cphoenix@csli.Stanford.EDU (Chris Phoenix) (04/11/91)
I recently started speculating about non-genetic mutations in cells, as a result of speculating about non-software mutations in self-replicating machines. This led to wondering whether it is really possible to reconstruct a cell from only its DNA. I'm sure everyone's read about recreating dinosaurs by finding dinosaur DNA in tar pits or ice and injecting it into a chicken embryo (or something like that). My question is whether DNA really completely determines what a cell "grows up" to be. I recently heard of an experiment in which a paramecium (I think) had part of its cell membrane removed, rotated, and put back, so that some of the spines were backwards. Its "descendants" also had backwards spines. This implies the possibility of one DNA set (sorry, I forget the fancy word) being consistent with several different cells. This would mean that: 1) Dinosaurs could not be replicated from their DNA alone, since there might be differences between dinosaur cells and host cells that the dinosaur DNA could not fix. 2) Evolution of cells might take place by other mechanisms than just mutation of DNA. Non-genetic changes in cells might be hereditary. In the case of sexual reproduction, I would expect that such changes would be passed on only through the mother. Is this totally off base, or is non-genetic evolution a possibility? (Yes, I know mitochondria have their own DNA, and they evolve (or at least mutate. I'm more interested, though, in changes to the cell that do not involve any change in nucleic acid.) Chris Phoenix cphoenix@csli.stanford.edu
eesnyder@boulder.Colorado.EDU (Eric E. Snyder) (04/11/91)
cphoenix@csli.Stanford.EDU (Chris Phoenix) writes: >I recently started speculating about non-genetic mutations in cells.... >I'm sure everyone's read about recreating dinosaurs by finding dinosaur DNA >in tar pits or ice and injecting it into a chicken embryo (or something like >that). My question is whether DNA really completely determines what a cell >"grows up" to be. This is a small departure from you question but it is an interesting case of 'non-genetic' inheritance. In most organisms, there are 'strict maternal effect' genes/mutations. In these cases, the viability of the progeny is independent of the progeny's genotype and determined by the genotype of the mother. Thus, a dinosaur will probably be in need of more than a few 'maternal effect gene-products' which the chicken egg into which it was injected would be lacking. --------------------------------------------------------------------------- TTGATTGCTAAACACTGGGCGGCGAATCAGGGTTGGGATCTGAACAAAGACGGTCAGATTCAGTTCGTACTGCTG Eric E. Snyder Department of MCD Biology ...making feet for childrens' shoes. University of Colorado, Boulder Boulder, Colorado 80309-0347 LeuIleAlaLysHisTrpAlaAlaAsnGlnGlyTrpAspLeuAsnLysAspGlyGlnIleGlnPheValLeuLeu ---------------------------------------------------------------------------
winalski@psw.enet.dec.com (Paul S. Winalski) (04/14/91)
In article <18637@csli.Stanford.EDU>, cphoenix@csli.Stanford.EDU (Chris Phoenix) writes: |>I recently heard of an experiment in which a paramecium (I think) had part of |>its cell membrane removed, rotated, and put back, so that some of the spines |>were backwards. Its "descendants" also had backwards spines. This implies |>the possibility of one DNA set (sorry, I forget the fancy word) being |>consistent with several different cells. Remember that paramecia are single-celled and reproduce mainly asexually (although they do conjugate as well) by simple cell division. The most likely explanation is that the line of cell division crossed the rotated portion of membrane, so that each of the descendents got part of the area with backwards cilia (paramecia don't have spines). |> This would mean that: |>1) Dinosaurs could not be replicated from their DNA alone, since there might |>be differences between dinosaur cells and host cells that the dinosaur DNA |>could not fix. This is quite likely, since it is the machinery of the egg's cytoplasm that drives the early stages of development. Incompatibilities between the proteins coded for by the dinosaur DNA and the cytoplasmic machinery of the host egg cell might well render the combintation non-viable. |>2) Evolution of cells might take place by other mechanisms than just |>mutation of DNA. Non-genetic changes in cells might be hereditary. |>In the case of sexual reproduction, I would expect that such changes |>would be passed on only through the mother. |>Is this totally off base, or is non-genetic evolution a possibility? |>(Yes, I know mitochondria have their own DNA, and they evolve (or at |>least mutate. I'm more interested, though, in changes to the cell |>that do not involve any change in nucleic acid.) I don't think that there are any cases of non-nucleic-acid-based hereditary mechanisms in any known organisms. Such mechanisms might be possible, but living organisms don't make use of them. --PSW
