Stem cell niche exit in C. elegans via orientation and segregation of daughter cells by a cryptic cell outside the niche

Version of Record

Accepted for publication after peer review and revision.

Download
Cite
Share
CommentOpen annotations (there are currently 0 annotations on this page).

Version of Record published: September 2, 2020 (This version)
Accepted Manuscript published: July 21, 2020 (Go to version)
Accepted: July 17, 2020
Received: February 25, 2020

1. Built upon by
Innexin function dictates the spatial relationship between distal somatic cells in the Caenorhabditis elegans gonad without impacting the germline stem cell pool

Theadora Tolkin, Ariz Mohammad ... David Greenstein

Research Advance Sep 13, 2022
Further reading

Abstract
eLife digest
Introduction
Results
Discussion
Materials and methods
Appendix 1
Data availability
References
Decision letter
Author response
Article and author information
Metrics

Abstract

Stem cells reside in and rely upon their niche to maintain stemness but must balance self-renewal with the production of daughters that leave the niche to differentiate. We discovered a mechanism of stem cell niche exit in the canonical C. elegans distal tip cell (DTC) germ stem cell niche mediated by previously unobserved, thin, membranous protrusions of the adjacent somatic gonad cell pair (Sh1). A disproportionate number of germ cell divisions were observed at the DTC-Sh1 interface. Stem-like and differentiating cell fates segregated across this boundary. Spindles polarized, pairs of daughter cells oriented between the DTC and Sh1, and Sh1 grew over the Sh1-facing daughter. Impeding Sh1 growth by RNAi to cofilin and Arp2/3 perturbed the DTC-Sh1 interface, reduced germ cell proliferation, and shifted a differentiation marker. Because Sh1 membrane protrusions eluded detection for decades, it is possible that similar structures actively regulate niche exit in other systems.

eLife digest

Stem cells have the rare ability to divide and specialize into the many different types of cells necessary for an organism to survive. For instance, germ stem cells can multiply to produce precursor cells that go on to become eggs or sperm needed for reproduction.

When a stem cell divides, the daughter cells can either remain ‘naïve’, or start to specialize into a given cell type. In many cases, this decision is strongly influenced by the properties of the environment that surrounds the stem cell. However, in the microscopic worm Caenorhabditis elegans, how the daughters of germ stem cells specialize was thought to be a random process, with nearby cells equally likely to specialize or remain naïve.

In this animal, germ stem cells reside in tube-shaped structures called gonads, which are enclosed by a large ‘distal tip’ cell. In addition, cells known as Sh1 surround the gonad. Here, Gordon et al. tracked dividing germ stem cells in the gonads of live worms. This revealed that both the distal tip cell and Sh1 cells have finger-like extensions that form contacts with the germ stem cells. The fate of dividing germ stem cells is shaped by these interactions. If they touch only the distal tip cell, they remain in a naïve state. However, if they contact both the distal tip cell and an Sh1 cell, one daughter of the stem cell becomes an egg precursor – with the daughter closest to the distal tip cell staying naïve. In fact, germ stem cells that are prevented from contacting Sh1 cells divide less often.

Many other types of stem cells, for example in human skin, are believed to make the decision to remain naïve or undergo specialization randomly. The results from Gordon et al. could provide a roadmap to discover hidden layers of control in other organisms, some of which may be potentially relevant in health and disease.

Introduction

Stem cells and their associated supportive niche cells are centrally important to development (Murry and Keller, 2008; Zhu and Huangfu, 2013), homeostatic tissue maintenance (Biteau et al., 2011; Blanpain and Fuchs, 2009; Snippert et al., 2010; Wilson et al., 2008), aging (Rao and Mattson, 2001), regeneration (Sánchez Alvarado and Yamanaka, 2014), cancer (Reya et al., 2001), and tissue engineering (Ruder et al., 2011). For this reason, the ways in which a niche anchors and signals to its stem cells is an area of active inquiry. Balance must be struck between retaining a sufficient pool of stem cells in the niche and generating a robust population of stem cell descendants that go on to differentiate. Therefore, the question of how stem cell progeny leaves the niche in a regulated fashion is likewise important, but has been less intensely studied. Most of what is known about stem cell niche exit is understood in the context of how the niche retains the self-renewing stem cell daughter rather than how the differentiating daughter escapes the influence of the niche. Examples include stochastic displacement in the mammalian intestine (van der Flier and Clevers, 2009), migration in the larval tracheal system of Drosophila (Chen and Krasnow, 2014), oriented division to a basal lamina in the mammalian epidermis (Poulson and Lechler, 2010), and oriented division to the niche cells in the Drosophila ovary (Casanueva and Ferguson, 2004).

The C. elegans germ line is supported by a canonical stem cell niche (Hubbard, 2007; Lander et al., 2012) called the distal tip cell (DTC). Because of the ease of visualization and experimental manipulation, many general principles have been obtained by analysis of this simple system, including the first demonstration of stem cell niche properties (Kimble and White, 1981). The genetics controlling stemness and differentiation are very well understood (Hubbard, 2007; Kimble and Seidel, 2013), and a cell biological understanding of the system is growing (Amini et al., 2014; Byrd et al., 2014; Linden et al., 2017). The adult DTC has a jellyfish-like appearance, with a flattened cell body at the distal end and long trailing processes that extend proximally and enwrap germ cells, including the presumptive stem cells (Byrd et al., 2014; Crittenden et al., 2006; Gordon et al., 2019). The germ line is partially syncytial, with membrane-bound germ cell bodies connected to a common cytoplasmic core (the rachis) by narrow bridges of cytoplasm (Hirsh et al., 1976; Seidel et al., 2018). Despite the cytoplasmic connections that facilitate the sharing of intracellular fate determinants (Lee et al., 2016), the germ line segregates cell fates across its distal-proximal axis, with germ cells undergoing meiosis proximal to the undifferentiated germ cells dividing stochastically in the distal ‘progenitor zone’. The progenitor zone is approximately 20 germ cell diameters long (~100 μm) and contains 243 + / - 25 cells in one-day adult animals (Crittenden et al., 2006). A subset of the progenitor zone germ cells makes up the germ stem cell pool. The DTC niche expresses the Notch ligands LAG-2 and APX-1 that activate Notch signaling in the germ stem cells (Henderson et al., 1994; Nadarajan et al., 2009). It has been hypothesized that divisions within the stem cell population simply push daughters out of the niche to eventually differentiate (Rosu and Cohen-Fix, 2017), however stem cell progeny breaking contact with the niche have not been visualized. Previous work by our group (Linden et al., 2017) suggests that a simple distal-to-proximal model of stem cell position does not take into account the effect of DTC geometry on Notch activation. When the DTC is asymmetrically shaped, only the germ cells closest to it express a Notch reporter, and other equally distal germ cells lack reporter expression, suggesting that close proximity to the DTC rather than distal position defines stem cells (Linden et al., 2017). Thus, while downstream effects of localized Notch signaling on germ cell stemness vs. differentiation are well understood, how the niche-stem cell association is organized and how it terminates and releases germ cells to differentiate given the complex and varied niche geometry is not known.

Proximal to the DTC, the remainder of the somatic gonad comprises five pairs of gonadal sheath cells that lie between the germ cells and the gonadal basement membrane (Hall et al., 1999). Sheath cell pair 1 (Sh1) covers germ cells in the distal gonad, with one cell on its superficial surface closer to the cuticle and the other on the deep surface closer to the gut (Hall et al., 1999). Between the distal boundary of Sh1 and the DTC, a ~ 20 germ cell diameter-long (~100 μm) ‘germ line bare region’ has been described, in which germ cells are thought to contact the gonadal basement membrane directly, not somatic cells. This germ line bare region makes up the entire progenitor zone outside of the distalmost cells under the DTC. Evidence for the bare region comes from scanning EM of dissected gonads and cytoplasmic GFP expressed in the gonadal sheath (Hall et al., 1999). How the differentiating germ cells navigate the putative somatic cell void between the DTC and Sh1 cells is unclear.

We set out to investigate the process by which germ cells break their contacts with the DTC niche and interact with the other cells of the somatic gonad. We used CRISPR/Cas9 genome editing, live cell time-lapse imaging, and RNAi-mediated gene knockdown to probe the relationships among germ cells, the DTC, and Sh1 sheath cells. Unexpectedly, we discovered that thin membranous processes from the Sh1 cells extend distally and intercalate among undifferentiated germ cells and the processes of the DTC, with soma contacting nearly every germ cell. There is no germ cell bare region. As a result, germ cell divisions take place in three anatomical compartments—in contact with the DTC niche alone, at the DTC-Sh1 boundary, and in contact with Sh1 alone. Divisions on the elaborate interface between the DTC and Sh1 are more frequent and are polarized, with daughters segregating between them, suggesting a mechanism for germ cell niche exit via the first documented asymmetric cell divisions in the C. elegans germ line. Sh1 actively grows over and between daughter cells that divide at the interface in an Arp2/3- and cofilin-dependent manner, which strengthens the association between one daughter cell and the sheath. When Sh1 was impeded from growing towards the DTC by RNAi against actin regulators, the dividing populations at the boundary and under Sh1 were reduced and a marker of differentiation was shifted in the same direction as the Sh1 cell, suggesting that distal Sh1 cell processes assist stem cell niche exit. We conclude that Sh1, a cell neighboring the niche, is an overlooked regulator of adult stem cell niche dynamics.

Results

The stem cell niche is closely apposed to Sh1

To investigate how the elaborate DTC germ line stem cell niche interacts with other cells in the gonad we used CRISPR/Cas9-mediated genome editing (Dickinson et al., 2015; Dickinson et al., 2013) to endogenously tag cell membrane-localized gap junction proteins (INX-8 and INX-9) that were previously found to be expressed in the DTC and the gonadal sheath cells (Starich et al., 2014). The inx-8 and inx-9 genes are genetically redundant with both single mutants fertile and the double mutant sterile (Starich et al., 2014). As innexins have not been endogenously tagged with genetically encoded fluorophores previously, we used this inx-9(ok1502) mutant allele to ascertain function of the inx-8 genome-edited alleles. We found that a homozygous C-terminal-tagged inx-8::mCherry in an inx-9(ok1502) mutant background was sterile. However, a homozygous N-terminal-tagged mKate2::inx-8(qy102) in this mutant background was fertile (Figure 1—figure supplement 1A). We used the sequence similarity between the genes to tag the inx-9 locus in the same N-terminal position. The atomic structure of another C. elegans innexin, INX-6, has been determined (Oshima et al., 2016); the N-terminus of C. elegans INX-6 is extracellular, flexible, and located in the pore of the gap junction. If INX-8 and INX-9 have the same structure, this means the fluorescent tags (from multiple monomers in a hexameric hemichannel) are linked to extracellular extensions from the pore of the gap junctions formed by tagged INX-8 or tagged INX-9.

Upon coexpression of each of these tagged innexins with a DTC membrane marker and examination by live-animal confocal microscopy, we observed expression in the DTC and in the gonadal sheath. Unexpectedly, we found that the long cellular processes of the DTC niche cell were directly bordered by membranous extensions from Sh1 (Figure 1A and B), contrary to the accepted model of gonad anatomy with a bare region separating DTC and Sh1 (Figure 1A, Lints and Hall, 2018). We verified this pattern with another endogenously tagged protein that localizes to the membrane, integrin α subunit INA-1::mNeonGreen (Jayadev et al., 2019; Figure 1—figure supplement 1B) and by crossing our mKate::INX-8 strain to a strain expressing cytoplasmic GFP (tnIs6(lim-7p::GFP) [Hall et al., 1999]) to reveal that the Sh1 membrane protrusions can extend farther distal than the brightest cytoplasmic GFP signal in the cell (Figure 1—figure supplement 1C). A revised model positions the DTC and Sh1 intercalating among the distal germ cells in the stem cell zone (Figure 1C). Unlike other cells in C. elegans, including even elaborately branching neurons (Chisholm et al., 2016; Smith et al., 2010; Zou et al., 2015), the extensive DTC-Sh1 interface is variable among individuals (Figure 1—figure supplement 1D and Figure 1—figure supplement 1E) and uneven around the circumference of the gonad (Figure 1D shows superficial and deep Sh1 cells above and below the germ line, respectively). Given its juxtaposition with the DTC, we reasoned that the cryptic neighboring Sh1 cells could be in a position to interact with the stem cell niche and potentially with the stem cells themselves.

Figure 1 with 1 supplement see all

Download asset Open asset

Sh1 is present in the distal gonad and intercalates with the processes of the DTC.

(A) Schematic depiction of the previous model of gonad morphology of *C. elegans* hermaphrodite adults with a germ cell ‘bare region’ (after [Lints and Hall, 2018]). Germ cell nuclei shown in pink, DTC in yellow, sheath in cyan (with Sh1 proximal boundary represented by line near bend). Regions of germ line shown in blow-up below, after (Crittenden et al., 2006). (B) Micrographs showing the gonad in DIC with fluorescent overlay (top) of the DTC (yellow, *lag-2p::mNeonGreen::PLC^δPH*) and Sh1 (cyan, *mKate2::inx-8*), fluorescence only (middle, projection from rachis up through surface of gonad), and DIC only (bottom, single slice through rachis). Images taken at 400x magnification, composite of two tiled fields of view. Scale bar 20 μm. (C) New model based on our observation that Sh1 and the DTC intercalate in the distal gonad. With the DTC closely bordered by an adjacent somatic Sh1 cell, the delineation of the stem cell population is unclear. DTC in yellow, Sh1 in cyan, and germ cell nuclei in pink. (D) Model of cross section of gonad, showing deep Sh1 cell in dark turquoise in addition to superficial Sh1 cell in cyan, DTC processes both superficial and deep in yellow, and germ cell membranes and nuclei in pink. Central void is the rachis. See also Figure 1—figure supplement 1.

Sh1 migrates with the DTC during gonad development

We next examined the relationship of the DTC with the Sh1 cell during development of the mature niche to determine how this elaborate interface is formed. C. elegans gonad arm formation entails long-distance postembryonic migration of the two DTCs as the germ cells proliferate behind them, with the entire gonad encased in a growing basement membrane (Sherwood and Plastino, 2018). The anterior and posterior DTCs are born from the second division of the gonad primordium cells Z1 and Z4. These migratory leader cells initiate gonad outgrowth in the L2 larval stage (21 hr post-hatching at 20°C). The DTCs migrate in opposite directions along the ventral body wall while the Z2 and Z3 germ line progenitors proliferate. During the L3 larval stage (27 hr post-hatching), the two pairs of Sh1 cells are born (Kimble and Hirsh, 1979; McCarter et al., 1997). Shortly after, the DTCs turn (30–33 hr) first to the dorsal body wall and then back to migrate to the midline where they come to rest in L4 animals (45 hr, see Figure 2A, timing after [Lints and Hall, 2018]). The cell body of Sh1 has not been observed during development, owing to its extreme thinness and uneven borders, but it is thought either to migrate over or be dragged by the growing germ line (Kimble and Hirsh, 1979). In the adult, its edge is thought to fall approximately midway down the distal branch of the gonad arm (Hall et al., 1999; Killian and Hubbard, 2005).

Figure 2 with 2 supplements see all

Download asset Open asset

The DTC and Sh1 cells are closely apposed throughout development.

(A) Schematic depiction of *C. elegans* DTC (yellow) migration path (orange arrows) during development, timing after (Lints and Hall, 2018): ventral migration begins at 21 hr post hatching; first turn at 30 hr post hatching, second turn at 33 hr post hatching, and end migration at 45 hr post hatching. (B) Time course analysis of somatic gonad elongation. DTC (yellow, *lag-2p::mNeonGreen::PLC^δPH*) and Sh1 (cyan, *mKate2::inx-8*, also detected in larval DTC). The two cells are in contact or within one germ cell diameter throughout development (white arrows). (C) New model of gonad development including Sh1 cell (only one of pair shown). (D) Schematic showing of the DTC-Sh1 boundary transformation between the end of migration and beginning of elaboration. (E) Stills from a time-lapse movie (Figure 2—video 1) of same strain shown above. Dashed white line shows position of distal Sh1 at time 0 in all subsequent panels; note remodeling of Sh1 back and forth to extend and close this distance (arrows). All scale bars 10 μm. See also Figure 2—video 1 and Figure 2—figure supplement 1.

To visualize the Sh1 cell during gonad formation, we examined mKate2::INX-8, which was expressed in the DTC and Sh1 cell membranes during gonadogenesis, along with a DTC membrane marker (Figure 2B). We found that the larval Sh1 cells were thin, elongated, irregular cells that extended away from the midline over the germ cells and reached towards the DTC with at least one long, thin projection (Figure 2B, arrows); the DTC sent at least one reciprocal projection back to touch Sh1 (n = 27/28 gonad arms have DTC and Sh1 in apparent contact). These thin projections persisted as the forming gonad made its first (n = 16/17) and second (n = 10/10) turns (Figure 2B). By the end of L4, the Sh1 came to rest with a smooth distal boundary approximately one germ cell diameter from the DTC, which typically extended one or more short, thin projections that touched or nearly touched the distal Sh1 boundary (n = 14/17 within one germ cell diameter at the end of migration). We conclude that the DTC and Sh1 share an intimate association during gonad migration in the larva (schematic Figure 2C).

During the L4/Adult Transition, the DTC and Sh1 Elaborate Dramatically

We continued our time-course analysis to determine when the DTC and Sh1 intercalate. The sharp distal boundary between the DTC and Sh1 was dramatically remodeled beginning at the L4/adult transition, coincident with the elaboration of the DTC (n = 4/4, Figure 2—figure supplement 1). The DTC transformed from its high, narrow larval cap morphology to a broad, thin, distended structure with proximal projections. As the DTC elaborated, Sh1 filled in the gaps between extending DTC processes (n = 13/14 young adults, Figure 2B and Figure 2D) instead of residing in the middle of the distal arm of the gonad as was previously thought (Hall et al., 1999; Killian and Hubbard, 2005; Korta and Hubbard, 2010). It has been hypothesized that the distal boundary of Sh1 would need to treadmill over germ cells to counter the distal-to-proximal movement of the underlying germ cells from the distal stem cell zone towards the bend (Hall et al., 1999). We found support for this hypothesis from time-lapse imaging of Sh1 and the DTC (Figure 2—video 1, Figure 2E). While the DTC was sometimes observed to stretch proximally, Sh1 almost always (n = 12/13) appeared to rapidly remodel its edges, moving both towards and away from the DTC on a scale of minutes (Figure 2—figure supplement 1). These observations suggest Sh1 is a more active cell than the DTC, which changes shape on a slower time scale (Wong et al., 2013). Time course analysis of the same cells revealed that Sh1 persisted in the distal zone well into adulthood (n = 18/18 animals 73 hr post L1 arrest, Figure 2B bottom). Taken together, these results indicate that cellular extensions from the Sh1 cells interdigitate with the processes from the DTC as the DTC elaborates and are then present in this interlaced pattern during adulthood.

Nearly every germ cell contacts the DTC, Sh1, or Both

Because Sh1 extensions are interlaced among DTC processes, we postulated that they likely interact with the actively dividing germ cells in the progenitor zone, including perhaps even germ stem cells in the DTC niche. We hypothesized that progenitor zone cells reside in four different anatomical compartments defined by their somatic cell contacts—DTC-associated, Sh1-associated, DTC-Sh1 interface, or not touching either somatic cell. Since the DTC is necessary and sufficient to maintain a mitotic germ cell population (Austin and Kimble, 1987; Kimble and White, 1981), these contacts could directly inform the regionalization of the mitotic zone into stem cells and mitotic cells that have lost their stemness.

