Ancestral pattern of the Xenopus germ cell cyst: From the germ stem cells to the oocytes and nurse cells

For many decades, biologists have used the invertebrate and vertebrate model organisms to study various biological processes. These model organisms are easy to maintain and reproduce in the laboratory, and they provide abundant, easily manipulated embryos and other experimental materials. Accordingly, Allan Spradling and his collaborators used Drosophila (flies), mice, and Xenopus (frogs) to decipher the cellular and molecular mechanisms of female germ cell formation in invertebrate and vertebrate embryos.

Davidian and Spradling’s paper “Early female germline development in Xenopus laevis: Stem cells, nurse cells and germline cysts,” published in the current issue of PNAS (1), describes studies on Xenopus oogenesis and discusses the results in the context of knowledge acquired from Drosophila and mouse model systems.

Across the animal kingdom—from fruit flies and frogs to mice—oogenesis occurs within clusters of sibling cells called cysts or nests, all connected by intercellular bridges known as ring canals (RC) (2–5). These clusters form when a progenitor cell, the cystoblast, divides several times with incomplete cytokinesis. Drosophila cystoblasts arise from the asymmetrical division of germline stem cells (GSCs). In Drosophila, each cystoblast undergoes four consecutive divisions, giving rise to the cyst of 16 cystocytes connected by intercellular bridges.

Davidian and Spradling’s paper “Early female germline development in Xenopus laevis: Stem cells, nurse cells and germline cysts,” published in PNAS (1), describes studies on Xenopus oogenesis and discusses the results in the context of knowledge acquired from Drosophila and mouse model systems

Germline cysts can be branched or linear. Examples of branched cysts are found in Drosophila, Xenopus, and the mouse. Examples of linear cysts are found in some beetles and some annelids (Fig. 1). The branched cyst of Drosophila has 16 cystocytes (2–5). Although it was believed that Xenopus cysts also form thorough four consecutive divisions and contain maximum 16 cells, now, Davidian and Spradling identified Xenopus cysts containing not only 16- but also 32 cells and they showed that ~17% of cysts consists of 6-, 19-, or 25 cells (i.e. does not follow the 2ⁿ cells rule), which suggests that similar to mice, Xenopus cyst may fragment (6, 7).

Fig. 1.

Diagram of the germline cyst in invertebrates and vertebrates. (A) Linear cysts form the telotrophic ovary of the rove beetle Creophilus maxillosus (8). As described previously (8), the cystoblast division creates two cystocytes: the pro-oocyte and the pronurse cell. They are connected by a RC (marked by a black rectangle). The fusome (not shown) penetrates the RCs. The next several synchronous divisions generate a chain of cells. The pro-oocyte transforms into oocyte, while the other cystocytes develop into the nurse cells. Cystocyte is dark green, pro-oocyte or oocyte is medium green, nurse cells or pronurse cells are light green. Researchers presume the existence of GSCs in the telotrophic ovary, and Rübsam and Büning mentioned them vaguely (9), but they have not been microscopically or molecularly identified. (B) Branched cysts are mutual for Drosophila, Xenopus, and mouse gametogenesis. The GSCs identified in Xenopus by Davidian and Spradling (1) divide asymmetrically, producing a GSC and a cystoblast. The cystoblast undergoes four divisions, producing a 16-cell cyst in which the two oldest cystocytes have 4 RCs each. Out of these two cystocytes, the one that contains the most of the fusomal material becomes the oocyte, while the remaining cystocytes become nurse cells. In addition to 16-cell cysts, 32-cell cysts and smaller cysts were observed in Xenopus and mouse, it is probable that the cystoblast undergoes more or fewer than four divisions, or that the 16-cell or 32-cell cysts split into smaller ones (1). Davidian and Spradling discovered that Xenopus GCSs share common structural and signaling molecules with neural cells, suggesting that they can interact reciprocally with the brain and/or pituitary gland circuits. A part of the figure was modified from Kloc (10), DOI: https://doi.org/10.1387/ijdb.240064mk; Figure was created in https://BioRender.com. Modified from ref. 10, which is licensed under CC BY 4.0.

The cystocyte bridges are penetrated by the fusome, which is a microtubule-rich, germline-specific structure described in invertebrate and vertebrate cysts (11–14). The fusome derives from the remnants of the mitotic spindle and segregates asymmetrically at each division (with its larger portion remaining in one of the daughter cells), establishing cyst polarity and determining which cystocytes become oocytes and which nurse cells (13, 14). In Drosophila, the two oldest cystocytes (containing four RCs) acquire the pro-oocyte fate. However, only one pro-oocyte transforms into oocyte, while the other, like the remaining 14 cystocytes, form the nurse cell. While mammalian cystocytes also exhibit this dual fate (15), the existence of nurse cells in Xenopus cysts has been questioned, leaving it uncertain whether frog oogenesis follows the fly and mammalian consensus, where some of the cytocytes become the nurse cells or whether all cystocytes become the oocytes (16). There was also a knowledge gap regarding the earliest steps in Xenopus cyst formation, and the identity of the cells that precede and give rise to the cystoblast was unknown. The groundbreaking work of Davidian and Spradling finally fills these gaps, showing that Xenopus cysts indeed follow the shared blueprint seen in both invertebrates and higher vertebrates, identifying GSCs, and revealing the inception of Xenopus cysts (1).

