In basal metazoans, embryos only form from oocytes fertilized with spermatozoa. This is a clear manifestation of the link between embryonic formation and sexual differentiation of gametes. This is important to emphasize, given that we are discussing a cellular biological system, that is, zygotes produced by cell fusion, and not simply a genetic system. Parthenogenesis, on the other hand, is an embryological process acquired a little later in animal evolution (Simon et al., 2003; Jarne and Auld, 2006) and based on some ecological requirements (Cuellar, 1977).

It is crucial to search for conditions for the occurrence of meiosis among protists. A vast number of protists are known to undergo sexual reproduction (Grell, 1973; Heywood and Magee, 1976; Raikov, 1995). Meiosis and/or fertilization have been described in the lobose testate amoeba Arcella, the filose naked vampyrellid amoeba Lateromyxa gallica, chrysophytes, prymnesiophytes, xanthophytes, dinoflagellates, trypanosomatids, piroplasmids, microsporidia, Myxosporidia, and Actinomyxidia, as amply described by Raikov in 1995 (Raikov, 1995). Furthermore, a cryptic sexual cycle and a meiosis-like recombination was inferred in Leishmania from population genetic studies and by laboratory crosses (Inbar et al., 2019; Louradour et al., 2022). Of the groups closest to animals, the choanoflagellate Monosiga brevicollis has meiotic genes (Carr et al., 2010) and the choanoflagellate Salpingoeca rosetta has sexual reproduction with meiotic recombination (Levin and King, 2013). The taxonomic names of higher protists are those adopted by Margulis et al. (1989) and thoroughly reviewed by Adl et al. (2019).

Raikov classifies sexual protists into three categories: i) haplonts that undergo zygotic meiosis (Figure 2A), ii) diplonts with gametic meiosis, and iii) heterophasic forms with intermediate meiosis. Zygotic meiosis occurs during the first divisions of the zygote, and only the zygote, which is often encysted, is diploid. The vegetative stages are haploid and produce (or become) gametes without meiosis. Gametic meiosis occurs during gamete formation; vegetative cells are diploid and only gametes are haploid. Forms with intermediate meiosis show alternation of two generations, a haploid that produces gametes and ends with fertilization (karyogamy) and a diploid that ends with meiosis, restoring haploidy. Both generations are represented by vegetative forms (Raikov, 1995). In any case, meiosis linked to karyogamy occurs in most groups of protists, having been deeply studied. These observations support the hypothesis that sexuality is a basic characteristic of all protists and has been reduced in some specialized groups (Grell, 1973). In most taxa, meiosis falls into a standard basic type, with pre-meiotic DNA synthesis, initial parallel chromosome pairing, crossing over (which is sometimes lost secondarily, as in achiasmatic meiosis), and synaptonemal complex formation (Grell, 1973; Raikov, 1995).

The meiosis genes Spo11 and Hop1 are present in all major lineages of protists (Weedall and Hall, 2015), having been detected in Apicomplexa (Malik et al., 2008), Euglenozoa (Malik et al., 2008), Fornicata (Ramesh et al., 2005), Parabasalia (Malik et al., 2008), and Amoebozoa (Ehrenkaufer et al., 2013). On the other hand, the presence of meiosis genes does not necessarily prove that a species is sexual; orthologs may have more than one function or may adapt to perform new functions in asexual species and thus be maintained, even in the absence of sex (Schurko and Logsdon, 2008). Nevertheless, association of genes with a given function is very recurrent in molecular genetics based on neo-Darwinism, representing a very ineffective thinking strategy for understanding biological processes. It is worth remembering that Drosophila melanogaster does not have many meiosis genes, and this has not led developmental biologists to question the sexuality of arthropods (Malik et al., 2008).

