Gnathostomata

All living vertebrates (Craniata) divide into two sister superclasses: Cyclostomata (the ~120 living species of jawless agnathans: lampreys [Petromyzontiformes] and hagfishes [Myxiniformes]) and Gnathostomata (the >65,000 living species of jawed vertebrates, encompassing cartilaginous fishes, ray-finned fishes, lobe-finned fishes, and tetrapods).

Bridging the evolutionary gap between the earliest Early Cambrian unmineralized jawless vertebrates (Myllokunmingia and Haikouichthys from the 518-million-year-old Chengjiang biota of China, which already possessed W-shaped myomeres, paired camera eyes, otic and nasal capsules, gill pouches, and cartilaginous arcualia) and the first Silurian jawed fishes is a rich Ordovician-to-Devonian fossil radiation of armored jawless fishes—collectively termed ostracoderms—followed by the armored jawed placoderms.

For nearly 40 years (1970s–2000s), morphological cladistics placed lampreys closer to jawed vertebrates ("Vertebrata" sensu stricto) while excluding hagfishes to a basal sister branch ("Craniata"), because lampreys possess rudimentary cartilaginous vertebral arches (arcualia), two semicircular canals, extrinsic eye muscles, and osmoregulatory kidneys, whereas adult hagfishes appeared to lack vertebral elements, lens-bearing eyes, and blood-pressure osmoregulation (remaining iso-osmotic with seawater).

However, every nuclear and mitochondrial molecular dataset since Mallatt & Sullivan (1998), Kuraku et al. (2009), Heimberg et al. (2010, microRNAs), and chromosome-level genome assemblies of the sea lamprey (Petromyzon marinus) and inshore hagfish (Eptatretus burgeri) by Juan Pascual-Anaya, Shigeru Kuratani, Jeramiah J. Smith, Marlétaz et al. (2024), and Yu et al. (2024) proved that lampreys and hagfishes form a strictly monophyletic clade: Cyclostomata ("round mouths"):

1. Embryological Discovery of Hagfish Vertebrae: In 2011–2013, Kinya G. Ota, Shigeru Kuratani, and colleagues bred embryos of the inshore hagfish Eptatretus burgeri and discovered that hagfishes do possess ventral cartilaginous vertebral elements (arcualia) in their post-cloacal tail, derived from a Pax1/9⁺ and Twist⁺ sclerotome—proving that the apparent absence of vertebrae in the hagfish trunk is a secondary evolutionary reduction, not a primitive pre-vertebrate state!
2. Genome Duplication Timing (1R-WGD Shared, Followed by Cyclostome vs. Gnathostome Divergent Polyploidies): Chromosome-scale synteny analyses (Simakov et al., 2020; Marlétaz et al., 2024) established that the first whole-genome duplication (1R-WGD) occurred in the stem ancestor of all vertebrates prior to the Cyclostomata–Gnathostomata split (~530 Ma), after which gnathostomes underwent a second allotetraploidization (2R-WGD) while cyclostomes underwent their own independent hexaploidization (triplication) event!

How did an ancestral jawless vertebrate—whose pharynx was supported by an unjointed series of cartilaginous gill bars—transform its first pharyngeal arch (the mandibular arch, PA1) into a dorsal upper jaw (palatoquadrate) and a ventral lower jaw (Meckel's cartilage) hinged together at a central joint?

1. The Hox-Negative Status of the First Pharyngeal Arch: As demonstrated by Drew M. Noden, Hunt, Lumsden, and Rijli, cranial neural crest cells migrating from the midbrain and rhombomeres r1–r2 into the first pharyngeal arch (PA1) are strictly Hox-negative (whereas PA2 expresses Hoxa2, and PA3+ express Hox3/4). Ectopic expression of Hoxa2 in PA1 transforms the jaw into a second hyoid arch, whereas deleting Hoxa2 in mice causes PA2 to duplicate a mirror-image set of jaws.
2. Liberation of the Mandibular Arch via Nasohypophyseal Separation (Galeaspida & Kuratani's Heterotopy Hypothesis): Why couldn't lampreys evolve jaws even though their PA1 is also Hox-negative? As shown by Shigeru Kuratani, Gai Zhi-Kun, and Philip C. J. Donoghue (2004–2011) through synchrotron X-ray microtomography of the 435-million-year-old Silurian jawless galeaspid Shuyu zhejiangensis, in cyclostomes the median nasohypophyseal placode sits directly beneath the forebrain, physically blocking premandibular/mandibular neural crest cells from migrating forward between the eyes and nose. In Galeaspida and Osteostraci, the nasal sacs separated laterally while the hypophyseal duct opened into the roof of the mouth—opening a central craniofacial corridor through which neural crest cells migrated forward to build the trabeculae cranii and free the mandibular arch to form jaws.
3. The Dorsoventral Dlx1–6 Nested Homeobox Code and the Bapx1 (Nkx3.2) Jaw Joint: Within thegnathostome mandibular arch, John L. R. Rubenstein, Michael J. Depew, and colleagues (2002) discovered that three tandem pairs of Distal-less homeobox genes establish a nested dorsoventral code driven by ventral Endothelin-1 (Edn1 → Ednra → Hand2) signaling:
   • Dorsal Maxillary/Palatoquadrate Domain: Expresses only Dlx1 and Dlx2;
   • Intermediate Domain: Expresses Dlx1/2 + Dlx5/6;
   • Ventral Mandibular (Meckel's Cartilage) Domain: Expresses Dlx1/2 + Dlx5/6 + Dlx3/4 + Hand2.