To examine germ cell-somatic cell associations in the progenitor zone, we generated worms with fluorescent cell membrane markers of the germ cells, DTC, and mKate2::INX-8 marking Sh1. We classified the anatomical positions of germ cells in distal gonads (Figure 3A). We focused only on the superficial portion (~10 μm deep) of each gonad arm because resolution and brightness diminish when imaging deeper. Viewed at 1000x magnification, we observed an average of 70 undifferentiated germ cells in the superficial distal gonad (n = 26 distal gonad arms). The DTC alone contacted an average of 17 germ cells in this region (Figure 3B). Sh1 alone contacted an average of 40 germ cells (Figure 3B). An average of 10 germ cells were found along the DTC-Sh1 interface (Figure 3B). Finally, very few germ cells (fewer than two germ cells per animal) lacked close proximity to either somatic cell (Figure 3B). Thus, Sh1 is intimately associated with the majority of germ cells at the distal end of the gonad, and practically all germ cells in the progenitor zone contact either the DTC, Sh1, or both (Figure 3B).

Figure 3 with 2 supplements see all

Download asset Open asset

Most germ cells contact one or both somatic cells, and divisions are enriched at the DTC-Sh1 interface.

(A) Schematic depicting germ cells in four somatic cell compartments defined by contact with the DTC, Sh1, DTC-Sh1, or neither. (B) Box and whisker plot showing the number of germ cells in each somatic compartment in 26 one-day adult animals (n = 1843 germ cells counted Figure 3—source data 1.). Very few germ cells contact neither somatic cell, while a minority contact both the DTC and Sh1. (C) Example micrograph showing that all germ cell membranes (magenta, *mex-5p::GFP:: PLC^δPH*) in the distal gonad contact the soma (DTC: yellow, *lag-2p::mTag:BFP::PLC^δPH* and/or Sh1: cyan, *mKate2::inx8*) with no germ cell bare region. Scale bar 10 μm, and images are to scale with distance axes in panels D and E. (D) Distance from the distal tip of n = 25 DTC-associated divisions, n = 29 DTC-Sh1 divisions, and n = 28 Sh1-associated divisions in 38 adult animals with germ cell nuclei (*mex-5p::H2B::mCherry*) and membranes (*mex-5p::GFP:: PLC^δPH*), DTC membrane (*lag-2p::mTag:BFP::PLC^δPH*) and Sh1 membrane (*mKate2::inx8*) labeled Figure 3—source data 2. No divisions of a ‘bare’ germ cell were observed. Shaded box shows range of distances in which the DTC and Sh1 were both sometimes observed. (E) The same 82 germ cell divisions as in D, plotted by binned distance from the distal tip, and colored by somatic cell contact with DTC (yellow), DTC-Sh1 (magenta), and Sh1 (cyan). (F) Relative proportions of germ cell division to total germ cells among 25 time-lapsed cell divisions in nine animals Figure 3—source data 3. These data are concordant with proportions for all measured divisions in each compartment over all counted cells in each compartment Figure 3—source data 2. The percentage of germ cell divisions in contact with the DTC alone is proportional to the percentage of all germ cells in contact with the DTC alone (yellow, p=1.0). Divisions in contact with Sh1 alone are slightly underrepresented (cyan p=0.13), likely because some of the germ cells in contact with Sh1 are already in meiotic S phase and therefore are not able to undergo mitosis (Fox and Schedl, 2015). Divisions at the interface (pink) are overrepresented threefold relative to the percentage of DTC-Sh1 associated germ cells (prop.test, Bonferroni corrected p=0.0012327, see Materials and methods). (G) Schematic depicting progenitor zone regionalization by somatic cell contacts. The territory in which dividing germ cells are found (the progenitor zone) contains three groups of dividing cells (uncovered germ cells in 3A and 3B were not found in substantial numbers and were not observed dividing). Previous research suggests the cells directly in contact with the DTC are stem cells, but we now can resolve two groups—those contacting the DTC alone (yellow) and DTC-Sh1 interface cells (magenta). The remaining dividing cells are outside the niche and in contact with Sh1 alone (cyan). Based on frequent divisions that segregated one daughter cell under the Sh1 in DTC germ cells that divided at the DTC-Sh1 boundary, we hypothesized that DTC-Sh1 interface cells might divide asymmetrically and thereby allow stem cell daughters to leave the niche. See also Figure 3—video 1 and Figure 3—figure supplement 1.

Figure 3—source data 1 Somatic cell contacts of germ cells in superficial portions of 26 young adult hermaphrodite gonads.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig3-data1-v2.xlsx
Download elife-56383-fig3-data1-v2.xlsx
Figure 3—source data 2 Somatic cell contacts of 82 dividing cells.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig3-data2-v2.xlsx
Download elife-56383-fig3-data2-v2.xlsx
Figure 3—source data 3 Somatic cell contacts of 25 dividing cells in 9 young adult hermaphrodite gonads with germ cell nuclei and membranes labeled.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig3-data3-v2.xlsx
Download elife-56383-fig3-data3-v2.xlsx

A disproportionate number of divisions occur at the DTC-Sh1 interface

Germ cell division dynamics are nonuniform across the progenitor zone, with a peak in the incidence of divisions between 20–40 μm from the distal tip (Hansen et al., 2004; Maciejowski et al., 2006). Notably, this is where the DTC-Sh1 interface falls (Figure 3C and D), so we next investigated germ cell divisions in the context of their somatic cell compartments.

We used live in-animal imaging to measure both the distance from the distal tip of the gonad and the somatic cell contacts of 82 cell divisions in 40 young adult animals in a worm strain in which the DTC, Sh1, germ cell membranes, and germ cell nuclear histones were fluorescently highlighted. The absolute positions of DTC-associated, Sh1-associated, and DTC-Sh1 interface cell divisions overlap in the distal 10–50 μm of the gonad (Figure 3D, shaded box), reflecting the variable, uneven, intercalating boundary between these somatic cells. While a minority of germ cells lie on the DTC-Sh1 interface (fewer than 15% of all cells, Figure 3B), these cells made roughly one third of total germ cell divisions observed (n = 29/82, Figure 3E, pink). Another third occurred in DTC-associated germ cells (n = 25/82, Figure 3E, yellow), and a third occurred in Sh1-associated germ cells (n = 28/82, Figure 3E, cyan). The distance distribution of these divisions roughly recapitulates the mitotic index peak pattern previously observed between ~20–40 μm from the distal end (Hansen et al., 2004; Maciejowski et al., 2006), and the extent of Notch target transcription among distal germ cells, with no transcripts being made beyond 40 μm from the distal end (Lee et al., 2019; Lee et al., 2016). This correspondence suggests that the stem cell zone extends to the DTC-Sh1 interface, and that combined DTC and Sh1 contacts promote cell division.

We next collected a dataset to explicitly test the proportionality of divisions across compartments. We analyzed ten time-lapse movies for which we could ascribe germ cell-somatic cell contacts for 25 cell divisions over about half an hour each (see example in Figure 3—figure supplement 1). We found that the number of divisions occurring at the DTC-Sh1 interface is disproportionate to the number of cells in that compartment (Figure 3F pink, prop.test, Bonferroni corrected p=0.0012327, see Materials and methods). In contrast, the shares of divisions under the DTC alone and Sh1 alone are proportional if not somewhat underrepresented (Figure 3F, corrected p values for DTC p=1.0, yellow and Sh1 p=0.13, cyan). We concluded that germ cell divisions are overrepresented on the DTC-Sh1 boundary.

Sh1 actively segregates germ cell daughters at the DTC-Sh1 interface

During our live-cell imaging, we noted that germ cell divisions at the DTC-Sh1 interface appeared to be polarized between the two somatic cells, with each condensation of chromatin noticeably closer to either the DTC or Sh1 cell (Figure 3—video 1, and Figure 3—figure supplement 1). We were intrigued by the possibility that external somatic cell contacts might provide a mechanism that dictates stem cell retention vs. exit from the niche (see schematic Figure 3G). The Notch ligand LAG-2 is localized to the DTC membrane (Gordon et al., 2019; Henderson et al., 1994), and losing active Notch signaling is necessary for germ cells to differentiate (Kershner et al., 2014; Maciejowski et al., 2006). We next asked how germ cells could escape this contact-mediated signal from the DTC niche and become able to differentiate. We hypothesized that these cells might either lose DTC contact and gain Sh1 associations by moving, initiate an exclusive association with Sh1 at birth via polarized cell division, or both. We examined how germ cells form associations with Sh1 by time-lapse imaging a strain that expressed GFP::INX-9 to mark the membrane of Sh1 (it is also expressed in puncta in the DTC), as well as a DTC membrane marker and germ cell nuclear histone marker (Figure 4). We focused on cell divisions associated with the DTC niche, including those at the DTC-Sh1 interface (that is, not cells dividing under Sh1 alone).

Figure 4 with 1 supplement see all

Download asset Open asset

Sh1 separates daughter cells that divide at interface.

(A) DTC membrane (yellow, *lag-2p::mTag:BFP::PLC^δPH)*, germ cell nuclei (magenta, *mex-5p::H2B::mCherry),* and Sh1 membrane (cyan, *GFP::inx9*, also expressed in puncta on DTC processes). Dashed circles in top and bottom rows marks focal germ cells enlarged in intervening timepoints; dashed outline in second and sixth rows denote starting position of Sh1 membrane. Solid outline in sixth row indicates Sh1 membrane that has extended beyond its starting position to cover dividing cell. (B) Germ cell nuclei (magenta, *mex-5p::H2B::mCherry)* channel only. Asterisks mark nuclei of focal cell(s). (C) DTC membrane (yellow, *lag-2p::mTag:BFP::PLC^δPH)* and Sh1 membrane (cyan, *GFP::inx9)* channels with germ cell asterisks from (B) and annotations from (A). 1000x magnification. Scale bar 10 μm. See also Figure 4—video 1.

We collected a total of 48 time-lapse movies of 15 min or longer in which a total of 126 germ cells in contact with the DTC divided. About half of these divisions (69/126) occurred at the distal edge of Sh1, at the DTC-Sh1 interface. Because some of these divisions did not resolve during the time-lapse interval or sunk out of the plane of focus, 64 were scored further (Supplementary file 1). All 64 divisions ended with at least one daughter cell directly beside or under the Sh1 cell. In most of these cases (56/64) one daughter remained under the DTC or at the interface and the other ended up under or beside Sh1, suggesting a polarized division. In the rare instances in which both daughter cells stayed at the DTC-Sh1 interface (8/64 divisions), the divisions still appeared polarized between the DTC and Sh1, but irregular cellular projections from these cells extended the DTC-Sh1 interface beyond the division plane such that both cells remained at the interface. Intriguingly, in 26/64 cases, Sh1 actively grew during the time-lapse interval to cover the closest daughter or grew into the cytokinetic furrow between the daughters to enwrap the closest daughter (see example in Figure 4). These dividing interface cells were the only cells (out of 468 total interface cells examined) that changed their germ cell-somatic cell contacts during observation. Similarly, out of the 55 scored cell divisions that occurred in contact with the DTC alone (Supplementary file 1), none of the daughters appeared to leave the niche, so division alone is not sufficient to displace a daughter cell out of the niche. These results offer compelling evidence that DTC-Sh1 interface germ cell division, often followed by Sh1 growth over one daughter cell, is the primary mechanism of stem cell niche exit of germ cells.

Germ stem cell and differentiation markers segregate at DTC-Sh1 interface

If germ cells leave the stem cell niche at the DTC-Sh1 interface, then we would expect DTC-associated germ cells to express a stemness marker and Sh1-associated germ cells to express a marker of meiotic differentiation, and for the boundary between these populations to coincide with the DTC-Sh1 interface. We used a sgyl-1p::H2B::gfp transgene (Kershner et al., 2014) as a direct target of active Notch signaling in the germ stem cells (Figure 5A). The promoter comes from the gene sygl-1, which acts directly downstream of active Notch signaling (Kershner et al., 2014) to promote stemness in the presence of FBF proteins (Shin et al., 2017). We use a gld-1::gfp transgene (ozIs5; Brenner and Schedl, 2016; Schumacher et al., 2005; Seidel et al., 2018) as a marker of differentiation (Figure 5A). The gld-1 gene encodes an RNA binding protein that is not abundant in distal mitotic cells (Jones et al., 1996); it is post-transcriptionally repressed by sygl-1 and FBF in the stem cell zone (Brenner and Schedl, 2016; Crittenden et al., 2002; Shin et al., 2017). GLD-1 RNA-binding protein promotes meiotic entry (Biedermann et al., 2009; Jungkamp et al., 2011; Kadyk and Kimble, 1998). We combined these transgenes with markers of Sh1 and the DTC from this study (Figure 5A).

Figure 5

Download asset Open asset

Genetic markers of germ cell fate segregate across DTC-Sh1 interface.

(A) Schematic depicting cross section of *C. elegans* gonad with regionalization of germ cell fate markers. A *sgyl-1p::H2B::gfp* transgene (Kershner et al., 2014) is a direct target of active Notch signaling in germ stem cells (magenta). *gld-1::gfp* transgene (*ozIs5*) is post-transcriptionally repressed in the stem cell zone (Brenner and Schedl, 2016) and its protein is found at low baseline levels there;it becomes de-repressed and protein accumulates in mitotic germ cells farther from the DTC (light and medium orange shading), and is highly expressed in germ cells committed to meiosis (dark orange shading). Previously reported regions of expression roughly correlate with our observed locations of the DTC (yellow) and Sh1 (cyan) cells. (B) Expression of *sgyl-1p::H2B::gfp* (magenta) with *lag-2p::2xmKate::PLC^δPH* (cyan/white) marking DTC and *mKate::inx-8(qy78)* (cyan) marking Sh1. Surface z-projection (top left) and cross section (bottom left), with white dashed line marking the boundary of brightest *sgyl-1p::H2B::gfp* expression. Grayscale images at right. (C) Expression of *gld-1::GFP(ozIs5)* (yellow) with *lag-2p::mNeonGreen::PLC^δPH* (yellow/white) marking the DTC and Sh1 (cyan, *mKate2::inx-8*) marking Sh1. Surface z-projection (top left) and cross section (bottom left), with white arrows indicating the distalmost edge of Sh1. Grayscale images at right, with magenta dashed line showing distalmost boundary of detectable GLD-1::GFP. Note correlation between Sh1 and GLD-1::GFP. (D) Plot of positions in 31 animals of proximal extent of DTC (empty circles, regression in red) or distal extent of Sh1 (filled black dots, regression in black) vs. distal extent of GLD-1::GFP Figure 5—source data 2. R and p values given on graph; confidence intervals shaded around regression lines. Dashed line indicates x = y; data points above this line have a distal boundary of GLD-1::GFP expression that is closer to the distal end than the somatic cell, while data points below the line have a distal boundary of GLD-1::GFP that is farther from the distal end than the somatic cell.

Figure 5—source data 1 sygl-1p:: H2B::GFP::sygl-1 3'UTR expression relative to somatic cells.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig5-data1-v2.xlsx
Download elife-56383-fig5-data1-v2.xlsx
Figure 5—source data 2 gld-1::GFP expression and somatic cell boundaries.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig5-data2-v2.xlsx
Download elife-56383-fig5-data2-v2.xlsx

The sygl-1p::H2B::gfp transgene is prone to silencing, and is quite variable among animals. However, we observed 31/36 adult animals had the region of brightest GFP expression limited to the cells that fall distal to the DTC-Sh1 interface, inclusive of interface cells Figure 5—source data 1. We found that the long DTC processes in older adults were not correlated with GFP expression in the underlying germ cells (Figure 5B); this is consistent with previous reports showing that sygl-1p::H2B::GFP expression localizes to germ line regions with dense, short DTC processes (Linden et al., 2017). We now know that the distal-most density of DTC processes is directly bordered by Sh1. This finding is consistent with cells under the DTC and at the DTC-Sh1 interface having a molecular signature of stemness, and cells outside of this domain under Sh1 alone losing that signature.

The GLD-1 protein and GLD-1::GFP expressed by a transgene have apparent ‘steps’ in intensity (Brenner and Schedl, 2016; Seidel et al., 2018), which were recently shown to reflect the folding of the germ cell syncytium, along which there is a smooth increase in GLD-1 intensity along the plane of the rachis from distal (low/absent) to proximal (high) (Seidel et al., 2018). In 31 young adult animals, we saw the distalmost GLD-1::GFP signal correlated with the distal edge of Sh1 (Figure 5C and D), supporting our hypothesis that distal Sh1-associated cells are in the early stages of differentiation. In fact, they seem to correspond with the position in the gonad at which germ cells are irreversibly committed to meiosis (Brenner and Schedl, 2016; Cinquin et al., 2010)—proximal to the densest DTC processes, but distal to the transition zone in which nuclear morphology is clearly meiotic. In most animals, processes of Sh1 extended just distal beyond the distal extent of GLD-1::GFP signal, meaning we observed GLD-1::GFP in only Sh1-associated cells (the data fall beneath the x = y line shown in gray dashes Figure 5D). On the other hand, while the DTC’s proximal extent is more weakly correlated with the distal extent of GLD-1::GFP (Figure 5D), an anticorrelation that was previously observed (Byrd et al., 2014), long DTC processes typically run deep into the region of active GLD-1::GFP (the data consistently fall above the x = y line shown in gray dashes). Taken together, the patterns of sygl-1p::H2B::GFP and GLD-1::GFP expression strongly support cell fate asymmetry across the DTC-Sh1 boundary that segregates stem cells from their differentiating progeny.

Germ cell spindles orient to the somatic cells

Given that germ cell fates segregate across the DTC-Sh1 boundary, we next asked how germ cells traverse the boundary. Because most (56/64 interface divisions, see above) of the DTC-Sh1 interface divisions we observed segregated their daughters asymmetrically (as in Figure 4), we hypothesized that the somatic cells orient the germ cell divisions at the interface. Oriented cell division is a common feature of stem cells (Morrison and Kimble, 2006) but has not been observed in C. elegans germ cells, which are instead thought to divide symmetrically and without orientation (Crittenden et al., 2006; Crittenden et al., 2003; Morrison and Kimble, 2006).

To resolve whether germ cells made oriented divisions at the DTC-Sh1 boundary, we performed time-lapse imaging and recorded 97 mitoses in a strain expressing beta-tubulin::GFP (Galy et al., 2003; Gerhold et al., 2015; Strome et al., 2001), a marker of the mitotic spindle. The 29 germ cell divisions at the DTC-Sh1 interface had spindles that were more aligned to the shortest line connecting the DTC to Sh1 than to the distal-proximal axis of the gonad (the null hypothesis, Figure 6B–E, one sided Wilcoxon rank sum test W = 141, p-value=2.537e-06). When cells divided on the DTC-Sh1 boundary, they oriented their spindles toward these two cells and divided asymmetrically (with one daughter associated with the DTC and the other toward Sh1, Figure 6A). Spindle orientation was observed across the DTC-Sh1 interface prior to division, which is consistent with a model in which the DTC-Sh1 interface dictates the asymmetric division (as opposed to the completed cell division later triggering remodeling at the DTC-Sh1 interface).

Figure 6 with 2 supplements see all

Download asset Open asset

Tubulin spindles polarize to the DTC and Sh1 as germ cells divide at the interface.

