Toolkit Parts for Multicellularity

by Techne

The article about the evolutionary history of body size on Earth has raised some interest. Words like "latent evolutionary potential was realized", "realize preexisting evolutionary potential" and" a major innovation in organismal complexity—first the eukaryotic cell and later eukaryotic multicellularity" seem to have raised a few eye brows. Are "latent", "pre-existing", "innovation" and "potential" the appropriate words?

From the following figure, the earliest multicellular (Grypania spiralis) eukaryotic fossil dates back ±1.6 billion years ago (bya) and at present the earliest evidence for eukaryotic cells is posited to have existed 1.68-1.78 bya , (perhaps 1.8 bya or possibly even 2.1 bya) (Figure 1).

Figure 1: From the article.

The following tree is adapted from discoverlife.org with the tentative dates for the origins of archaea and bacteria, eukaryotes, as well as the origins of multicellular body plans (>3 cell types) (Figure 2).*

Figure 2: Tree of life (Addapted from discoverlife.org)*

As suggested by the paper, increases in cellular sizes roughly coincide with the alleviation of at least one environmental constraint, namely low atmospheric oxygen pressure. The origin of body plans (±0.6 bya) also seem to coincide with an increase in atmospheric pressure. Why could that be?

A look at hedgehogs

With about 6, 000 spines on their back, an excellent sense of smell, a running speed of 4.5 mph a normal heart rate of up to 19o bps and 10 bps during hibernation, hedgehogs are interesting little animals. Hedgehog (hh) genes are equally fascinating. The reason for the name of this gene is that a malfunctioning hh gene often results in the formation of small pointy projections on embryos, similar to that of a hedgehog. So what does it do?

Functions

The hh signaling pathway plays a fundamental role in cell patterrning, cell proliferation and participates in the development of tissues and organs during the stages of animal development. It exerts its effect by influencing the transcription of many target genes in a concentration dependent manner.

Mechanism of action and signal transduction: Hints from hedgelings and hoglets

The hh protein comprises of two domains, namely the hedge domain (hedgling) and the hog domain (hoglet). The hedge domain acts as a ligand after processing and binds to a set of conserved receptors to activate downstream signal transduction pathways [1]. After transcription, the hh-gene undergoes a post-translational autocatalyzing editing process initiated by the hoglet resulting in the formation of the hedgling protein. Further processing of the hedgling occur and include the palmitoylation and sterolation (addition of cholesterol) of the ligand (Figure 3). Interestingly, hh proteins are the only examples of sterolation in contempory biology (more on that later) [2]. After processing, the hedgling ligand is transported through the Dispatched receptor where it binds to a specific lipid transport molecule (different in invertebrates and vertabrates) and is transported and binds to the 12-transmembrane protein called Patched. Internalisation of Patched alleviates the inhibitory effect of Patched on the 7-transmembrane protein Smoothened. This in turn activates the hedghog-related transcription factors (Gli in vertebrates and Ci in invertabrates) (Figure 3) [2]. This relatively simple pathway plays a crucial role in the unfolding of the developmental program in vertebrates and invertebrates.

Figure 3: Hedgehog signal transduction. The hedgehog protein is post-translationally modified through autocatalyzation and palmitate and cholesterol addition. Processed hedgelings are transported to the extracellular matrix through dispatched receptors and in turn transported by lipid transport molecules to bind to patched receptors. Binding of hedgling molecules to Patched receptor results in the subsequent activation of hedgehog mediated transcrition factors e.g. Gli in vertebrates and Ci in invertebrates.

