One of my favorite signal transduction pathways (what? You didn’t know that true nerds had favorite molecular pathways?) is the one mediated by the receptor Notch. Notch is one of those genes in the metazoan toolkit that keeps popping up in all kinds of different contexts—it’s the adjustable wrench of the toolbox, something that handles a general problem very well and therefore gets reused over and over again, and the list of places where it is expressed in Drosophila is impressive.
The general problem that Notch solves is the resolution of a binary decision in cell fate, one where a few dispersed cells in a field of otherwise undifferentiated cells (called an equivalence group) must acquire a new fate, and further, must communicate to neighboring cells that they should follow a different fate. It’s like the problem of selecting a team captain in sports: there might be different strategies for picking such an individual, but one thing that must be done is to inform everyone on the team when the choice has been made, and suppress competing captains from emerging. Notch does that job of spreading the news.
The diagram below illustrates the process. In an equivalence group of cells (all yellow), one cell acquires a unique fate (red), and subsequently enforces the general fate on all of its neighbors by a process called lateral inhibition. Notch is the receptor molecule at the heart of that lateral inhibition. It does its job by interacting with ligands, such as Delta, Serrate or Lag2 (DSL for short).
(click for larger image)
Notch receptor signaling and architecture. (A) All cells within an equivalence group are in a particular state (yellow) and possess similar amounts of Notch and Delta. A combination of intrinsic and extrinsic signals induces a new cell state (light red) and breaks this balance by increasing the abundance of Delta in one or a few cells (dark red). Consequently, Notch signaling is activated in neighboring cells, suppressing the alternative fate and reverting these neighboring cells to their original fate through lateral inhibition. (B) Domain architecture of the Notch receptor family. The organization of human Notch1 is shown as an example. All Notch receptors consist of two polypeptides at the cell surface, except Drosophila, in which a linear form of the receptor might exist at the cell surface. RAM, recombination signal-binding protein for J-κ (RBPJκ)associated molecule region; ANK, ankyrin; NLS, nuclear localization sequence; Pro-Glu-Ser-Thr (PEST), degradation motif.
How does Notch do this? Remember, initially all of the cells in this tissue are an equivalence group—they’re all expressing the same genes, including Notch and, for instance, Delta. All of the cells are signaling to each other with Notch and Delta.
There is a kind of tension here, though, a dynamic instability. When Notch is activated by binding Delta, what it does is stimulate a pathway (Suppressor-of-hairless (Su(H))→Enhancer-of-split (E(Spl))) that turns off the Achaete-Scute gene complex (AS-C)—by the way, another reason I like this pathway so much is the baroque names of the elements) which regulates the expression of Delta. In a field of perfect, uniform, determinate cells, what ought to happen then is this: all the cells would interact with each other with Notch and Delta, Notch would turn off Delta production in all of the cells, all of the Notch receptors would then be unbound, releasing the cells from inhibition, Delta would be produced, all of the cells would interact with each other with Notch and Delta, and the cycle would continue. Boring!
Of course, cells are not perfect, uniform, and determinate. In real life, something like the situation diagramed below arises.
The first panel is the situation in the equivalence group: all of the cells are expressing both Notch and Delta.
The second panel is where things get interesting. The cell on the top has acquired a slight edge, and is expressing its Achaete-Scute complex more strongly than the cell on the bottom. This could be a result of some entirely random phenomenon—the cell started expressing those genes a little earlier, a few extra molecules made it through processing, by chance its Notch receptors were bound to slightly fewer ligands from its neighbors—or perhaps some other more deterministic process upregulated Delta in this particular cell. Whatever the difference, the cell on top is signaling via Delta more strongly than its neighbors, which means the neighbors activate the Su(H)→E(Spl) pathway more strongly, and therefore produce less Delta ligand. This produces a runaway loop, so that cells that initially had a slight edge in Delta signaling produce a stronger and stronger signal, and all the cells that were initially at a slight disadvantage are more and more suppressed.