(A) Stills from a time-lapse movie (Figure 6—video 1) of a germ cell dividing (dashed outline in second row) at the DTC-Sh1 interface in a young adult animal. DTC (yellow, *lag-2p::mTag:BFP::PLC^δPH*), Sh1 (cyan, *mKate2::inx8*), germ nuclei (cyan, *mex-5p::H2B::GFP*), and tubulin (pink, *β-tub::GFP*). Tubulin spindles are transiently visible (arrowheads in third row). Yellow and cyan overlays in the bottom row mark cell outlines of DTC and Sh1, respectively. Scale bar is 10 μm. (B) Schematic depicting how angle of division to gonad was calculated. For each cell dividing at the interface between the DTC and Sh1 (n = 29) Figure 6—source data 1. Line one was drawn through the spindles into the neighboring soma. Line two was drawn along the distal-proximal axis of the gonad. The angles between Line one and Line two reflect how aligned the spindles are to the axis of the gonad, and are plotted in (C). (D) Schematic depicting how angle of division to DTC-Sh1 interface was calculated. Line one again was drawn through the spindles and into the soma (same as in B). Line three was drawn through the shortest path between DTC and Sh1 that passes through at least one of the chromatin condensations at anaphase; this is the DTC-Sh1 axis. The angles between Line one and Line three reflects how aligned the spindles are to the DTC-Sh1 axis, and are plotted in (E). Note that every angle was expressed as an acute angle for the sake of consistency and to increase power. A Wilcoxon rank sum test was performed on the angles shown in C and E, and the result shows that they are significantly different from each other with W = 91 and p=1.967e-05, with the division angles being more aligned to the DTC-Sh1 axis. See also Figure 6—video 1 and Figure 6—figure supplement 1.

Figure 6—source data 1 Division angles of 97 germ cell mitoses.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig6-data1-v2.xlsx
Download elife-56383-fig6-data1-v2.xlsx

The absence of orientation with respect to the distal-proximal gonad axis was also observed for cells that divided under the DTC alone, and cells that divided under Sh1 alone (each Wilcoxon test for division angle to gonad between pairs of the three compartments has a p>0.1, Figure 6—figure supplement 1). The absence of orientated division to the gonad axis has been construed by others as evidence against oriented divisions of adult C. elegans germ cells (Crittenden et al., 2006; Morrison and Kimble, 2006). Our results, however, offer compelling evidence that the geometrically complex and variable DTC-Sh1 interface has obscured a clear view of strongly oriented divisions along this cryptic anatomical boundary.

Distal sheath extension requires the F-actin regulators Arp2/3 and Cofilin

Spindle orientation is a key feature of germ cell division at the DTC-Sh1 interface; the other key feature is the ‘grabbing’ of one daughter at cytokinesis by growth of the Sh1 cell. We hypothesized that the dynamic reaching of Sh1 may be actin-dependent, and that disrupting the actin cytoskeleton during adulthood may cause Sh1 to fail to maintain distal membrane extensions over the proliferating germ cells as their progeny are displaced proximally.

To test if actin remodeling was required for Sh1 to maintain its distal position near the DTC, we devised a developmentally timed RNAi treatment to knock down two crucial F-actin regulators, a branched actin nucleator Arp2/3 subunit encoded by arx-2 and the C. elegans ortholog encoding the F-actin disassembly protein cofilin (unc-60). Both arx-2 (Roh-Johnson et al., 2012; Sawa et al., 2003) and unc-60 (Ono et al., 2003) are essential genes required for embryonic development. To avoid developmental defects caused by genetic loss or early RNAi knockdown (Ono et al., 2008), we fed early L4 larval animals bacteria expressing RNAi targeting arx-2 (Figure 7—figure supplement 1) and unc-60. Consistent with our hypothesis, in adults two days after RNAi exposure there was a gap between the DTC and Sh1 (Figure 7A), and Sh1 took on a stringy, fragmented form at its distal end while the DTC structure was not similarly perturbed. While worms fed empty RNAi vector had an average distal position of Sh1 that fell 44 μm from the distal end of the gonad, unc-60 RNAi-treated worms had an average distal position of 84 μm (Figure 7B). A linear mixed model revealed that the positions in the RNAi treated worms are different from the control worms (linear mixed model: F_1,62 = 40.00675 Tukey post-hoc test p<0.0001). Further supporting the notion that F-actin-mediated dynamic membrane extensions promote maintenance of Sh1 position, we found that an endogenously tagged Arp2/3 subunit encoded by arx-2::gfp (Kelley et al., 2019; Zhu et al., 2016) localized dynamically to the edge of the Sh1 cell during membrane extensions over the dividing daughter cell closest to the Sh1 cell (Figure 7—video 1, Figure 7—figure supplement 1D and Figure 7—figure supplement 1E, and n = 10/11 time-lapse recordings of interface divisions). We conclude that continuous and sporadic distal sheath growth requires actin remodeling and is required to maintain the Sh1 interface with the DTC.

Figure 7 with 2 supplements see all

Download asset Open asset

The distal sheath maintains proximity to the DTC via cofilin-dependent growth, and loss of proximity correlates with diminished germ cell proliferation and regulated niche exit.

(A) The DTC-Sh1 interface (DTC in yellow, *lag-2p::mTag:BFP::PLC^δPH*; Sh1 in cyan, *GFP::inx-9* bottom) seen in in the control (left) is disrupted by RNAi (right) against gene encoding worm ortholog of actin binding protein cofilin (*unc-60*). Germ cell nuclei (in magenta, *mex-5p::H2B::mCherry*) can be observed in the undifferentiated (distal, smaller) and meiotic (crescent and larger nuclei) states. Crescent shaped nuclei mark the meiotic transition zone. (B). Neither the DTC length (yellow arrows in A) nor the length of the progenitor zone (magenta arrows in A) are affected by RNAi treatment, while the distal boundary of Sh1 (cyan arrows in A) is perturbed 0: F_1,62 = 40.00675 Tukey post-hoc test p<0.0001. In all graphs, boxes show median and second and third quartiles, while whiskers show minimum and maximum values. (C) Binned distances from the distal tip of germ cell divisions in three somatic cell compartments: DTC-alone (yellow), DTC-Sh1 (magenta), Sh1-alone (cyan) for control (solid fill) and *unc-60* RNAi-treated (hashed) worms 48 hr after exposure beginning in the L3/L4 stage Figure 7—source data 1. Note that this is ~1 day later than the animals analyzed in Figure 3E, and the DTCs are substantially more elongated by this stage. Control n = 31 worms, *unc-60* RNAi n = 35 worms. Representative images of (D) control and (E) *unc-60* RNAi-treated worms expressing DTC membrane GFP, GLD-1::GFP transgene, and mKate2::INX-8 to mark Sh1. Yellow asterisk marks autofluorescent gut granules in neighboring tissue. Arrowheads mark distal strings of retracted Sh1. (F) Boxplots showing distance from distal end of Sh1 in control and *unc-60* RNAi-treated worms Figure 7—source data 2. Welch two-sample t-test p<0.001. Note concordant results to those shown in (B). (G) Boxplots showing distance from distal end of GLD-1::GFP expression Figure 7—source data 2. Welch two-sample t-test p<0.001.

Figure 7—source data 1 Gonad feature measurements, positions and somatic cell contacts of dividing cells in control and RNAi-treated animals.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig7-data1-v2.xlsx
Download elife-56383-fig7-data1-v2.xlsx
Figure 7—source data 2 Position of Sh1 and GLD-1::GFP boundaries in control and RNAi-treated animals.: https://cdn.elifesciences.org/articles/56383/elife-56383-fig7-data2-v2.xlsx
Download elife-56383-fig7-data2-v2.xlsx

Distal Sh1 has a Pro-proliferative effect

The Sh1 cells are important for larval germ line mitotic proliferation (Killian and Hubbard, 2005; Killian and Hubbard, 2004; Pepper et al., 2003), but were thought not to interact with proliferating germ cells in the adult. Given that our results suggested that the DTC-Sh1 interface promoted proliferation, we next assessed the effect of Sh1 and the underlying germ cells by perturbing its position with unc-60 RNAi. The undifferentiated germ cells at the distal end of the gonad in the progenitor zone are sometimes called ‘mitotic’ germ cells because of their cell fate (they are not in meiosis, as differentiated germ cells are). Mitotic germ cell nuclei are smaller and less bright by histone fluorescence than meiotic germ cell nuclei, and reside distal to the crescent-shaped nuclei of the meiotic ‘transition zone’ (Hubbard, 2007; Figure 7A magenta arrows). However, the mitotic fate must not be conflated with active mitotic events; germ cells can fail to proliferate but also remain undifferentiated (or ‘mitotic’) (Seidel and Kimble, 2015). We hypothesized that if distal Sh1 is pro-proliferative (due to its association with most observed cell divisions, Figure 3), its mispositioning by RNAi would cause fewer active divisions in the progenitor zone. Indeed, the average number of actively dividing cells in the progenitor zone in age-matched control animals was 2.29, while the average number of dividing cells in unc-60 RNAi animals was 1.37 at the endpoint 48 hr after RNAi exposure beginning in the early L4 (n = 31 control, n = 35 experimental, linear mixed model: F_{1, 62} = 5.46813, Tukey post-hoc test p<0.05). While just 5/31 control animals lacked active cell division, 14/35 unc-60 RNAi animals lacked active cell division (chi-squared = 4.5694, p<0.05). The lengths of the progenitor zones were not different between the unc-60 RNAi-treated animals and control. However, because Sh1 was shifted so profoundly (Figure 7B), the amount of the mitotic germ line covered by Sh1 was dramatically reduced (linear mixed model: F_1, 62 = 42.566, Tukey post-hoc test p<0.005). While the control animals had about 29 μm of undifferentiated germ cells in contact with Sh1, after treatment with unc-60 RNAi on average no undifferentiated germ cells contact Sh1. RNAi against unc-60 can cause germ cell defects under certain exposure conditions (Ono et al., 2008). However, the RNAi-treated animals still had normally sized progenitor zones, normally elaborated DTCs (Figure 7B), and a number of visibly dividing germ cells.

We assessed the positions and somatic cell contacts of these dividing cells as in Figure 3E. The numbers and positions of actively dividing germ cells in contact with the DTC alone are similar in the control and unc-60 RNAi-treated animals (Figure 7C, yellow solid and hashed), indicating that the RNAi treatment did not dramatically affect divisions of germ cells in the DTC plexus. Among the animals with active divisions (ignoring the 40% of RNAi-treated animals that lack dividing cells entirely), the control (n = 46/26 or 1.77 DTC divisions/animal) and treatment (n = 38/21 or 1.81 DTC divisions/animal) groups had roughly equivalent average numbers of DTC-associated divisions. However, the number of DTC-Sh1 interface divisions was dramatically different between control (n = 23/26 or 0.88 DTC-Sh1 divisions/animal) and treatment (n = 5/21 or 0.24 DTC-Sh1 divisions/animal), since the interface was greatly reduced. We never observed a germ cell dividing without contact to a somatic cell.

Finally, we examined GLD-1::GFP expression in animals after unc-60 RNAi treatment (Figure 7D–G). In this experiment, we again observed a stringy Sh1 phenotype (Figure 7E, compare to control in 7D) and a robust proximal displacement of the distal Sh1 boundary (Figure 7F, compare to Sh1 distance in 7B). Additionally, we observed a small (~8 μm) but significant (Welch two-sample t-test, p<0.001) shift of the distal boundary of GLD-1::GFP expression in the direction that Sh1 shifted. From this we infer that Sh1, while not required for differentiation, helps germ cells differentiate by maintaining its distal position, possibly by competing with the DTC niche for germ cell contact (Figure 8). Taken together, these experiments indicate that the dividing population of undifferentiated, Sh1-associated germ cells are nearly eliminated when Sh1 is mispositioned, leading to an overall deficit of germ cell proliferation and regulated exit from the niche, suggesting that Sh1 is an important but overlooked regulator of germ cells in the progenitor zone.

Figure 8

Download asset Open asset

Schematic depicting DTC-Sh1 interface germ cell division.

(A) DTC (yellow) and Sh1 (cyan) contact germ cell membranes (magenta lines) in the distal gonad. Germ cell nuclei (dark turquoise) are visible during mitosis as round nucleus (Step 1), metaphase plate (Step 2), anaphase (Step 3), cytokinesis (Step 4), and in two segregated daughter cells (Step 5). Tubulin spindles (red, Step two and Step 3) and Arp2/3 (orange, Step 2–4) transiently localize in germ cells and Sh1, respectively. As the germ cell divides in a polarized fashion between the DTC and Sh1, Sh1 grows at its distal edge into the cytokinetic furrow to separate the two daughter cells. From birth, these two daughter cells have asymmetric contacts with the soma, and one has exited the niche.

Discussion

Stem cells rely on their niches to maintain stemness, but their progeny must leave the niche in order to differentiate. Only by studying stem cells and their associated support cells in vivo can we understand how this balance is regulated. Challenges of in vivo imaging have recently been overcome in the zebrafish larval hematopoietic niche, mammalian epidermal stem cell niche, and mouse intestinal crypt, revealing important behaviors such as ‘endothelial cuddling’ of stem cells (Tamplin et al., 2015), different rates of division across stem cells and their daughters, (Rompolas et al., 2012), and different trajectories for stem cells depending on their position within the niche (Ritsma et al., 2014). By developing genome-edited alleles to act as in vivo cellular markers in the C. elegans gonad, we made the surprising discovery that some cells in the C. elegans germ line stem cell niche divide in an asymmetric manner towards a niche-adjacent cell. Germ cells that contact this cell instead of the niche have begun to differentiate. This discovery raises the possibility that facilitated niche exit is part of the repertoire of cellular behaviors that regulate the balance between stem cell renewal and differentiation and a new player to a canonical stem cell niche system.

It was previously thought that a germ line bare region extended over most of the progenitor zone, with the Sh1 cells contacting only differentiating germ cells in the adult. By generating endogenously tagged alleles of somatic gonad-expressed innexins, we have demonstrated that adult Sh1 cells extend much farther distal than was previously understood. They contact putative stem cells in the progenitor zone and grow distally into this zone against the proximal flow of germ cells in an Arp2/3- and cofilin-dependent manner. The sheath is known to interact with and promote proliferation of larval germ cells (Killian and Hubbard, 2005; McCarter et al., 1997), to interact with differentiating adult germ cells to facilitate exit from meiotic pachytene (McCarter et al., 1997), and to phagocytose physiologically apoptotic cells (Li et al., 2012). It was not thought to contact the progenitor zone in the adult. Our results demonstrate not only that Sh1 processes extend into the progenitor zone, but also that Sh1 plays an important role regulating adult stem cells and their progeny. A somatic gonad cell directly bordering the DTC is also observed in the filarial nematode Brugia malayi (Landmann et al., 2012; Foray et al., 2018), raising the possibility that this cellular arrangement and function are widespread within the clade. During larval development, Sh1 develops in close contact with the DTC niche, implying an intimate relationship between the niche and its neighboring somatic cells as well.

By performing live-cell imaging of cell divisions at the border of the DTC and Sh1 we revealed that while this population comprises a minority of progenitor zone cells, it is more proliferative than cells under the DTC alone or Sh1 alone. As we observed this excess of DTC-Sh1 interface divisions, we noted that dividing cells polarized their spindles between the DTC and Sh1 and divided asymmetrically between them, often with Sh1 actively extending over the nearest daughter. Such divisions were the only times we observed germ cells lose contact with the DTC to exit the niche—we never saw spontaneous growth of Sh1 over a non-dividing cell, nor displacement of germ cells from the niche by distal-to-proximal pushing produced by other cells dividing entirely under the DTC. Previous lineage tracing of germ stem cells and their progeny as they leave the niche (Rosu and Cohen-Fix, 2017) revealed that progenitor zone cells move very slowly, with photoconverted populations being displaced about five germ cell diameters over 8 hr. Coupled with the slow rate of division in this population (about 1% of adult germ cells dividing at any given time point [Roy et al., 2016]), we think it is reasonable that our time-lapse imaging captured the full range of movements made by the cells in the progenitor zone. Based on these observations, we propose a simple model (Figure 8) for germ cell exit from the stem cell niche: germ cells at the DTC-Sh1 interface divide asymmetrically, with one daughter remaining anchored to the DTC and the other becoming enwrapped by Sh1. Those stem cell daughters that maintain their connections to the DTC would comprise an anchored, sygl-1p::H2B::GFP positive, asymmetrically dividing stem cell population, while those associating with Sh1 are set on a path to differentiation and dramatically increase levels of GLD-1::GFP. Our results show that Sh1 is not required for germ cell differentiation—GLD-1::GFP is visible in germ cells in the gap between the DTC and Sh1 when Sh1 is shifted proximally—much as we would expect given that differentiation is still observed in the reduced germ cell pools of sheath-ablated animals (McCarter et al., 1997 and Killian and Hubbard, 2005). Instead, we hypothesize that Sh1 promotes differentiation indirectly through its positional exclusion of the DTC from germ cells born from polarized interface divisions (thus allowing these daughter cells to escape the influence of the DTC, which maintains stem cell fate, Figure 8). Since cells born from polarized divisions that contact Sh1 lack DTC-contact-mediated Notch signaling from birth, they can then derepress meiosis-promoting factors and initiate differentiation.

Our findings also have implications for the organization of the progenitor zone of the C. elegans germ line. The three somatic cell-associated compartments—DTC, DTC-Sh1 interface, and Sh1—may host three types of progenitor cell divisions: symmetric-stem cell renewing (under the DTC), asymmetric (at the DTC-Sh1 interface), and symmetric-differentiating (or transit-amplifying, under Sh1 alone), which might be comparable to the ‘mixed mode’ of stem cell division that acts in the mammalian epidermis (Doupé et al., 2010; Rompolas et al., 2016; Yang et al., 2015). This layer of Sh1 control needs to be studied in the context of other factors that influence germ cell proliferation and their increasingly appreciated regulation by age (Cinquin et al., 2016; Kocsisova et al., 2019; Seidel and Kimble, 2015) and environment (Aprison and Ruvinsky, 2016; Roy et al., 2016).

Most of what we know about how stem cell progeny leave the niche is understood as the absence of the mechanisms that retain the self-renewing stem cell daughter in the niche, be that oriented division to a basal lamina in the mammalian epidermis (Poulson and Lechler, 2010) or to niche cells in the Drosophila ovary, (Casanueva and Ferguson, 2004) or stochastic displacement in mammalian intestine (Snippert et al., 2010; van der Flier and Clevers, 2009) and epidermis (Doupé et al., 2010). One study that specifically addresses how stem cell progeny leave a niche found they do so by migration in the larval tracheal system of Drosophila (Chen and Krasnow, 2014), a very different mechanism to the one we observe in this study. Cells outside the niche had previously been known to guide stem cell progeny only after they leave the niche. For example in the larval trachea of Drosophila, stem cell progeny follow external cues as they migrate from the niche (Chen and Krasnow, 2014), and in adult mammalian neurons, niche-adjacent cells send differentiation cues to stem cell progeny (Ming and Song, 2011). Some niche-adjacent cells limit the range of niche cues to restrict the stem cell population, like the Drosophila testis cyst cells (Fairchild et al., 2016; Fairchild et al., 2015). Active processes like Sh1 segregating the nearest daughter may take place in other cases of niche exit that appear stochastic, like the mammalian intestinal niche in which ‘border cells’ are thought to be passively displaced from the niche (Ritsma et al., 2014). Similar niche-infiltrating structures may not have been discovered because of the difficulty of observing thin, membranous cellular structures deep inside tissues in vivo that also may not survive dissection or fixation necessary for observing them by immunostaining or electron microscopy, as in the case of cytonemes (Kornberg and Roy, 2014). Given the consequences of failing to maintain stem cells and failing to provide sufficient progeny for homeostatic tissue maintenance, mechanisms that balance niche exit with niche retention may be important but overlooked features of many stem cell niches.