With the knowledge of some of the proteins that play a part in hh-signal control, let's look at the evolution and origin of some of the components. The following proteins can be used for BLAST.
Hedgling (Amphimedon Queenslanica)
Hoglet (Monosiga Ovata)
Patched (Ciona Intestinalis)
Dispatched (Ciona Intestinalis)
Suppresor of Fused (Sufu) (Ciona Intestinalis)
Smoothened (Ciona Intestinalis)
Fused (Drosophila)
Gli1 (Human) or Ci (Drosophila)
Kif27 (vertebrate) or Cos2 (Drosophila)

Using the InterProScan Tool with these sequences, the following results were obtained:
Hedgling: The oldest (phylogenetically) bona fide hedgeling found so far is in the genome of the sponge, Amphimedon Queenslanica. However, the structure of this domain is structurally homologous to the zinc-binding motif in bacterial D-alanyl-D-alanine carboxypeptidases (the same motif found in beta-lactamases and the various nylonase genes).
Hoglets: Hoglets are typical Intein (internal protein) proteins also known as HINTs (hedgehog inteins) [3]. Inteins are selfish DNA elements that are distributed accross all the domains of life [4].
Patched: Patched is a transmembrane protein with a sterol sensing domain (SSD) and is also distributed in all the domains of life.
Dispatched: Dispathed is also a transmembrane protein with a SSD and forms a subfamily of the sterol sensing receptors. Also present in all the domains of life.
Fused: Fused is kinase conserved in all the domains of life.
Suppresor of Fused (Sufu): Sufu yielded an interesting result. Acting as a suppressor of the hh-signaling pathway, it is limited to the bilaterians and cnidaria and bacteria. It seems to have been lost in other linages.
Smoothened (Frizzled domain, G-protein-coupled receptor (GPCR) domain): Smoothened contains a frizzled domain and a GPCR domain. The frizzled domain is limited to eukaryotes, while the GPCR domain is conserved in all the domains of life.
Gli1: This protein (and cos2) is a transcription factor and hh-signaling converges to control the activity of this protein. It is a zinc-finger protein. While zinc-finger proteins are conserved in all domains of life, this particular protein seems to be limited to eukaryotes.
Kif27: Kif27 (and Cos2) is a kinesin-related protein (KRP). Kif27 appears to be functional molecular motor while Cos2 seems to have lost the ability to function as a motor protein. KRPs however are conserved accross all domains of life [5]. A conserved function of KRPs is to facilitate movement of vesicle along microtubules and one of the functions of Cos2 seems to be just that [6].

From the above, the following picture of the components of the hh-signaling tool kit can be drawn.

Figure 4: Origins of the parts in the hedgehog signaling pathway. (Red = absent, Orange = reasonable sequence and/or structural simlarity, Green = present, Graded green = part of the same family).*

Note that many of the components of the signaling pathway are present in various bacterial and archaeal lineages. Also note that the origin of multicellular body plans roughly coincide with an increase in atmospheric oxygen pressure as well as the first bona fide hedgling. Remember, hedglings are the only examples of post-translational sterolation (addition of cholesterol) of proteins in contempory biology. Why is this interesting? Well, oxygen is needed for cholesterol synthesis, more importantly, oxygen is needed for placing the hydroxyl group in the 3-position of cholesterol which plays a crucial role in subsequent transformations (including sterolation). Thus, while large parts of the hh-signaling pathway was present, a little extra oxygen was needed to unlock multicellular signaling capabilities of hedglings.

Therefore, words like "pre-existing", "latent" and "potential" seem apt in describing the hedghog signaling pathway and the unfolding of multicellular body plans in relation to the increase in atmospheric oxygen pressure. "Innovation" perhaps not so much, seeing that only real innovation was bought on about by life itself namely the increase in atmospheric oxygen. This increase in atmospheric oxygen in turn seemed to have unlocked the pathways to multicellular body plans (>3 cell types).

Gene loss vs Innovation
Looking at the hh-signaling pathway, there seem to be very little innovation, and a lot of co-option of pre-existing information into new functions. Sufu was an interesting example of gene loss only to be co-opted later into a role in the hh-signaling pathway. With this in mind, what can one expect to find in the Last Universal Common Ancestor (LUCA)? Also consider the following. The Tetrahymena thermophila (alveolate) genome has been sequenced, and a number of genes that are absent in yeast (fungi), are found in amoebas, vertebrates, invertebrates as well as in the Tetrahymena genome. It paints the following picture (Figure 5) [7].