mroussel@alchemy.chem.utoronto.ca (Marc Roussel) (04/15/91)
In article <1991Apr13.203239.22379@hollie.rdg.dec.com> winalski@psw.enet.dec.com (Paul S. Winalski) writes: >I don't think that there are any cases of non-nucleic-acid-based hereditary >mechanisms in any known organisms. Such mechanisms might be possible, but >living organisms don't make use of them. I don't think that right, unless you want to extend the meaning of "hereditary mechanisms" to the point of meaninglessness. Most biological morphogenesis in multi-celled organisms is now known to be the result of complex spatio-temporal interactions (chemical waves and the like) between embryo gene products and the chemical environment provided by the egg and/or uterus. The number of digits that a person (or other mammal) develops is now known to be determined by chemical waves at certain stages of development rather than by direct coding. (I'm afraid that I can't produce any references at the moment due to a library strike at the University of Toronto.) Marc R. Roussel mroussel@alchemy.chem.utoronto.ca
rowe@pender.ee.upenn.edu (Mickey Rowe) (04/15/91)
In article <1991Apr14.211248.2072@alchemy.chem.utoronto.ca> mroussel@alchemy.chem.utoronto.ca (Marc Roussel) writes: > The number of digits that a person (or other >mammal) develops is now known to be determined by chemical waves at >certain stages of development rather than by direct coding. (I'm afraid >that I can't produce any references at the moment due to a library >strike at the University of Toronto.) How well is this known? I'm aware of the affects that retinoic acid has on developing and regenerating limbs, and I know that most developmental biologists anticipate that we will find morphogens that orchestrate such development in situ, but I think your statement might be a little strong. I'm only dimly aware of it, but there seems to be a storm brewing about retinoic acid, and it seems most people are backing away from the position that it does what you say. Is there perhaps something else that you had in mind? If there is or if my understanding is inaccurate, I'd love to hear about it. > Marc R. Roussel > mroussel@alchemy.chem.utoronto.ca Mickey Rowe (rowe@pender.ee.upenn.edu)
winalski@psw.enet.dec.com (Paul S. Winalski) (04/16/91)
In article <1991Apr14.211248.2072@alchemy.chem.utoronto.ca>, mroussel@alchemy.chem.utoronto.ca (Marc Roussel) writes: > |>In article <1991Apr13.203239.22379@hollie.rdg.dec.com> |>winalski@psw.enet.dec.com (Paul S. Winalski) writes: |>>I don't think that there are any cases of non-nucleic-acid-based hereditary |>>mechanisms in any known organisms. Such mechanisms might be possible, but |>>living organisms don't make use of them. |> |> I don't think that right, unless you want to extend the meaning of |>"hereditary mechanisms" to the point of meaninglessness. Most |>biological morphogenesis in multi-celled organisms is now known to be |>the result of complex spatio-temporal interactions (chemical waves and the like) |>between embryo gene products and the chemical environment provided by |>the egg and/or uterus. The number of digits that a person (or other |>mammal) develops is now known to be determined by chemical waves at |>certain stages of development rather than by direct coding. Granted. However, the effects of these morphogenetic mechanisms are not inheritable. To use your example, suppose that we fiddle with the uterine chemistry in such a way as to alter the number of digits on the developing embryo's hands. The baby will duly be born with excess or missing digits, but that morphological characteristic will not be inherited by any of the child's offspring. See the history of thalidomide for proof that this is the case. Now, if we instead affect the spatio-temporal-chemical interactions involving the development of the digits by introducing a point mutuation in the DNA encoding a certain protein critical to the process, we can get a change in the number of digits that *is* inherited by future generations (e.g., polydactyly in domestic cats). By "hereditary mechanisms" I meant means of transmitting persistent (i.e., inheritable by future generations) changes to morphogenesis. I know of no instances of such mechanisms that are not nucleic-acid-based. This includes some examples where the mechanism of the developmental characteristic involves the chemical environment of the egg. Consider the left-hand-spiral mutuation in the pear whelk (genus Busycon). The shell of this gastropod normally has a clockwise spiral