Materials and methods

Contact for reagent and resource sharing

Request a detailed protocol

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Kacy L. Gordon (kacy.gordon@unc.edu). Strains containing mNeonGreen can only be distributed to labs with a mNeonGreen license from Allele Biotechnology. Strains sourced from Caenorhabditis Genetics Center (CGC) are to be requested directly from CGC. Please see Appendix 1: Key Resources Table for more information.

Experimental model and subject details

Request a detailed protocol

C. elegans strains were maintained on standard NGM media at 20°C and fed E. coli OP50. For RNAi experiments, animals were fed E. coli HT115(DE3) containing the L4440 plasmid with or without dsRNA trigger insert (see RNAi experiments, Materials and methods). All animals scored were hermaphrodites (as males have structurally different gonads) at the ages specified in the text.

Strains

In strain descriptions, we designate linkage to a promoter with a p following the gene name and designate promoter fusions and in-frame fusions with a double semicolon (::). Some integrated strains (xxIs designation) may still contain unc-119(ed4) mutation and/or the unc-119 rescue transgene in their genetic background, but these are not listed in the strain description for the sake of concision, nor are most transgene 3’ UTR sequences. Further details available upon request. Strains are as follows:

PS3460 (unc-119(ed4)), NK2569 (xnSi1(mex-5p::GFP::PLCδ^PH::nos-2 3’UTR) II; inx-8(qy78(mKate::inx-8)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH)), NK2570 (naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II/xnSi1(mex-5p::GFP::PLCδ^PH::nos-2 3’UTR) II; inx-8(qy78(mKate::inx-8)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH)), NK2571 (inx-8(qy78(mKate::inx-8)) IV; (cpIs122(lag-2p::mNeonGreen:: PLC^δPH)), NK2572 (inx-9(qy79(GFP::inx-9)) IV; cpIs91(lag-2p::2x mKate2::PLC^δPH::3xHA::tbb-2 3'UTR LoxN) II), NK2573 (naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II; inx-9(qy79(mKate::inx-8)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH)), NK2574 (naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II; inx-8(qy78(mKate::inx-8)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH); ojIs1 [pie-1p::GFP::tbb-2 + unc-119(+)]), NK2575 (naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II; inx-8(qy78(mKate::inx-8)) IV; arx-2(cas607[arx‐2::gfp knock‐in]) V), NK2576 (inx-9(ok1502) IV; inx-8(qy102(mKate::inx-8)) IV; cpIs122(lag-2p::mNeonGreen::PLC^δPH)), NK2324 (qy23 [ina-1::mNG] III), KLG004 (cpIs91(lag-2p::2x mKate2::PLC^δPH::3xHA::tbb-2 3'UTR LoxN) II); qSi26(sygl-1p::H2B::GFP::sygl-1 3UTR) II; (inx-8(qy78(mKate::inx-8)) IV; teIs1(oma-1::GFP) IV), KLG005 (cpIs122(lag-2p::mNeonGreen:: PLC^δPH)); inx-8(qy78(mKate::inx-8)) IV; ozIs5(gld-1::GFP)), KLG006 tnIs6(lim-7p::GFP); inx-8(qy78(mKate::inx-8) IV; (cpIs122(lag-2p::mNeonGreen:: PLC^δPH)). All strains with multiple genetic elements were generated for this study by crossing the cited strains with genetic elements created for this study (nasi2 transgene from Roy et al., 2018; xnSi1 parent from Chihara and Nance, 2012; cpIs91 transgene from Gordon et al., 2019, qSi26 and teIs1 from Kershner et al., 2014, ozIs5 transgene from Brenner and Schedl, 2016; Schumacher et al., 2005; Seidel et al., 2018, ojIs1 transgene from Galy et al., 2003 via (Gerhold et al., 2015, arx-2(cas607) from Zhang et al., 2017, inx-9(ok1502) allele from Consortium TC elegans DM, 2012).

Molecular biology

Request a detailed protocol

We cloned mTagBFP and a PLC^δPH membrane-localization domain into the plasmid containing the DTC-expressing lag-2 promoter fragment generated in Linden et al., 2017, and injected this construct into unc-119(ed4) mutants with the unc-119 rescue plasmid and EcoRI-digested salmon sperm DNA. The resulting extrachromosomal array was integrated by standard gamma irradiation protocol.

Genes inx-8 and inx-9 are neighboring duplicate genes on Chromosome IV (Scheme 1A). The alleles inx-8(qy78) and inx-8(qy102) were generated using Cas9-triggered homologous recombination with a self-excising selection cassette (Dickinson et al., 2015). The target sequence was 5’- GCATCTTCACTCGGGTTCGAAGG-3’ with PAM site shown in bold; the guide RNA was cloned into the eft-3p::Cas9+sgRNA expression vector pDD162 using this same primer sequence (excluding the PAM). The PAM is seven nucleotides upstream of the start codon of inx-8. Homology arms were amplified by PCR from N2 genomic DNA (5’ homology arm forward primer inx-8 sequences are 5’-tgtacgactgtaggcaggcaggtag-3’ and reverse primer 5’-tctgcaattcgaacccgagtgaagatg-3’ (with mutation in PAM site shown underlined); 3’ homology arm forward primer inx-8 sequences 5’-ATGTTTTCTGTTCCATTTCTTACCTC-3’ and reverse primer 5’-CTGTACATCTCCACGGCAACCTCCG-3’) and cloned into a plasmid modified from Dickinson et al., 2015 containing a nine amino acid N-terminal leader sequence (with an ATG codon added during assembly), an mKate2 coding gene, a 15 amino acid linker sequence (with TEV site), and a self-excising selection cassette (SEC) flanked by LoxP sites in an intron of mKate2 (Scheme 1B). The repair plasmid was coinjected with the sgRNA+Cas9 plasmid into N2 animals, and genome-edited animals were selected by hygromycin B treatment and phenotypic identification (roller). The selection cassette was excised by heat shock as described in Dickinson et al., 2015. This construct was also injected into inx-9(ok1502) and fertility verified (Figure 1—figure supplement 1A), suggesting it is a functional protein fusion.

Scheme 1

Download asset Open asset

Gene models of the *inx-8/inx-9* locus with alleles annotated.

Related to Materials and methods. (A) Genomic locus from Wormbase legacy genome browser. Change of function allele *inx-9(ok1502)* (see Figure 1—figure supplement 1) annotated with yellow bar. (B) Annotation of genome-edited allele *inx-8(qy78) mKate::inx-8* with *mKate* exons shown with red boxes, introns shown with peaked lines, and linker amino acid sequences shown with white boxes. (C) Annotation of genome-edited allele *inx-9(qy79) GFP::inx-9* with GFP exons shown with green boxes, introns shown with peaked lines, and linker amino acid sequences shown with white boxes. Endogenous UTRs and regulatory sequences, and their relative positions to their ends of the genes, are preserved.

The allele inx-9(qy79) was generated using Cas9-triggered homologous recombination with a self-excising selection cassette (Dickinson et al., 2015). The target sequence was 5’- CTTTCAGAGCATTGTCACTTTGG-3’ with PAM site shown in bold; the guide RNA was cloned into the eft-3p::Cas9+sgRNA expression vector pDD162 using this same primer sequence (excluding the PAM). The PAM is 16 nucleotides upstream of the start codon of inx-9. Homology arms were amplified by PCR (from N2 genomic DNA, 5’ homology arm forward primer inx-9 sequences are 5’-gaaataatcgagatgaaactgtcg-3’ and reverse primer 5’-CATtctgtccctttgaacgaaagtg-3’ (with mutation in PAM site shown underlined); 3’ homology arm forward primer inx-9 sequences 5’-ATGTTTTCTGTTCCATTTCTTACC-3’ and reverse primer 5’-tgagttggactgacatcgag-3’). Note that the beginning of inx-8 and inx-9 coding sequences are identical (this is why both sgRNAs target sequences upstream of the coding region). To selectively amplify from inx-9 for the 3’ homology arm, we first used as template a PCR amplicon from N2 genomic DNA including 4.3 kb of the inx-9 locus using primers that exclude the region of sequence identity at the beginning of the inx-8 gene (forward primer in inx-8 downstream exon 5’-GTGACTTCCAAGTTCGTGAGATGGC-3’ and reverse primer 5’-gcttgaaaacggtgcggatccagc-3’>800 nt downstream of inx-9 stop codon). These homology arms were cloned into a plasmid modified from Dickinson et al., 2015 containing a GFP coding gene, a 54 amino acid linker sequence (including TEV site and ZF1 tag) and a self-excising cassette (SEC) flanked by LoxP sites in an intron of GFP (Scheme 1C). The plasmid was coinjected with sgRNA+Cas9 plasmid into N2 animals, and genome-edited animals were selected by hygromycin B treatment and phenotypic identification (roller). The selection cassette excised stochastically and non-rollers were recovered. This tagged protein was not assessed for its function in an inx-8 mutant background.

The strain containing the lag-2p::mTagBFP:: PLC^δPH to mark the DTC, the mKate2::inx-8(qy78) to mark Sh1 cells, the germ cell nuclear histone tag mex-5::H2B::mCherry, and the germ cell membrane marker mex-5p::GFP:: PLC^δPH was maintained as heterozygotes for the germ cell nuclear and germ cell membrane markers, as both are carried in the same Mosci site on chromosome II. While the germ cell membrane maker was helpful in determining whether a germ cell bare region exists, it was ultimately dispensable for inferring germ cell-somatic cell associations in the presence of a germ cell nuclear marker due to the tight packing of germ cells and the close intercalation of the somatic cells.

Time-lapse imaging

Request a detailed protocol

Worms were anesthetized with 0.02% tetramisole in M9 buffer for 15 min prior to mounting on 4% noble agar pads with coverslips sealed with valap and flooded underneath with 0.02% tetramisole to prevent drying (adapted from Gerhold et al., 2015, which used 0.04% tetramisole and grooved agarose pads). We found that tricaine and sodium azide, two commonly used worm anesthetics, caused germ cell divisions to cease. Acquisition intervals were set for every five minutes, but the relatively light anesthetic meant that some worm movement still occurred, and pauses to reposition the field of view were occasionally necessary. We found that lowering the laser power of the 402 nm and 488 nm lasers helped decrease spontaneous worm movement and photobleaching. Animals with cell divisions that did not progress were not analyzed.

Microscopy and image acquisition, processing, and analysis

Request a detailed protocol

Confocal DIC and fluorescent images were acquired on an AxioImager A1 microscope (Carl Zeiss) equipped with an EMCCD camera (Hamamatsu Photonics), a 100x or 40x Plan-Apochromat (1.4 NA) objective, and a spinning disc confocal scan head (CSU-10; Yokogawa Electric Corporation) driven by μManager software (Edelstein et al., 2010) at 20°C, except Figure 7D and E, which were acquired on a Leica DMI8 with an xLIGHT V3 confocal spinning disk head (89 North) with a 63x Plan-Apochromat (1.4 NA) objective and an ORCA‐Fusion Gen‐III sCMOS camera (Hamamatsu Photonics). Worms were mounted on 4% noble agar pads containing 0.01 M sodium azide for imaging during endpoint experiments. Images were processed with FIJI 2.0 and Photoshop CC (Adobe Systems Inc). Images are displayed as single confocal z-slices or maximum intensity projections generated in FIJI, as noted. Supplemental videos were generated with FIJI. Graphs generated by R and MS Excel were refined using Illustrator CC (Adobe Systems Inc).

Timed RNAi

Request a detailed protocol

We performed timed RNAi by feeding on worms carrying a DTC membrane marker (lag-2p::mTagBFP::PLC^δPH) the inx-9(qy79) GFP-tagged allele and the nasi2(mex-5p::H2BmCherry) transgene marking germ cell nuclear histones. We did not expose animals to L1 larval RNAi for fear that early knockdown of actin associated proteins might lead to arrest of development (Ono et al., 2008; Ono et al., 2003). To circumvent the early defects caused by their loss of function, we applied post-embryonic RNAi by feeding starting at the early L4 larval stage. By the time RNAi knockdown is achieved, gonad migration is complete. Worms were terminally anesthetized in 0.01 M sodium azide and imaged after 48 hr on RNAi (in the second day of adulthood). Note that these animals are ~24 hr older than most of the other worms shown in the study.

Measurements of RNAi-treated gonads

Request a detailed protocol

Gonad images were measured in FIJI for the length of the longest continuous DTC process, length of the mitotic zone, and distance of distal Sh1 from the distal end of the gonad. Mitotic figures and their positions relative to the fluorescently labeled somatic cells were noted. Germ cells were assessed for their general condition (germ cell number, large germ cells, gaps among germ cells, etc.) and the Sh1 cell was assessed for its condition (normal or stringy). We noted variation in the extent of germ line damage due to unc-60 RNAi knockdown, with one replicate showing highly abnormal distal germ lines (as noted in Ono et al., 2008) with gaps among the misshapen germ cell nuclei presumably caused by the collapse of the rachis and the failure of cytokinesis, which both require proper actin remodeling (Dorn et al., 2016; Priti et al., 2018). In this replicate, germ cell nuclear size was a more reliable marker of the end of the proliferative zone than nuclear crescent shape. The results shown in Figure 7 are pooled from three replicates of exposures starting on three consecutive days for a total of 31 control animals and 35 unc-60 RNAi-treated animals.

Quantification

Request a detailed protocol

Dividing germ cells were identified by chromatin condensations at metaphase and anaphase; their distance from the distal tip of the gonad was measured in FIJI. Distal gonads encompassed by the field of view at 1000x magnification do not always contain the entire progenitor zone, so our estimates of the numbers of undifferentiated germ cells in contact with Sh1 alone is conservative. This distal field of view did contain the DTC-Sh1 interface in every non-RNAi-treated animal in the study. Germ cell-somatic cell contacts were scored by eye, and typically only germ cells in the superficial layer were counted, as deeper contacts are more ambiguous. Angles of dividing cells, the gonad axis, and the shortest path through the dividing cell from DTC to Sh1 were measured in FIJI and analyzed as shown in Figure 6 schematic. The angles were expressed as the acute angle between the lines shown. RNAi scoring could not be meaningfully blinded because of the strength of the phenotypes, although the dataset shown in Figure 7—figure supplement 1 was nominally blinded before acquisition and unblinded after scoring. Sample sizes were evaluated a posteriori for statistical significance.

Scoring of differentiation markers

Request a detailed protocol

Genomically encoded fluorescent markers were used to analyze cell fate asymmetry across the DTC-Sh1 boundary. The marker of a stem-like fate is sygl-1p::H2B::gfp (Kershner et al., 2014), and we set a threshold in the images to exclude all residual GFP below the expression level of the distal most cells, based on findings that the sygl-1 gene is actively transcribed in a distal subpopulation but its transcripts perdure in more cells further proximal (Lee et al., 2016). We analyzed whether every part of the DTC was in contact with sygl-1p::H2B::GFP (+) cells—in most cases, DTCs had long processes that extended proximal to the proximal boundary of sygl-1p::H2B::GFP expression. Then we analyzed the region of overlap between Sh1 and sygl-1p::H2B::GFP (+) cells, and found it confined to the DTC-Sh1 interface. The marker of the differentiating fate is a gld-1p::gld-1::GFP transgene (ozIs5; Brenner and Schedl, 2016; Schumacher et al., 2005; Seidel et al., 2018). Looking only at the green channel with the DTC marker and GLD-1::GFP, we examined the farthest distal perinuclear GFP speckles that we could detect in the germ line, and marked their position in the image. We also marked the length of the longest DTC process. We then turned on the red channel and noted the position of the distal extent of Sh1. These data are plotted in Figure 5D. The measurements for Sh1 and GLD-1::GFP were made the same way for the RNAi experiment shown in Figure 7F and G.

Statistical analysis

Request a detailed protocol

Linear model followed by the Tukey-Kramer HSD test (nmle package in R), prop.test (stats v3.6.0 package), Welch two-sample t-test (t.test, stats v3.6.2 package), Pearson’s correlation test (cor.test, stats v3.6.2), or one tailed Wilcoxon Signed Rank tests (stats v3.6.0 package) were performed in R as noted in figure legends and text.