Figure 5: Genes present in Tetrahymena thermophila but absent in yeast indicate either convergnece in higher organisms or that the genes were present in the eukaryote common ancestor.

Possible points for discussion (but not limilted to them):
1) Why does an increase in atmospheric oxygen seem to have the effect of driving eukaryotic multilcellular life but not bacteria and archaea? Is an intrinsic and latent property present in this domain?
2) Gene loss vs innovation: How much gene loss and how much innovation (not just co-option) has occured from the LUCA? (Speculating)
3) Why did all the tool kit parts for the hh-pathway converge on a single sterolation pathway when so many other possibilities are available? Or is it the optimal possibility and random variation and selection processes used by life hit a global optimum?

References
1. Matus DQ, Magie CR, Pang K, Martindale MQ, Thomsen GH. The Hedgehog gene family of the cnidarian, Nematostella vectensis, and implications for understanding metazoan Hedgehog pathway evolution. Dev Biol 2008; 313: 501-518.
2. Bijlsma MF, Spek CA, Peppelenbosch MP. Hedgehog: an unusual signal transducer. Bioessays 2004; 26: 387-394.
3. Perler FB. Protein splicing of inteins and hedgehog autoproteolysis: structure, function, and evolution. Cell 1998; 92: 1-4.
4. Pietrokovski S. Intein spread and extinction in evolution. Trends Genet 2001; 17 465-472.
5. Varjosalo M, Taipale J. Hedgehog: functions and mechanisms. Genes Dev 2008; 22: 2454-2472.
6. Ogden SK, Ascano M Jr, Stegman MA, Robbins DJ. Regulation of Hedgehog signaling: a complex story. Biochem Pharmacol 2004; 67: 805-814.
7. Eisen JA, Coyne RS, Wu M, Wu D, Thiagarajan M, Wortman JR. et al. Macronuclear genome sequence of the ciliate Tetrahymena thermophila, a model eukaryote. PLoS Biol 2006; 4: e286.

*-corrections are welcomed.

This entry was posted on Sunday, January 18th, 2009 at 12:19 pm and is filed under Biology, Cell, Evolution, Front-loading, Proteins. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site. The trackback link is: http://telicthoughts.com/toolkit-parts-for-multicellularity/trackback/

12 Responses to “Toolkit Parts for Multicellularity”

Bradford Says:
January 18th, 2009 at 12:57 pm
Techne:

1) Why does an increase in atmospheric oxygen seem to have the effect of driving eukaryotic multilcellular life but not bacteria and archaea? Is an intrinsic and latent property present in this domain?
2) Gene loss vs innovation: How much gene loss and how much innovation (not just co-option) has occured from the LUCA? (Speculating)
3) Why did all the toolkit parts for the hh-pathway converge on a single sterolation pathway when so many other possibilities are available? Or is it the optimal possibility and random variation and selection processes used by life hit a global optimum?

Good work Techne. I have not given prior consideration to #s 1 and 3. Never even thought to ask the questions. The fact that you raise the questions speaks well of your analytical bent. I have no answers yet but do think it appropos to quote a genius crackpot. If the great ocean of truth lays before us then it behooves critics to refrain from dismissive postures vis a vis ID. Isaac Newton:

I seem to have been only like a boy playing on the seashore, and diverting myself in now and then finding a smoother pebble or prettier shell than ordinary, whilst the great ocean of truth lay all undiscovered before me.

Comment by Bradford — January 18, 2009 @ 12:57 pm

Raevmo Says:
January 18th, 2009 at 1:18 pm
Techne:

Why does an increase in atmospheric oxygen seem to have the effect of driving eukaryotic multilcellular life but not bacteria and archaea?

Who says there was no effect on non-eukaryotes?

