Listing selected discoveries that paved the way for the current model of the inherited strategy of germ cell specification.
Multicellular species use gametes for their propagation. Gametes are formed from primordial germ cells (PGCs), which develop during embryogenesis. In some species, PGCs are specified by the inheritance of a RNA granule known as germ plasm. During germ cell specification, the germ plasm conveys a unique set of properties, e.g. the germ cell specific meiotic cell cycle to the PGCs. Germ plasm assembly is controlled by independently evolving organizer proteins like Oskar in Drosophila or Bucky ball in zebrafish. These organizers are intrinsically disordered proteins, which rapidly changed their amino acid sequence during evolution. A common recipe has emerged by studies on organizer proteins for animals that use germ plasm to specify their germline. Investigating the nature of these organizers might therefore provide a clue to germ cell specification in other species, which are less accessible to molecular-genetic and embryological approaches. Moreover, we might understand how the first metazoans modified their existing cellular structures from unicellular eukaryotes to ensure their reproduction.
- germ plasm
- primordial germ cell
- Bucky ball
- intrinsically disordered protein
- stem cells
Germ cells are precursors to animal gametes. After fusion, gametes have the impressive capacity to develop into a new organism. As all cells of this organism are descendants of PGCs, they are considered totipotent. Interestingly, gametes are also formed in every subsequent generation from the same germ cell. These features identify germ cells as a truly immortal cell line, whereas somatic cells die at the end of life. These are the same characteristics seen in stem cells, thus making germ cells the superior stem cell.
Germline development has to be tightly regulated and controlled to ensure the development of a fertile adult organism. Any misregulation in the pathway would affect fertility and might lead to no offspring. Eventually, sterility might therefore result in the end of that lineage and ultimately in the extinction of the species. Hence, any errors in the germ cell program could have disastrous consequences for a species compared to mistakes in a somatic cell program like forming an organ.
Compared to somatic tissue, very little is known about the critical period of PGC specification. Understanding the biochemical activity of all germ plasm components could help us to grasp, how germ cells get specified. Furthermore, it could identify how “stemness” is achieved at the molecular level. This knowledge might help to treat many degenerative Wof new drug targets for therapy.
2. Mechanisms of germ cell specification
Two different modes of germ cell specification have been described.
2.1. Inductive mode
Germ cell specification by induction is often described as the ancestral or more prevalent mode (Figure 1A) . In the induction mode, germ cell fate is specified through external signals from developing embryonic cells. Induction was described in some invertebrates and in some vertebrates like mammals [3, 4]. The most studied example is the mouse [5, 6, 7]. One of the signals inducing germ cells is BMP4 . However, it is currently not clear how conserved this signal is during germ cell specification in other species of the animal kingdom.
Regardless whether PGCs are specified by induction or inheritance, they show several commonalities at the molecular level. In most species, numerous proteins and mRNAs like Vasa, Piwi, and Nanos are conserved [9, 10]. In spite of two different modes of specification, they activate common downstream components. We will address the evolutionary conservation of germ plasm again at the end of this chapter, when we describe a potential origin of germ plasm in unicellular organisms.
PGCs adopt different lineages, if transplanted to different parts of the embryo. In the mouse, which uses the induction mode, transplanted PGCs later on colocalize with neural plate and surface ectoderm cells . In
The key to understanding the specification of PGCs is to separate species-specific adaptations from a core program of germ cell formation. As information about the initial phase of germ cell specification is still quite fragmentary in different organisms, the core program of germ cell specification is unclear. For instance, the molecule that acts as a master or “kick starter” for the germ plasm or PGC program appears to be different in each organism. Therefore, in the rest of this chapter, we will concentrate on the inherited mechanism of germ cell specification.
2.2. Inherited mode
Inheritance of cytoplasmic determinants represents the second mode, by which germ cells are specified (Figure 1B). This mechanism of germ cell specification is described amongst others in dipteran insects (e.g.