Appendix 1

Key Resources Table

Appendix 1—key resources table

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Gene (C. elegans)	inx-8	wormbase.org	Sequence CELE_ZK792.2	Encodes gap junction hemichannel subunit
Gene (C. elegans)	inx-9	wormbase.org	Sequence CELE_ZK792.3	Encodes gap junction hemichannel subunit
Gene (C. elegans)	tbb-2	wormbase.org	Sequence CELE_C36E8.5	Encodes worm beta tubulin ortholog
Gene (C. elegans)	unc-60	wormbase.org	Sequence CELE_C38C3.5	Encodes worm cofilin ortholog
Gene (C. elegans)	arx-2	wormbase.org	Sequence CELE_K07C5.1	Encodes Arp2/3 subunit
Strain, strain background (Escherichia coli)	HT115(DE3)	Caenorhabditis Genetics Center (CGC)	HT115(DE3)	RNAi feeding strain
Genetic reagent (Escherichia coli)	Ahringer RNAi Library	Source Bioscience	C. elegans RNAi Collection (Ahringer)
Chemical compound, drug	hygromycin B from Streptomyces hygroscopicus	Millipore Sigma	CAS 31282-04-9
Genetic reagent (C. elegans)	unc-119(ed4)	Paul Sternberg	PS3460	for transgenic rescue
Genetic reagent (C. elegans)	xnSi1(mex-5p::GFP::PLCδ^PH::nos-2 3’UTR) II; inx-8(qy78(mKate::inx-8)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH)	xnSi1 parent from Chihara and Nance, 2012 doi:10.1242/dev.079863	NK2569	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II/xnSi1(mex-5p::GFP::PLCδ^PH::nos-2 3’UTR) II; inx-8(qy78(mKate::inx-8)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH)	nasi2 transgene from Roy et al., 2018 doi:10.1534/g3.118.200511; xnSi1 parent from Chihara and Nance, 2012 doi:10.1242/dev.079863; others this study	NK2570	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	inx-8(qy78(mKate::inx-8)) IV; cpIs122(lag-2p::mNeonGreen:: PLC^δPH)	This study	NK2571	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	inx-9(qy79(GFP::inx-9)) IV; cpIs91(lag-2p::2x mKate2::PLC^δPH::3xHA::tbb-2 3'UTR LoxN) II	cpIs91 transgene from Gordon et al., 2019 doi: 10.1016/j.cub.2019.01.056	NK2572	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II; inx-9(qy79(GFP::inx-9)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH)	nasi2 transgene from Roy et al., 2018 doi: 10.1534/g3.118.200511; others this study	NK2573	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II; inx-8(qy78(mKate::inx-8)) IV; qyIS546(lag-2p::mTagBFP::PLC^δPH); ojIs1 [pie-1p::GFP::tbb-2 + unc-119(+)]	nasi2 transgene from Roy et al., 2018 doi:10.1534/g3.118.200511; ojIs1 transgene from Gerhold et al., 2015 doi: 10.1016/j.cub.2015.02.054	NK2574	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	naSi2(mex-5p::H2B::mCherry::nos-2 3′UTR) II; inx-8(qy78(mKate::inx-8)) IV; arx-2(cas607[arx‐2::gfp knock‐in]) V	nasi2 transgene from Roy et al., 2018 doi:10.1534/g3.118.200511; arx-2(cas607) from Zhang et al., 2017 doi: 10.1242/bio.026807	NK2575	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	inx-9(ok1502) IV; inx-8(qy102(mKate::inx-8)) IV; cpIs122(lag-2p::mNeonGreen::PLC^δPH)	inx-9(ok1502) allele from Consortium TC elegans DM, 2012 doi: 10.1534/G3.112.003830	NK2576	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	cpIs91(lag-2p::2x mKate2::PLC^δPH::3xHA::tbb-2 3'UTR LoxN) II); qSi26(sygl-1p::H2B::GFP::sygl-1 3UTR) II; (inx-8(qy78(mKate::inx-8)) IV; teIs1(oma-1::GFP) IV,	qSi26 and teIs1 from Kershner et al., 2014 doi: 10.1073/pnas.1401861111	KLG004	Can be obtained from K. Gordon lab
Genetic reagent (C. elegans)	(cpIs122(lag-2p::mNeonGreen:: PLC^δPH)); inx-8(qy78(mKate::inx-8)) IV; ozIs5(gld-1::GFP)	ozIs5 from Brenner and Schedl, 2016; Schumacher et al., 2005 doi: 10.1534/genetics.115.185678	KLG005	Can be obtained from K. Gordon lab
Sequence-based reagent	inx-8 5’ arm Forward primer	This study	5_Nterminx8F	5’-TGTACGACTGTAGGCAGGCAGGTAG-3’
Sequence-based reagent	inx-8 5’ arm Reverse primer	This study	5_Nterminx8R	5’-tctgcaattcgaacccgagtgaagatg-3’ mutation in PAM site underlined
Sequence-based reagent	inx-8 3’ arm Forward primer	This study	3_Nterminx8F	5’-ATGTTTTCTGTTCCATTTCTTACCTC-3’
Sequence-based reagent	inx-8 3’ arm Reverse primer	This study	3_Nterminx8R	5’-CTGTACATCTCCACGGCAACCTCCG-3’
Sequence-based reagent	inx-8 sgRNA primer	This study	sg_Nterminx8	5’- GCATCTTCACTCGGGTTCGA-3’
Sequence-based reagent	inx-9 5’ arm Forward primer	This study	5_Nterminx9F	5’-gaaataatcgagatgaaactgtcg-3
Sequence-based reagent	inx-9 5’ arm Reverse primer	This study	5_Nterminx9R	5’-CATtctgtccctttgaacgaaagtg-3’ mutation in PAM site underlined
Sequence-based reagent	inx-9 3’ arm Forward primer	This study	3_Nterminx9F	5’-ATGTTTTCTGTTCCATTTCTTACC-3’
Sequence-based reagent	inx-9 3’ arm Reverse primer	This study	3_Nterminx9R	5’-tgagttggactgacatcgag-3’
Sequence-based reagent	inx-9 sg RNA primer	This study	sg_Nterminx9	5’- CTTTCAGAGCATTGTCACTT-3’
Sequence-based reagent	inx-8 downstream exon Forward primer	This study	Inx8exonF	5’-GTGACTTCCAAGTTCGTGAGATGGC-3
Sequence-based reagent	downstream of inx-9 stop codon Reverse primer	This study	Inx9downstreamR	5’-gcttgaaaacggtgcggatccagc-3’
Recombinant DNA reagent	arx-2 feeding RNAi clone	Ahringer RNAi library Kamath and Ahringer, 2003 doi: 10.1016/S1046-2023(03)00050–1	HGMP_Location V-7M13
Recombinant DNA reagent	unc-60 feeding RNAi clone	Vidal RNAi library Rual et al., 2004 doi:10.1101/gr.2505604	RNAi well, GHR-11003@H04	RNAi well, GHR-11003@H04
Recombinant DNA reagent	modified plasmid for SEC CRISPR strategy	Dickinson et al., 2015 doi: 10.1534/genetics.115.178335	RRID:Addgene_132523	pDD268
Recombinant DNA reagent	eft-3p::Cas9+sgRNA expression vector	Dickinson et al., 2015 doi: 10.1534/genetics.115.178335	RRID:Addgene_47549	pDD162, Addgene plasmid #47549
Recombinant DNA reagent	RNAi empty vector control	Andrew Fire	RRID:Addgene_1654	L4440, Addgene plasmid #1654
Recombinant DNA reagent	unc-119 rescue plasmid	Morris Maduro	pPD#MM016B
Software, algorithm	μManager software v1.4.18	Edelstein et al., 2010 doi: 10.1002/0471142727.mb1420s92	RRID:SCR_016865	https://micro-manager.org/
Software, algorithm	FIJI 2.0	Schindelin et al., 2012 doi:10.1038/nmeth.2019	RRID:SCR_002285	https://fiji.sc/
Software, algorithm	Adobe Photoshop CC	Adobe Systems Inc	RRID:SCR_014199
Software, algorithm	Adobe Illustrator CC	Adobe Systems Inc	RRID:SCR_010279

Data availability

Source files for all figure graphs have been provided.

References

1. Amini R
2. Goupil E
3. Labella S
4. Zetka M
5. Maddox AS
6. Labbé J-C
7. Chartier NT
(2014) C. elegans Anillin proteins regulate intercellular bridge stability and germline syncytial organization
Journal of Cell Biology 206:129–143.
https://doi.org/10.1083/jcb.201310117
- Google Scholar
1. Aprison EZ
2. Ruvinsky I
(2016) Sexually antagonistic male signals manipulate germline and soma of C. elegans Hermaphrodites
Current Biology 26:2827–2833.
https://doi.org/10.1016/j.cub.2016.08.024
- PubMed
- Google Scholar
1. Austin J
2. Kimble J
(1987) glp-1 Is required in the germ line for regulation of the decision between mitosis and meiosis in C. elegans
Cell 51:589–599.
https://doi.org/10.1016/0092-8674(87)90128-0
- Google Scholar
1. Biedermann B
2. Wright J
3. Senften M
4. Kalchhauser I
5. Sarathy G
6. Lee MH
7. Ciosk R
(2009) Translational repression of cyclin E prevents precocious mitosis and embryonic gene activation during C. elegans meiosis
Developmental Cell 17:355–364.
https://doi.org/10.1016/j.devcel.2009.08.003
- PubMed
- Google Scholar
(2011) Maintaining tissue homeostasis: dynamic control of somatic stem cell activity
Cell Stem Cell 9:402–411.
https://doi.org/10.1016/j.stem.2011.10.004
- PubMed
- Google Scholar
1. Blanpain C
2. Fuchs E
(2009) Epidermal homeostasis: a balancing act of stem cells in the skin
Nature Reviews Molecular Cell Biology 10:207–217.
https://doi.org/10.1038/nrm2636
- Google Scholar
1. Brenner JL
2. Schedl T
(2016) Germline stem cell differentiation entails regional control of cell fate regulator GLD-1 in Caenorhabditis elegans
Genetics 202:1085–1103.
https://doi.org/10.1534/genetics.115.185678
- PubMed
- Google Scholar
1. Byrd DT
2. Knobel K
3. Affeldt K
4. Crittenden SL
5. Kimble J
(2014) A DTC niche plexus surrounds the germline stem cell pool in Caenorhabditis elegans
PLOS ONE 9:e88372.
https://doi.org/10.1371/journal.pone.0088372
- PubMed
- Google Scholar
1. Casanueva MO
2. Ferguson EL
(2004) Germline stem cell number in the Drosophila ovary is regulated by redundant mechanisms that control Dpp signaling
Development 131:1881–1890.
https://doi.org/10.1242/dev.01076
- Google Scholar
1. Chen F
2. Krasnow MA
(2014) Progenitor outgrowth from the niche in Drosophila Trachea is guided by FGF from decaying branches
Science 343:186–189.
https://doi.org/10.1126/science.1241442
- PubMed
- Google Scholar
1. Chihara D
2. Nance J
(2012) An E-cadherin-mediated hitchhiking mechanism for C. elegans germ cell internalization during gastrulation
Development 139:2547–2556.
https://doi.org/10.1242/dev.079863
- Google Scholar
(2016) The genetics of axon guidance and axon regeneration in Caenorhabditis elegans
Genetics 204:849–882.
https://doi.org/10.1534/genetics.115.186262
- PubMed
- Google Scholar
(2010) Progression from a stem cell–like state to early differentiation in the C. elegans germ line
PNAS 107:2048–2053.
https://doi.org/10.1073/pnas.0912704107
- Google Scholar
1. Cinquin A
2. Chiang M
3. Paz A
4. Hallman S
5. Yuan O
6. Vysniauskaite I
7. Fowlkes CC
8. Cinquin O
(2016) Intermittent stem cell cycling balances Self-Renewal and senescence of the C. elegans germ line
PLOS Genetics 12:e1005985.
https://doi.org/10.1371/journal.pgen.1005985
- PubMed
- Google Scholar
1. Consortium TC elegans DM
(2012) Large-scale screening for targeted knockouts in the Caenorhabditis elegans Genome
G3: Genes, Genomes, Genetics 2:1415.
https://doi.org/10.1534/g3.112.003830
- Google Scholar
(2002) A conserved RNA-binding protein controls germline stem cells in Caenorhabditis elegans
Nature 417:660–663.
https://doi.org/10.1038/nature754
- Google Scholar
(2003) Regulation of the mitosis/meiosis decision in the Caenorhabditis elegans germline
Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 358:1359–1362.
https://doi.org/10.1098/rstb.2003.1333
- Google Scholar
(2006) Cellular analyses of the mitotic region in the Caenorhabditis elegans adult germ line
Molecular Biology of the Cell 17:3051–3061.
https://doi.org/10.1091/mbc.e06-03-0170
- PubMed
- Google Scholar
(2013) Engineering the Caenorhabditis elegans genome using Cas9-triggered homologous recombination
Nature Methods 10:1028–1034.
https://doi.org/10.1038/nmeth.2641
- Google Scholar
(2015) Streamlined genome engineering with a Self-Excising drug selection cassette
Genetics 200:1035–1049.
https://doi.org/10.1534/genetics.115.178335
- PubMed
- Google Scholar
1. Dorn JF
2. Zhang L
3. Phi T-T
4. Lacroix B
5. Maddox PS
6. Liu J
7. Maddox AS
(2016) A theoretical model of cytokinesis implicates feedback between membrane curvature and cytoskeletal organization in asymmetric cytokinetic furrowing
Molecular Biology of the Cell 27:1286–1299.
https://doi.org/10.1091/mbc.E15-06-0374
- Google Scholar
1. Doupé DP
2. Klein AM
3. Simons BD
4. Jones PH
(2010) The ordered architecture of murine ear epidermis is maintained by progenitor cells with random fate
Developmental Cell 18:317–323.
https://doi.org/10.1016/j.devcel.2009.12.016
- PubMed
- Google Scholar
1. Edelstein A
2. Amodaj N
3. Hoover K
4. Vale R
5. Stuurman N
(2010) Computer control of microscopes using µmanager
Current Protocols in Molecular Biology 92:mb1420s92.
https://doi.org/10.1002/0471142727.mb1420s92
- Google Scholar
(2015) A somatic permeability barrier around the germline is essential for Drosophila spermatogenesis
Development 142:268–281.
https://doi.org/10.1242/dev.114967
- Google Scholar
(2016) Occluding junctions maintain stem cell niche homeostasis in the fly testes
Current Biology 26:2492–2499.
https://doi.org/10.1016/j.cub.2016.07.012
- PubMed
- Google Scholar
(2018) Wolbachia control stem cell behavior and stimulate germline proliferation in filarial nematodes
Developmental Cell 45:198–211.
https://doi.org/10.1016/j.devcel.2018.03.017
- PubMed
- Google Scholar
1. Fox PM
2. Schedl T
(2015) Analysis of germline stem cell differentiation following loss of GLP-1 notch activity in Caenorhabditis elegans
Genetics 201:167–184.
https://doi.org/10.1534/genetics.115.178061
- PubMed
- Google Scholar
(2003) Caenorhabditis elegans nucleoporins Nup93 and Nup205 determine the limit of nuclear pore complex size exclusion in vivo
Molecular Biology of the Cell 14:5104–5115.
https://doi.org/10.1091/mbc.e03-04-0237
- PubMed
- Google Scholar
(2015) Investigating the regulation of stem and progenitor cell mitotic progression by in situ imaging
Current Biology 25:1123–1134.
https://doi.org/10.1016/j.cub.2015.02.054
- PubMed
- Google Scholar
(2019) Ectopic germ cells can induce Niche-like enwrapment by neighboring body wall muscle
Current Biology 29:823–833.
https://doi.org/10.1016/j.cub.2019.01.056
- PubMed
- Google Scholar
1. Hall DH
2. Winfrey VP
3. Blaeuer G
4. Hoffman LH
5. Furuta T
6. Rose KL
7. Hobert O
8. Greenstein D
(1999) Ultrastructural features of the adult hermaphrodite gonad of Caenorhabditis elegans: relations between the germ line and soma
Developmental Biology 212:101–123.
https://doi.org/10.1006/dbio.1999.9356
- PubMed
- Google Scholar
(2004) Multi-pathway control of the proliferation versus meiotic development decision in the Caenorhabditis elegans germline
Developmental Biology 268:342–357.
https://doi.org/10.1016/j.ydbio.2003.12.023
- Google Scholar
1. Henderson ST
2. Gao D
3. Lambie EJ
4. Kimble J
(1994)
lag-2 may encode a signaling ligand for the GLP-1 and LIN-12 receptors of C. elegans

Development 120:2913–2924.
- Google Scholar
(1976) Development of the reproductive system of Caenorhabditis elegans
Developmental Biology 49:200–219.
https://doi.org/10.1016/0012-1606(76)90267-0
- Google Scholar
1. Hubbard EJ
(2007) Caenorhabditis elegans germ line: a model for stem cell biology
Developmental Dynamics 236:3343–3357.
https://doi.org/10.1002/dvdy.21335
- PubMed
- Google Scholar
1. Jayadev R
2. Chi Q
3. Keeley DP
4. Hastie EL
5. Kelley LC
6. Sherwood DR
(2019) α-Integrins dictate distinct modes of type IV collagen recruitment to basement membranes
Journal of Cell Biology 218:3098–3116.
https://doi.org/10.1083/jcb.201903124
- Google Scholar
(1996) GLD-1, a cytoplasmic protein essential for oocyte differentiation, shows stage- and sex-specific expression during Caenorhabditis elegans germline development
Developmental Biology 180:165–183.
https://doi.org/10.1006/dbio.1996.0293
- PubMed
- Google Scholar
1. Jungkamp AC
2. Stoeckius M
3. Mecenas D
4. Grün D
5. Mastrobuoni G
6. Kempa S
7. Rajewsky N
(2011) In vivo and transcriptome-wide identification of RNA binding protein target sites
Molecular Cell 44:828–840.
https://doi.org/10.1016/j.molcel.2011.11.009
- PubMed
- Google Scholar
1. Kadyk LC
2. Kimble J
(1998)
Genetic regulation of entry into meiosis in Caenorhabditis elegans

Development 125:1803–1813.
- Google Scholar
1. Kamath RS
2. Ahringer J
(2003) Genome-wide RNAi screening in Caenorhabditis elegans
Methods 30:313–321.
https://doi.org/10.1016/S1046-2023(03)00050-1
- PubMed
- Google Scholar
1. Kelley LC
2. Chi Q
3. Cáceres R
4. Hastie E
5. Schindler AJ
6. Jiang Y
7. Matus DQ
8. Plastino J
9. Sherwood DR
(2019) Adaptive F-Actin Polymerization and Localized ATP Production Drive Basement Membrane Invasion in the Absence of MMPs
Developmental Cell 48:313–328.
https://doi.org/10.1016/j.devcel.2018.12.018
- Google Scholar
1. Kershner AM
2. Shin H
3. Hansen TJ
4. Kimble J
(2014) Discovery of two GLP-1/Notch target genes that account for the role of GLP-1/Notch signaling in stem cell maintenance
PNAS 111:3739–3744.
https://doi.org/10.1073/pnas.1401861111
- Google Scholar
1. Killian DJ
2. Hubbard EJA
(2004) C. elegans pro-1 activity is required for soma/germline interactions that influence proliferation and differentiation in the germ line
Development 131:1267–1278.
https://doi.org/10.1242/dev.01002
- Google Scholar
1. Killian DJ
2. Hubbard EJA
(2005) Caenorhabditis elegans germline patterning requires coordinated development of the somatic gonadal sheath and the germ line
Developmental Biology 279:322–335.
https://doi.org/10.1016/j.ydbio.2004.12.021
- Google Scholar
1. Kimble J
2. Hirsh D
(1979) The postembryonic cell lineages of the hermaphrodite and male gonads in Caenorhabditis elegans
Developmental Biology 70:396–417.
https://doi.org/10.1016/0012-1606(79)90035-6
- Google Scholar
1. Kimble J
2. Seidel H
(2013)
StemBook