Knockout of Dlx5 and Dlx6 in mice completely abolishes ventral lower-jaw identity, transforming the lower jaw into a mirror-image duplicate upper jaw (maxilla) complete with upper-jaw rugae and vibrissae! At the exact boundary between the dorsal Dlx1/2 domain and the ventral Dlx5/6 domain, the transcription factor Bapx1 (Nkx3.2) is activated to specify the primary palatoquadrate–Meckel's jaw joint (which later evolved into the mammalian malleus–incus middle-ear joint).

A second landmark discovery of comparative genomics is that all vertebrates possess an adaptive immune system with clonally diverse B-like and T-like lymphocytes, yet cyclostomes and gnathostomes built their antigen receptors from completely unrelated protein folds!

• Gnathostomes (From Sharks to Humans): Assemble immunoglobulin-superfamily (IgSF) antibodies (BCR) and T-cell receptors (TCR) via RAG1/RAG2 transposase-mediated V(D)J recombination.
• Cyclostomes (Lampreys and Hagfishes): Completely lack RAG1/2, Ig, TCR, and MHC. In 2004, Max D. Cooper, Zeev Pancer, and colleagues discovered that lampreys and hagfishes instead assemble Variable Lymphocyte Receptors (VLRs: VLR-A, VLR-B, and VLR-C) out of solenoidal leucine-rich repeat (LRR) modules (structurally similar to innate Toll-like receptors)! Using APOBEC/AID-family cytidine deaminases (CDA1 in thymus-equivalent gill-tip VLR-A/C T-like cells and CDA2 in hematopoietic VLR-B B-like cells), developing lamprey lymphocytes copy flanking LRR cassettes sequentially into an incomplete germline VLR gene via gene conversion, generating >10¹⁴ distinct antigen-binding concave LRR surfaces.

How does a Pacific hagfish (Eptatretus stoutii) turn 20 liters of seawater into solid slime in less than a second?

When bitten at its gill pouches by a shark or conger eel, a hagfish fires milky exudate from ~150 ventrolateral slime glands, each containing two specialized cell types: gland mucous cells (which release negatively charged mucin vesicles that rupture via Ca²⁺/aquaporin influx) and gland thread cells. Each single gland thread cell is packed with a single, tightly coiled, 15-centimeter-long intermediate filament protein skein (α- and γ-biopolymer threads) engineered by Douglas S. Fudge and colleagues: upon contact with seawater, the skein unravels hydrodynamically in <100 milliseconds, and the threads trap mucin strands and water across a volume 10,000 times that of the initial exudate—instantly clogging the predator's gills while the toothless hagfish ties its own body into an overhand knot and slides the knot forward to wipe the slime off its own skin.

How did Silurian fossils like Entelognathus and Qilinyu rewrite the origin of bony fishes vs. sharks?

For a century, sharks (Chondrichthyes, which lack macromeric dermal skull bones) were assumed to represent the primitive condition of jawed vertebrates, whereas bony fishes (Osteichthyes, with maxilla, premaxilla, and dentary jaw bones) were considered derived. In 2013–2016, Min Zhu, Per E. Ahlberg, and colleagues discovered the 419-million-year-old Silurian armored placoderms Entelognathus primordialis and Qilinyu rostrata in Yunnan, China: despite having a typical placoderm braincase and trunk armor, both fossils possess authentic osteichthyan-style marginal jaw bones (premaxilla, maxilla, and dentary)—proving that bony jaws evolved in stem-gnathostomes before the split between sharks and bony fishes, and that sharks secondarily lost their ancestral dermal skull bones!

1. Depew, M. J., Lufkin, T., & Rubenstein, J. L. (2002). Specification of jaw subdivisions by Dlx genes. Science, 298(5592), 381–385. doi:10.1126/science.1075703
2. Pancer, Z., Amemiya, C. T., Ehrhardt, G. R., Ceitlin, J., Gartland, G. L., & Cooper, M. D. (2004). Somatic diversification of variable lymphocyte receptors in the agnathan sea lamprey. Nature, 430(6996), 174–180. doi:10.1038/nature02740
3. Gai, Z., Donoghue, P. C., Zhu, M., Janvier, P., & Stampanoni, M. (2011). Fossil jawless fish from China foreshadows early jawed vertebrate anatomy. Nature, 476(7360), 324–327. doi:10.1038/nature10276
4. Zhu, M., Yu, X., Ahlberg, P. E., Choo, B., Lu, J., Qiao, T., ... & Zhu, Y. A. (2013). A Silurian placoderm with osteichthyan-like marginal jaw bones. Nature, 502(7470), 188–193. doi:10.1038/nature12617
5. Marlétaz, F., Timoshevskaya, N., Timoshevskiy, V. A., Parey, E., Simakov, O., Gavriouchkina, D., ... & Rokhsar, D. S. (2024). The hagfish genome and the evolution of vertebrates. Nature, 627(8005), 811–819. doi:10.1038/s41586-024-07070-3