Embryology   Biology 441   Vertebrate Embryology, Spring 2018   Albert Harris


April 27, 2018

Final Lecture

Concluding Thoughts

Remember that "Star Wars" scrolling screen on the first day of classes? Let's ask ourselves if what it said was true.

And also ask what surprised you; what facts could have easily been otherwise?

I) Subdivision into differentiated cell types could have occurred all at once, instead of sequentially. All 250 cell types might have differentiated simultaneously.

Instead of branching like a railway, with sub-branches, and sub-sub branches.

e.g. no germ layers, no subdivisions of germ layers, no somites

Cells here, become heart; cells there, become muscle.

II) Different branching pathways could be used by different kinds of animals. For example, fish could have used a different branching patterns than salamanders, or frogs, reptiles, mammals birds. At least insects could be different.

What logical necessity is there for skin and nerves to share the same subdivision; or for heart and uterus, etc. and what's a connection between those cell types that develop from neural crest: sensory nerve ganglia, post-ganglionic autonomic ganglia, pigmented mesenchymal cells, Schwann cells, skeleton of the face?

Do these have something special in common?

For example, it was seriously proposed that this connection is biochemical synthesis of melanin and synthesis of adrenaline, which are biochemically related.

III) Notice that fundamentally different extra-embryonic membranes occur in teleost fish, versus reptiles, birds and mammals. The enveloping layer is not homologous to the chorion or amnion; the yolk syncytial layer is not homologous to the yolk sac. But these do serve similar functions.

IV) Notice superficial difference between invagination versus forming solid rod and cavitating.
Not just difference between teleost fish versus mammals and amphibia, but also difference between anterior and posterior of bird neural tube.

From the point of view of cell motility, invagination and cavitation are much less different than our intuition.

V) Locations of eye formation of vertebrates versus eyes of insects are controlled by very similar transcription factors. (Which was unexpected based on the optics of camera eyes being so different from compound eyes of arthropods)

Likewise, anticipate different transcription factors wnt, distal-less, hox genes
So little evolutionary change in transcription factors, versus such big change in what organs these proteins control. The major families of transcription factors, discovered in the genetic screens in Drosophila, also control development in vertebrates.

VI) Active cell rearrangement & migration creates many/most anatomical patterns.
(Unlike higher plants, where each cell type differentiates at the same location where it will remain during life). That includes plant gametes; plants don't have primordial germ cells.

VII) Regeneration of salamander limbs is by rearrangement and reshaping of skeletal and muscle cells (also skin, pigment nerves, blood vessels).
(Not by "archeocytes" or undifferentiated stem cells;
and NOT by real dedifferentiation and switching between differentiated cell types.
(which are what most embryologists had assumed).

Notice the parallel history of H. V. Wilson trying to explain cell sorting by switching from one cell type to another, based on location; or if not that, then archeocytes.

Notice that many theories of regeneration and embryonic development assume switching from one differentiated cell type to another, according to location.
Turing's reaction-diffusion system and Wolpert's "French Flag Theory" both assume that anatomical patterns are created by spatial control of differentiation, and ignore rearrangements of cells (like H.V. Wilson did).

VIII) A consensus is now developing among researchers on embryonic stem cells that mechanical forces, and resistance to elastic forces, are normal parts of the system for controlling cell differentiation

Richard Gordon and Antone Jacobson were the first to propose this, in the 1970s. James Murray, George Oster and I invented a mathematical system for compresssion of mesenchyme to stimulate cartilage differentiation; analogous to Turing, but used forces & cell population densities of crawling cells instead of "A" & "B" etc and differences of diffusion rates.

Compression -> Skeleton (which resists compression)
Tension -> tendons, ligaments, muscles (which resist tension)

To emphasize this idea: imagine if we had a cell type that resisted twisting;
then differentiation of twist-resistant cells would be stimulated by twisting.

Ossification is stimulated by stresses on bone: strengthening of muscles in response to exercise. These types of phenomena occur in embryonic structure development.

Notice the analogy to homeostasis: getting hot induces sweating; getting cold induces shivering. (Analogy to plants sprouting when pruning induces sprouting.)

The newest breakthrough is this direction is that stem cell differentiation can be changed by differences in elasticity of rubber or gel substrata.

IX) What causes the near-irreversibility of cell differentiation?
Changes in chromatin itself? Self-perpetuation processes, like cytosine methylation?
Now learning how to make stem cells, analogous to "rooting" plant cuttings.