C. elegans germline stem cells and their niche, StemBook, Harvard Stem Cell Institute, 10.3824/stembook.1.95.1.
- Google Scholar
1. Kimble JE
2. White JG
(1981) On the control of germ cell development in Caenorhabditis elegans
Developmental Biology 81:208–219.
https://doi.org/10.1016/0012-1606(81)90284-0
- Google Scholar
(2019) Rapid population-wide declines in stem cell number and activity during reproductive aging in C. elegans
Development 146:dev173195.
https://doi.org/10.1242/dev.173195
- PubMed
- Google Scholar
1. Kornberg TB
2. Roy S
(2014) Cytonemes as specialized signaling filopodia
Development 141:729–736.
https://doi.org/10.1242/dev.086223
- Google Scholar
1. Korta DZ
2. Hubbard EJA
(2010) Soma-germline interactions that influence germline proliferation in Caenorhabditis elegans
Developmental Dynamics 239:1449–1459.
https://doi.org/10.1002/dvdy.22268
- PubMed
- Google Scholar
1. Lander AD
2. Kimble J
3. Clevers H
4. Fuchs E
5. Montarras D
6. Buckingham M
7. Calof AL
8. Trumpp A
9. Oskarsson T
(2012) What does the concept of the stem cell niche really mean today?
BMC Biology 10:19.
https://doi.org/10.1186/1741-7007-10-19
- Google Scholar
1. Landmann F
2. Bain O
3. Martin C
4. Uni S
5. Taylor MJ
6. Sullivan W
(2012) Both asymmetric mitotic segregation and cell-to-cell invasion are required for stable germline transmission of wolbachia in filarial nematodes
Biology Open 1:536–547.
https://doi.org/10.1242/bio.2012737
- PubMed
- Google Scholar
1. Lee C
2. Sorensen EB
3. Lynch TR
4. Kimble J
(2016) C. elegans GLP-1/Notch activates transcription in a probability gradient across the germline stem cell pool
eLife 5:e18370.
https://doi.org/10.7554/eLife.18370
- PubMed
- Google Scholar
1. Lee C
2. Shin H
3. Kimble J
(2019) Dynamics of Notch-Dependent transcriptional bursting in its native context
Developmental Cell 50:426–435.
https://doi.org/10.1016/j.devcel.2019.07.001
- PubMed
- Google Scholar
1. Li X
2. Johnson RW
3. Park D
4. Chin-Sang I
5. Chamberlin HM
(2012) Somatic gonad sheath cells and Eph receptor signaling promote germ-cell death in C. elegans
Cell Death & Differentiation 19:1080–1089.
https://doi.org/10.1038/cdd.2011.192
- Google Scholar
1. Linden LM
2. Gordon KL
3. Pani AM
4. Payne SG
5. Garde A
6. Burkholder D
7. Chi Q
8. Goldstein B
9. Sherwood DR
(2017) Identification of regulators of germ stem cell enwrapment by its niche in C. elegans
Developmental Biology 429:271–284.
https://doi.org/10.1016/j.ydbio.2017.06.019
- Google Scholar
1. Lints R
2. Hall DH
(2018) Reproductive system
Wormatlas 1:2.
https://doi.org/10.3908/wormatlas.2.1.2018
- Google Scholar
1. Maciejowski J
2. Ugel N
3. Mishra B
4. Isopi M
5. Hubbard EJA
(2006) Quantitative analysis of germline mitosis in adult C. elegans
Developmental Biology 292:142–151.
https://doi.org/10.1016/j.ydbio.2005.12.046
- Google Scholar
1. McCarter J
2. Bartlett B
3. Dang T
4. Schedl T
(1997) Soma-germ cell interactions in Caenorhabditis elegans: multiple events of hermaphrodite germline development require the somatic sheath and spermathecal lineages
Developmental Biology 181:121–143.
https://doi.org/10.1006/dbio.1996.8429
- PubMed
- Google Scholar
1. Ming GL
2. Song H
(2011) Adult neurogenesis in the mammalian brain: significant answers and significant questions
Neuron 70:687–702.
https://doi.org/10.1016/j.neuron.2011.05.001
- PubMed
- Google Scholar
1. Morrison SJ
2. Kimble J
(2006) Asymmetric and symmetric stem-cell divisions in development and cancer
Nature 441:1068–1074.
https://doi.org/10.1038/nature04956
- Google Scholar
1. Murry CE
2. Keller G
(2008) Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development
Cell 132:661–680.
https://doi.org/10.1016/j.cell.2008.02.008
- PubMed
- Google Scholar
(2009) MSP and GLP-1/Notch signaling coordinately regulate actomyosin-dependent cytoplasmic streaming and oocyte growth in C. elegans
Development 136:2223–2234.
https://doi.org/10.1242/dev.034603
- Google Scholar
1. Ono K
2. Parast M
3. Alberico C
4. Benian GM
5. Ono S
(2003) Specific requirement for two ADF/cofilin isoforms in distinct actin-dependent processes in Caenorhabditis elegans
Journal of Cell Science 116:2073–2085.
https://doi.org/10.1242/jcs.00421
- Google Scholar
1. Ono K
2. Yamashiro S
3. Ono S
(2008) Essential role of ADF/cofilin for assembly of contractile actin networks in the C. elegans somatic gonad
Journal of Cell Science 121:2662–2670.
https://doi.org/10.1242/jcs.034215
- Google Scholar
(2016) Atomic structure of the innexin-6 gap junction channel determined by cryo-EM
Nature Communications 7:13681.
https://doi.org/10.1038/ncomms13681
- Google Scholar
1. Pepper AS-R
2. Lo T-W
3. Killian DJ
4. Hall DH
5. Hubbard EJA
(2003) The establishment of Caenorhabditis elegans germline pattern is controlled by overlapping proximal and distal somatic gonad signals
Developmental Biology 259:336–350.
https://doi.org/10.1016/S0012-1606(03)00203-3
- Google Scholar
1. Poulson ND
2. Lechler T
(2010) Robust control of mitotic spindle orientation in the developing epidermis
The Journal of Cell Biology 191:915–922.
https://doi.org/10.1083/jcb.201008001
- Google Scholar
1. Priti A
2. Ong HT
3. Toyama Y
4. Padmanabhan A
5. Dasgupta S
6. Krajnc M
7. Zaidel-Bar R
(2018) Syncytial germline architecture is actively maintained by contraction of an internal actomyosin corset
Nature Communications 9:4694.
https://doi.org/10.1038/s41467-018-07149-2
- Google Scholar
1. Rao M
2. Mattson MP
(2001) Stem cells and aging: expanding the possibilities
Mechanisms of Ageing and Development 122:713–734.
https://doi.org/10.1016/S0047-6374(01)00224-X
- Google Scholar
(2001) Stem cells, cancer, and cancer stem cells
Nature 414:105–111.
https://doi.org/10.1038/35102167
- Google Scholar
(2014) Intestinal crypt homeostasis revealed at single-stem-cell level by in vivo live imaging
Nature 507:362–365.
https://doi.org/10.1038/nature12972
- Google Scholar
1. Roh-Johnson M
2. Shemer G
3. Higgins CD
4. McClellan JH
5. Werts AD
6. Tulu US
7. Gao L
8. Betzig E
9. Kiehart DP
10. Goldstein B
(2012) Triggering a cell shape change by exploiting preexisting actomyosin contractions
Science 335:1232–1235.
https://doi.org/10.1126/science.1217869
- PubMed
- Google Scholar
1. Rompolas P
2. Deschene ER
3. Zito G
4. Gonzalez DG
5. Saotome I
6. Haberman AM
7. Greco V
(2012) Live imaging of stem cell and progeny behaviour in physiological hair-follicle regeneration
Nature 487:496–499.
https://doi.org/10.1038/nature11218
- Google Scholar
1. Rompolas P
2. Mesa KR
3. Kawaguchi K
4. Park S
5. Gonzalez D
6. Brown S
7. Boucher J
8. Klein AM
9. Greco V
(2016) Spatiotemporal coordination of stem cell commitment during epidermal homeostasis
Science 352:1471–1474.
https://doi.org/10.1126/science.aaf7012
- PubMed
- Google Scholar
1. Rosu S
2. Cohen-Fix O
(2017) Live-imaging analysis of germ cell proliferation in the C. elegans adult supports a stochastic model for stem cell proliferation
Developmental Biology 423:93–100.
https://doi.org/10.1016/j.ydbio.2017.02.008
- Google Scholar
1. Roy D
2. Michaelson D
3. Hochman T
4. Santella A
5. Bao Z
6. Goldberg JD
7. Jane E
8. Hubbard A
(2016) Cell cycle features of C. elegans germline stem/progenitor cells vary temporally and spatially HHS public access
Developmental Biology 1:261–271.
https://doi.org/10.1016/j.ydbio.2015.10.031
- Google Scholar
1. Roy D
2. Kahler DJ
3. Yun C
4. Hubbard EJA
(2018) Functional interactions between rsks-1 /s6k, glp-1 /notch, and regulators of Caenorhabditis elegans fertility and germline stem cell maintenance
G3: Genes, Genomes, Genetics 8:3293–3309.
https://doi.org/10.1534/g3.118.200511
- Google Scholar
1. Rual JF
2. Ceron J
3. Koreth J
4. Hao T
5. Nicot AS
6. Hirozane-Kishikawa T
7. Vandenhaute J
8. Orkin SH
9. Hill DE
10. van den Heuvel S
11. Vidal M
(2004) Toward improving Caenorhabditis elegans phenome mapping with an ORFeome-based RNAi library
Genome Research 14:2162–2168.
https://doi.org/10.1101/gr.2505604
- PubMed
- Google Scholar
1. Ruder WC
2. Lu T
3. Collins JJ
(2011) Synthetic biology moving into the clinic
Science 333:1248–1252.
https://doi.org/10.1126/science.1206843
- PubMed
- Google Scholar
1. Sánchez Alvarado A
2. Yamanaka S
(2014) Rethinking differentiation: stem cells, regeneration, and plasticity
Cell 157:110–119.
https://doi.org/10.1016/j.cell.2014.02.041
- PubMed
- Google Scholar
1. Sawa M
2. Suetsugu S
3. Sugimoto A
4. Miki H
5. Yamamoto M
6. Takenawa T
(2003) Essential role of the C. elegans Arp2/3 complex in cell migration during ventral enclosure
Journal of Cell Science 116:1505–1518.
https://doi.org/10.1242/jcs.00362
- Google Scholar
1. Schindelin J
2. Arganda-Carreras I
3. Frise E
4. Kaynig V
5. Longair M
6. Pietzsch T
7. Preibisch S
8. Rueden C
9. Saalfeld S
10. Schmid B
11. Tinevez JY
12. White DJ
13. Hartenstein V
14. Eliceiri K
15. Tomancak P
16. Cardona A
(2012) Fiji: an open-source platform for biological-image analysis
Nature Methods 9:676–682.
https://doi.org/10.1038/nmeth.2019
- PubMed
- Google Scholar
1. Schumacher B
2. Hanazawa M
3. Lee MH
4. Nayak S
5. Volkmann K
6. Hofmann ER
7. Hofmann R
8. Hengartner M
9. Schedl T
10. Gartner A
(2005) Translational repression of C. elegans p53 by GLD-1 regulates DNA damage-induced apoptosis
Cell 120:357–368.
https://doi.org/10.1016/j.cell.2004.12.009
- PubMed
- Google Scholar
1. Seidel HS
2. Smith TA
3. Evans JK
4. Stamper JQ
5. Mast TG
6. Kimble J
(2018) C. elegans germ cells divide and differentiate in a folded tissue
Developmental Biology 442:173–187.
https://doi.org/10.1016/j.ydbio.2018.07.013
- Google Scholar
1. Seidel HS
2. Kimble J
(2015) Cell-cycle quiescence maintains Caenorhabditis elegans germline stem cells independent of GLP-1/Notch
eLife 4:e10832.
https://doi.org/10.7554/eLife.10832
- Google Scholar
1. Sherwood DR
2. Plastino J
(2018) Invading, leading and navigating cells in Caenorhabditis elegans: Insights into Cell Movement in Vivo
Genetics 208:53–78.
https://doi.org/10.1534/genetics.117.300082
- PubMed
- Google Scholar
1. Shin H
2. Haupt KA
3. Kershner AM
4. Kroll-Conner P
5. Wickens M
6. Kimble J
(2017) SYGL-1 and LST-1 link niche signaling to PUF RNA repression for stem cell maintenance in Caenorhabditis elegans
PLOS Genetics 13:e1007121.
https://doi.org/10.1371/journal.pgen.1007121
- PubMed
- Google Scholar
1. Smith CJ
2. Watson JD
3. Spencer WC
4. O'Brien T
5. Cha B
6. Albeg A
7. Treinin M
8. Miller DM
(2010) Time-lapse imaging and cell-specific expression profiling reveal dynamic branching and molecular determinants of a multi-dendritic nociceptor in C. elegans
Developmental Biology 345:18–33.
https://doi.org/10.1016/j.ydbio.2010.05.502
- Google Scholar
(2010) Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells
Cell 143:134–144.
https://doi.org/10.1016/j.cell.2010.09.016
- PubMed
- Google Scholar
(2014) Two classes of gap junction channels mediate soma-germline interactions essential for germline proliferation and gametogenesis in Caenorhabditis elegans
Genetics 198:1127–1153.
https://doi.org/10.1534/genetics.114.168815
- PubMed
- Google Scholar
1. Strome S
2. Powers J
3. Dunn M
4. Reese K
5. Malone CJ
6. White J
7. Seydoux G
8. Saxton W
(2001) Spindle dynamics and the role of gamma-tubulin in early Caenorhabditis elegans embryos
Molecular Biology of the Cell 12:1751–1764.
https://doi.org/10.1091/mbc.12.6.1751
- PubMed
- Google Scholar
1. Tamplin OJ
2. Durand EM
3. Carr LA
4. Childs SJ
5. Hagedorn EJ
6. Li P
7. Yzaguirre AD
8. Speck NA
9. Zon LI
(2015) Hematopoietic stem cell arrival triggers dynamic remodeling of the perivascular niche
Cell 160:241–252.
https://doi.org/10.1016/j.cell.2014.12.032
- PubMed
- Google Scholar
1. van der Flier LG
2. Clevers H
(2009) Stem cells, self-renewal, and differentiation in the intestinal epithelium
Annual Review of Physiology 71:241–260.
https://doi.org/10.1146/annurev.physiol.010908.163145
- PubMed
- Google Scholar
1. Wilson A
2. Laurenti E
3. Oser G
4. van der Wath RC
5. Blanco-Bose W
6. Jaworski M
7. Offner S
8. Dunant CF
9. Eshkind L
10. Bockamp E
11. Lió P
12. Macdonald HR
13. Trumpp A
(2008) Hematopoietic stem cells reversibly switch from dormancy to self-renewal during homeostasis and repair
Cell 135:1118–1129.
https://doi.org/10.1016/j.cell.2008.10.048
- PubMed
- Google Scholar
1. Wong BG
2. Paz A
3. Corrado MA
4. Ramos BR
5. Cinquin A
6. Cinquin O
7. Hui EE
(2013) Live imaging reveals active infiltration of mitotic zone by its stem cell niche
Integrative Biology 5:976–982.
https://doi.org/10.1039/c3ib20291g
- Google Scholar
(2015) The role of symmetric stem cell divisions in tissue homeostasis
PLOS Computational Biology 11:e1004629.
https://doi.org/10.1371/journal.pcbi.1004629
- PubMed
- Google Scholar
1. Zhang Y
2. Yang Y
3. Zhu Z
4. Ou G
(2017) WASP-Arp2/3-dependent actin polymerization influences fusogen localization during cell-cell fusion in Caenorhabditis elegans embryos
Biology Open 6:1324–1328.
https://doi.org/10.1242/bio.026807
- Google Scholar
1. Zhu Z
2. Chai Y
3. Jiang Y
4. Li W
5. Hu H
6. Li W
7. Wu JW
8. Wang ZX
9. Huang S
10. Ou G
(2016) Functional coordination of WAVE and WASP in C. elegans Neuroblast Migration
Developmental Cell 39:224–238.
https://doi.org/10.1016/j.devcel.2016.09.029
- PubMed
- Google Scholar
1. Zhu Z
2. Huangfu D
(2013) Human pluripotent stem cells: an emerging model in developmental biology
Development 140:705–717.
https://doi.org/10.1242/dev.086165
- Google Scholar
1. Zou W
2. Yadav S
3. DeVault L
4. Nung Jan Y
5. Sherwood DR
(2015) RAB-10-Dependent membrane transport is required for dendrite arborization
PLOS Genetics 11:e1005484.
https://doi.org/10.1371/journal.pgen.1005484
- PubMed
- Google Scholar

Decision letter

Yukiko M Yamashita

Reviewing Editor; University of Michigan, United States
Utpal Banerjee

Senior Editor; University of California, Los Angeles, United States
Yukiko M Yamashita

Reviewer; University of Michigan, United States
Dave Hansen

Reviewer

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

This manuscript by Gordon et al., describes germline stem cell divisions in C. elegans and provides compelling evidence that GSCs, which have been believed to be maintained at a population level, may divide asymmetrically to create one stem cell and one differentiating cell. This changes a long standing paradigm in the field. The study is well conceived and executed beautifully. The authors addressed the reviewer comments well, and I believe the manuscript is ready for publication.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "Stem cell niche exit in C. elegans via orientation and segregation of daughter cells by a cryptic cell outside the niche" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Yukiko M Yamashita as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Senior Editor, Utpal Banerjee. The following individual involved in review of your submission has agreed to reveal their identity: Dave Hansen (Reviewer #3).

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we decided that we cannot consider the publication of your manuscript in its current form.

However, all the reviewers are very enthusiastic about your fascinating observation, and see the potential of breakthrough in the field. Converged concern of reviewers is that the current manuscript does not provide the evidence for 1) Sh1 promoting germ cell differentiation, 2) germ cell division at the border of DTC-Sh1 being truly asymmetric (with respect to the cell fate). Whereas the reviewers felt that the claim can be solidified by toning down the text as well, the overall conclusion was that the manuscript would still require some evidence to support fate asymmetry and fate determination by Sh1 processes.

Whereas these issues can be well addressed within the time frame of normal revision, eLife's policy is to invite revisions only when the revision experiments are expected to be straightforward (unlikely changing major conclusions). Therefore, our decision comes in the form of rejection. However, we would like to convey that the reviewers are very enthusiastic about your work, therefore, if you can address the major comments and submit it as new submission, we would be happy to re-review your paper with the same sets of reviewers.

Specific comments are listed below, but as summarized above, the key point for the revision is to provide the evidence for asymmetric cell fates. As far as the main points are addressed, other points may be addressed simply by modifying the text for accuracy.

Reviewer #1:

In this paper, Gordon and colleagues find that Sh1, a somatic cell population, is located right after the distal tip cell, and encapsulate differentiating germ cells to promote their differentiation. They provide series of very intriguing observations describing how undifferentiated cells physically interact with DTC and Sh1, leading to their proposal that Sh1 creates the microenvironment that promote differentiation. They found that Sh1 cells form thin protrusions that interdigitate with DTC protrusions, where germ cells divide at a high frequency in an oriented manner. This is potentially a significant advancement to the field, but I felt that certain claims are not well supported at this point. The following are the main points that I feel need to be more strongly supported.

Although the authors postulate that Sh1 contact stimulates “niche exit” and thus differentiation of germ cells, they show no evidence that Sh1 indeed promotes differentiation. Although it has been well established that DTC is the niche component, and thus by the extension of logic, Sh1, which exhibits mutually exclusive positional pattern, would correlate with differentiation. However, no evidence is presented that Sh1 indeed plays any roles in differentiation. I don't necessarily think that they have to go too far showing signaling mechanism etc that is elicited by Sh1, but they need to show something that demonstrates Sh1's role in differentiation.

Germ cells that are contacting DTC and Sh1 have disproportionally high incidence of division. They further observed that division orientation of these cells is biased such that one daughter will remain attached to the DTC, and the other attached to Sh1. This is certainly a very interesting observation but it remains unclear why division must be promoted there, and why they divide asymmetrically (promotion of cell proliferation and potentially asymmetric division are two entirely distinct process, which do not have to be tied together). Also, other than the asymmetry in the cell contact (DTC vs. Sh1), no other asymmetry was shown (such as one keeps expressing high Notch vs. the other one low Notch), leaving the functional significance of their observation unclear.

Their only attempt to show the significance of Sh1 encapsulation is in Figure 6 by perturbing actin cytoskeleton. The only outcome they show is reduced proliferation of germ cells at the interface. Again, this echoes with the lack of functional significance in early figures and readers would be left wondering what is the evidence for their conclusion that Sh1 promotes differentiation.

May I also point out to a series of nice work by Tanentzapf lab that showed encapsulation of differentiating cells by somatic cell population in Drosophila male germline?

https://www.ncbi.nlm.nih.gov/pubmed/27546574

https://www.ncbi.nlm.nih.gov/pubmed/25503408

In summary, I really like their observations, which potentially provide a breakthrough to the field. However, at this stage, their claims are not strongly supported, I am not asking for whole another paper worth of data (such as which signaling pathway regulates what aspects of their observation etc), but some core data to show the functional significance of Sh1 should be provided.

Reviewer #2:

The manuscript by Gordon et al. addresses the relationship between the spatial structure of the somatic cells interacting with C. elegans germline stem cells on the one hand, and the characteristics of germline mitotic divisions on the other hand. Two observations are particularly compelling: 1) the distal tip cell (DTC) and sheath cell pair 1 (Sh1) are much closer together and may interact more actively than realized; 2) the orientation of germ cell division is less random than previously thought, and tends to correlate with the local edges of the somatic cells. The study relies on live imaging that is beautifully executed and still novel in this field, and opens up an important new line of investigation for the field. Weaknesses are that few mechanistic insights are gained, the functional significance is unclear, and the model oversells the results.