To support our hypothesis, it is important to search for an association between meiosis and the formation of different cells (anisogamy) in protists. A link would therefore be established between the first animal zygote and differentiation of oocytes and spermatozoa of the first metazoan embryo that would have been hermaphrodite. Protists have extensive biological experience in the formation of singular potentials that result in gametogenesis (sperm and oocyte phenotypes) and, particularly, differentiation of vegetative cells into gametes. Gametogenesis in Chlamydomonas, for example, is a phenomenon that allows the study of sexual differentiation in a single cell at the molecular level. Such differentiation is easily controlled by environmental signals, including lack of nitrogen and light (Beck and Haring, 1996). Some anisogamous species, such as Chlamydomonas suboogama, produce large immobile macrogametes and small mobile microgametes (a clear case of anisogamy) (Tschermak-Woess, 1959). Microgametes merge with macrogametes.

Some other anisogametic genera of flagellated protists include Eudorina and Volvox, which are either dioecious or monoecious. In monoecious species, each individual colony is either female or male, whereas, in dioecious species, a clone produces either female or male colonies (Tarín and Cano, 2000). Another example of anisogamy can be found in the Apicomplexa group. Infectious sporozoites invade host cells and then produce large and small gamonts. Large macrogamonts turn into macrogametes, and small gamonts undergo fissions, resulting in many small microgametes. Fertilization occurs by penetration of the microgamete into the macrogamete (Tarín and Cano, 2000). Among unicellular holozoans, Salpingoeca rosetta is the only species known to have a sexual life cycle, transitions between haploid and diploid states, and anisogamous mating, during which small flagellated cells fuse with larger flagellated cells (Levin and King, 2013).

Although there are currently no studies on the sexual reproduction of some groups of unicellular holozoans, such as Ichthyosporea, Filasterea, and Pluriformea, we believe that they have some cellular characteristics and that their life cycles are of extreme relevance in the context of our hypothesis. The amoeboid forms of Corallochytrium limacisporum, Creolimax fragrantissima, Sphaeroforma arctica, Capsaspora owczarzaki, Ministeria vibrans, and Syssomonas multiformis are of interest for the development of the hypothesis on the formation of the first embryo (Mendoza et al., 2002; Paps and Ruiz-Trillo, 2010; Sebé-Pedrós et al., 2013; Suga and Ruiz-Trillo, 2015; Hehenberger et al., 2017). Also of interest to our proposal are the flagellated forms of Rhinosporidium seeberi, Dermocystidium sp., Dermocystidium salmonis, Dermocystidium percae, Sphaerothecum destruens, M. vibrans, Pigoraptor vietnamica, and Pigoraptor chileana (Mendoza et al., 2002; Marshall et al., 2008; Torruella et al., 2012; Hehenberger et al., 2017; Mylnikov et al., 2019), as well as the flagellated amoeboid form of S. multiformis (Hehenberger et al., 2017). The extremely complex life cycles of these groups are noteworthy, with emphasis on the colonial forms of C. fragrantissima, S. arctica, P. vietnamica, and P. chileana (Marshall et al., 2008; Paps and Ruiz-Trillo, 2010; Hehenberger et al., 2017) and the “syncytial” form of C. fragrantissima and S. arctica (Suga and Ruiz-Trillo, 2015) (syncytium-like structure may be a more appropriate term, given that it does not involve a fusion process), all of which demonstrate a successful attempt of unicellular organisms to remain together with a cellular structure that has numerous genes associated with Neoplasia (Domazet-Lošo and Tautz, 2010; Chen et al., 2015; Grau-Bové et al., 2017; Bartas et al., 2019).

Fertilization is highly conserved in eukaryotes and animals (Cavalier-Smith, 1978; Carvalho-Santos et al., 2011). We understand that flagella were co-opted to gamete fusion and therefore to the germline formation. Flagella are sensory structures in protists (Marshall and Nonaka, 2006; Singla and Reiter, 2006; Brunet and Arendt, 2016; Hilgendorf et al., 2016; Smith et al., 2020). Some of their original characteristics were conserved in metazoans, such as chemotaxis (Miller and Brokaw, 1970; Ward et al., 1985). For movement, M. vibrans does not use flagella but rather microvilli (Mylnikov et al., 2019). As will be discussed in a future article, multiflagellate fusion might have been decisive in the evolution of metazoans. From the point of view of our hypothesis, the recruitment of cilia/flagella to NFM was crucial for animal evolution.