The end result is in panel 3. All of the cells are producing Notch receptor, but only one is now producing the Delta ligand, and that ligand keeps all the neighbors in a Delta-free state. This is the outcome of lateral inhibition—in this case, a field of cells with a scattered subset of non-adjacent cells with a different pattern of gene expression.
One specific example where this takes place is in the insect neurectoderm, illustrated below, where Notch is essential in mediating the early fate decision in whether a cell will become a neuroblast, the progenitor to the neural elements of the central nervous system, or an epidermoblast, which will make the structural and supporting cells of the tissue. This is a situation where the desired end result is scattered neuroblasts surrounded by supporting cells, and is the classic place for Notch to play a role.
A recent short review on Science gives a summary of the details of the signal transduction pathway for Notch. It’s delightfully elaborate, and has a nice diagram I’ll be using in my class today…and now you know why I’m suddenly babbling about this particular receptor.
In order to do its job, Notch is being constantly churned over in the membrane. The response to binding the Delta/Serrate/Lag-2 ligand is to be cleaved by an extracellular enzyme, releasing a fragment, the Notch intracellular domain (NICD), which is then further processed by presenilin, ubiquitinated, and transported into the nucleus, where it cooperates with a family of transcription factors (CBF1, Su(H), and Lag1, or CSL) to regulate gene activity. As you can see in the diagram, there’s also continuous recycling and degradation of all of these proteins—cells are not static!
(click for larger image)
Ligand binding to the Notch receptor triggers signaling through successive proteolytic cleavages. The ligand (DSL) binds to the EGF repeats (orange) in the extracellular domain of the Notch receptor. A protease cleaves Notch at the S2 site, removing the bulk of the extracellular domain. The identity of this protease is not clear; it may be a disintegrin and metalloprotease protein (ADAM), tumor necrosis factor-α converting enzyme (TACE), or Kuzbanian (Kuz). The membrane-tethered intracellular domain is then cleaved by the Presenilin complex at site S3, either in the plasma membrane or after endocytosis, freeing the NICD. NICD then moves to the nucleus, where it forms a binary complex with CSL, which recruits coactivators (CoA) to form the transcription-activating complex. Notch signaling can be regulated by endocytosis. NICD becomes monoubiquitylated (Ub), targeting the receptor to the lysosome for degradation. NICD in the nucleus is degraded in a proteasome-dependent manner, although the location of degradation is not clear. Solid arrows indicate demonstrated routes; dashed arrows indicate routes tentatively inferred from data.
Despite the complexity of this pathway, though, this is not at all unusual, and isn’t even the most elaborate pathway I’ve seen—this is typical. And before anyone rushes to argue that this kind of complexity is an obstacle to evolution, I’ll mention that signal transduction elements that modify patterns of gene expression are not unique to animals, and that this general kind of activity is common in the bacterial world, too. The toolkit our cells use evolved first in single-celled organisms, and these are universal solutions to the problem of modifying gene activity in response to conditions in the environment.
In addition, welcome to the strange world of the cell, which IDists to the contrary, is not a world full of analogs to the machines of human culture. Look at the constant breakdown and turnover and synthesis of this signal. Michael Behe likes to compare this sort of thing to traffic lights, but in our macro world, traffic lights don’t work by constantly glowing red, being smashed to signal ‘go’, and stopping traffic again by getting re-manufactured and hung back out in the intersection. That is a standard way of managing signaling in a cell, however.
Ehebauer M, Hayward P, Arias AM (2006) Notch, a universal arbiter of cell fate decisions. Science 314(5804):1414-1415.
Now that takes me back. My work during the Drosophila portion of my very non-brilliant academic career was on pecanex,a maternal-effect gene the lack of whose gene product in the oocyte produces a Notch-like neuralization of the embryonic neurectoderm. AFAIK to this day nobody’s been motivated to find out much about its gene product and what it actually does (which I guess shows what a brilliant choice of research problem it was, sigh) though I’m aware there’s a mammalian homologue.