when viewed from the top. There is a mutation that results in a counterclockwise spiral. It behaves as a simple Mendelian recessive gene, but with one difference. The factor that determines the direction of the shell spiral comes from the egg cytoplasm and therefore is dictated by the genotype of the female parent, not by that of the zygote. Thus, if you mate two snails heterozygous for the shell spiral gene (Ss x Ss) to produce a female snail homozygous for the recessive left-hand spiral (ss), that snail's shell will have a right-hand spiral, as determined by the mother's genotype. However, all offspring of this snail will have counterclockwise-spiralling cells, regardless of their genotype (Ss or ss). Here we have a case where cytoplasmic factors result in a phenotype that seemingly contradicts the genotype, but it is only the DNA-encoded information about this characteristic that is passed on to other generations. I am not aware of any exceptions to this principle. --PSW
mroussel@alchemy.chem.utoronto.ca (Marc Roussel) (04/16/91)
In article <1991Apr15.185049.3959@hollie.rdg.dec.com> winalski@psw.enet.dec.com (Paul S. Winalski) writes: >By "hereditary mechanisms" I meant means of transmitting persistent (i.e., >inheritable by future generations) changes to morphogenesis. I hate to do this, but with the libraries off limits to me right now I have no choice but to try to reproduce from memory an example that I heard several months ago. The organism in question is called spiralea (I think) and seems to engage in non-genetic phenotype transmission. I think that this organism is multicellular, although it doesn't reproduce sexually (if that makes a difference to you). The thing has some kind of curly tail. Clockwise tailed parents produce only clockwise tailed offspring. By changing the handedness of the tail surgically, one changes the phenotype of all successive generations of offspring in the same way that the parent's was changed. No genetic changes are involved. The only thing I remember clearly about this organism is that B.C. Goodwin has suggested it as an important counterexample to the genotype --> phenotype paradigm, so clearly I'm not just hallucinating (unless B.C.G. is also a hallucination...). I may be able to dig out a reference if anyone is interested, but I'm afraid that I can't for the moment do more than that. Marc R. Roussel mroussel@alchemy.chem.utoronto.ca
rs54@cunixf.cc.columbia.edu (Richard Sucgang) (04/17/91)
In article <1991Apr14.211248.2072@alchemy.chem.utoronto.ca> mroussel@alchemy.chem.utoronto.ca (Marc Roussel) writes: >In article <1991Apr13.203239.22379@hollie.rdg.dec.com> >winalski@psw.enet.dec.com (Paul S. Winalski) writes: >>I don't think that there are any cases of non-nucleic-acid-based hereditary >>mechanisms in any known organisms. Such mechanisms might be possible, but >>living organisms don't make use of them. > > I don't think that right, unless you want to extend the meaning of >"hereditary mechanisms" to the point of meaninglessness. Most >biological morphogenesis in multi-celled organisms is now known to be >the result of complex spatio-temporal interactions (chemical waves and the like) >between embryo gene products and the chemical environment provided by >the egg and/or uterus. The number of digits that a person (or other >mammal) develops is now known to be determined by chemical waves at >certain stages of development rather than by direct coding. (I'm afraid >that I can't produce any references at the moment due to a library >strike at the University of Toronto.) > > Marc R. Roussel > mroussel@alchemy.chem.utoronto.ca I take it you refer to the Turing models of pattern formation. The problem is, what encodes for the source and sinks of the gradient? This is still a genetic factor. In drosophila, for example, proper morphogenesis is dependent on how the maternal RNA is distributed in the initial embryo. That remains quite essentially, genetic. -rich Richard Sucgang : Dept. of Anatomy and Cell Biology Columbia University (sucgang@cuhhca.hhmi.columbia.edu; de slime god rs54@cunixf.cc.columbia.edu)
mroussel@alchemy.chem.utoronto.ca (Marc Roussel) (04/17/91)