|Weismann (1893)||Inheritance depends on germ cells. Postulates that germ plasm localizes to the nucleus.|
|Hegner (1911), Boveri (1910)||Germline determinants (germ plasm) localize to the cytoplasm. Germ plasm is necessary (Hegner) and sufficient (Boveri) for germline development.|
|Bounoure (1934)||Germ plasm for the first time visualized in a vertebrate egg.|
|Smith (1966)||UV-irradiation of |
|Illmensee and Mahowald (1977)||Ectopic germ plasm is sufficient for PGC formation.|
|Heasman (1984)||The Balbiani body of |
|Ephrussi and Lehman (1992)||Ectopic expression of a single protein termed Oskar gives rise to functional PGCs in |
|Hashimoto (2004)||Ablation of germ plasm in zebrafish reduces PGCs.|
|Bontems (2009)||Ectopic Expression of a single protein termed Bucky ball induces PGCs in zebrafish.|
|Brangwynne (2009)||Biophysical studies on embryonic germ plasm reveal a liquid-like hydrogel in |
|Tada (2012)||Germ plasm transplantation in Xenopus induces ectopic germ cells.|
|Boke (2016)||The |
3. Germ plasm
Germ plasm is a collection of maternally provided RNAs, proteins, and organelles like mitochondria and endoplasmic reticulum [ER]. The entire assembly forms a cytoplasmic structure in the oocyte named Balbiani body . Sometimes it is also referred to as the mitochondrial cloud in
Loss of germ plasm leads to a decrease or no germ cells, whereas in gain of function experiments more germ plasm leads to more germ cells  (Table 1). Germ plasm components are believed to act in stem cells to convey longevity and totipotency, similar to the magic substances
As several germ plasm components have a role in stem cells, it should have a much greater effect in maintaining “stemness” and increased longevity than their somatic stem cell counterparts. As germ plasm conveys a high degree of longevity to germ cells, it would be of stupendous importance to further dissect the germ plasm and study this network of protein and RNA to get further insights into these stemness features.
In the section below, we will concentrate on the two organizer proteins Oskar in invertebrates and Bucky ball in vertebrates that are involved in germ plasm assembly. Both molecules specify germ cells indicating that their biochemistry and mode of action is similar.
4. Oskar in invertebrates
Oskar protein acts as a master regulator of germ plasm assembly . In
Mislocalization of Oskar protein at the anterior end of the embryo leads to ectopic germ cells and a second abdomen . Oskar was the first protein, which is both necessary and sufficient to assemble germ plasm. Increasing the amount of Oskar protein in the fly embryo causes an increase in activity of the Nos protein. Thus, the amount of Osk protein and the level of Nos protein accumulation are related. Possibly the heightened expression of Nos represses the somatic cell fate pushing it to a germ cell lineage [36, 37]. Such an activity supports the role of Oskar as a master regulator of PGC specification in invertebrates.
|Long Oskar||Short Oskar|
|606 amino acids long||467 amino acids long|
|Anchoring germ plasm||Assembling germ plasm|
|Associated with endosomes||Associated with RNA granules|
|Interacts with Lasp to be tethered to posterior pole||Interacts with Lasp to be tethered to posterior pole|
|Not essential for patterning and germ cell formation||Necessary for germ cell formation and posterior patterning|
5. Germ cell specification by Oskar
Fascinating insight into sOsk function was recently gathered by crystallizing two of its domains. These were a domain at the N-terminus of sOsk [139–240aa], which was termed LOTUS domain and previously predicted to be involved in RNA-binding. The second structure described the C-terminal “OSK” domain, which resembles a SGNH hydrolase [40, 41] (Figure 2). However, looking carefully at the biochemical interactions and crystallizing sOsk with these binding partners revealed some unexpected information.
sOsk directly interacts with Vasa , which is an ATP-dependent helicase [41, 46]. Interesting biochemical and biophysical studies show that the eLOTUS domain of Oskar does not interact with RNA, but in fact binds to the RNA helicase Vasa, which is an important component of germ plasm. Surprisingly, the extension of the LOTUS domain (eLOTUS) encodes an intrinsically disordered motif, which forms a structured domain upon Vasa binding. This stretch of 18 amino acids outside of the LOTUS domain is essential for the Vasa interaction. Moreover, binding the eLOTUS domain increases the ATPase activity of Vasa. This is the first time an instructive role was assigned to Oskar, which was previously regarded as a scaffold protein aggregating germ plasm components within the
The OSK domain shows a lot of similarity to a SGNH hydrolase, but lacks three of the four residues of the SGNH motif, as well as the serine triad to be an active hydrolase . The C-terminal OSK-domain forms a globular structure, which carries several basic, positively charged residues at its surface suggesting it could interact with nucleic acids. Indeed, this domain binds in
Taken the interaction data of sOsk together, a modified picture of germ cell specification emerges. sOsk initiates the assembly of germ plasm by binding to Vasa and mRNA. This interaction activates Vasa and might sterically bring it in proximity with specific RNA(s). This could regulate translation or stability of the RNA(s) involved in specifying PGCs . Hence, Vasa and Osk seem to act in a co-operative manner to specify germ cells.
