A neuron is a very fascinating cell. It is a cell that is specialized to detect changes in the environment, translate that environmental information into the language of membrane potential changes (electrical signals), and engage in long-distance communication by transmitting such electric signals to distant targets in a body. The key to this transmission is the synapse, where the axon of one neuron uses exocytosis to release neurotransmitters that can diffuse and bind to receptors on the dendrite of an adjacent neuron.
In essence, the synapse is a ‘decision point’ for determining whether or not the signal will proceed. It is the synapse which confers immense plasticity and potential for control to the whole circuit.
We have just seen that at least one single-celled organism contains most of the calcium toolkit that would be exploited in the release of neurotransmitters from the pre-synaptic neuron. But what about the other side of the synapse?
Over a year ago, researchers found that the sponge, which has no nervous system and does not form neurons, contains most of the machinery needed to complete the synapse. According to this news article:
Considered among the most primitive and ancient of all animals, sea sponges have no nervous system (or internal organs of any kind, for that matter), notes Todd Oakley, assistant professor in the Department of Ecology, Evolution and Marine Biology at the University of California, Santa Barbara. But, he adds, they “have most of the genetic components of synapses.”
He, Oakley and the rest of the team listed all the genes known to be operative in synapses in the human nervous system. They then examined the sponge genome. “That was when the surprise hit,” said Kosik. “We found a lot of genes to make a nervous system present in the sponge.”
Thus is not surprising from the perspective of front-loading evolution. On the contrary, this is the very type of data we would expect if neurons and nervous systems were front-loaded.
“We found this mysterious unknown structure in the sponge, and it is clear that evolution was able to take this entire structure and, with small modifications, direct its use toward a new function,” said Kosik. “Evolution can take these ‘off the shelf’ components and put them together in new and interesting ways.”
Yes, evolution works quite well if it has handy tools on the shelf whose functional potential can be more readily exploited in a multicellular context. After all, evolution largely depends on the tools and material it is handed.
According to the research article itself:
The data presented here support the presence of a proto-post-synaptic scaffold in the last common ancestor to all living animals. The presence of a large number of post-synaptic genes in the genome of demosponge Amphimedon, the nearly absolute conservation of binding domains and ligands between this sponge and animals with neurons, as well as the expression of a set of post-synaptic mRNAs in the same cell type, suggest the proto-post-synaptic scaffold existed as an assembled functional structure very early in animal evolution.
Another way of putting this is as follows: if we rewound the tape of life just prior to the origin of metazoa, it would not be surprising to see neurons evolve all over again. The stage was set.
But the research paper also notes there are certain key synaptic genes missing from the sponge genome. This raises some interesting questions from an FLE perspective. As we sequence more genomes from other sponges and protozoa, we may find these other key components. Also, to what degree would this toolkit act as bait for fishing out the key genes (the mechanism of baiting is explained in The Design Matrix)?
Moving from the synapse, we can now turn our attention to the fact that hugely significant chunks of the circuitry needed to form neurons during embryological development are part of the toolkit possessed by organisms that do not make neurons.
The title of this latest FLE-friendly study is “Sponge Genes Provide New Insight into the Evolutionary Origin of the Neurogenic Circuit” and was published in the Aug 5th issue of Current Biology. Let’s have a look.
Given that sponges contain many genes needed to construct a synapse, these researchers attempted to determine “whether homologs of genes that direct the primary specification of neurogenic fields in bilaterian embryos—basic helix loop helix (bHLH) proneural genes—and the selection of neural precursors within these fields—the Notch signaling pathway—are present in the Amphimedon genome.”
And they hit pay dirt!
Amphimedon possesses 16 bHLH genes, one of which, AmqbHLH1, appears to be derived from a single ancestral gene that generated, via gene duplications, the atonal-neurogenin-related bHLH gene families (the ‘‘Atonal superfamily’’) that include most bilaterian proneural genes  (Figure S2). Elements of the Notch signaling pathway are also present in the Amphimedon genome, which encodes a single Notch receptor, AmqNotch, and five Notch ligands, AmqDelta1– AmqDelta5, as well as a suite of proteins that direct the initiation, transduction, and deactivation of Notch signaling in bilaterians (Table S1).
They then reached the following conclusion:
In Amphimedon, the expression of these neurogenic molecules is associated with globular cells, which also expresscomponents of what became the eumetazoan postsynapse , demonstrating that the molecular preadaptations for nerve cell formation can be traced back to an ancestor that existed prior to the evolution of true neurons. (emphasis added)
Such preadaptation suggests that neurons may have indeed been front-loaded, especially when viewed in light of the other independent preadaptations – the calcium toolkit and the post-synaptic scaffold. And this leads us to two important questions.