It is important to reflect on the reasons for excluding choanoflagellates as the original group of metazoans. Our line of reasoning is based on the contributions of Manuel Maldonado (Maldonado, 2005)^, who attributed numerous reductions and losses of this protist group and whose line of thought was confirmed in a recent phylogenetic study that placed choanoflagellates as the most distant group, outside the direct line of metazoans (Arroyo et al., 2018). On the other hand, reassessment of molecular and histological evidence demonstrated that choanocytes are specialized cells that develop from non-collared ciliated cells during sponge embryogenesis, discarding choanocytes as primitive cells in animal evolution (Maldonado, 2005). Thus, the connection of choanoflagellates with metazoan emergence is a completely unrealistic idea.

We also cannot disregard the important contribution of Alexander Eresvkovski in expanding the knowledge about sponges (Ereskovsky, 2010; Eresksluovsky et al., 2013). As co-author of a recent publication, Eresvkovski described the differences between the flagellar apparatus (or kinetid) of sponges and that of choanoflagellates. According to the authors, analysis of these structures was inexplicably neglected (Pozdnyakov et al., 2017). The kinetid of choanocytes contains more features considered plesiomorphic for opisthokonts than the kinetid of choanoflagellates. Therefore, the hypothesis that Porifera, and consequently all metazoans, originated directly from unicellular choanoflagellates does not seem plausible (Pozdnyakov et al., 2017).

Thus, according to our hypothesis and phylogenetic studies, the unicellular holozoans Ichthyosporea, Filasterea, and Pluriformea would be closer and in the direct line of metazoans (Suga et al., 2013; Grau-Bové et al., 2017; Hehenberger et al., 2017). Considering these groups brings an innovative element that reinforces the idea of the link between fungi and animals (Maldonado, 2005). We are aware that yet undiscovered intermediate organisms could be in the direct line of metazoans; however, our hypothesis was developed on the basis of current data, and we await confirmation of meiosis associated with anisogamy in other groups of unicellular holozoans. In any case, we consider anisogamy as a condition for the biological prototype that would come to contribute to the establishment of the first animal embryo. Anisogamy is quite primitive among protists (Tschermak-Woess, 1959; Tarín and Cano, 2000) and was conserved in metazoans.

Finally, if we accept the premise of an NP at the heart of animal embryogenesis, we can explain the rapid evolution of forms of sex determination by contradicting one of the main myths of animal evolution (that it is slow and gradual) (Bachtrog et al., 2014). NP consolidation might have had effects at the cellular level, including gene regulation, or might have provided for broad genomic recombination, thereby explaining why, sometimes, closely related species or populations of the same species have different modes of sex determination (Charlesworth, 1996). On the other hand, the NP at the core of animal embryos may help explain the mysterious preponderance of sexual reproduction between species (the “sex paradox”), revealing that sexually reproducing organisms always proliferate faster (Chen and Wiens, 2021). As evidenced by the conditions for the formation of the first embryo, the association of multicellularity and sexual reproduction would make it possible to explain the rapid evolutionary radiation (Chen and Wiens, 2021). In our view, multicellularity and sexual reproduction can only explain the rapid formation of metazoans if strictly and intimately linked to the context of animal embryogenesis.

Gametogenesis plays a unique role in gamete production and has a great impact on the heredity of embryogenesis, whose goal is the reconstruction of the embryo with each generation. Because it is a process of rebuilding the structural coherence of the embryo, it also plays a key role in evolution. Therefore, understanding the mechanisms of germ cell specification is a central challenge in developmental biology and evolution research.

During animal embryogenesis, the germline is segregated from somatic cells, but this dichotomy is not universal in animals. Recent findings in ctenophores (Fierro-Constaín et al., 2017), poriferans, and cnidarians suggest that the germline appears to specify from a multipotent cell line during embryogenesis, whose fates include somatic cells and the germline (Juliano and Wessel, 2010). Thus, the boundary between the germline and the Soma is fluid and may have a broader significance for development than initially predicted, being important, for instance, in active regeneration processes (Edgar et al., 2021).