Major criticisms:

1) I would expect the manuscript to build on the observation that germ cell divisions are oriented with respect to the local DTC and Sh1 edges, and to provide substantial tests of the regulation and functional significance of this orientation. This is partially done with unc-60 and arx-2 RNAi, which extend the gap between the DTC and Sh1 and reduce cell division (as scored in a way that is unclear). A major concern with this experiment is that reduced proliferation could be a result of RNAi directly affecting the germ cells themselves (or a subset of those cells), rather than a result of the enlarged DTC-Sh1 gap – especially since the authors note in the Materials and methods that unc-60 RNAi leads to severe germ cell morphological defects. It is not clear why soma-specific (or even DTC- or sheath-cell-specific) RNAi was not performed. Furthermore, the molecular changes in germ cells that underlie reduced proliferation and randomized division planes, or potential downstream consequences such as decreased fertility, are not addressed.

2) The story built by the manuscript around the interesting observations is somewhat tautological, does not distinguish cause from effect, and assigns functional purpose that is not substantiated. This is a prominent problem in the following part of the manuscript: "We next tested the hypothesis that by initiating an exclusive association with Sh1 at birth via polarized cell division, a germ cell could directly escape the influence of the DTC niche and embark upon the path to differentiation. […] DTC-Sh1 interface germ cell division, often followed by Sh1 growth over one daughter cell, is the primary mechanism of stem cell niche exit of germ cells."

a) How is "embarking on differentiation" defined? This study does not include markers of germ cell differentiation

b) It appears implied that germ cells need a specific mechanism to exit the stem cell niche. But given the overall movement of germ cells in the distal to proximal direction, why can a germ cell not just passively exit the niche irrespective of what is happening at the DTC-Sh1 interface?

c) The manuscript assumes that germ cell division is regulated by the Sh1-DTC interface. But why could it not be the other way around? (Or could both be downstream of something else?) The Sh1-DTC interface is a very dynamic structure, as reported in this manuscript. Could Sh1 and the DTC not actively remodel their interface in response to germ cell changes that lead up to division?

d) "In most of these cases (56/64) one daughter remained under the DTC or at the interface and the other ended up under or beside Sh1, suggesting a polarized division." If cell division was randomly oriented, would it not often be the case that a division at the Sh1-DTC interface results in one cell closer to the DTC and one cell closer to Sh1? In other words, how strongly do the reported numbers suggest a "polarized division"?

e) Various places where functional purpose is assumed without substantiation: "enwraps the closest daughter to remove it from the niche" (it is not clear why a cell would need to be enwrapped to leave the niche, or that enwrapping plays any role in that exit), "facilitated niche exit", "Sh1 grows over the Sh1-facing daughter, removing it from the niche"

Reviewer #3:

In this manuscript, Gordon et al. describe their analysis of a pair of somatic cells and the effect that these cells have on the stem cell population in the C. elegans germ line. This stem cell system is intensively studied and has contributed significantly to our general understanding of how the balance between stem cell self-renewal and differentiation is controlled. This manuscript challenges our understanding of the stem cell population in two significant ways. First, that the first pair of somatic sheath cells are quite far away from the somatic distal tip cell that caps the distal end of the gonad. There has long been thought to be a bare region in between the DTC and sheath cells. Gordon et al., through the use of new markers, demonstrate that the sheath cells are present much more distally than previously appreciated, and that the sheath cells appear to coordinate with the DTC as to the extent of their position. Second, Gordon et al. demonstrate that the plane of germ cell division for the cells at the interface between the DTC and sheath cells is largely perpendicular to the shortest distance between the DTC and sheath cells. Previously, it had been reported and accepted that the orientation of germ cell division is largely random. That it may not be random brings up the possibility of asymmetric cell division, whereas previously it was largely thought to be symmetric. These finding significantly change our understanding of the stem cell population in the C. elegans gonad, and stem cell populations in general.

My primary concerns do not have to do with the experiments themselves, but rather with some of the conclusions that were reached. In the Discussion, the authors state "our findings demonstrate that cells outside the niche can play an equal and opposite role to the niche in the balance that maintains stem cell pools with the production of differentiating progeny". At the end of the Introduction, they state "We conclude that Sh1… is an orchestrater of stem cell exit from the niche". While the authors have nicely demonstrated that dividing cells in the DTC-SH1 interface often have one daughter in contact with SH1, and the other with the DTC, they have not directly demonstrated that SH1 helps to regulate the balance between self-renewing and differentiating cells, or that Sh1 helps stem cells exit from the niche. For example, they have not shown that in the absence of SH1 that this balance is affected. Indeed, when they shortened SH1 through RNAi “…the lengths of the progenitor zones were not different between the unc-60 RNAi-treated animals and control”, suggesting that cells were able to enter meiosis (differentiation) normally. Indeed, McCarter et al., 1997, demonstrated that in the absence of any sheath cells, while proliferation levels are lower and cells do not progress through meiosis normally, cells do enter meiosis (differentiate). Therefore, at a minimum, the sheath cells are not “equal” to the DTC (niche) in balancing maintaining stem cell pools and production of differentiating progeny-ablation of DTC renders cells incapable of self-renewal, while cells are still able to enter meiosis in the absence of sheath cells. The authors can easily address this concern by limiting their conclusions and discussing more explicitly the limitations of their data. A brief description of the McCarter results in the Discussion may be of help.

The authors make a very significant observation in that the orientation of cell division does not appear random in the Sh1-DTC interface. From this they conclude that the cell divisions are asymmetric, which I take to mean that the two daughters are different (perhaps one remaining a stem cell, and the other entering a differentiation pathway). I wonder if this is an accurate conclusion. They show that these daughter cells differ with respect to which somatic cells contact the daughters, and perhaps their position in the gonad arm; however, there isn't really any data that the two daughter cells are different (markers, entered a differentiation pathway, etc.). Perhaps the authors need to be more less strong in the conclusion that this is asymmetric cell division. This could be easily addressed.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for submitting your article "Stem cell niche exit in C. elegans via orientation and segregation of daughter cells by a cryptic cell outside the niche" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Utpal Banerjee as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Yukiko M Yamashita (Reviewer #1); Dave Hansen (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission. Please aim to submit the revised version within two months, but we are happy to extend this timeframe if needed.

This manuscript by Gordon et al., describes germline stem cell divisions in C. elegans and provide a compelling evidence that GSCs, which have been believed to be maintained at a population level, may divide asymmetrically to create one stem cell and one differentiating cell. This changes a long standing paradigm in the field. The study is well conceived and executed beautifully.

The reviewers raised several minor points that can all be addressed by textual changes. We anticipate that the manuscript will be accepted without further review. Please provide point-by-point responses together with your revised manuscript.

Reviewer #1:

The manuscript by Gordon et al. reports an important discovery regarding the mode of stem cell division in C. elegans germline. It is well established that the distal tip cell (DTC) serves as a niche component of C. elegans germline stem cells (GSCs): any germ cells that are associated with the DTC maintains GSC identity, and those that are “pushed away” from DTC-covered region will initiate meiotic program. It has been long believed that GSCs do not divide asymmetrically, and are maintained as a population. This paper fundamentally changes this view by finding that sheath cell (Sh1) generates an alternate pattern as DTC (DTC's cellular processes and Sh1's cellular processes intercalate), and along this boundary between DTC and Sh1, GSCs likely divide asymmetrically.

Their live imaging provides convincing case that GSC divisions, which happens most frequently at the border of DTC and Sh1, are mostly oriented such that one daughter remain associated with DTC, and the other associated with Sh1.

In the previously-reviewed version, the authors did not provide the evidence that this oriented division and daughter cells' association with DTC vs. Sh1 indeed correlates with asymmetric cell fate. In the revised version, they provide convincing evidence that indeed DTC association vs. Sh1 association nicely correlates with stem cell fate (monitored by sygl-1 expression) vs. differentiating cell fate (monitored by gld-1 expression).

Together, this study significantly contributes to our understanding of how the stem cell niche is formed in C. elegans gonad. Also, the study reveals that C. elegans GSCs divide asymmetrically, completely flipping the previous notion (and the evidence is very strong). I do not have any major concerns at this point, and I predict that this study will be a landmark paper that will be cited for a long time.

Reviewer #2:

I appreciate the efforts the authors made to address the reviewers' comments and strengthen their conclusions. This manuscript reports beautifully-executed experiments, and makes a significant contribution to the C. elegans germline field and, more broadly, to the stem cell field. I am not completely sold in on the implications of the term "facilitated niche exit", but the data are solid, the observations are highly novel and impactful, and the authors are well within their right to propose this interpretation.

There is just one major point left that I cannot wrap my head around. The authors state: "we never saw spontaneous growth of Sh1 over a non-dividing cell, nor displacement of germ cells from the niche by distal-to-proximal pushing produced by other cells dividing entirely under the DTC". If that is correct, and if cells can only leave the "DTC region" by a mitotic division with one daughter cell that remains in the DTC region, then division of cells "entirely under the DTC" should cause a continuous increase in the number of cells in that region. Is that really compatible with the data? I think it would be useful for the authors to resolve this point.

Reviewer #3:

As I stated when I first reviewed this paper, Gordon et al. describe their analysis of a pair of somatic cells and the effect that these cells have on the stem cell population in the C. elegans germ line. This manuscript challenges our understanding of the stem cell population in two significant ways. First, that the first pair of somatic sheath cells are quite far away from the somatic distal tip cell that caps the distal end of the gonad. There has long been thought to be a bare region in between the DTC and sheath cells. Gordon et al., through the use of new markers, demonstrate that the sheath cells are present much more distally than previously appreciated, and that the sheath cells appear to coordinate with the DTC as to the extent of their position. Second, Gordon et al. demonstrate that the plane of germ cell division for the cells at the interface between the DTC and sheath cells is largely perpendicular to the shortest distance between the DTC and sheath cells. Previously, it had been reported and accepted that the orientation of germ cell division is largely random. That it may not be random brings up the possibility of asymmetric cell division, whereas previously it was largely thought to be symmetric. Since the last submission, the authors have significantly improved the manuscript. Importantly, they have more accurately described the conclusions that can be made based on their data. They have also added more data that supports the model of daughters of asymmetrically divided cells are different. They do this through the use of GLD-1 and SYGL-1. This is a valuable experiment and supports their model; however, it does have its limitations. For example, I am somewhat confused as to their model for GLD-1 expression. As they mention, previous experiments have shown that the sometimes observed steps of increasing GLD-1 accumulation could be due to folds. Other than these dramatic steps, I thought the increase in GLD-1 accumulation is gradual. Therefore, I am not clear as to the boundary of GLD-1 accumulation that they are showing on Figure 5. Is this where GLD-1 is first detected, or is it at one of the steps? If it is where it is first detected, then this seems to contradict previous studies that show GLD-1 accumulation all the way to the distal end of the gonad, though at low levels. If it is at a step, and if these steps are due to folds, then this increase may not be due to the position of SH1. Is there some way that the folds are affected by SH1 position, or vice versa? Therefore, the use of GLD-1 as a marker could use some additional explanation. The interpretation of this data was made more difficult by the absence of a figure legend for Figure 5. It seems like the labeled Figure legend 5 corresponds to new Figure 6.

https://doi.org/10.7554/eLife.56383.sa1

Author response

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #1:

In this paper, Gordon and colleagues find that Sh1, a somatic cell population, is located right after the distal tip cell, and encapsulate differentiating germ cells to promote their differentiation. They provide series of very intriguing observations describing how undifferentiated cells physically interact with DTC and Sh1, leading to their proposal that Sh1 creates the microenvironment that promote differentiation. They found that Sh1 cells form thin protrusions that interdigitate with DTC protrusions, where germ cells divide at a high frequency in an oriented manner. This is potentially a significant advancement to the field, but I felt that certain claims are not well supported at this point. The following are the main points that I feel need to be more strongly supported.

Although the authors postulate that Sh1 contact stimulates “niche exit” and thus differentiation of germ cells, they show no evidence that Sh1 indeed promotes differentiation. Although it has been well established that DTC is the niche component, and thus by the extension of logic, Sh1, which exhibits mutually exclusive positional pattern, would correlate with differentiation. However, no evidence is presented that Sh1 indeed plays any roles in differentiation. I don't necessarily think that they have to go too far showing signaling mechanism etc that is elicited by Sh1, but they need to show something that demonstrates Sh1's role in differentiation.

We appreciate that all reviewers made a similar recommendation, and have since added a new figure (Figure 5) showing the positional relationships among DTC, Sh1, and two germ cell fate markers of stem-like cells (sygl-1p::H2B::GFP (Kershner et al., 2014)) and differentiating cells (gld-1::GFP (Brenner and Schedl, 2016)), which correlate with the distal and proximal sides of the DTC-Sh1 boundary, respectively. We also show that when we perturb Sh1 position by unc-60 RNAi, the distal boundary of GLD-1::GFP expression is shifted proximally (Figure 7), suggesting that Sh1 promotes entry into the differentiation program. As Sh1 is not absolutely required for differentiation (see discussion of McCarter et al., 1997 and Killian and Hubbard, 2005), what our data suggest is that Sh1may act to remove germ cells more effectively from the influence of the DTC/Notch signaling through promoting asymmetric cell division and subsequently its enwrapping Sh1-facing daughters might contribute to preventing DTC contact and Notch signaling. We have also now modified the Discussion as follows:

“Instead, we hypothesize that Sh1 promotes differentiation indirectly through its positional exclusion of the DTC from germ cells born from polarized interface divisions (Figure 8). Since these cells lack DTC-contact-mediated Notch signaling from birth, they can derepress mitosis-promoting factors and eventually differentiate (Brenner and Schedl, 2016).”

Germ cells that are contacting DTC and Sh1 have disproportionally high incidence of division. They further observed that division orientation of these cells is biased such that one daughter will remain attached to the DTC, and the other attached to Sh1. This is certainly a very interesting observation but it remains unclear why division must be promoted there, and why they divide asymmetrically (promotion of cell proliferation and potentially asymmetric division are two entirely distinct process, which do not have to be tied together).

We agree that increased proliferation and asymmetry do not need to be tied together, but nevertheless we observe these two phenomena concurrently at the DTC-Sh1 interface. In our field generally there is a conflation of mitotic fate and mitotic division. While the DTC promotes the mitotic fate through active Notch signaling, it is not yet known what factors stimulate proliferation in germ cells. As it is clear from our data that both the DTC and Sh1 cells are associated with proliferating germ cells, perhaps proliferative signals from both of these somatic cells have an additive effect on the germ cells at the interface. The interface is the only part of the gonad where the soma can create asymmetric cell contacts (Figure 6), and we have now shown that this asymmetry correlates with cell fate asymmetry (Figure 5).

Also, other than the asymmetry in the cell contact (DTC vs. Sh1), no other asymmetry was shown (such as one keeps expressing high Notch vs. the other one low Notch), leaving the functional significance of their observation unclear.

The new Figure 5 with germ cell fate markers helps address this question of asymmetry. We show that the marker of active Notch signaling sygl-1p::H2B::GFP (Kershner et al., 2014) is found under the DTC and the marker of early differentiation gld-1::GFP (Brenner and Schedl, 2016) is found under Sh1.

Their only attempt to show the significance of Sh1 encapsulation is in Figure 6 by perturbing actin cytoskeleton. The only outcome they show is reduced proliferation of germ cells at the interface. Again, this echoes with the lack of functional significance in early figures and readers would be left wondering what is the evidence for their conclusion that Sh1 promotes differentiation.

In addition to discovering that markers of germ cell fate sygl-1p::H2B::GFP (Kershner et al., 2014) and gld-1::GFP (Brenner and Schedl, 2016) segregate at the DTC-Sh1 boundary, we also tested the effect of unc-60 RNAi on the boundary of GLD-1::GFP. This experiment is technically challenging, as early RNAi treatments with unc-60 cause more severe systemic defects in the worm, but late exposure to the RNAi was not sufficient to cause any defect by the 24 hours after L4 time point that we characterized the GLD-1::GFP boundary for. Nevertheless, we present data in Figure 7 that shows that the Sh1 phenotype caused by 40+ hours on unc-60 RNAi—a stringy and proximally displaced distal edge—is correlated with a proximal displacement of GLD-1::GFP, the differentiation marker, which provides evidence that the Sh1 cell promotes germ cell differentiation.

May I also point out to a series of nice work by Tanentzapf lab that showed encapsulation of differentiating cells by somatic cell population in Drosophila male germline?

https://www.ncbi.nlm.nih.gov/pubmed/27546574

https://www.ncbi.nlm.nih.gov/pubmed/25503408

Thank you for pointing out this important literature. It is certainly relevant to the question of how niche-adjacent cells can limit the influence of the stem cell niche. We have now cited this work:

“Some niche-adjacent cells limit the range of niche cues to restrict the stem cell population, like the Drosophila testis cyst cells (Fairchild et al., 2016, 2015).”

In summary, I really like their observations, which potentially provide a breakthrough to the field. However, at this stage, their claims are not strongly supported, I am not asking for whole another paper worth of data (such as which signaling pathway regulates what aspects of their observation etc), but some core data to show the functional significance of Sh1 should be provided.

Thank you for recognizing the potential of our findings to open up new areas of inquiry in a canonical stem cell niche model system. We took to heart the suggestion to pursue markers of germ cell fate as a way to directly interrogate the relationship between germ cell differentiation and somatic cell contact (see Figures 5 and 7, and their associated Results sections).

Reviewer #2:

The manuscript by Gordon et al. addresses the relationship between the spatial structure of the somatic cells interacting with C. elegans germline stem cells on the one hand, and the characteristics of germline mitotic divisions on the other hand. Two observations are particularly compelling: 1) the distal tip cell (DTC) and sheath cell pair 1 (Sh1) are much closer together and may interact more actively than realized; 2) the orientation of germ cell division is less random than previously thought, and tends to correlate with the local edges of the somatic cells. The study relies on live imaging that is beautifully executed and still novel in this field, and opens up an important new line of investigation for the field. Weaknesses are that few mechanistic insights are gained, the functional significance is unclear, and the model oversells the results.

We thank the reviewer for identifying the most important findings of our study, and for the kind words about the live imaging experiments. We especially are grateful for the moderation the reviewer urges in not over-interpreting the findings we have made. We have now toned-down our interpretation (see following responses) and provided new experiments that more strongly support the relationship among the DTC, Sh1, and the proliferative germ cells.

Major criticisms:

1) I would expect the manuscript to build on the observation that germ cell divisions are oriented with respect to the local DTC and Sh1 edges, and to provide substantial tests of the regulation and functional significance of this orientation. This is partially done with unc-60 and arx-2 RNAi, which extend the gap between the DTC and Sh1 and reduce cell division (as scored in a way that is unclear). A major concern with this experiment is that reduced proliferation could be a result of RNAi directly affecting the germ cells themselves (or a subset of those cells), rather than a result of the enlarged DTC-Sh1 gap – especially since the authors note in the Materials and methods that unc-60 RNAi leads to severe germ cell morphological defects. It is not clear why soma-specific (or even DTC- or sheath-cell-specific) RNAi was not performed. Furthermore, the molecular changes in germ cells that underlie reduced proliferation and randomized division planes, or potential downstream consequences such as decreased fertility, are not addressed.