1. If we replayed the Gouldian tape of life back to the first metazoans devoid of neurons, would we expect neurons to evolve all over again?
2. If these three preadaptations did not exist, would neurons have ever likely evolved?
The hypothesis of FLE would have us answer the questions as yes and no. Can the non-teleological viewpoint provide answers to these questions? I think not.
Let’s now dig a little deeper into the rabbit hole where sponges contain significant chunks of the developmental circuitry needed for neuron formation. Stick with me and I’ll show ya something pretty cool (albeit with some over-simplification).
What is truly neat about this study is that the researchers did not merely find gene sequences, they actually tested the function of one of the key sponge genes (AmqbHLH1) in a deeply significant manner.
First, they noted, “Phylogenetic analysis suggests that AmqbHLH1 is homologous to all the atonal-related bHLH genes found in bilaterians, including the proneural genes atonal and neurogenin.Accordingly, the bHLH domain of AmqbHLH1 shows a mixture of conserved amino acid residues observed in the bHLHs encoded by the different atonal-related bHLH genes but only a limited similarity to any one gene in particular. The atonal genes are important proneural genes in Drosophila, and the neurogenin genes are important proneural genes in vertebrates.”
In other words, atonal is the fruit fly version of the sponge gene and neurogenin is the vertebrate version of the sponge gene. And when we look at the sequence of the sponge gene, it is a mosaic of the fly and frog gene.
When neurogenin (the vertebrate version) is artificially added to fruit fly embryos, it fails to trigger any significant nervous tissue development. And when atonal (the invertebrate version) is injected into frog embryos, it fails to trigger any significant nervous tissue development. Yet here’s the key point. When the sponge version is injected into either frog or fly embryos, it stimulates nervous tissue development in both. In other words, the sponge gene acts like neurogenin in vertebrates, yet acts like atonal in flies.
The researchers describe it as follows:
Taken together, our data indicate thatAmqbHLH1has strong proneural effects when it is misexpressed in either Xenopus or Drosophila. In Xenopus, AmqbHLH1 behaves like neurogenin genes, inducing the formation of numerous ectopic sensory neurons and synergizing with XMyT1. In Drosophila, AmqbHLH1 behaves like atonal genes, inducing a high number of ectopic sensory organs. AmqbHLH1 thus possesses structural/functional properties that are found across a range of bilaterian proneural proteins and that are required to promote neural development, including the ability to set up a Notch-Delta-mediated lateral inhibition process.
I find all this to be significant for several reasons. First, it ties into the section of my book entitled, Unpredictably Predictable, where I note, “While a particular course of evolution may be inherently unpredictable, not everything about evolution is unpredictable.” As one example, I go on to note, “Even though yeast and humans have diverged from each over a billion years ago, where each lineage has experienced its own unique history of contingent evolutionary challenges, much of the basic cell cycle machinery from human cells can replace that found in yeast.”
In this case, this theme of modularity and deep homology is further emphasized in that we have a sponge gene that is able to trigger the development of neurons in two distantly related organisms, flies and frogs. But this time it is more interesting, in that the fly gene does not function in frogs and the frog gene does not function in flies. Since the sponge gene can function in both, it is as if both the fly and frog lineages independently lost information that was housed in the sponge gene.
And this takes us to another theme from The Design Matrix:
Allan Force, from the Virginia Mason Research Center, adds another twist to the story of gene duplication that reinforces its great potential for front loading……Subfunctionalization, through gene duplication, clearly is a process quite friendly to front-loading, as it is consistent with an originally designed complex state that can be teased apart over time in a manner that channels evolution.
And in fact, this point is echoed by the researchers:
Phylogenetic analysis indicates thatAmqbHLH1is neither a neurogenin nor an atonal gene but derives from a single ancestral gene that produced both families (as well as others, such as the Net family) by gene duplications in the eumetazoan lineage. Taken together, our data suggest that the Atonal superfamily’s last common ancestor already possessed ‘‘proneural-like’’ capabilities, and although the expansion of this family in eumetazoans resulted in the partitioning of these capabilities among descendent paralogs, no such expansion occurred in the poriferan lineage, in which there remains a single ortholog.
Another way of saying this is that gene duplication allowed the sponge gene, which contains both fly and frog activity, to tease these dual preadaptations apart from each other so they could flourish in their respective, future lineages. Subfunctionalization thus untapped the dual potential that was inherent in the ancestral. Put simply, two front-loading events built into a single sequence.