In article <1991Apr16.235422.20331@cunixf.cc.columbia.edu> rs54@cunixf.cc.columbia.edu (Richard Sucgang) writes: >In drosophila, for example, >proper morphogenesis is dependent on how the maternal RNA is distributed >in the initial embryo. That remains quite essentially, genetic. Anything to do with biochemistry is ultimately genetic. I don't think there's any denying that. There is a question however which needs to be raised which isn't often enough asked: by focussing so much on genes, are we missing parts of the big picture? Let's go back to the original question of this thread. If I remember correctly, someone asked if a dinosaur could be produced from fossilized DNA. Just for the sake of argument, let's replace the dinosaur with something more exotic. (It doesn't matter what exactly. I'll let you fill in the blanks.) If DNA were the blueprint which many claim it to be, then certainly we could reproduce an entire organism from its DNA. However it is my contention that unless we know something about the environment in which development of the original owner of the DNA occured, we would have no hope of reproducing the organism from its genetic material alone. Thus we might be able to get a dinosaur by guessing that a modern reptile would provide an appropriate environment for dinosaur genetic material, but we would have no hope of getting a functioning organism from an unidentified piece of DNA. Marc R. Roussel mroussel@alchemy.chem.utoronto.ca
szabo@crg5.UUCP (Nick Szabo) (04/18/91)
In article <1991Apr17.013059.7708@alchemy.chem.utoronto.ca> mroussel@alchemy.chem.utoronto.ca (Marc Roussel) writes: >...if a dinosaur could be produced from fossilized DNA.... >If DNA were the blueprint which many claim it to be, then certainly we >could reproduce an entire organism from its DNA. However it is my >contention that unless we know something about the environment in which >development of the original owner of the DNA occured, we would have no >hope of reproducing the organism from its genetic material alone. Thus >we might be able to get a dinosaur by guessing that a modern reptile >would provide an appropriate environment for dinosaur genetic material, >but we would have no hope of getting a functioning organism from an >unidentified piece of DNA. The question is, what conditions in the surrounding zygote must be present in order for the DNA to run its embryological course and produce fertile offspring? Tbe answer is, a whole bunch of stuff -- from the immediate transcription RNA to the whole "digestive system" of the zygote. Much of this is based on that or similar DNA from the _previous_ generation. Thus, it is somewhat of a recursive problem. Dare I say "chicken and egg" problem? :-) BTW, what is the current status of research on using host mothers of common species for endangered species? I seem to remember some zoo had a common tiger female giving birth to an endangered tiger a few years back, but I don't remember any of the details... -- Nick Szabo szabo@sequent.com "The biscuits and the syrup never come out even" -- Robert A. Heinlein The above opinions are my own and not related to those of any organization I may be affiliated with.
tomh.bbs@shark.cs.fau.edu (Tom Holroyd) (04/18/91)
> >The number of digits that a person (or other > >mammal) develops is now known to be determined by chemical waves at > >certain stages of development rather than by direct coding. > > The problem is, what encodes for the source and sinks of the > gradient? This is still a genetic factor. In drosophila, for example, > proper morphogenesis is dependent on how the maternal RNA is distributed > in the initial embryo. That remains quite essentially, genetic. I disagree. There are, initially, no sources or sinks, only a homogenous field with within-field negative feedback. This leads to a symmetry breaking dynamic which, in a totally self- organized fashion - nothing to do with genetics here, only chemical dynamics, *develops* sources and sinks. The precise locations are determined by distance dependent mechanisms and the wavelengths of the self-organized chamical waves. This type of mechanism is much more flexible than a 'hard coded' system. It self-adjusts for small differences between embryos and doesn't beg the question of 'who decides where the fingers go'. It's also non-genetic. But to put it another way: The genes regulate the purely physical process - physical law determines what happens, but the genes have evolved to control these processes. Philosophical point: Fighting against nature is harder than getting nature to do the work for you. Tom Holroyd Florida Atlantic University Center for Complex Systems tomh@bambi.ccs.fau.edu
colby@bu-bio.bu.edu (Chris Colby) (04/21/91)
In article <18637@csli.Stanford.EDU> cphoenix@csli.Stanford.EDU (Chris Phoenix) writes: >2) Evolution of cells might take place by other mechanisms than just >mutation of DNA. Evolution is defined as a change in the gene pool. So, changes in a heritable material must occur of evolution is to occur. So far DNA (and RNA) are the only two known genetic materials. >Non-genetic changes in cells might be hereditary. This is like saying non-blue balls might be blue. If something is hereditary, it must be genetically based (although it would not neccesarily have to be genetic material in the form of nucleic acid). >In the case of sexual reproduction, I would expect that such changes >would be passed on only through the mother. Why? >Is this totally off base, or is non-genetic evolution a possibility? It's totally off base ;-) Non-genetic evolution is an oxymoron. >Chris Phoenix cphoenix@csli.stanford.edu Chris Colby email: colby@bu-bio.bu.edu
