Re: COMPARATIVE VERIFICATION OF A GENETIC AND AN EPIGENETIC MODEL

On Mar 4, 4:29 pm, CNCa...@xxxxxxx wrote:
On Mar 3, 12:48 pm, "Perplexed in Peoria" <jimmene...@xxxxxxxxxxxxx>
wrote:

"John Harshman" <jharshman.diespam...@xxxxxxxxxxx> wrote:
CNCa...@xxxxxxx wrote:
 I have shown you, in numerous examples, that changes in
regulatory networks and the CNS of the offspring are induced
by signals from the parental central nervous system.

Actually, I don't recall any example of that. You have shown some cases
in which parental transcripts are deposited in the ovum under CNS
stimulus.

Let me  show you again some cases when  the maternal
central nervous system, without changes in genes, induces
in the offspring phenotypic changes and the CNS of the offspring
induces in the next generation the same new chararcter.

You are going to again point out what are called maternal effects in
insects, which have a syncitial form of mosaic development, in which
*gradients* of factors are laid down in the egg cytoplasm. These
materials that are laid down as gradients during oogenesis are
*genetic* factors, in that they are sequenced material ultimately due
to genes. The genes that determine these factors are maternal genes
because these are factors laid down by the mother's cells. But,
again, maternal effect genes are perfectly normal mendelian genes,
with the genotype of the mother being due to normal mendelian genes
from both mother and father. Like many other mendelian genes, some of
these genes production are *regulated* by environment.

Daphnia magna is  an all-female parthenogenetically reproducing
crustacean species. This state is determined by a neural
mechanism: normal favorable, nonstressful environmental
conditions  received by the sensory neurons, are converted
into electric  message, processed in specific neural circuits,
which generate a chemical signal (serotonin?).

Again, we are supposed to be shocked, shocked that CNS is involved in
"sensing" an organism's environment.

The chemical
signal stimulates neurons of the X organ/sinus gland to secrete
the neurohormones MOIHs (mandibular organ-inhibiting
hormones), which prevents secretion of the juvenoid hormone
methyl farnesoate, which is known to induce transformation of
this species into a sexually (male + female offspring) reproducing
organism.

Simple regulation. Sex can certainly be determined by environmental
conditions in many, *but not all* organisms. This is no different
than temperature determining sex in crocodilians, or the presence or
absence of a female determining sex in groupers, or the absence of a
male in some coral fish, or age in birds, or many other examples of
*regulatory* sex-determination (which are still due to the action of
genes) and the more familiar (to us) *determinative* sex determination
where the presence or absence of a particular gene determines sex.

By contrast, environmental stimuli presaging deterioration of
environmental conditions, such as crowding and shortening of
 photoperiod are received by sensory neurons, converted into
a different electrical message, which is processed in the same
neural circuit, but generates a different output, a chemical signal
(dopamine?), which stimulates neurons in the X-organ/sinus
gland to produce a different neurohormone, CHH (crustacean
hyperglycemic hormone). This neurohormone stimulates the
mandibular gland to secrete the juvenoid hormone which itself
or one of its products is  deposited in the  egg and induces
production of male and female individuals (Gilbert, J.J. 2003.
Evolution & Development 5: 19-24). This transgenerational
plasticity is also experimentally transmitted to the offspring
by simply exposing females with maturing oocytes in the ovary

"exposing" them to what?

(Rider, C.V. et al. 2005. The Journal of Experimental Biology.
208: 15-23).