Comment by Raevmo — January 18, 2009 @ 1:18 pm

Techne Says:
January 18th, 2009 at 1:33 pm
Hi Bradford,

Nice quote. Guess if you are not a crackpot, you are not good at what you do.

Hi Raevmo,

Raevmo: Who says there was no effect on non-eukaryotes?

It didn't. From:
Rokas A. The origins of multicellularity and the early history of the genetic toolkit for animal development. Annu Rev Genet. 2008; 42: 235-251.

Organism vs Cell type number:
Actinobacteria 3
Cyanobacteria 3
Myxobacteria 3
Cellular slime molds 3
Animals 3–122
Fungi 3–9
Volvocine green algae 2
Plants 5–44
I suspect archaea do not consist of more than 3 types.

Also no body plans in bacteria or archaea.

Comment by Techne — January 18, 2009 @ 1:33 pm

Raevmo Says:
January 18th, 2009 at 4:41 pm
Techne, I agree that eukaryote evolution has been much more spectacular than prokaryote evolution since the jumps in O2-concentration. But I suspect it has also affected prokaryotes a lot, albeit in a less obviously visible way.

Comment by Raevmo — January 18, 2009 @ 4:41 pm

Unlocking a Buried Potential « The Design Matrix Says:
January 18th, 2009 at 7:53 pm
[...] 18, 2009 by Michael Over at Telic Thoughts, Techne has posted an excellent analysis that digs a little deeper into the hypothesis of front-loading evolution, adding to its [...]

Pingback by Unlocking a Buried Potential « The Design Matrix — January 18, 2009 @ 7:53 pm

fifth monarchy man Says:
January 18th, 2009 at 9:23 pm
Raevmo

But I suspect it has also affected prokaryotes a lot, albeit in a less obviously visible way.

Any suggestions on how to quantify this effect. Or is this an untestable hunch?

Peace

Comment by fifth monarchy man — January 18, 2009 @ 9:23 pm

KC Says:
January 19th, 2009 at 9:34 am
1) Why does an increase in atmospheric oxygen seem to have the effect of driving eukaryotic multilcellular life but not bacteria and archaea? Is an intrinsic and latent property present in this domain?

The article cites a paper that discusses one possible effect: the rising oxygen levels brought selection pressure in bacteria and archaea for transmembrane proteins which did not exclude oxygen:

Acquisti C, J Kleffe & S Collins (2007). Oxygen content of transmembrane proteins over macroevolutionary time scales. Nature 445: 47-52

From the abstract:

We observe that the time of appearance of cellular compartmentalization correlates with atmospheric oxygen concentration. To explore this correlation, we predict and characterize the topology of all transmembrane proteins in 19 taxa and correlate differences in topology with historical atmospheric oxygen concentrations. Here we show that transmembrane proteins, individually and as a group, were probably selectively excluding oxygen in ancient ancestral taxa, and that this constraint decreased over time when atmospheric oxygen levels rose. As this constraint decreased, the size and number of communication-related transmembrane proteins increased. We suggest the hypothesis that atmospheric oxygen concentrations affected the timing of the evolution of cellular compartmentalization by constraining the size of domains necessary for communication across membranes.

KC

Comment by KC — January 19, 2009 @ 9:34 am

Techne Says:
January 19th, 2009 at 10:36 am
Hi KC,

Thanks for the interesting article. From the article:

ConclusionIn this study, we have shown that transmembrane proteins can be divided into two groups according to their oxygen content. Independent topology prediction reveals these same two groups. We have shown that the proportion of receptors to channels increases over time and coincides with a change in cellular organization. In addition, older proteomes contain less oxygen per residue and produce fewer high-oxygen proteins. Taken together, this suggests that oxygen use was selected against in these proteomes. This constraint lessened over time as the concentration of atmospheric oxygen increased, which resulted in the extracellular domains of transmembrane proteins increasing in size over time faster than the internal domains. Consequently, we propose the following hypothetical mechanism: atmospheric oxygen concentration constrained the topology of ancient transmembrane proteins by limiting the number and size of external domains that could be formed. Any mechanistic explanation of how atmospheric oxygen concentration limited the number and size of external domains is necessarily speculative. One possibility is that it was simply futile to exude large, oxygen-rich domains in a reducing atmosphere where oxidized amino acids could have been rapidly reduced. In this case, the use of oxygen-rich amino acids would have been selected against by natural selection because protein structure would have been more robust when fewer oxidized residues were exuded.