We thank the reviewer for pointing out the need for improved quantification and description of the germ cell division phenotype that we describe upon unc-60 RNAi treatment in Figure 7. We have presented the data in a histogram like that shown in Figure 3. This analysis revealed that the total number and positions of the DTC-associated germ cells is relatively similar (among RNAi-treated animals with germ cell divisions). The DTC-Sh1 boundary divisions that are almost entirely absent after RNAi treatment account for the overall diminished proliferation observed after RNAi treatment. These results strongly supporting the notion that unc-60 RNAi is not generally eliminating cell division but rather that the reduction we are seeing is a result of the loss of Sh1 processes at the DTC-Sh1 boundary (see Results section):

“We assessed the positions and somatic cell contacts of these dividing cells as in Figure 3E. The numbers and positions of actively dividing germ cells in contact with the DTC alone are similar in the control and unc-60 RNAi-treated animals (Figure 7C yellow solid and hashed), indicating that the RNAi treatment did not dramatically affect divisions of germ cells in the DTC plexus. However, dividing germ cells at the interface were almost entirely absent because the interface was largely eliminated (compare solid pink to hashed pink, Figure 7C). If we focus on the animals with active divisions (ignoring the 40% of RNAi-treated animals that lack dividing cells entirely), the control (n = 46/26 or 1.77 DTC divisions/animal) and treatment (n = 38/21 or 1.81 DTC divisions/animal) groups have roughly equivalent average numbers of DTC-associated divisions. However, the number of DTC-Sh1 interface divisions is dramatically different between control (n = 23/26 or 0.88 DTC-Sh1 divisions/animal) and treatment (n = 5/21 or 0.24 DTC-Sh1 divisions/animal).”

We think the presence of mitotic figures and a normal sized mitotic zone suggests that germ cell autonomous unc-60 RNAi defects are not the driving force behind the defects we measured. Downstream fertility defects were not analyzed because of embryonic effects of maternal unc-60 RNAi.

2) The story built by the manuscript around the interesting observations is somewhat tautological, does not distinguish cause from effect, and assigns functional purpose that is not substantiated. This is a prominent problem in the following part of the manuscript: "We next tested the hypothesis that by initiating an exclusive association with Sh1 at birth via polarized cell division, a germ cell could directly escape the influence of the DTC niche and embark upon the path to differentiation. […] DTC-Sh1 interface germ cell division, often followed by Sh1 growth over one daughter cell, is the primary mechanism of stem cell niche exit of germ cells."

a) How is "embarking on differentiation" defined? This study does not include markers of germ cell differentiation

“Embarking on the path to differentiation” is a commonly used phrase in the C. elegans germ line literature (see Crittenden, 2006, Kimble and Crittenden, 2005 Wormbook chapter), but we thank the reviewer for pointing out that it risks being confusing to a broader audience and we have removed it. It refers to cells of the progenitor zone that are no longer stem-like (have lost contact with the DTC), but may still divide mitotically. These cells will go on to differentiate, but they are not differentiated yet:

“We next asked how germ cells could escape this contact-mediated signal from the DTC niche and become able to differentiate. We hypothesized that these cells might either lose DTC contact and gain Sh1 associations by moving, initiate an exclusive association with Sh1 at birth via polarized cell division, or both.”

Low levels of GLD-1::GFP (Brenner and Schedl, 2016) mark this cell fate, which we now show in Figure 5. The suggestion to look at markers of cell fate was very helpful.

b) It appears implied that germ cells need a specific mechanism to exit the stem cell niche. But given the overall movement of germ cells in the distal to proximal direction, why can a germ cell not just passively exit the niche irrespective of what is happening at the DTC-Sh1 interface?

The reviewer is correct that it is theoretically possible that germ cells are able to leave the niche passively. However, over a combined ~12 hours of time-lapse imaging of 468 cells at the DTC-Sh1 interface, we never saw a cell leave contact with the DTC without dividing at the interface and one daughter becoming associated with Sh1. Furthermore, we never saw a cell divide in contact with the DTC alone and give rise to a daughter cell that ended up out of DTC contact. This is described in the Results section:

“These dividing interface cells were the only cells (out of 468 total interface cells examined) that changed their germ cell-somatic cell contacts during observation. Similarly, out of the 55 scored cell divisions that occurred in contact with the DTC alone (Supplementary file 1), none of the daughters appeared to leave the niche, so division alone is not sufficient to displace a daughter cell out of the niche.”

We hypothesize that niche exit is one of many biological events that is overdetermined, or for which there is an added layer of regulatory control.

c) The manuscript assumes that germ cell division is regulated by the Sh1-DTC interface. But why could it not be the other way around? (Or could both be downstream of something else?) The Sh1-DTC interface is a very dynamic structure, as reported in this manuscript. Could Sh1 and the DTC not actively remodel their interface in response to germ cell changes that lead up to division?

We thank the reviewer for this very thoughtful comment. Indeed, we favor the hypothesis that Sh1 growth is stimulated by the precursors of cytokinesis. The fact that a germ cell’s spindles orient to the DTC and Sh1 points of contact before it divides suggests that the interaction between the germ cells and Sh1 initiates before division. We present these data and their interpretation in the following Results section:

“Spindle orientation was observed across the DTC-Sh1 interface prior to division, which is consistent with a model in which the DTC-Sh1 interface dictates the asymmetric division (as opposed to the completed cell division later triggering remodeling at the DTC-Sh1 interface).”

d) "In most of these cases (56/64) one daughter remained under the DTC or at the interface and the other ended up under or beside Sh1, suggesting a polarized division." If cell division was randomly oriented, would it not often be the case that a division at the Sh1-DTC interface results in one cell closer to the DTC and one cell closer to Sh1? In other words, how strongly do the reported numbers suggest a "polarized division"?

We thank the reviewer for this probing question, and agree that the positional data alone is suggestive but not definitive evidence for oriented division. Stronger evidence comes from investigating the spindle positions of dividing cells at the interface with marked DTC, Sh1, germ cell histones, and tubulin (Figure 6). If the spindles were not polarized between the DTC and Sh1, then we would expect our rosette plot in Figure 6E to resemble the plot in Figure 6C, which shows no relationship between the spindle orientation and the gonad axis. Instead, the division angles to the DTC-Sh1 axis are significantly smaller (more aligned).

e) Various places where functional purpose is assumed without substantiation: "enwraps the closest daughter to remove it from the niche" (it is not clear why a cell would need to be enwrapped to leave the niche, or that enwrapping plays any role in that exit), "facilitated niche exit", "Sh1 grows over the Sh1-facing daughter, removing it from the niche"

Thank you for this helpful observation. Our fate markers give us additional evidence niche exit occurs when a germ cell is in contact with Sh1 and not the DTC, we have modified the language to remove any implication of goal-directed behavior by cells, or to reach beyond the functions we have evidence to support.

The first quoted phrase has been replaced:

“…we made the surprising discovery that some cells in the C. elegans germ line stem cell niche divide in an asymmetric manner towards a niche-adjacent cell. Germ cells that contact this cell instead of the niche have begun to differentiate”

The last quoted phrase has been replaced:

“Sh1 actively grows over and between daughter cells that divide at the interface in an Arp2/3- and cofilin-dependent manner, which strengthens the association between one daughter cell and the sheath.”

Additionally, we changed the Abstract that read: “and Sh1 grew over the Sh1-facing daughter, removing it from the niche,” to read as follows: “…and Sh1 grew over the Sh1-facing daughter.”

The word “remove” no longer appears in the manuscript.

“facilitated niche exit” appeared two times in the Discussion section. We deleted one, and the other has been changed as follows:

“This discovery raises the possibility that facilitated niche exit is part of the repertoire of cellular behaviors that regulate the balance between stem cell renewal and differentiation and a new player to a canonical stem cell niche system.”

We were surprised to find that this concept—a cell or structure that removes differentiating progeny from a stem cell niche—is not represented in the literature. While this first paper may not prove that the Sh1 cell is performing this role, we do feel it is a justified point of discussion to name the phenomenon.

Reviewer #3:

In this manuscript, Gordon et al. describe their analysis of a pair of somatic cells and the effect that these cells have on the stem cell population in the C. elegans germ line. This stem cell system is intensively studied and has contributed significantly to our general understanding of how the balance between stem cell self-renewal and differentiation is controlled. This manuscript challenges our understanding of the stem cell population in two significant ways. First, that the first pair of somatic sheath cells are quite far away from the somatic distal tip cell that caps the distal end of the gonad. There has long been thought to be a bare region in between the DTC and sheath cells. Gordon et al., through the use of new markers, demonstrate that the sheath cells are present much more distally than previously appreciated, and that the sheath cells appear to coordinate with the DTC as to the extent of their position. Second, Gordon et al. demonstrate that the plane of germ cell division for the cells at the interface between the DTC and sheath cells is largely perpendicular to the shortest distance between the DTC and sheath cells. Previously, it had been reported and accepted that the orientation of germ cell division is largely random. That it may not be random brings up the possibility of asymmetric cell division, whereas previously it was largely thought to be symmetric. These finding significantly change our understanding of the stem cell population in the C. elegans gonad, and stem cell populations in general.

My primary concerns do not have to do with the experiments themselves, but rather with some of the conclusions that were reached. In the Discussion, the authors state "our findings demonstrate that cells outside the niche can play an equal and opposite role to the niche in the balance that maintains stem cell pools with the production of differentiating progeny".

We appreciate greatly that the reviewer recognizes the importance of our work to the field of stem cell biology. In this statement, they are correct that “equal” is an overstatement and this sentence has been removed.

At the end of the Introduction, they state "We conclude that Sh1… is an orchestrater of stem cell exit from the niche".

We thank the reviewer for highlighting this language and in the interest of not overinterpreting our results (also mentioned by the other reviewers), we have removed the reference to “orchestration” at the end of the Introduction.

While the authors have nicely demonstrated that dividing cells in the DTC-SH1 interface often have one daughter in contact with SH1, and the other with the DTC, they have not directly demonstrated that SH1 helps to regulate the balance between self-renewing and differentiating cells, or that Sh1 helps stem cells exit from the niche. For example, they have not shown that in the absence of SH1 that this balance is affected. Indeed, when they shortened SH1 through RNAi “…the lengths of the progenitor zones were not different between the unc-60 RNAi-treated animals and control”, suggesting that cells were able to enter meiosis (differentiation) normally. Indeed, McCarter et al., 1997, demonstrated that in the absence of any sheath cells, while proliferation levels are lower and cells do not progress through meiosis normally, cells do enter meiosis (differentiate). Therefore, at a minimum, the sheath cells are not “equal” to the DTC (niche) in balancing maintaining stem cell pools and production of differentiating progeny-ablation of DTC renders cells incapable of self-renewal, while cells are still able to enter meiosis in the absence of sheath cells. The authors can easily address this concern by limiting their conclusions and discussing more explicitly the limitations of their data. A brief description of the McCarter results in the Discussion may be of help.

We were inspired by this reviewer’s thoughtful comment to directly test GLD-1::GFP expression (a meiosis-promoting factor that is repressed in stem cells by active Notch signaling) in the unc-60 RNAi-treated worms. We found a significant but minor shift away from the distal end (~8 μm) in treated vs. control animals. The magnitude of the change may explain why the overall length of the progenitor zone is not different between controls and RNAi treated animals.

We agree that McCarter et al., 1997 is the most comprehensive previous work done on Sh1 and that it demonstrates germ cells are able to differentiate in the absence of Sh1, meaning Sh1 is not required for differentiation as the DTC is for the mitotic fate. We have removed the phrasing “equal” and have expanded the Discussion of McCarter et al., 1997:

“Our results show that Sh1 is not required for germ cell differentiation—GLD-1::GFP is visible in germ cells in the gap between the DTC and Sh1 when Sh1 is shifted proximally—much as we would expect given that differentiation is still observed in the reduced germ cell pools of sheath-ablated animals (McCarter et al., 1997 and Killian and Hubbard, 2005). Instead, we hypothesize that Sh1 promotes differentiation indirectly through its positional exclusion of the DTC from germ cells born from polarized interface divisions (Figure 8).”

The authors make a very significant observation in that the orientation of cell division does not appear random in the Sh1-DTC interface. From this they conclude that the cell divisions are asymmetric, which I take to mean that the two daughters are different (perhaps one remaining a stem cell, and the other entering a differentiation pathway). I wonder if this is an accurate conclusion. They show that these daughter cells differ with respect to which somatic cells contact the daughters, and perhaps their position in the gonad arm; however, there isn't really any data that the two daughter cells are different (markers, entered a differentiation pathway, etc.). Perhaps the authors need to be more less strong in the conclusion that this is asymmetric cell division. This could be easily addressed.

We appreciate this comment, and it agrees with the other reviewers’ comments that more experimental support was needed to make this conclusion. We have now included two germ cell fate markers, one for stem-like cells (sygl-1p::H2B::GFP, Kershner et al., 2014) and one for cells that will differentiate (gld-1::GFP, Brenner and Schedl, 2016). The data are shown in Figure 5, and are consistent with our cell biological studies of asymmetry across the DTC-Sh1 interface. In animals in the first day of adulthood, all of the GLD-1::GFP positive germ cells are associate with Sh1, suggesting germ cells associating with the Sh1 cell have left the niche and begun to differentiate. Additionally, all of the sygl-1p::H2B::GFP positive cells are limited to the region distal to the DTC-Sh1 interface, inclusive, and do not correlate with long DTC processes.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Reviewer #2:

I appreciate the efforts the authors made to address the reviewers' comments and strengthen their conclusions. This manuscript reports beautifully-executed experiments, and makes a significant contribution to the C. elegans germline field and, more broadly, to the stem cell field. I am not completely sold in on the implications of the term "facilitated niche exit", but the data are solid, the observations are highly novel and impactful, and the authors are well within their right to propose this interpretation.

There is just one major point left that I cannot wrap my head around. The authors state: "we never saw spontaneous growth of Sh1 over a non-dividing cell, nor displacement of germ cells from the niche by distal-to-proximal pushing produced by other cells dividing entirely under the DTC". If that is correct, and if cells can only leave the "DTC region" by a mitotic division with one daughter cell that remains in the DTC region, then division of cells "entirely under the DTC" should cause a continuous increase in the number of cells in that region. Is that really compatible with the data? I think it would be useful for the authors to resolve this point.

Thank you for the astute observation—the number of DTC-alone associated germ cells does indeed increase dramatically from the end of larval gonad elongation through the first days of egg laying.

The reason we do not directly address this question is that there is still somewhat surprising disagreement about the cell cycle length of mitotic germ cells in the adult. The cell cycle length in the adult proliferative zone has been estimated at 5-6 h (Rosu and Cohen-Fix, 2017; Chiang et al., 2015), to 9-12 h (Fox and Schedl, 2015), up to 16-24 hours (Crittenden et al., 2006). My data do not directly bear on this question, so I decided to avoid commenting on the issue in the manuscript.

However, I can take these published estimates and some data I have at home on hard drives to try to answer the question.

I find an average of 10 germ cells under the DTC alone in animals of the last larval stage, L4 (this is unpublished data from a strain imaged at the stage shown in Figure 2 “end migration”).

A day later an average of 34 germ cells are under the DTC alone (This number comes from doubling the 17 germ cells on average shown in Figure 3B “DTC-alone”, since we imaged only the superficial half of the gonad; it’s a rough estimate admittedly).

If we have 10 cells under the DTC alone x doubling rate of 2/12 hours x 24 hours = 40, which is close enough to an average of 34.

We can apply the doubling rate again.

34 cells under the DTC alone x doubling 2/12 hours x 24 hours = 136.

Counting the number of germ cells under the DTC of worms aged 48 hours after the L4 stage for two different strains (one with the DTC and germ cell histones marked, and one with GFP::INX-9 and germ cell histones marked), we get averages of 145 and 125 germ cells, respectively (this is also unpublished data, counted in a dozen full-thickness Z-stacks I made as in Author response image 1; two different projections were made through the top and bottom halves of the gonad to avoid double counting or losing cells by projecting them onto one another). The estimate of 136 falls right into this observed range.

Author response image 1

Download asset Open asset

Of course these calculations assume all germ cells divide at the same rate, which we do not believe is true, and it assumes a division rate in the middle of the pack of estimates, which may or may not ultimately be accurate. Actual cell cycles may also vary among the DTC-associated germ cells either over time or spatially within the niche. In fact, as we note in the manuscript, the peak in cell division frequency observed by Maciejowski et al., 2006, overlaps the region of the DTC-Sh1 interface, with a lower rate of division distal to this peak, so cells under the DTC alone may correspond to those that were found in a row-by-row analysis by Maciejowski et al., 2006, to have a lower frequency of division/longer cell cycle length.

Additionally, Chiang et al., 2015, find evidence for a population that balances longer lived, slower cycling distal stem cells with faster dividing progeny that subsequently differentiate, with about a ~1.5-2 fold difference between the slower-cycling cells at the distal end and those more proximal mitotic cells closer to ~10 germ cell diameters from the distal end of the gonad (this corresponds to the neighborhood of the DTC-Sh1 interface).

Finally, remodeling at the distal edge of Sh1 combined with proliferation of DTC-associated germ cells must sometimes lead to a daughter of a DTC-only-associated germ cell landing at the DTC-Sh1 interface, where its subsequent divisions will result in daughters leaving the niche. Longer-term time-lapsing will eventually shed light onto these dynamics.

Taken together, it does appear feasible that germ cells born under the DTC-alone can remain associated with the DTC for the medium and slow estimates of germ cell proliferation rate. Future studies will determine whether these cells ever contribute to gamete formation, if there are certain environmental conditions that “tap into” this reservoir of distal germ cells, or if this is perhaps a hedge against germ cell death of progenitors outside the niche. Alternatively, as we expand our long-term imaging technique, we may see that some germ cells do indeed leave the niche without dividing, albeit at a rate that is too infrequent to have captured with our time-lapsing so far.

https://doi.org/10.7554/eLife.56383.sa2

Article and author information

Author details

Kacy L Gordon

Department of Biology, The University of North Carolina at Chapel Hill, Chapel Hill, United States

Contribution
Conceptualization, Data curation, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing

For correspondence
kacy.gordon@unc.edu

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0003-0967-4020
Jay W Zussman

Department of Biology, Duke University, Durham, United States

Contribution
Investigation, Writing - review and editing

Competing interests
No competing interests declared
Xin Li

Department of Biology, The University of North Carolina at Chapel Hill, Chapel Hill, United States

Contribution
Validation, Investigation, Writing - review and editing

Competing interests
No competing interests declared
Camille Miller

Department of Biology, The University of North Carolina at Chapel Hill, Chapel Hill, United States

Contribution
Investigation, Writing - review and editing

Competing interests
No competing interests declared
David R Sherwood
1. Department of Biology, Duke University, Durham, United States
2. Regeneration Next, Duke University, Durham, United States
Contribution
Conceptualization, Resources, Supervision, Funding acquisition, Writing - review and editing

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0003-2245-2334

Funding

National Institute of General Medical Sciences (R01 GM079320)

David R Sherwood

National Institute of General Medical Sciences (R35 MIRA GM118049)

David R Sherwood

National Institute of General Medical Sciences (GM121015-01)

Kacy L Gordon

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We would like to thank members of the Sherwood Lab, especially Lara Linden-High and Sara Payne (Duke), Ari Pani (UVA), Andres Collazo (Cal Tech), David Greenstein (University of Minnesota), Sarah Crittenden (UW-Madison), Judith Kimble (UW-Madison), and our three reviewers for helpful feedback. We thank Abigail Gerhold (McGill University) and Jane Hubbard (NYU) for sharing strains. Some strains were provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440). KLG was supported by postdoctoral fellowship GM121015-01 from the NIGMS. This work was supported by NIGMS R01 GM079320 and NIGMS R35 MIRA GM118049 to DRS.

Senior Editor

Utpal Banerjee, University of California, Los Angeles, United States

Reviewing Editor

Yukiko M Yamashita, University of Michigan, United States

Reviewers

Yukiko M Yamashita, University of Michigan, United States
Dave Hansen

Version history

Received: February 25, 2020
Accepted: July 17, 2020
Accepted Manuscript published: July 21, 2020 (version 1)
Version of Record published: September 2, 2020 (version 2)

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.