bill@ut-emx.uucp (Bill Jefferys) (04/21/91)
In article <79788@bu.edu.bu.edu> colby@bu-bio.UUCP (Chris Colby) writes: #In article <18637@csli.Stanford.EDU> cphoenix@csli.Stanford.EDU (Chris Phoenix) writes: #>Non-genetic changes in cells might be hereditary. # # This is like saying non-blue balls might be blue. If something #is hereditary, it must be genetically based (although it would not #neccesarily have to be genetic material in the form of nucleic acid). # #>In the case of sexual reproduction, I would expect that such changes #>would be passed on only through the mother. # # Why? # #>Is this totally off base, or is non-genetic evolution a possibility? # # It's totally off base ;-) Non-genetic evolution is an #oxymoron. Actually, Chris P. is not so far off as one might think. _The Economist_ had a story this week (in the science section) about a certain species of wasp that reproduced parthenogenetically; all of the offspring are female. Yet, if the wasps are fed antibiotics, they will produce males and females in 50% ratio. Why? It appears that a bacterium has co-evolved with the wasp. It is transmitted via the eggs (see Chris P's speculation above!!!), and has evolved the ability to supress the production of male offspring. Presumably this is because it provides an advantage to the bacterium! Yet the wasp retains the ability to produce males, so it is the _conditions_ under which the embryo develops that determine the sex. A non-genetic "evolution" that "breeds true" as long as the bacterium is present. Bill Jefferys -- If you meet the Buddha on the net, put him in your kill file --Robert Firth
andrewt@cs.su.oz (Andrew Taylor) (04/21/91)
In article <21567@crg5.UUCP> szabo@crg5.UUCP (Nick Szabo) writes: > BTW, what is the current status of research on using host mothers of common > species for endangered species? I seem to remember some zoo had a common > tiger female giving birth to an endangered tiger a few years back, but > I don't remember any of the details... There is only one species of tiger. The zoo was presumably breeding a particular subspecies. Sounds likes it was transfering embryos to other tigers. Most captive tigers are of mixed lineage. Cross-species fostering has been used to good effect in several breeding programs for endangered birds. Re-introduction currently seems to me the biggest technical obstacle to captive breeding programs. Still more difficult usually is removing the causes for the decline of the species. Andrew Taylor
cphoenix@csli.Stanford.EDU (Chris Phoenix) (04/25/91)
In article <79788@bu.edu.bu.edu> colby@bu-bio.UUCP (Chris Colby) writes: >In article <18637@csli.Stanford.EDU> cphoenix@csli.Stanford.EDU (Chris Phoenix) writes: >>2) Evolution of cells might take place by other mechanisms than just >>mutation of DNA. > Evolution is defined as a change in the gene pool. So, changes >in a heritable material must occur of evolution is to occur. So far >DNA (and RNA) are the only two known genetic materials. >>Non-genetic changes in cells might be hereditary. > This is like saying non-blue balls might be blue. If something >is hereditary, it must be genetically based (although it would not >neccesarily have to be genetic material in the form of nucleic acid). OK, I used the wrong words. I should have said "transmissible to offspring" instead of "hereditary" and "change in characteristics of cell" rather than "evolution". That said, my question remains: Is it possible to have a transmissible change take place in a cell that does not involve nucleic acid? Here's a possible example: Prions do not contain nucleic acid, but can make more prions inside a cell. If a prion is introduced to a cell, and the cell then produces offspring, the offspring will presumably have the prion in them. From then on, the descendants would all have the prion (plus whatever effects the prion had), whereas the ancestors didn't. But the cell's nucleic acid would be unchanged. BTW, can you give me an example of genetic material that is not in the form of nucleic acid? >>In the case of sexual reproduction, I would expect that such changes >>would be passed on only through the mother. > Why? Because the (egg) cells the mother contributes are much larger and more complex than the (sperm) cells the father contributes. Eggs would be more likely than sperm to preserve non-nucleic changes. Chris Phoenix cphoenix@csli.stanford.edu