Vasa is also involved in piRNA processing. The amount of Vasa in the germ plasm, therefore, prevents the degradation of the germ cell genome by transposon activity, but piRNAs could also play an undiscovered early role in germ cells . Aubergine, a well-known component of the piRNA pathway, is needed for Osk translation, which also needs Vasa to localize. This could indicate a feedback mechanism ensuring all the downstream germ plasm members are expressed . Figuring out the biochemical process, which is initiated by sOsk/Vasa, is probably the key to understand the molecular mechanism of the germ cell specification program.
6. Zebrafish as a model organism to study germ cell specification in vertebrates
Compared to invertebrates such as
7. Bucky ball in zebrafish
To identify maternal factors controlling early vertebrate development, a maternal-effect mutant screen was carried out in zebrafish . Among the mutants with a defect prior to midblastula transition (MBT), one line produced embryos with radial segregation of cytoplasm instead of animal pole aggregation. In addition, the fertilized embryo from the mutant mother does not show cellular cleavages and hence does not develop beyond the 1-cell stage. As the mutant embryo lacks polarity similar to Buckminsterfullerenes, it was referred to as
In the oocyte, Buc mutants fail to assemble germ plasm into a Balbiani body (Bb) (Figure 3A). Instead, germ plasm components like
7.1. The conservation of Buc across the vertebrate kingdom
Buc is present in vertebrates; however, across its homologs in the vertebrate phylum, the sequence changes quite rapidly . Zebrafish has two paralogs of Buc in its genome, whereas the salmon has three . Currently, the function of the other paralogs is not clear. The
8. Similarities between Oskar and Buc
Buc and sOsk show a striking homology at the genetic level regarding germ plasm formation. Both mutants show a defect in polarity and a failure of germ plasm aggregation [54, 58]. Remarkably, ectopic overexpression of sOsk and Buc induces the formation of additional germ cells [32, 54]. To this end, no other proteins have been described, which can induce PGC formation in an organism.
Fascinatingly, ectopic expression of
sOsk was shown to interact with Vasa, Valois, and Lasp [45, 53, 59]. For example, if Buc also binds to zebrafish Vasa, it could mean that Buc uses a similar set of germ cell core factors like Osk to specify germ cells.
8.1. Conservation between Oskar and Buc
According to the sequence-structure-function paradigm, proteins with a conserved activity contain homologous sequence motifs to interact with similar binding partners. Conserved sequences were previously not identified between sOsk and Buc [41, 54, 60]. Buc does not have a visible LOTUS domain, which is required for multimerization and takes part in the interaction with Vasa . Moreover, Buc has no motif with homology to any known RNA binding domain. However, the OSK RNA-binding domain was also not described previously in other proteins and many RNA binding motifs do not show conserved domains . Presently, none of the published bioinformatic analysis detected sequence similarities between the two germ plasm organizers Osk and Buc. Hence, their conserved activity remains a mystery. Overall, this would suggest that the structure or biophysical nature of both proteins might be similar in order to accomplish the same activity by which both would give rise to the “core” RNA-protein complex. sOsk and Buc might, therefore, represent the first protein pair of a frequently postulated phenomenon: Two proteins with similar function without sequence similarity .
9. Vasa: the ubiquitous germ cell marker
Vasa seems to be the most widely used molecular marker to identify germ cells [63, 64, 65, 66, 67]. Vasa is well conserved during evolution and required for germline development. Vasa is a member of the DEAD-box protein family of RNA helicase suggesting that it resolves duplex RNA or RNA-protein hybrids. Mutations in Vasa show defects in posterior patterning and in germ cell specification in the
Vasa RNA or protein expression is frequently used to label PGCs in animals. As at least one homolog seems to be present in all metazoans, Vasa is also an easily accessible marker across the animal kingdom . However, the restriction of Vasa at the blastula stage to the germ plasm and prospective PGCs varies across species. In some species like the zebrafish, Vasa protein is ubiquitous at early stages and later gets restricted into PGCs , which raised concerns about the role of Vasa during germ cell specification.
Exciting results from
10. Low complexity proteins
|Structure||Low complexity regions form beta sheets.||Very low complexity with FG or FXXG repeats, in most cases with no secondary structure formation.|
|Chemical||Aggregates are resistant to SDS and high salt concentrations.||Aggregates are dissolved by SDS or high salt concentrations.|
|Aggregation||Aggregates are resistant to 1,6 hexanediol.||1,6-hexanediol dissolves hydrogels formed by IDPs.|
|Staining||Stain positively with Thioflavin S and T.||No accumulation of Thioflavin.|
Both Buc and sOsk have been suggested to have low complexity regions [41, 75, 77]. Indeed, it was shown that sOsk contains an intrinsically disordered region critical for Vasa binding. In Buc and Velo1, it was shown that parts of the conserved BUVE-motif form prions or amyloid-like aggregates. IDPs frequently evolve faster than structured proteins [74, 82]. This feature might hide conserved motifs in both proteins, which are critical to interact with the same biochemical network.