Another fundamental aspect is the exact moment of germline segregation. In animals, it seems to occur along a continuum, shortly after embryogenesis or at the very beginning of it, with the possibility of intermediate points during embryogenesis (Seervai and Wessel, 2013). In all cases, germline segregation requires that a population of cells be established in the embryo, whether multipotent or not. Given that germline segregation from a multipotent precursor occurs after embryogenesis in ctenophores, poriferans, cnidarians, and even in lophotrochozoans and echinoderms, it is inevitable to conclude that this is truly an ancestral mechanism (Agata et al., 2006; Juliano et al., 2010).

Late segregation after embryogenesis is completely expected and necessary for the beginning of animal life, being one of the conceptual bases of our hypothesis. The first embryo and its germ cells must have first received the mechanical and physical stimuli of embryogenesis to trigger germline segregation. Therefore, the philosophical question should be, “What came first, the egg or the beroid-like ctenophore?” The answer is unequivocal; the egg came first, but with one caveat: only after it had been completely impregnated by its surroundings and physical trajectory inside the embryo. This first embryonic trajectory is fundamental: by walking, the path is made (Machado, 2021). This rule is absolute for the first embryo, which would come to establish a topological map of physics in the genetic material (Uhler and Shivashankar, 2017a; Uhler and Shivashankar, 2017b; Uhler and Shivashankar, 2018; Tsai and Crocker, 2022).

Some clues of this physical map indicate a primitive chromatin state after fertilization (Xia and Xie, 2020) that was highly organized, structured (Kaaij et al., 2018), and conserved in evolution (Hug et al., 2017; Kaaij et al., 2018). Such a state seems to be important for the embryo to reach totipotency (Flyamer et al., 2017). This organization occurs through topological associating domains (TADs) and chromosomal loops present in germ cells and animal zygotes (Flyamer et al., 2017; Collombet et al., 2020). Such a process is compatible with the physical impregnation of the embryo and its impact on germ cells. Another clue from the physical map is the extensive remodeling of chromatin after fertilization, which involves de novo labeling/trimethylation of histone H3 at lysine 9 (H3K9me3) by the G9a/GLP complex. This labeling facilitates subsequent establishment of a mature constitutive chromatin (Burton et al., 2020). In other words, they seem to be “bookmark promoters for future compaction” (Xia and Xie, 2020), thereby “creating a less constrained epigenetic environment for subsequent zygotic genome activation” (Wang C. et al., 2018). Once heterochromatin is established, H3K9me3 plays an important role in genome stability and cell differentiation fidelity (Nicetto et al., 2019). We understand that this ability of H3K9me3 to anticipate what comes next in embryonic development (mechanical memory) is only possible because it recreates the physical map of the germline after fertilization. Finally, an important clue to the physical impact of embryogenesis lies in the cohesins that mediate the formation of chromosomal loops and TADs and are sensitive to mechanical force (Kim et al., 2019). Polycomb repressive complex 2 (PRC2), which interacts physically and functionally with G9a/GLP (Mozzetta et al., 2014), participates in a mechanosensory mechanism dependent on F-actin and the protein emerin (Le et al., 2016).

There is clear evidence that TADs and, therefore, chromatin architecture are deeply interrelated with gene regulation (Dixon et al., 2012; Nora et al., 2012; Rao et al., 2014; Fraser et al., 2015; Wang H. et al., 2018; Esposito et al., 2020; Belyaeva et al., 2022). The idea that the genome may be represented, in part, as a series of chromosomal blocks that can be opened or closed for transcription under specific conditions (Hurst et al., 2004), among them physical ones, seems to be a new, useful, inspiring model. This is consistent with the notion that hematopoietic stem cells have the potential to generate non-hematopoietic tissues (Akashi et al., 2003) and with labor division of yeast histone deacetylases in domains other than the chromatin (Robyr et al., 2002). In mouse NIH 3T3 cells, actomyosin contractility regulates the spatial distribution of histone deacetylase-3 (Nikhil et al., 2013). Therefore, physical forces can alter gene expression profiles and differentiation programs (Tajik et al., 2016). Recent studies have shown that chromosomal configurations are altered in response to changes in nuclear mechanical properties following cues from the mechanical microenvironment (Farid et al., 2019; Carthew et al., 2021).