I will say this is a peculiar process for Notch, if only because I didn’t know about it when I was taking Cell Biology.
On the one hand, it is nice, if only for our own sake of convience, to be able to present the macabre world of cell to cell signalling in terms the layman can digest, but, like you say, it is naive, if not missing the point entirely, to translate everything in cell biology into cute analogues…
As such, would it be wrong of me to think of Notch as being akin to an inter-office telegram system where once the telegram has been received, the delivery device then gets a “Return to Sender” stamp and then is recycled?
If it’s an interoffice memo that reiterates something basic like “keep working”, and you get it, tear it up, send it off to be pulped and remade into paper that the boss then uses to send you more “keep working” memos every half hour, and if you don’t get the memo, you go home and bake cookies instead.
I think the point is that trying to find real-world analogs to these processes quickly generates over-stretched, inappropriate analogies and the total collapse of the argument.
Yeah, but your analogy is a great one to trash ID arguments by analogy, because as dysfunctional as offices are, we don’t design them to work the way that cell signaling does.
Stretching your analogy further to cover crosstalk between signaling pathways, as the “keep working” memos pile up on your desk, some of them might float over to your colleagues’ desks and remind them to work, too.
Just a quick question – in (A) of the first figure legend, should ‘light red’ and ‘dark red’ be the other way around?
Paul wrote:
“The toolkit our cells use evolved first in single-celled organisms, and these are universal solutions to the problem of modifying gene activity in response to conditions in the environment.
In addition, welcome to the strange world of the cell, which IDists to the contrary, is not a world full of analogs to the machines of human culture.”
I was reading Frank Rich’s column in the New York Times yesterday and I couldn’t help noticing the similarities between you and George W. Bush. Rich says that Bush “isn’t merely in a state of denial but is completely untethered from reality.”
You are the George W. Bush of molecular biology and evolution.
You continue to cling to the myth that there was no intelligent input involved in the evolution of these highly organized systems and processes even when the preponderance of evidence clearly demonstrates that they could not have emerged by any kind of random, accidental or non-directed process.
You can’t handle the truth about evolution. In fact, you don’t even know what the truth is.
Your stubborn resistance to any hint of intelligent input and your continued insistence that these structures and processes emerged by random chance clearly parallels President Bush’s insistence that “Al Qaeda is responsible for Iraq’s spiraling violence”.
You, like President Bush, cling to your delusions with a near-rage that can’t be explained away by sheer stubbornness or misguided principles. You’re not stupid by any measure, but you are a cynic and you’ve allowed that cynicism and your contempt for creationists to cloud your scientific judgement.
Charlie, speaking as a trained molecular geneticist, you’re comprehensively ignorant of science and you wouldn’t know what “scientific judgement” was if it reared up and bit you in the ass. You’ve wasted your own time and ours for too long around here. Grow up, get a life, and go away.
Charlie, speaking as another trained molecular geneticist, I find your buffoonery to be incredibly dishonest, particularly because PZ’s post here is based on actual evidence, the polar opposite of Bush’s wishful thinking.
Now, if you see some evidence that suggests intelligent input into the baroque mechanisms of Notch signaling, please offer it up, as well as some predictions that flow from your hypothesis.
Since that’s beyond you, how about answering one question: does the presence of intelligent input into Notch signaling predict that we would have learned more about the function of presenilin by studying the development of the vulva in a hermaphroditic nematode than we did by studying null presenilin mutants in mice?
No. I briefly wondered that too before realising they had skipped out a diagram. All the cells in an area reach a stage where they go to light red (more delta) first; and then one of them wins that cascade race to go dark/bright red (the middle diagram) while suppressing the surrounding cells. The “losers” then revert back to not being delta producers.
John wrote:
“how about answering one question: does the presence of intelligent input into Notch signaling predict that we would have learned more about the function of presenilin by studying the development of the vulva in a hermaphroditic nematode than we did by studying null presenilin mutants in mice?”