X) Dolly the sheep, in contrast, was not a surprise to embryologists.
Already, for decades, since 1952, researchers had been producing animals by nuclear transplantation. (Briggs and King, 1952), Gurdon in the 1970s.)
All the cells have all the genes.

If Dolly and other animals produced by nuclear transplantation into oocytes turn out not to age sooner that animals born the normal way, what implications would that have for the idea that aging results from shortening of telomeres?
What about theories that aging results from accumulations of somatic mutations?
What about getting the nuclei from cells that have been living in tissue culture for various lengths of time? What ought to happen?
Hint: Maybe being put into an oocyte causes telomeres to increase greatly in length?

XI) There is surprisingly little use of growth to create anatomical shapes.
Mitosis doesn't push. The real physical forces inside developing embryos are water pressure, osmotic pressure, electro-osmotic.

XII) Scalars, vectors, tensors, & other engineering concepts are needed by embryologists, but are seldom mentioned in any biology textbooks.

XIII) Symmetry concepts are also very important for explaining patterns:

    spherical symmetry,
    axial symmetry,
    dilation (scaling)

"...if there is something you can do to it, such that it looks the same afterward as it did before." Weyl's definition of curvature. Using this definition allows one to invent many new kinds and combinations of symmetry.

For example, certain embryology textbooks might be regarded as being symmetrical with respect to transcription factors. Every 100 pages, replace wnt with notch, etc.

Curie's principle:
Symmetries of causes produce the same combinations of symmetries in the results.

Why the spherical shapes of most oocytes & early stages of development?
Are they recapitulating spherical early life forms?
Or do their internal pressures and tensions have spherical symmetry?

Also notice that reduced symmetry ("symmetry breaking") gets produced by random fluctuation --> reaction diffusion

    Point of sperm entry
    direction of gravity
    rotational symmetry of flagellar basal body

XIV) P = CT + ct If genes can control P, T and t, then they control C and c

P = pressure difference; C = curvature in one direction, c is curvature in the direction perpendicular to the direction of C. T and t are tensions in the directions of C and c which means controlling shape.

D'Arcy Thompson mistakenly used a similar equation:

P = T (1/R = 1/r) "R" and '"r" are radii of curvature: R=1/C r=1/c

Thompson treats tension as if it were a scalar, not capable of varying with direction. That greatly decreased the number of geometric shapes he could explain in terms of the combinations of forces that cause them.

a quote from D'Arcy Thompson with my illustration

Differential geometry is the appropriate math for explaining embryology, because it defines long-range shapes in terms of local properties, especially rates of angular changes of tangent directions, per unit distance along a surface (Which is the definition of what?)

XV): Primordial germ cells: Why don't oocytes and sperm differentiate from the cells of the germinal ridges? Is it to protect them from inductive signals that would stimulate them to differentiate irreversibly? By what experiments could you prove or disprove that explanation?

XVI): Bioassays: How almost every important biological chemical got discovered
But notice some "blind-spots" of bioassays: Lack of specificity. Multiple Alternative Signals, any one of which can give a positive result. Think of other reasons why a bioassay might either fail to identify an actual chemical, or give a false positive, in the sense of seeming to discover some substance that doesn't actually exist.

XVII): Genetic screens: Becoming the more powerful "discovery machinery"
Find mutant animals that are abnormal in some particular properties.
Identify all the different genes that affect that process.

Which of these genes code for enzymes, which code for transcription factors?
signaling proteins, receptor proteins, structural proteins, motor proteins.

If auxin and serotonin hadn't been discovered by bioassays, what sort of genetic screens would have discovered the same set of phenomena? (Plus the receptor proteins, the transport method, how roots are stimulated, etc.

A major "blind-spot" of genetic screens, is lethal mutations. If even the smallest change in a protein causes the animal to die, then will you be able to study that protein genetically? But temperature sensitive mutation can often get around that problem.

Consider the many advantages of using one of the model organisms: flies, mice, zebra-fish etc.

XVIII) Regeneration of legs: Why only salamanders?
Also frog legs before these are completed.
Also human finger-tips, especially in childhood.

Nobody understands why. Discovering the reasons will be a big step toward discovering how to stimulate regeneration of human fingers, hands, legs or arms. Conversely, can we hope to control regeneration until we understand these fundamental aspects of its mechanism?