We speculate that mechanical changes (tension, substrate stiffness, or geometric constraints) are fundamental for the recruitment of what we call NFM (Figure 1). The module undergoes some co-optations and exaptations throughout the first animal embryogenesis, following the mechanical cues of embryogenesis. With the formation of the first embryo and its germline, there will be possibilities in the next-generation to modify the trajectory built in the first embryo; thus, innovations of this topological map and NFM may arise at the beginning or middle of embryogenesis, producing a promising monster (Goldschmidt, 1982). Viruses (Breitbart et al., 2015), symbionts (Margulis and Fester, 1991; Ohtsuka et al., 2009), and physical or epigenetic disruptions can produce evolutionary innovations. But, for this to happen, physical phenomena must first have impregnated embryology.

One of the most fascinating current concepts is that of mechanical memory. The biophysical regulation of chromatin architecture produces stable remodeling and long-term changes of cellular behavior instigated by mechanical signals (Balestrini et al., 2012; Heo et al., 2015; Li et al., 2017). Mesenchymal stem cells cultured for 3 weeks on a soft substrate produced a persistent neural differentiation resistant to other stimuli produced by cell differentiation factors (Engler et al., 2006). Short-term mechanical memory is dependent on actomyosin contractility. Long-term increases in deformation that persistently affect chromatin condensation do not seem to depend on actin contractility but on calcium ion (Heo et al., 2015). This phenomenon should be analyzed with care, as actin filaments gained reputation as bionanowires capable of conducting calcium waves (Hunley and Marucho, 2022). An important review study analyzed the different mechanisms responsible for storing mechanical memory in the cell nucleus (Dai et al., 2020), including loops stabilized by CCCTC-binding factor (CTCF) or cohesins (Rao et al., 2014; Kim et al., 2019) and the epigenome (Le et al., 2016), both of which are sensitive to mechanical force (Le et al., 2016; Kim et al., 2019). In vivo models of mechanical memory have been little explored, being used only in cancer research (Paszek et al., 2005; Northey et al., 2017). In this context of cellular memory, the order of NFM expression must apparently be determined by the 3D structure of the nucleus (Wang H. et al., 2018), which in turn is influenced by the 3D structure of the embryo.

In relation to germ cells, special emphasis should be given to ctenophores of the order Beroida. The recognizable germ cells of ctenophores are closely associated with the endoderm of the meridional canals and always underlie the comb rows (Pianka, 1974). In beroids, gonadal tissues either form simple tracts in proper meridional canals, extend into the lateral branches that are characteristic of this order, or develop into separate sexual diverticula external to lateral branches. Thus, the presence of a multipotent cell line and a fluid relationship between Soma and germline (Edgar et al., 2021) may be explained by organization or physical construction of the embryo. Germ cells (established within the gonads) pass on or receive directly or indirectly most of the physical impacts and tension of the embryo (e.g., epiboly, invagination, and EMT, among others, as will be discussed below) (Figure 1). Luke Parry’s analysis provides insight about a sensory network in the first beroid-like ctenophores (Parry et al., 2021). The main highlights regard the nerve tracts of comb rows and polar fields, showing a much more complex sensory and nervous organization than that of living ctenophores (Parry et al., 2021). This type of network can reflect on the physical, mechanical (tensional actin networks), and electrical (nerve impulses) organization of the embryo. The implications of such a network may be depicted in the ability to regenerate. Surprisingly, Beroe ovata lost nerve endings in comb rows as a byproduct of evolution (Parry et al., 2021), thereby losing the ability to regenerate (Edgar et al., 2021). Mechanical forces may be morphogenic and organize how cells regenerate. There is strong evidence that regeneration capacities are related to the apical (sensory) organ, when present (Martindale, 2016). Mark Martindale suggested that nervous system involvement may be crucial for ctenophore regeneration and proposed experimental manipulation to confirm or rule out neural involvement (Edgar et al., 2021).