colby@bu-bio.bu.edu (Chris Colby) (04/29/91)
In article <18841@csli.Stanford.EDU> cphoenix@csli.Stanford.EDU (Chris Phoenix) writes: >Here's a possible example: Prions do not contain nucleic acid, but can >make more prions inside a cell. If a prion is introduced to a cell, and >the cell then produces offspring, the offspring will presumably have >the prion in them. From then on, the descendants would all have the >prion (plus whatever effects the prion had), whereas the ancestors didn't. >But the cell's nucleic acid would be unchanged. I assume you are talking about the putative cause of mad cow disease (I just read in Discover about this). Do you have any scientific references that confirm they are autonomously replicating? (BTW, it's a question not a challenge) >BTW, can you give me an example of genetic material that is not in the >form of nucleic acid? Not unless prions check out. I only read about them in Discover so it's hard to tell what the scoop is. >Chris Phoenix cphoenix@csli.stanford.edu Chris Colby email: colby@bu-bio.bu.edu
cphoenix@csli.Stanford.EDU (Chris Phoenix) (05/01/91)
In article <80395@bu.edu.bu.edu> colby@bu-bio.UUCP (Chris Colby) writes: >In article <18841@csli.Stanford.EDU> cphoenix@csli.Stanford.EDU (Chris Phoenix) writes: > I assume you are talking about the putative cause of mad >cow disease (I just read in Discover about this). Do you have any >scientific references that confirm they are autonomously replicating? >(BTW, it's a question not a challenge) Yes, that's what I had in mind. I think they also cause some Alzheimer's- like disease in sheep. No, I don't have any scientific references, but I did read a model of how it might work on the net a while back, in sci.chem I think: The prion protein is a trimer, but I'll give the analogous model for a dimer. (A dimer is made from two proteins that fit together, and a trimer from three (as I understand it)). So: You have two proteins [ and ], and normally they come together to create []. But when they are "prionized" they create [ which of course can lengthen itself, break apart to ] form two smaller chains and so replicate itself, [ and tie up lots of protein that should be making []. ] >>control of fetal growth with interest. But I haven't seen anyone >>address the question of where the hormones come from. I think all of >>them will be coded for by the DNA of the original fertilized egg. > > Hormones are proteins coded for by genes. I know *what* they are... I was wondering if some hormones might be put in the [macro] egg by the chicken, and so be part of the environment no matter what DNA is put into the egg [cell]. -- Chris Phoenix cphoenix@csli.stanford.edu #insert <funnyquote.h> #insert <graphic.h> #insert <stddisclaimer.h>
bill@ut-emx.uucp (Bill Jefferys) (05/03/91)
In article <21630@crg5.UUCP> szabo@crg5.UUCP (Nick Szabo) writes: #In article <47570@ut-emx.uucp> bill@ut-emx.uucp (Bill Jefferys) writes: # #>...a bacterium has co-evolved #>with the wasp. It is transmitted via the eggs... and #>has evolved the ability to supress the production #>of male offspring. Presumably this # #Since when did bacterium lose their genes? Parasatism is #every bit as much genetic evolution as more "normal" methods #of making a living. And it is well known that predator and parasite Sure, I know that bacteria contain genes. And (as I pointed out to Chris C. in E-mail and as others have pointed out here) the acquisition of organelles by eukaryotic cells is an example of "infection" of an organism by outside genetic material that changes the genetic complement of the organism. What is perhaps unique, and certainly different, about this example is the fact that the condition is reversible when the wasp is exposed to antibiotics. If you will reread my original posting, you will note that I did not posit this as a case of non-genetic evolution. I did point out that the example shows that the idea is not as far-fetched as others were making it out to be. In biology, it is frequently difficult to draw clear lines of demarcation. Bill Jefferys -- If you meet the Buddha on the net, put him in your kill file --Robert Firth
ireland@ac.dal.ca (05/06/91)
Some time ago in this discussion of non-genetic evolution someone mentioned cortical inheritance in ciliates. If I recall correctly, it was stated that the cortical changes are not inherited over many generations, which was not how I remembered these experiments. I was looking through some old papers recently and came across a nice paper which discusses this work and I thought some of you might like to have a look at it. The reference is: Nanney, D. L., "Heredity without genes: ciliate explorations of clonal heredity". Trends in Genetics (November 1985) pp.295-298. Keith Conover ireland@ac.dal.ca