No. The mechanism by which the pathway emerged during evolution is not related in any way to the structure or function of the process as it appears in any number of different extant biological systems. Similarly, one can study the electrical pathways in complex electrical systems without concerning themselves about the mechanisms by which the system was designed.
Similarly, one can study the electrical pathways in complex electrical systems without concerning themselves about the mechanisms by which the system was designed.
but without understanding the mechanisms, do you actually have a realistic grasp of how it works?
what if something goes wrong?
I can study the circulatory system and understand how blood flows through it by simple observation. I would be totally clueless as to how disease affects it, or why, or what to do about it without having some knowledge of the mechanisms involved in its production.
your analogy, if it was intended to indicate a distinction between function, development, and mechanism was flawed at best.
evoltionary science exists to answer “why” questions, not “how” questions, but to think the two are entirely unrelated is entirely simplistic.
Ichthyic wrote:
“evoltionary science exists to answer “why” questions, not “how” questions, but to think the two are entirely unrelated is entirely simplistic.”
Nevertheless, that’s how it is.
Science does not concern itself with “why” questions at all. That is the realm of philosophy. Science should only and does only concern itself with “how”. Evolution and a knowledge of evolutionary mechanisms is totally unnecessary for biological science to function. The mechanism by which a structure or process emerged is irrelevant to elucidating how it functions.
The stupidest words ever uttered by an alleged “scientist” were the claim that “nothing in biology makes any sense except in the light of evolution.” (Dobzhansky) In fact, we can completely understand the structures, processes and systems that are incorporated into living creatures without knowing diddlysquat about their origins.
John:
Just in case you didn’t know:
Specie: Confrontational drive by troll appearing with long intervals.
Individual characteristics: ID, frontloaded information, atheist, panspermia, no big bang, infinitely old universe, … I think, it is hard to keep up with the specifics of all buffoons and cranks. Perhaps I should start keeping score cards.
John:
Just in case you didn’t know:
Specie: Confrontational drive by troll appearing with long intervals.
Individual characteristics: ID, frontloaded information, atheist, panspermia, no big bang, infinitely old universe, … I think, it is hard to keep up with the specifics of all buffoons and cranks. Perhaps I should start keeping score cards.
Better make that claims to be agnostic, IIRC.
Better make that claims to be agnostic, IIRC.
I always though there was something interesting about the Notch pathway that no one seems to talk about: no amplification. Because the receptor (or part of it) IS the signal, there is a 1:1 relationship between events at the membrane and in the nucleus. This is very peculiar compared to, for example, kinase cascades, activation of adenylyl/guanyl cyclases, calcium release, etc. Even in TGFbeta signaling, one activated receptor can give you a lot of phosphorylated SMADs.
I’ve always wondered if some clever mathematical biologist could model this and show that this type of linear dynamic is particularly well-suited to lateral inhibition and other Notchy functions.
Nevertheless, that’s how it is.
wrong again, dufous.
Science does not concern itself with “why” questions at all.
charlie, you and a knowledge of science are apparently as far apart as New Zealand and California on the globe.
the idea that science never addresses why type questions is so wrong, i don’t even know where to begin.
go home and read your bible, would ya? stop wasting your time trying to miseducate folks; you aren’t very good at it.
…and yeah, now i remember how pathetic your attempts were on PT too. Haven’t seen you around for a while though.
what were you doing? sulking?
The stupidest words ever uttered by an alleged “scientist” were the claim that “nothing in biology makes any sense except in the light of evolution.” (Dobzhansky)
Oh, i dunno, I’d be willing to be that I easily could find something far stupider said by yourself on any given day.
today even.
Ichthyic wrote:
“the idea that science never addresses why type questions is so wrong, i don’t even know where to begin.”
Many “how” questions are improperly framed as “why” questions. For example, a young child might ask “why does the moon rise in the east and set in the west”? This is not really a “why” question but a “how” question. It involves learning about how the moon revolves in relation to the earth and how these movements cause the effect of rising and setting. The true “why” question here is “why is there a moon at all and why does it rise in the east rather than the west?