IDPs are also known to act as hubs for supra-molecular complexes and are also more prevalent in RNA-binding proteins. As sOsk fits this profile, it would be interesting to know whether Buc binds RNA to explain their conserved activities. Moreover, IDPs form liquid-liquid phase separations such as RNA-granules, which were also described for the germ plasm in
Interestingly, Buc has been discussed to have both amyloid and IDP regions. In
Overall the aggregation of IDPs emerge as a central theme in germ cell specification. Just like Vasa, which is also intrinsically disordered region  and like the polymerizing substrates of P-granules which are the MEG1 and MEG 3 proteins in
11. A common recipe to make germ cells
If Osk and Buc have diverged from a common ancestor, their precursor would have been an ancient protein of low complexity, which induces germ cell formation. Both proteins probably have unrelated sequences as consequence of their role as intrinsically disordered scaffolds. This structural role releases the constraints to maintain a defined protein structure as described for other IDPs . This divergence probably hides conserved motifs, which bind to a similar interactome such as Vasa, Valois, and probably other common mRNA binding partners (Figure 5). Finding interaction partners and mapping the interaction motifs like for the sOsk-Vasa interaction will determine, to which level interaction motifs are conserved between sOsk and Buc.
Describing the Balbiani body, a picture of the popular “bubble tea” comes to mind. In this picture, the organizer proteins form a scaffold probably via self-aggregation or upon binding with their interactors similar to the chewy alginate balls, which form during polymerization. During this process, germ plasm assembles and thereby integrates RNA and proteins into this 3D liquid lattice. The assembly also initiates Vasa’s activity to start the downstream program, e.g. to protect RNAs and proteins from degradation . The germ plasm also exchanges components with the cytoplasm similar to those spheres floating in the bubble tea. When inherited into a cell, the germ plasm probably releases some proteins whose translation and stability is tightly controlled. Once these factors are unleashed from the bubble spheres, they change the transcriptional program to specify the maturation of a PGC to a gamete.
Why should germ cell specification be conserved? Reproduction is a conserved feature of all biological systems and must have been, therefore, be present in the first metazoans before other cell types like neurons, muscle or a vascular system. Germ cell specification was, therefore, present before the formation of an eye or even a nervous system. Nonetheless, the conservation of the master regulator Pax6/Eyeless showed that light sensing organs were already present at the base of metazoan evolution . Although this hallmark finding is currently accepted in the scientific literature, the insect compound eye and the vertebrate camera-eye were regarded as a paradigm for convergent adaptations. We, therefore, speculate that germ cell formation is the more ancient tissue compared to eyes, would use an even more conserved molecular regulation than Pax6/Eyeless.
When animals started to become multi-cellular, they could no longer continue to reproduce by simple cell cleavage. They needed to set the germline apart from the soma for their reproduction . For this task, they had to evolve proteins, which served as master switches for germ cell specification. Any changes to the function of these proteins could have lasting consequences on the propagation of that species. However if these proteins were IDPs, they could still perform their function, despite of rapid (localized or random) changes. These changes could have roles in speciation or better coordinated control of specification. Whatever the case, if they still aggregated and setup the “core” complex, a germ cell would have still formed.
13. Future directions and recommendations: back to the future
Ciliates form a cytoplasmic aggregate called the conjusome . This structure is present only during sexual reproduction. Similar to the Balbiani body in
Expanding on this hypothesis, protein phase transition might have been present before the first unicellular organisms. If the beginning of life was an RNA world  and formation of a cell was needed to protect the genetic material, it would have been easier to have a hydrogel aggregate of slime or protein lock the RNA into an RNA granule than to establish a lipid bilayer with an internal framework. Indeed if that was the case, this structure would have been more similar to the germ plasm that we see today than to a membrane-bound cell. Thus the origin of life would have been from a germ plasm ancestor similar to a drop of Amrit or Ambrosia spilled from the heavens.
Pritesh Krishnakumar was supported by the German Academic Exchange Service (DAAD). Research in the Dosch lab is supported by the Deutsche Forschungsgemeinschaft, the GGNB Junior Group Stipend and the “Forschungsförderungsprogramm” of the University Medical Center Goettingen.
|C. elegans||Caenorhabditis elegans|
|IDPs||Intrinsically disordered proteins|
|Pax6||Paired box protein 6|
|PGCs||Primordial germ cells|
|Xvelo-1||Xenopus Vegetal localized 1|