Another cyclically parthenogenetic rotifer, Brachionus angularis,
in response to crowding and a kairomone released by conspecifics,
gives birth to haploid  eggs developing into males that, when
fertilized develop into diapausing eggs, which hatch into female
individuals incapable of sexual reproduction for several generations.
Here is the investigator's interpretation of
the mechanism of the epigenetic inheritance:
"The proposed chemical responsible for the crowding stimulus could
diffuse into the maternal body cavity and directly affect the oocyte
such that the mictic-female phenotype is expressed later on in
development....Alternatively, the chemical could directly affect the
physiology of the mother. For example, it could target the nervous
system via chemoreceptors, and some secreted factor could then affect
the growing oocyte (Gilbert, J.J. 2003. Evolution & Development 5:
19-24).
Phase transition in Schistocerca gregaria is associated by
transmission of a number of changed morphological, physiological, and
behavioral characters to the offspring.
"This process is epigenetic, being transgenerational, inducible and
persistent for some duration in the absence of inducing
stimuli." [Jablonka and Lamb, 2002; Jablonka and Lamb, 2005; Jablonka
et al., 2002 (according to Miller, G.A. et al. 2008. Journal of
Experimental Biology 211: 370-376)]
All the phase change-inducing factors act via the insect central
nervous system [crowding in this insect is a stressor that also acts
via the CNS as is concluded by the experimental evidence that
antennectomized locusts do not change phase under conditions of
crowding (Applebaum and Heifetz, 1999)].
The transfer of phase state across generations in locusts is clearly
epigenetic  (Simpson, S.J. and Miller, G.A. 2007. Journal of Insect
Physiology 53: 869-876)

"Epigenetic transfer of phase state depends upon low molecular
mass, water-soluble chemicals within foam secreted by the
reproductive accessory glands at the time of oviposition...a specific
chemical agent responsible for transmission of gregarious behaviour
between locust generations." ( Miller, G.A. et al. 2008. Journal of
Experimental Biology 211: 370-376) "and perhaps also from what appears
to be identical glandular tissue lining the oviduct. Exposure to
this material soon after ovulation, in the oviduct fluid and/
or in the egg foam after oviposition, results in the graded
development of gregarious behaviour and colouration in
the resulting hatchlings. " (Simpson, S.J. and Miller, G.A. 2007.
Ibidem)

Intense changes are observed in the levels of numerous
neurotransmitters in the locust brain (Rogers, S.M. et al., 2004. The
Journal of Experimental Biology 207:3603-3617). At a neuroendocrine
level this transformation is related to an elevated level of JH
(juvenile hormone) under stimulation of neurohormones allatotropins
and nerves innervating the corpora allata as well as cerebral
secretion of [His7]-corazonin (Grach, C. et al., 2003. Journal of
Peptide Research 62: 135-142), also known as DCIN (dark-color-inducing
neurohormone). A  5-10-fold increase of the amount of a cytoplasmic
factor, the neurally regulated ecdysteroid hormones was found in the
eggs of gregarious locusts, than in the eggs of the solitarious
locusts (Tawfik, A. I. et al., 1999. Proceedings of the National
Academy of Sciences USA 96:7083-7087; Tawfik, A.I. and Sehnal, F.
2003. Physiological Entomology 28: 19-24; Hagele, B.F.  et al. 2004.
Journal of Insect Physiology 50: 621-628 ).
It is interesting to  point out that transition from the solitarious
to
gregarious phase is immediate, reversion from the gregarious
to the solitary phase is slower and may take generations to occur.

These examples of multigenerational plasticity

Sorry. Where is there information of multigenerational transfer (as
opposed to one generation maternal transfer) of environmentally
induced information to determine the reproductive pattern of that next
generation? Genes do not encode results. They encode capacities.

clearly show
that under the influence of signal cascades and regulatory
networks started by the maternal CNS,  the offspring CNS
has changed so that it induces e the same new characters
in the next generation.

Where was that claim made? The offspring's CNS *also* has the
capacity to record information about its environment (just like its
mother did) and, depending on the state of that environment,
determines, to the extent that it is genetically identical to its
mother, the reproductive mode of *it's* progeny. What you need is a
change in the CNS so that it *no longer* responds to environmental
conditions (which is what its mother did), but transmits its inherited
condition (the one it got from its mother) intact to its progeny. And
it must do so regardless of what alleles it gets from mother and
father in the genome (leaving out maternal effect genes). And it must
do so regardless of changes in the environment.