Your example above of the circulatory system is clearly a “how” question. In fact, you say it yourself. “How does disease affect the circulatory system”?
Also, don’t confuse the mechanisms involved with the functioning of a system with the mechanism that created the system. Clearly, it is important to study the mechanisms of blood clotting and disease in order to explain how the system works and how we can correct problems when things go wrong. But this has no relationship whatsoever to the mechanism the created the system in the first place nor does it tell us anything about why such a system exists. That is the realm of philosophy and it carries with it a large telelogical component and a strong hint of purpose.
I have a favorite supernova (SN1987A) and a favorite mathematical proof (Goedel’s Incompleteness Theorem). Does that count?
charlie, you presuming to lecture me on the definitions of how vs. why questions is laughable, considering your poor understanding of science to begin with.
it’s patently obvious to everybody here that it’s just wishful thinking on your part when you project that science never deals with “why” questions.
In fact, just like all creobots, you try to redefine the purview of science to fit your own cognitive malfunctions.
It simply don’t work that way, dufous.
for example, even with your celestial analogy, we see you project what you want onto it, rather than the reality:
The true “why” question here is “why is there a moon at all and why does it rise in the east rather than the west?
for you see, if you define that as your “why” question, the answers are ENTIRELY within the purview of physics and cosmology to answer.
Indeed, there has been a lot of research on planet spins, gravity, the origin of the moon, etc., all considered with evidence in support.
no matter how far you try to stretch your false analogies to represent your concept that philosophy is the realm of the “why”, you will still be wrong.
It really is pathetic you, and so many like you, are mentally incapable of cognitive recognition of this, as obviously you have completely succumbed to your own unconscious defense mechanisms to support your warped compartmentalization.
remarkable on the face of it, how obvious this is to those not suffering from the extreme level of cognitive dissonance you seem to be.
Ichthyic wrote:
“It simply don’t work that way, dufous.”
I’ve seen this word spelled “dufus” and “doofus”, but I’ve never seen it spelled “dufous”
If you’re going to insult me, at least make an effort to spell the insult correctly.
If you’re going to insult us, you could at least do better than “all algorithms required intelligent input to be created, LOLs!!!1!!one!”
Charlie Wagner, you seem to have made a few errors passed over in this thread.
First, you said “could not have emerged by any kind of random, accidental or non-directed process.”
While certainly non-directed, natural selection is not random, chance or accidental. It operates through a mechanical algorithm much like a sieve. Population goes in, smaller population comes out after running the gauntlet.
That’s not random or accidental. But it is an extremely simple algorithm.
Secondly, “Your stubborn resistance to any hint of intelligent input”. The point is that stubbornness is definitely
called for. No there is no reason intelligent intervention could not occur in any process at any time. Its just that
if you find a mechanism to explain a complex process, it makes such an intervention unnecessary and arbitrary.
Since these points are pretty obvious, I have to ask, is there a bias involved in your argument? I think there must be.
I can understand the frustration expressed by many regarding Charlie Wagner’s position.
If you’re going to insult me, at least make an effort to spell the insult correctly.
THAT’S your response?
ROFLMAO!
Rick, trust me when i say you are giving CW way too much credit.
he lost it years ago.
Rick wrote:
“While certainly non-directed, natural selection is not random, chance or accidental.”
I’ve dealt with this issue many times in the past. There is a non-random component to natural selection in that not all members of a population living under the same conditions have equal ability to survive. But natural selection can only act on adaptations and variations that are already present. It has no ability to create these variations or to assemble them into new structures, processes and systems.
“there is no reason intelligent intervention could not occur in any process at any time. Its just that
if you find a mechanism to explain a complex process, it makes such an intervention unnecessary and arbitrary.”
Indeed, if you could find a mechanism to explain a complex process such an intervention would be unnecessary and arbitrary. But no such mechansm has been found for evolution. There is not a shred of empirical evidence, either observational or experimental, that establishes a clear nexus betwwen the process of natural selection and the emergence of highly organized structures, processes and systems.