Although the so-called germline multipotency program is conserved in ctenophores (Fierro-Constaín et al., 2017), which involve, among other genes, vasa, piwi, and nanos, these genes show little or no expression in germ cells or gametogenic regions (Reitzel et al., 2016). Restricted expression in the apical organ and tentacles of Mnemiopsis leidyi, which are areas of high cell proliferation, suggests that these genes could be involved in the specification and maintenance of stem cells (Reitzel et al., 2016). Similarly, members of the gene family Dmrt, essential components of sex determination and differentiation in bilateral animals, were not expressed in adult gametogenic regions.

What becomes obvious from these analyses is that vasa, nanos, and piwi genes do not participate in the establishment or maintenance of the germline of animals. Ctenophores, for example, produce haploid gametes and reproduce sexually; thus, their germline has the same function as that of bilaterian animals. That is, it is not this set of genes that determines germline development and maintenance but rather the appropriate cellular context in which the genes function. The appropriate context for vasa, nanos, and piwi was established a little later in animal evolution and thus co-opted by embryonic germ cells for more specialized functions (Juliano and Wessel, 2010). The genes responsible for maintaining and establishing the germline in ctenophores remain unknown, but evolution has left some traces and clues of their origin.

Some evidence on the establishment of the germline in basal metazoans emerged with the identification of a highly conserved system in invertebrates and vertebrates related to the protein Mos and its function in meiosis (Amiel et al., 2009). Mos, curiously a protooncogene, would prevent meiotic/mitotic conversion (meiotic to mitotic conversion) after meiosis I, directing cells to meiosis II and ensuring ploidy reduction (Tachibana et al., 2000), thereby preventing undesirable DNA replication or parthenogenetic activation prior to fertilization (Furuno et al., 1994; Tachibana et al., 2000; Amiel et al., 2009). Mos participates in induction of oocyte maturation (Freeman et al., 1990) and migration of the meiotic spindle (Verlhac et al., 2000). These functions are crucial for two very unique moments of embryogenesis: i) fertilization (ensuring ploidy reduction before the onset of embryogenesis) and ii) separation of the germline from the somatic lineage, when meiosis begins (soma-to-germline transition). This leads us to speculate on the role of Mos, or a similar protein, in establishing the germline in the first embryo. Unfortunately, Mos has not yet been studied in the cell cycle or meiosis of ctenophores but was characterized in a genomic analysis of Pleurobrachia pileus (Amiel et al., 2009). Cytoplasmic tyrosine kinases (among them Mos/MAPK) seem to have been established before divergence of Metazoa (Suga et al., 2012).

The protein Mos serves to remind us that meiosis evolved from mitosis and that it could also alleviate the polyploidy instability (Cleveland, 1947) (Figure 2E). In fact, endomitosis is a form of meiosis but without karyogamy (Raikov, 1995). In the flagellated protists Trichonympha and Barbulanympha (Cleveland, 1947) and the radiolarian Aulacantha scolymantha (Grell and Ruthmann, 1964), there is evidence of a prototype of chromosomal pairing in endomitosis (Figure 2D). As reported by Lemuel Cleveland, polyploidy in Barbulanympha is invariably reduced by meiosis (Figure 2E), and it is neither preceded nor followed by any type of syngamy (Cleveland, 1947). It is also recognized that, for endomitosis to occur, chromosome condensation and telomere duplication are important, which clearly distinguishes it from polyteny (Erenpreisa et al., 2005). It has already been hypothesized that Mos would be part of the molecular basis for somatic reduction and a return to the mitotic cycle of endopolyploid tumor cells (Erenpreisa et al., 2005). This proposal by Jekaterina Erenpreisa and Mark Cragg is consistent with results in two-cell embryos from Clytia hemisphaerica, in which “lower concentrations of Mos RNA caused cleavage arrest with multiple nuclei or spindles within a common cytoplasm,” demonstrating the occurrence of Mos-mediated endomitosis in Cnidaria (Amiel et al., 2009), consistent with coenocytic division carried out by our closest relatives, ichthyosporeans (Ros-Rocher et al., 2021) (Figure 2C). Also, the blockade of Mos in oocytes produces a “parthenogenetic development with complete cleavages” but the presence of “multiple asters and spindle poles during mitotic cycling,” further corroborating the presence of endomitosis in Asterina pectinifera (starfish) (Tachibana et al., 2000). The multiple asters and spindle poles are also present in the megakaryocyte endomitosis and possibly arise due to alterations in the regulation of mitotic exit (Vitrat et al., 1998), known to be controlled by Mos. In our NFM model, we speculate the control of three types of division (mitosis, meiosis, and endomitosis) and the recruitment of Mos (or a similar protein) would have been essential for germline emergence and incorporation of polyploidy in animal embryogenesis. As will be discussed in another article, the control of these types of cell division, including the two types of meiosis (meiosis and endomitosis), was fundamental to the success of the multicellular embryo.