As I recall, the evidence only showed that the CNS controlled the timing
of the transcript production, and perhaps the location of the production
(i.e. "make the transcripts in the ovum, but not in other cells").  You did
not show that the CNS controlled the choice of which genes should be
transcribed, nor that it controlled how those transcripts should be
spliced.  Either could conceivably be part of an evolutionarily-relevant
mechanism.  The actual *sequences* of those transcripts is almost
certainly evolutionarily-relevant, but we know that those are genetic,
rather than epigenetic.

Let me first point out that there is no known genetic mechanism of
determination of the pattern of expression of genes for production
of maternal/ paternal factors in the oocyte, the follicle cells and
nurse
cells.

Sure there is. There certainly are mutations that cause abnormal
deposition of maternal factors or the absence of their depostion.
These have (usually) lethal effects, but sometimes results in
morphological differences (e.g. shell-coiling direction in snails).
That is why and how we know that there are maternal effect genes.

Genes do not know which genes are needed for the development
of the embryo. Right? They only form of information they contain is
for
determining the sequence of amino acids. Right?  

Yes. But how much of a particular protein a nurse cell makes and
deposits may be affected by, say, simple timing of synthesis, thus
laying down factors in a gradient. I have already mentioned how two
factors laid down in a gradient can lead to complex stripping patterns
(the original math ideas came from Turing of cryptology fame) and how
those are key features in development. Simple rules and simple
patterns can lead to complex results.

But that information
for orderly determining the deposition of cytoplasmic factors has to
be somewhere within not outside the organism. Right?

No. It doesn't. Simple rules for nurse cells (a positive feedback
loop leads to more synthesis of a substance as the egg matures,
leading to a gradient of deposition of that substance in the egg
toward the anterior end. For example, if there is a mutation in the
gene bicoid in Drosophila or if it is not expressed in the mother, the
egg will have two tails and no head. There are two other gene
products that are required to prevent the bicoid message from
diffusing from its mostly anterior position (swallow and
exuperantia). Mutation of these genes lead to bicoid extending
further posterior in the egg. If you artificially cause a leakage of
anterior plasm, the consequence is reduction of head structures.

And this is in an organism where that is a lot of "mosaic" regulation
due to deposited maternal factors. Keep in mind that in mammals,
there is very little deposition of maternal factors. Development
there is regulatory. And regulatory and the paucity of maternal
effect features certainly isn't because mammals have a poor CNS
unable to "sense" its environment.

I know that some easy off-the-shelf "solutions" such as cell-cell
interactions also exist. But this complex and ordered process
of deposition of thousands of  cytoplasmic factors (most of
them in strictly determined patterns) requires information for
both selective secretion and transport of these factors which
is impossible to imagine how this information for early
development and its appropriate deposition in the gametes
can be generated by any cell-cell interactions. This would
require the  presence of a vitalist principle in these cells,
which we all reject.

I am not so sure that you do reject vitalism. Your discussion of the
CNS verges on vitalist "information" being transmitted.

But crucially, you repeatedly fail to understand that there
essentially are no "maternal factors" in the early development of
mammals. That early embryonic development is regulatory and position
dependent and environment dependent. That is, all the early
differentiating factors are environmental and, usually, internal to
the developing embryo's genetic capacities and differences. Even the
small amount of difference seen in horse/ass hybrids (aka, mules and
hinnys) dependent on the mother the fetus is raised is probably in
part due to differential interaction between mitochondria (effectively
maternal) and genome (50:50 horse:ass). Or due to sex-related
differences in chromatin methylation (which is the sperm and which the
egg) that gets *reset* each generation.

New knowledge on expression of non-housekeeping genes
shows that what have been considered to be cell-cell
interactions are only parts of the full picture, only last elements
of
neuroendocrine signal cascades.

