibited a dot fusome (arrow), we have never observed dot fusomes in cells around the distal arc of _lines_-aggregates (B‴). This testis also contains a phenotypically normal germline clone (asterisk). Scale bar: 25 μm.
Fig. 4.
lines mutant aggregates are associated with Zfh1-expressing cells. (A,A′) Control clone analyzed 5-7 days after induction. Zfh1 accumulates among CySCs (two are indicated by arrows) flanking the hub (asterisk; DE-Cadherin, white) and decays thereafter. This testis also contained a marked GSC clone (not in focus; two spermatocyte descendants are visible, upper right). (B,C) Testis bearing two linesg2 aggregates, analyzed 5-7 days after induction. B was taken at a hub focal plane (asterisk; DE-Cadherin, white), while C was taken 15 μm deeper, with hub now out of focus (marked by parentheses). (B′,B″,C′,C″) Examples of _lines_− cells expressing Zfh1 and marked by GFP (arrowheads), confirming the somatic derivation of the mutant cells. However, far from endogenous hub there are also Zfh1-expressing cells flanking the aggregates (arrows in B-C″; bracket in C-C″). Scale bar: 25 μm.
Fig. 5.
Aggregates recruit wild-type cells that express Wg and are cycling. (A,A′) Arrows indicate Wg puncta (A′) found in GFP-negative (A), and therefore heterozygous, wild-type cells associated with lines mutant aggregates. The endogenous hub is not in this focal plane. Clones were analyzed 4-6 days after induction. (B-B‴) A testis bearing a lines g2 clone was pulse-labeled in vitro for 30 minutes 7 days post-induction. GFP-negative (B′), BrdU-positive (B″), Tj-positive (B‴) somatic cells are adjacent to the aggregate (arrows; for quantitation, see Table 1). The hub is in a different focal plane (B, asterisk). This testis also contains gonial cells in S-phase (upper right, B″). The lines mutant cells have withdrawn from the cell cycle, as we consistently observed no BrdU-positive GFP-positive cells. Why they arrest proliferation is unclear. Scale bar: 25 μm.
Fig. 6.
The recruited somatic cells chase into differentiating Eya+ cyst cells. (A-B″) An in vivo EdU pulse was analyzed 2 days after linesg2 clones were induced (A-A″), and after a further 2-day chase (B-B″). (A,A′) An EdU-labeled cell (white, arrow) is adjacent to distal side of GFP+ _lines_− aggregate. (A″) The EdU-positive cell is Eya negative. The endogenous hub is towards the left, out of focus. Other EdU+ cells in this image are germ cells. (B,B′) The EdU-labeled cell (white, arrows) is adjacent to distal side of GFP+ _lines_− aggregate. (B″) The EdU-positive cell is Eya positive. The recruited CySC that must have generated this EdU+ Eya+ cyst cell is no longer expected to be visible, as it would have diluted its EdU label through division. The endogenous hub is towards the left, out of focus. The other EdU signal (B) is in a germ cell. Quantitation can be found in Table 2. Scale bar: 25 μm.
Fig. 7.
CySCs rarely generate hub cells. (A,B) lacZ lineage tracing. (A) Two days after induction, several CySCs (arrows) and GSCs (arrowheads) are marked (hub, asterisk, DE-Cadherin). (B,B′) Six days after induction, differentiated cyst cells continue to express lacZ (arrows). No hub cells (asterisk, DE-Cadherin in B) are marked by lacZ (B′). (C) GFP driven by FngGAL4 accumulates in early tier somatic cells (arrows indicate some CySCs) and early cyst cells (one marked by an arrowhead). Asterisk indicates hub. (D,D′) Lineage tracing using Fng-GAL4 UAS-FLP Act5cFRTstopFRTGAL4. When Fng-expressing cells were indelibly marked, GFP accumulation is seen in the cytoplasm of cyst cells accompanying transit-amplifying spermatogonia and spermatocytes (arrows indicate a few of the labeled, late-stage cyst cells). No hub cells (D, asterisk, FascIII) are marked by GFP (D′). Scale bar: 25 μm.
Fig. 8.
Hub and CySCs derive from a common precursor pool. PS 11 SGPs were marked using Prd-GAL4, which is expressed in PS 11 mesodermal cells, to Flip out the transcriptional stop within Act5cFRTstopFRTGAL4; testes were analyzed from young adults. (A-A‴) Of two adjacent CySCs (white arrow and arrowhead, A,A′,A‴), one is marked by GFP while one is not (A‴; other CySCs in this focal plane are GFP+). Of two adjacent hub cells (yellow arrow and arrowhead, A,A′,A‴), one is marked by GFP (yellow arrow, A‴), while the other is not (yellow arrowhead, A‴). Similar data were obtained using Zfh1 as a CySC marker (data not shown). Scale bar: 25 μm.
Fig. 9.
lines mutants contain excess hub cells, while bowl mutants contain fewer. (A,B) Gonads from wild type (A; lines/CyO) and lines2f mutants (B). (A′,A″,B′,B″) Hub cells (Filamin, A′,B′), germ cells (Vasa, A″,B″), DNA (merge panels A, B). Hub cell number is increased in this focal plane (see Fig. S3 in the supplementary material for quantitation). Scale bar: 10 μm. (C) bowl mutants. Consistent with the reduction in hub cell number, there are significantly fewer STAT+ germline cells in bowl mutants; four other bowl− gonads exhibited no detectable STAT. Data are absolute values.
References
- Aboïm A. N. (1945). Dévelopement embryonnaire et postembryonnaire des gonades normales et agamétiques de Drosophila melanogaster. Rev. Suisse Zool. 52, 53-154
- Barroca V., Lassalle B., Coureuil M., Louis J. P., Le Page F., Testart J., Allemand I., Riou L., Fouchet P. (2009). Mouse differentiating spermatogonia can generate germinal stem cells in vivo. Nat. Cell Biol. 11, 190-196 - PubMed
- Bendall S. C., Stewart M. H., Menendez P., George D., Vijayaragavan K., Werbowetski-Ogilvie T., Ramos-Mejia V., Rouleau A., Yang J., Bosse M., et al. (2007). IGF and FGF cooperatively establish the regulatory stem cell niche of pluripotent human cells in vitro. Nature 448, 1015-1021 - PubMed
- Bokor P., DiNardo S. (1996). The roles of hedgehog, wingless and lines in patterning the dorsal epidermis in Drosophila. Development 122, 1083-1092 - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources
Molecular Biology Databases
Research Materials