Stated differently, eukaryotic proteomes were more receptive and robust in accepting oxygen-rich amino acids.

Linking this to the timing of appearance of eukaryotic cells implies that the oxygen content is preferentially increased in receptors, and that this increase affects receptor function. This makes intuitive sense because the external domains of receptors required for communication have specific secondary and tertiary structures, many of which have some minimum size [23]. This is consistent with the bias we found towards having both longer and more oxygen-dense external domains in receptors relative to channels, and with the fact that eukaryotic genomes encode more and larger receptors than do prokaryotes. This suggests that protein oxygen content itself is important, rather than being a proxy for some other property.

Thus, it seems reasonable to argue that eukaryote proteomes were more receptive of oxygen-rich amino acids, and this in turn aided in the development of secondary and tertiary structures needed for communication. A pre-existing property of eukaryotes was able to take more advantage of the change in atmospheric oxygen pressure. This lead to the unfolding of toolkits for multicellular body plans.

Comment by Techne — January 19, 2009 @ 10:36 am

Best Bunnah Pic Evah « The Design Matrix Says:
January 19th, 2009 at 12:22 pm
[...] For context, see here. [...]

Pingback by Best Bunnah Pic Evah « The Design Matrix — January 19, 2009 @ 12:22 pm

chunkdz Says:
January 19th, 2009 at 12:46 pm
Techne, thanks for a great post.

Some "priveleged planet" implications immediately come to mind. Namely that an oxygen rich atmosphere can only be expected to stabilize on a "Goldilocks" planet roughly the mass of Earth – not too small or too big.

http://researchpages.net/media...

Comment by chunkdz — January 19, 2009 @ 12:46 pm

JJS P.Eng. Says:
January 19th, 2009 at 5:31 pm
Good day Techne. Great post! Figure 1 really caught my eye (we engineers looove graphs ), and I could stop reading until the end.

…increases in cellular sizes roughly coincide with the alleviation of at least one environmental constraint, namely low atmospheric oxygen pressure. The origin of body plans (±0.6 bya) also seem to coincide with an increase in atmospheric pressure.

Those two points in time are remarkable and seem to suggest Mike's 4th expectation of FLE. It would seem to me that the task of reverse-engineering how the biodiversity of life came to be is way more complex than it ever was as it now appear that the environment affects biological entities, and vice-versa.

3) Why did all the toolkit parts for the hh-pathway converge on a single sterolation pathway when so many other possibilities are available? Or is it the optimal possibility and random variation and selection processes used by life hit a global optimum?

This seems to correlate (somewhat) to my post on Probabilistic Design. Were the initial conditions set so that the first life forms (single-celled organisms) could not only evolve using stochastic evolutionary mechanisms, but participate in terraforming the environment so that more complex forms could evolve in a more favourable setting. It would be interesting to see an in-depth analysis of this.

Comment by JJS P.Eng. — January 19, 2009 @ 5:31 pm

Placozoan Genome Joins the Party and Throws Its Support to Front-loading « The Design Matrix Says:
February 11th, 2009 at 11:50 pm
[...] Ah, but it is likely this is because Trichoplax was lost this circuit. See the analysis by Techne. [...]

Pingback by Placozoan Genome Joins the Party and Throws Its Support to Front-loading « The Design Matrix — February 11, 2009 @ 11:50 pm

You must be logged in to post a comment.

Toolkit Parts for Multicellularity

12 Responses to “Toolkit Parts for Multicellularity”

Leave a Reply

Featured Books