Again, it is not shocking, given that the CNS's function is
identifying and transmitting information about the local environment,
that the CNS is involved in transmitting that environmental
information in ways that activate or shuts down genes or otherwise
modulates genetic activity.

Let me illustrate this with a well known example of a "cell-cell
interaction", the synthesis of  estrogen by granulosa cells in
the ovary according to the 2-cell model. Behind the proximate
phenomenon of the cell-cell interaction is a full signal cascade
originating in the CNS with the cell-cell interaction being the
proximate link in a Via blood circulation the pituitary FSH (follicle
stimulating hormone) and LH (luteinizing hormone reach ovarian
theca cells and trigger  a complex  interaction between the
theca (interna and externa) and granulosa cells
Hormones diffuse in the granulosa layer where
FSH, via the enzyme aromatase induces  conversion of LDL-
bound cholesterol into progesterone. According to the 2-cell model,
paracrine secretion of estradiol by granulosa  cells stimulates
theca cells  to synthesize androgens, which diffuse into
granulosa cells to be converted there into progesterone. Now
 it is clear that the 2-cell model results from a signal cascade
involving the pituitary hormones, which in turn are induced by
 the hypothalamic GnRH (gonadotropin releasing hormone, which
 in turn  is secreted according to signals from other brain centers,
which in turn are the electric/chemical output of the processing of
numerous internal and external stimuli in neural circuits. It is
namely
 in these circuits that receive and process  various stimuli on the
state of the internal and external environment where the "decisions"
whether, when, and for how long the ovary  has to produce estrogens
No attempt has ever been to genetically explain this phenomenon.

Wrong. The typical way that homeostatic activity works requires the
presence of a "receptor" molecule on the surface of an appropriate
cell that recognizes and binds a specific "allosteric effector". In a
sense, the number of receptors and number of effectors measure levels
in the body. When the receptor binds the effector, a shape change is
transduced across the cell membrane and, typically, a modifying
function (phosphorylation, dephosphorylation, acetylation,
deacetylation etc.) is triggered, modifying a messenger (also often a
phosphorylating -- or whatever -- enzyme). We then get a cascade of
reactions that eventually leads to the activation or modification of
DNA-binding proteins that selectively turn on or turn off specific
genes, such as those making another hormone. There are genes embedded
and being turned on and off throughout the entire process, including
the process of how the CNS sends signals. That is how genes work.

Now, from an epigenetic point of view, the synthesis of all the
hormones synthesized  by the endocrine glands (ovary/ testicles,
adrenal, thyroid, thymus and pancreas) is ultimately under the
CNS control via the hypothalamus--> pituitary -->target endocrine
glands.

Which makes it ultimately under genetic contol. Every step in the
process involves a gene.

The fact that growth factors (of various families such as Wnt,
TGF-beta, EGFfamily, etc.) are also regulated by hormones, implies
that they are ultimately under neural (central and adjacent) control.
We know that all the signal transduction pathways that end with
specific and differential expression of genes in various cells are
extracellularly regulated by hormones and growth factors whose
expression is function of signal cascades starting in the CNS
and by chemicals released by nerve endings.

And the CNS doesn't work by magically producing "information". It
works by the same genetic process of turning specific genes on or off
under certain environmental conditions. There is no "bootstrapping"
of information by the action of skyhooks.

The maintenance of the integrity of the body and homeostasis in the
broadest meaning of the word could not be imagined without  a
central control system capable of monitoring the state of the
 system, detecting deviations from the norm assessing losses
of the order in the structure and function, computating decisions
and sending "instructions" for restoring the normal state of the
system.

Not really. And such a system need not be a CNS. There are many
parasites and essentially all plants that do not have a CNS, yet are
able to respond to changes in their environments.

A control system capable of these functions implies that
iit is in possession of information for the normal structure of the
living system.

No, it implies no such thing. All it implies is that the organism is
able to "sense" its environment and respond appropriately. That is
what CNSs are good for.