I’ve dealt with this issue many times in the past.
…and continue to get it wrong, every time.
There is a non-random component to natural selection in that not all members of a population living under the same conditions have equal ability to survive.
that would be entirely non random, not just a “component”, which is exactly what was pointed out to you. don’t forget (meh, who am i kidding), it’s about the ability to reproduce, not just survive.
Indeed, if you could find a mechanism to explain a complex process
complex according to whom? your subjective perceptions, which, like Behe’s produce idiotic and unsupportable concepts like “irreducible complexity”? which mechanisms are we talking about? the immune system that has literally hundreds of peer reviewed articles expounding on the evolution of just about every piece of it? the flagellum that we have even more for?
c’mon genius! let’s see those details of your “IC” publications showing exactly how these supposedly “complex” systems could not possibly have evolved utilizing the mechanisms and pathways already expounded on in great detail in the articles presented to Behe at Dover, for example.
..or do wish to blatantly continue to lie like you did here:
There is not a shred of empirical evidence, either observational or experimental, that establishes a clear nexus betwwen the process of natural selection and the emergence of highly organized structures, processes and systems.
You really need treatment for that cognitive dissonance, there, CW.
only someone like yourself would have seen so many instances of an insult thrown at him he would know multiple spellings of it.
Charlie,
You strain credibility.
“not all members of a population living under the same conditions have equal ability to survive. But natural selection can only act on adaptations and variations that are already present. It has no ability to create these variations or to assemble them into new structures, processes and systems.”
I am simply stunned by this remarkable collision of ideas. If not all members survive, and variation occurs all the time, how can it be said to be unable to assemble new expressions? Evolutionary theory postulates that small potentials in variation over vast amounts of time can create new structures, processes and systems. The literature is choked with examples from Drosophila to dairy cattle to T-Rex. Darwin accounts admirably for everything thus far investigated while we await any suggestion of evidence for an alternative theory within the bounds of reason?
I’m afraid that I must conclude that your desire for consolation exceeds your respect for evidence.
Rick wrote:
“Evolutionary theory postulates that small potentials in variation over vast amounts of time can create new structures, processes and systems.”
Evolutionary theory, if that’s what it says, is completely wrong. There is not a shred of empirical evidence, either observational or experimental, that establishes any kind of nexus between the trivial effects of mutation acted on by natural selection and the emergence of complex structures, processes and systems.
But don’t take my word for it. Investigate for yourself. The belief that natural selection has the power to create new structures, processes and systems is a myth. There is no evidence to support this audacious contention, none at all.
Charlie,
You are a man of faith with his hat pulled firmly down over his eyes (thus “blind faith”). You dwell in an inner world where belief is easy and well protected.
Those of this forum wear their hats cocked back and stare directly ahead in amazement and awe at the wonders of the world. Our beliefs are constantly being challenged by new discoveries. These views are very incompatible
as far as I can see. You would do better to spend time with those of your own kind reassuring each other than to fritter away your time desperately trying to convert
the unconvertible.
bye, charlie.
maybe, someday, you will learn to stop hitting yourself in the head with that hammer you carry around.
Late to the game again, but I want to comment on:
The toolkit our cells use evolved first in single-celled organisms
I do not see this point made often enough, maybe because it is so obvious. However, I am asked by evolution-questioners how these complex enzyme systems could evolve in humans in a few million years. Humans did not start from scratch! There were a couple billion years of evolution of single cell organisms that preceded the first multicellular organisms. Multicellularity is built on the systems evolved in single cell organisms. As a result, in individual cells many ‘housekeeping’ genes are similar across species (DNA or RNA polymerases, ribosomal proteins). Tissues, organs and other organized cell systems were derived from these unicellular systems.
It’s something I think needs more emphasis, or an appropriate inclusion as in this case. Thanks for listening.
And thanks for the post, PZ