According to our proposal, for the embryo to achieve reconstruction and structural coherence in following generations, it must necessarily incorporate NP (through NFM) into the cellular context of germ cell formation. In other words, the germline is closely linked to NP, because it is at the heart of animal formation. Therefore, germline and NP are impossible to separate in the context of metazoans. We may even predict that germline gene expression will manifest in the context of somatic cancers and that germline cancers may tend to produce arrangements that can be “interpreted as attempts at organization” (Pierce, 1967).

Consistent with our proposal, several studies on D. melanogaster (Janic et al., 2010), mice (Ma et al., 2012), and humans (Koslowski et al., 2004; Rosa et al., 2012; Rousseaux et al., 2013) revealed that tumors acquire a state very similar to that of the germline, indicating that tumorigenesis in metazoans involves a soma-to-germline transition, possibly contributing to the acquisition of neoplastic characteristics (Wang et al., 2011; Feichtinger et al., 2014). It is important to remember that the boundary between Soma and germline is fluid (Edgar et al., 2021) and associated with a neoplastic force for embryo construction. It was demonstrated that germline gene expression has the potential to be oncogenic in D. melanogaster (Janic et al., 2010) and is associated with more clinically aggressive tumors in humans (Rousseaux et al., 2013). Inactivation of some germline genes results in suppression of tumorigenesis, indicating that they play an essential role in tumor development (Janic et al., 2010). Finally, the soma–germline transition was also observed in Caenorhabditis elegans, suggesting a conserved functional relationship between tumorigenesis and germline gene expression in metazoans (Unhavaithaya et al., 2002; Wang et al., 2005). Such relationships are sensible in the context of our proposal of NP as an evolutionary engine.

Also consistent with our hypothesis is that, in germline cancers, two different forces oppose each other: a force of embryonic organization and differentiation contained in the germ cell structure (neoplastic process) and a disruptive force of cancer (disease). In this regard, the works of G. Barry Pierce are very inspiring. The author demonstrated that, in a small area of testicular teratocarcinoma, the mesenchyme and endoderm are often organized in patterns that resemble early stages of embryogenesis, having been called embryoid bodies (Dixon and Moore, 1953; Pierce et al., 1959; Pierce, 1967). In the works of Winston Evans, some embryoids “closely mimic the form of normal early human embryos” (Evans, 1957). A conceptual basis for the developmental model of cancer was well documented in Pierce’s works; thus, the dogmas that neoplastic forces are stable and irreversible were broken. It was hypothesized that cancer is controlled by intrinsic mechanisms of the embryo (embryonic induction) (Pierce, 1967; Pierce, 1983; Arechaga, 2003). In the context of this article, we extend the concept to propose that the embryo and cancer are the result of an equilibrium achieved within the NFM during embryogenesis and at the beginning of the evolution of metazoans.

Embryogenesis records will be restricted to a cellular set that incorporates NFM and can later reproduce and restart in a coherent and systematic way the construction of a new animal, with the same structural form as the reproduced one. This is a central aspect that clearly separates any multicellular organization from a true animal embryo.