Thus, I have shown that the nervous system determines expression of
non-housekeeping genes under normal conditions (for details
read my book Neural Control of Development  (2005) or read
Chapters 1 and 2  of Epigenetic Principles of Evolution(the latter
partly accessible in my websitehttp://www.epigeneticscomesofage.com

I don't have direct evidence that expression of All
(I don't believe we have to have it before we construct a model)
genes  for cytoplasmic factors is determined by the CNS.
One should however not forget that the role of he nervous system
in  production of cytoplasmic factors has never been object of
biological research. However, that evidence is not lacking.
We have evidence that expression and the plasma level of
many of these cytoplasmic factors (testosterone, estrogens,
methyl farnesoate, etc.) and differential uptake from the blood
in oocytes is determined centrally.
We have evidence that deposition of some important cytoplasmic
factors such as bicoid is  neurohormonally is regulated via
microtubules.

Bicoid is a genetic factor. Specifically, it is the bicoid mRNA that
is maternally deposited. That means that bicoid's presence or absence
in an egg is determined by the maternal *genome*.

We have evidence that neurally is regulated the squeezing of the
nurse cell cytoplasm into oocyte.

What "squeezing"? More like a charge differential that favors
material going into the egg.

We have evidence that endocytic deposition of some cytoplasmic
factors is neurally regulated.

Genes are regulated. Deposition of some cytoplasmic factors are
regulated. Regulation implies a "sensing" system of some sort (but
not necessary neural, and even neural becomes molecular at some point
in time and space).

We have evidence that the whole process of oogenesis and
spermatogenesis is under strict neural central and adjacent
(via local innervation)

It is hardly surprising that the CNS is involved in "sensing" the
environment and transferring that environmental information to the
ovary or testes to optimize reproductive success. But any information
the CNS transmits is simply environmental information. And no
evidence that that information gets transmitted in a quasi-genetic
fashion beyond the generation in the egg. Environmental information
gets reset each generation depending on local conditions.

Finally, we have the evidence that expression of non-housekeeping
genes before reproduction and after reproduction is centrally
regulated and there is no visible reason to believe that the
expression
of non-housekeeping genes for cytoplasmic factors has to be
different.

How do you think it is regulated *epigenetically* and *epigenetically*
transmitted to more than a single generation? The only mechanism I
can come up with is a particularly stable prion-like switch or non-
symmetric self-assembly of proteins. But then, that would change with
mutation to DNA.

As for the splicing of genes, this can rightly be considered
to be a "specialty" of the central nervous system.

I would pick the immune system.

In distinction
from other tissues in the nervous system the splicing is
manipulative rather than constitutive, in the meaning that the
CNS by using adaptively using different neurotransmitters
specifies particular pre-mRNAs out of the great number of
possible alternatives.
Manipulative splicing in the nervous system is regulated by a
neuron-specific system of splicing. The electrical activity
eems to be at the basis of production of different protein
iisoforms in the CNS.  

IOW, the differential splicing is the result and consequence of
environmental differences. And acts as one mechanism for transducing
that environmental information to the rest of the organism.

Calcium elevation resulting from the
electrical activity induces specific changes in splicing and
translation processes.

Neurexins, e.g. are surface cell proteins encoded by three
genes, but expressed exclusively in the brain. In the brain
these genes are subject to intense manipulative splicing for
producing specific neurexin isoforms that are involved in
synaptogenesis as it has been concluded from the fact
that their expression changes during synaptic remodeling
(Gorecki, D.C. et al., 1999. Molecular and Cellular
Neuroscience 13: 218-227). Estimates based on experimental
work with expression of neurexins in various parts of the
brain have shown that between 600 and 3000 distinct
splicing-generated neurexins are produced by three
eurexin genes in the brain of the developing Xenopus laevis
(Ullrich, B. et al. 1995. Neuron 14: 497-507).