Repair of scratches in children's skin is very rapid. Someday you will go blackberry-picking with some young children; notice that both you and they get scratches, and that three days later your scratches are still healing, but the children's scratches have completely healed in 1 or 2 days.

XIX) Oncogenes: Interesting that there are so few, only around a hundred.
Also interesting that cancer results more from over-activity than under-activity

Chain of signals causing growth
     Sis, receptor proteins, ras, src, myc, bcl2

XX) Can knowledge of oncogenes be used to invent better chemotherapy?
Maybe a chemical that gets converted into a poison when it is phosphorylated? How to kill just those cells whose Ras proteins can't convert ras-GTP to ras-GDP? The latter could cure a fifth of cancers; the answer will seem easy, once somebody figures out how to do it.

The goal is to kill the abnormal cells, not to inhibit their abnormal properties.

XXI) How do current chemotherapy drugs selectively kill cancer cells?
Why should faster-growing cells be selectively killed by drugs that block some stage of growth (e.g. DNA synthesis, mitotic cell division)?

True explanation that loss of growth control at check-points allows damaged chromosomes to pass check-points, mitosis, DNA synthesis.

What kills is not that they are fast, and not that the drugs inhibit their growth. (but the opposite) because even though damaged, cancer cells don't stop growing.

XXII) Which are better?
* Drugs that fix abnormalities of cancer cells, & make them more nearly normal.

* Drugs that selectively kill those cells with cancerous abnormalities.
(Getting rid of the cancer cells)

Gleevec...(tyrosine kinase inhibitor)?
Inhibitors of angiogenesis...?

XXIII) Why have the later stages of drug synthesis and testing been handed over to pharmaceutical companies?

    Why doesn't insulin cost a thousand dollars per day?
    Why doesn't snake venom anti-toxin cost a year's salary?
    Why don't vaccines cost as much as a house?
    Could such prices have been charged? Why weren't they?
Did people just not imagine the possibility of charging such high prices?

Can we predict that, from now on, ALL major new medical treatments will cost thousand$? (Like the bisphosphonate anti-osteoporosis drugs)

Would such large financial incentives have caused such treatments to be discovered sooner, or better versions to be developed?

Have discoveries been missed, of these levels of importance, for lack of large prices. For example, would cures already have been found for cancer and lupus, if even larger profits could be made; or if there were more research funding?

Who discovers which part:

The existence of auxin? Darwin.
The chemical structure of auxin? Dutch Professors.
Synthesis of auxin analogs, like 2-4 D? US Professors.
How to synthesize tons of 2-4 D? Chemical companies.

Who discovered uracil ribo-phosphate structure and methods of synthesis?
Who discovered that chemical analogs sometimes cure diseases?

Who discovered how to legalize patenting of chemicals, and for special uses?
Who invented the idea of persuading reporters that India is breaking patent laws, because they continue to apply the traditional laws against patenting chemicals?


Non-Hodgkins lymphoma has been cured by isolating the antibody binding sites of lymphoma B-cells, injecting these into young mice, and isolating monoclonal antibodies specific for killing that particular person's cancerous lymphocytes.
"But this method was not developed because treatments would be prohibitively expensive"

Translation? A treatment that actually costs a hundred thousand dollars, isn't worth it; because the sale price could only be ten times the cost of manufacture. (not 50 times)

But a treatment for which you can charge a hundred thousand dollars, but only actually costs a few hundred dollars, is profitable enough to be developed and sold.

If the new thousand dollar per pill hepatitis drug actually cost five hundred dollars per pill to make, (half the sale price) then would it be "prohibitively expensive"?

Is this going to be the "New Normal" of medical research? Thousand dollar pills!
Cures for cancer abandoned because of not being profitable enough.
Pasteur, Fleming. Waksman, and Salk could each have been multi-billionaires!

Each richer than Gates; Why didn't they? Why doesn't the news discuss reasons?

Revised from the 1960s folk-song: "All My Trials, Lord, Soon Be Over"

      If living was a thing that money could buy,
      Well, you know, the rich would get richer,
      and the un-insured would die...


There are opportunities for major new discoveries in your lifetimes, building on the facts and principles that you have learned in this course.

For example, improvements on cancer chemotherapy should take more advantage of the many abnormalities that cancer cells have in addition to their uncontrolled growth. For example, their high acidity and weakened adhesiveness and locomotion forces.