In their study, Davidian and Spradling generated transcriptomes from 8,544 germline cells (identified by the expression of conserved markers such as ddx4 (coding for RNA helicase, also known as DEAD-box polypeptide 4 or Vasa) and daz (coding for Deleted in Azoospermia RNA-binding proteins crucial for gametogenesis) and 9,866 somatic cells. Further analysis using the directionality-reducing technique UMAP identified previously uncharacterized GSC and nurse cell (NC) clusters. The GSC cluster expressed piwil4 (coding for piwi-like RNA-mediated gene silencing 4 Argonaute family protein that regulates gene expression in stem cell development) and ddx4. The NC cluster, like mouse nurse cells, showed downregulation of the meiotic genes: sycp1, sycp3, syce2, dmc1, spata22, cilia regulator arl3, hormad1, and the chromatin regulator macroh2a2. Davidian and Spradling’s research also found that in Xenopus, ~80% of cystocytes transiently serve a nurse-like role by transferring molecules and organelles to oocytes before being eliminated. Further, the authors showed that Xenopus GCSs also express spocd1 and morc1, which, in their role as effectors of the PIWI-piRNA pathway, participate in transposon silencing, which was found to be critical for mouse gametogenesis (17). The authors also found that the GCS population in adult Xenopus can reinitiate oogenesis after partial ovariectomy.

Interestingly and unexpectedly, the Xenopus GCSs also expressed genes associated with neuronal functions such as cytoskeletal and synaptic components (neurofilament medium and light chains, βIII-tubulin, neuroligin 1, neurexin 1), neuropeptide precursors and secretory factors (VGF, secretogranin II, proenkephalin, chromogranin, synuclein beta, neuropeptide y, synaptosome-associated protein), and genes responsible for neuronal identity (fam163, diencephalon/mesencephalon homeobox). Xenopus GSCs, like terminally differentiated neurons and Sertoli cells in testes, also express tubb3 (class III β-tubulin), which is regulated by androgens and affects the cytoskeleton and stress resistance.

The authors point out that Xenopus GSCs share properties and mechanisms with neural cells, including compartmentalized translational regulation, mRNA localization, and organelle movement along microtubules. Additionally, the authors postulate that because animals adjust ovarian functions, including ovulation, to external cues by modulating hormone production via the hypothalamus-pituitary-gonadal (HPG) axis, the neural genes expressed in Xenopus GSCs may be used to communicate with other ovarian cells and provide feedback to the HPG axis. The similarities between genes operating in neurons and GCSs discovered in this research open new avenues for the study of the molecular and cellular basis and exact role of similarities between GSC and neuronal pathways (Fig. 1).

These astonishing findings that the GCSs share structural features and molecular signaling circuits with the neural cells may also explain how invertebrate and vertebrate animals living in large groups or colonies, such as corals, bats, or birds can synchronize reproduction between all individuals or suppress the reproduction in the “slaves”/workers in naked mole rats, bees, or ant colonies, and how gametogenesis may be regulated by stress (18–22).

The study presented here also characterized and followed the function of the fusome-like structure (FLS), which resembles the fusome in Drosophila cysts. The authors showed that Xenopus FLS forms a rosette-like connection between cystocytes. Its composition of stable microtubules, Golgi vesicles, and ER suggests participation in polarized trafficking within the cyst participants. They also showed that FLS is necessary for cyst architecture and coordinated development of germ cells within the cyst. When FLS microtubules are experimentally depolymerized, the individual cystocytes drift apart, although they remain connected by the RC. Authors also indicate that the fragmentation of cysts observed in Xenopus occurs mainly at the cyst periphery. The FLS mechanically stabilizes the cyst. Because of its asymmetric distribution within the cyst, the cyst periphery contains the least of FLS material, thus these areas are more prone to fragmentation. In summary, through single-cell RNA sequencing and high-resolution imaging, Davidian and Spradling reconstructed the trajectory of germ cell differentiation within Xenopus laevis germ cell cyst, from the germ stem cells to the oocytes and nurse cells, uncovered common features shared with invertebrate and mouse cyst development, and described structural and signaling commonalities between germ stem cells and neurons. These new discoverers not only establish the existence of a common pattern of germ cell development in invertebrate and vertebrate oogenesis, but also open new avenues to study the effects of neuronal type signaling on germ line stem cells’ functions and fate.