For a long time biologists wondered (and still do) how the
synapses might generate the great molecular diversity
necessary for recognizing each of their potential target
neurons out of the myriad of neurons in the central
nervous system. Now we know that may be enabled
by the manipulative splicing, which generates numerous
protein isoforms (with different binding affinities) from a
single gene. So, e.g., it is found that three neurexin (nrxn)
genes in mammalian neurons potentially encode over
1,000 isoforms of neuron membrane proteins (Zeng, Z. et al.,
2006. International Journal of Developmental Biology 50: 39-46).

You have shown that these transcripts have developmental
effects. But I don't recall anything that carried the chain through to
causing the development of a system that would deposit identical
parental transcripts in the next generation of ova. Refresh my memory..
Briefly.

Three parts to this.  (1) Show that different CNS signals produce different
sets of transcripts.  

This is  a CNS routine. Look at the previous example, in response to
normal conditions, D. magna produces a neurotransmitter that
stimulates some neurons to produce the  neurohormone MOIH, which
prevents the mandibular gland from producing another hormone, methyl
farnesoate. In response to deteriorating conditions, it produces
another neurotransmitter that stimulates expression of another gene
whose product,  crustacean hyperglycemic hormone induces the
mandibular gland to secrete methyl farnesoate which via the egg cell
determines a transgenerational change.

(2) Show that different sets of transcripts produce

different CNSs.  (Yes, I realize that you have done this to some extent
with your evidence related to the short-term inheritance of insect
morphs.  But the next one is the tough one.)

Cases of transgenerational plasticity show that the nervous
system of the offspring has inherited the newly acquired ability
of the maternal CNS to start the same signal cascade (see the
case of Brachionus angularis when a kairomone, a neurally
perceived species-specific substance, induces  the brain to
secrete a factor affecting the oocyte, so that the brain of the
offspring induces secretion of the same factor in a number of
generations. Such also is the case with phase transition in S.
 gregaria, where the same changes in neural circuits and the
neurotransmitter content (and the resulting morphometric,
physiological and behavioral changes) induced by perception
in the  olfactory circuit are transmitted in a number of
generations.

 (3) Show that (across the

zoosphere) there is enough variation in transcript-sets and CNS
signals to account for the observed variation in CNS structures
(ie. There are millions of animal species.  So we need at least dozens
of bits of information to answer the question "Which CNS do we wish
to develop here?"  But it is not clear to me that there are even that
many different kinds of neurotransmitters.)

The central nervous system, as pointed out many times,
prenatally creates a myriad of connections (estimated in
quadrillions) and neural circuits based on its self-organizing
ability (this is what neurobiologists say), which implies
information-generating ability.

By a simple set of rules.

The variability and evolvability of the central nervous system
depends not on the number of neurotransmitters it is in
possession of or can produce but in the myriad of specific
neural connections and neural circuitries as well as in the
incomparable plasticity of these structures.

We need to also remember a number of facts when we
discuss about the CNS
Neural circuits are at the basis of all  animal behavior.
Neural circuits and the myriad neuronal connections are
established during the embryonic development, i.e experience-
independent (although the postnatal experience is used by the
CNS for fine-tuning these connections and circuits).
(Contrary to what  has been believed) a neural circuit
is not specialized in producing a specific
neurotransmitter/neuromodulator/neurohormonebut it
adaptively can switch to production of different neurotransmitters.
Neural circuits, by processing internal external stimuli, can
relate each of these stimuli (which per se do not and cannot
induce any gene) to adaptively induce any gene by activating
signal cascades that via segnal trsnsduction pathways
can adaptively induce expression of particular genes
even though the stimuli generally have no access to genes.
It is well known that changes in the structure of neural
circuits modify their computational properties and their
output and the potential of the central nervous system
to adaptively change these circuits and their output is
enormous.

Rather, it looks like a mechanism that sloppily generates many
connections randomly and then pares it down by the simple rule of "Use
it or lose it". Simple set of rules. Complex result. Just like
evolution.

Regards,

N.C.

.

Loading