Find ways to stimulate apoptosis of just those cells having one or more of these different abnormalities: For example, why not use anaerobic metabolism to activate caspases. Chemotherapy should be re-designed so that such abnormalities will cause cancer cells to die, instead of trying to repair their abnormal properties.


Caspases might also be selectively activated so as to fight autoimmune diseases if you arrange for apoptosis to kill just those lymphocytes whose binding sites are attacking myelin inside the central nervous system, etc.

We can do a lot better than the current non-specific inhibition of the whole immune system, by methods that are inherently impossible to be improved enough to cure MS, Lupus, Rheumatoid Arthritis and other autoimmune diseases. Billions of dollars are now being spent just delaying worsening of these diseases.


Maybe you can find more and different functions of the voltage differences between the inside and outside of plasma membranes, in addition to nerve impulses and blocks to polyspermy. Unless such additional functions exist, I doubt if cells would be using a third or more of their ATP energy pumping potassium ions inward and sodium ions outward. Steering of crawling tissue culture cells by voltage gradients got lots of attention for a couple decades, but was never really proven to be a normal function of voltages.

A major obstacle to progress is lack of understanding of electroosmosis as compared with piezoelectric voltages, and not enough scientists understand that both these phenomena produce only alternating voltages, never continuous unidirectional voltages, whereas galvanotaxis requires direct current.

Better treatments for bone weakening should be considered as important as cures for cancer. It is also a tantalizing intellectual question how bone cells and osteoclasts normally adjust the strengths of bones based on how much physical force gets exerted at different locations and directions.

Until this normal mechanism gets understood, what chance will there be to prevent failure to continue forming as much bone as young people do. Adding more calcium to diets is just a cop-out, in my opinion. What's the chance that more calcium will inhibit osteoclasts or stimulate bone formation. The means by which a tennis player's right arm bones become stronger is not because more calcium ions get sent there.


What causes walls of arteries to accumulate smooth muscle cells, vacuole-bloated macrophages, and insoluble amounts of cholesterol? Furthermore, why do these accumulations occur within artery walls, instead of inside the lumen.

Another kind of question is why the public was told Aesop's Fables children's stories for so long, about atherosclerosis misguiding almost everybody toward beliefs that atherosclerosis is just a matter of inert lipids precipitating out of solution, like minerals in a water pipe.

It can be a very bad thing to satisfy curiosity with fiction, especially in relation to diseases that kill millions. The main effect is to delay and misguide discoveries of cures.


Currently there is a lot of enthusiasm about possibilities of using undifferentiated "stem cells" to assist artificial regeneration of fingers and toes and arms and legs, and even kidneys and muscles. I share this enthusiasm. But progress will be delayed unless researchers realize that no anatomical structures are formed by accumulating disorganized masses of undifferentiated cells and then causing those cells at certain positions to change into muscle cells and those cells at other places to differentiate into bones.

The muscle cells and the bone cells both get put into proper geometric arrangements before they complete differentiation. That's how we make arms and legs, how salamanders make legs; it's even how salamanders regenerate legs, by moving cells around according to which cell type they are already committed to become.


Why must we continue to make that same false assumptions that H. V. Wilson made about his dissociated cells: guessing that cells switch from being one cell type to becoming another, based on the cell's location? Not only is that not what happens in the sorting out of sponge cells, it's also not what happens in regeneration of salamander legs.

Why should anyone expect that non-differentiated "stem cells" possess any means of re-creating legs or any other anatomical structures by means of spatial control of differentiation?

The reason, besides wishful thinking, is that they haven't yet bothered to learn central facts about either embryonic development or regeneration. If you go into this new field called "Regenerative Medicine", you could contribute a lot of much needed information.


For example, an important set of related facts is that all skeletal muscles have dozens or hundreds of nuclei per cell, and get this way by fusing one-celled myoblast cells. Furthermore, inside salamander blastemas the multinucleate muscle cells somehow cleave into many one celled cells, which then grow, divide and re-fuse to form new multinucleate cells.

No one knows what mechanism is used to cleave multinucleate muscle cells back to having just one nucleus per cell. Nobody even knows whether mammal and bird muscle cells are capable of cleaving into uni-nucleate cells. If they don't have that ability, that would be a major part of the reason we can't regenerate arms and legs.

"Regenerative Medicine" is never going to succeed at anything beyond raising tons of money, without collaboration with people like you, who have learned the basic facts of embryology.

The end.


back to syllabus