In this first part Art tackles the question of whether proteinoid microsphere
protocells have any relevance to the origin of life.
> The absurdity of the proteinoid microsphere route to a living cell begins
> not with the proteinoid microsphere for which Kevin claims living
> properties, but with the very idea that proteinoids could ever have been a
> part of the prebiotic world. In order to make proteinoids you must have
> essentially pure amino acids....
No, you don't; see Fox and Dose pg. 150-82. "The polymers may include such
added substances as iron or hemin [Dose K, Zaki L. _Z. Naturforsch_. 1971;
26b:144]. The reactions are very rugged; they have been carried out in the
presence of a wide variety and varied amounts of added terrestrial materials.
The condensations are not easily disturbed [pg. 153]." The authors go on to
explain that the molten polymer simply does not dissolve significant
proportions of other substances and that many minerals are simply not very
soluble in amino acid mixtures that produce proteinoids, so contamination is
not a problem. Pure amino acids were used simply for experimental
expediency, not to satisfy some theoretical or experimental requirement.
> ...with a great proponderance of Glu and Asp.
Untrue. Nonneutral amino acids such as aspartate, glutamate or lysine are
required (they appear to aid condensation through acid or base catalysis),
but you can get proteinoids from mixtures with as little as 15% molar
composition of nonneutral amino acids (Fox and Dose pg. 150, 151).
> Then you must heat the purified amino acids in the absence of water, then
> dump them into water to make the microspheres.
As Fox and Dose point out (pg. 150): "The process of heating amino acids is
of the utmost simplicity, and can properly be imputed to spontaneous
reactions on the Earth."
> Before you get too excited about the outcome, note:
> 1) Proteinoids are not proteins....
Fox and Dose point that out themselves (pg. 194), but they add: "Their role
is that of mother substance, or 'urprotein' [Alcock RS. _Physiol. Rev._
1936; 16:1]." However, proteinoids are polyaminoacids linked by the peptide
bond, thus qualifying them as polypeptides if not full proteins. Chemical
Abstracts classifies proteinoids as proteins, under the subheading "thermal".
They also share many characteristics with proteins, such as high molecular
weight (as opposed to simple peptides), qualitative and quantitative
compositions closely resembling those of proteins, sharply limited
heterogeneity in physical properties and composition and sequence, diverse
arrays of catalytic activities and the ability to react selectively with
polynucleotides (Fox and Dose pg. 154-78, 194).
> ...they contain many non-peptide bonds....
So do proteins: van der Waals bonds, electrostatic bonds and hydrophobic
bonds, as well as the disulfide covalent bond. You would probably call the
first three "interactions" or "associations", but they are classified by
chemists with covalent interactions as the four basic types of bonds.
Proteinoids are known to have other types of covalent bonds, but the first
three listed above, plus the peptide covalent bond, predominate.
> ...and unnatural cross-linkages.
"Unnatural" is a biased term; any chemical bond is natural. Of course, what
you mean is that they contain crosslinkages that have not been reported for
proteins. Fox and Dose discuss this (pg. 151-2) and point out that covalent
crosslinkages between certain amino acids (such as between aspartate and
lysine) have been postulated for certain kinds of proteins, and that their
presence could explain the resistence of some polypeptides to hydrolysis (the
postulated crosslinkages are known to be difficult to hydrolyze). However,
they also point out that with certain exceptions, proteinoid amino acids are
largely joined by hydrolyzable linkages, of which the peptide bond
However, proteins are now known to have a wide variety of covalent
crosslinkages which were not known about when Fox and Dose wrote their book.
Transglutaminases catalyze the formation of an epsilon-peptide bond between
glutamate and lysine to form fibrin clots during hemostasis, histidine and
lysine can photodynamically crosslink, hydroxylysine and lysine create the
crosslinks that hold collagen molecules together, certain cytochromes form
crosslinks between heme groups and both tyrosine and lysine residues, and
carbohydrates are used to crosslink proteins to form proteoglycans, that act
as structural materials in bone and cartilage. So in fact it would appear
that proteinoids are not all that different from proteins even in this
> 2) The peptide bonds they do contain are beta bonds....
As I explained above, peptide bonds predominate in proteinoids, so Art's
implication that peptide bonds are rare, or at least a distinct minority, is
unfounded (see Fox and Dose pg. 168). Imide linkages are found in
proteinoids that contain aspartic acid, but they are unstable and tend to
readily convert to amide bonds in water by dilute alkali or warming (Fox and
Dose pg. 168). As for peptide bonds, Art's implication that they are all
beta-peptide bonds is also unfounded. However, many proteinoids do contain
beta- and epsilon-peptide bonds as well as alpha-peptide bonds. The amount
depends upon the composition. Polyaspartate proteinoids contain roughly 50%
beta-peptide bonds, whereas polyglutamate proteinoids contain virtually none.
Polymers of alanine and lysine have been reported to have over half of the
lysines forming epsilon-peptide bonds. Even so, the backbone of any
proteinoid still consists of a polymer of amino acids linked by peptide bonds
and so still qualify as polypeptides; whether they are alpha-, beta- or
epsilon-linkages is irrelevant. The modern protein with its predominant
alpha-linkage is almost certainly a product of evolution, not some inherent
> ...whereas all biological peptide bonds are alpha.
Incorrect. Crosslinkages between amino acid residues often utilize
epsilon-peptide bonds, and since proline is an imino acid it forms an imide
bond rather than the normal amide (peptide) bond.
> 3) His starting materials are purified amino acids....
As I explained above, this factor is not a requirement, just an experimental
> ...bearing no resemblance
> to the materials available in the hypothetical "dilute soup."
This is also incorrect. See Fox and Dose pg. 69-100, especially pages
93-100. Fox and other researchers have shown that, regardless of whether you
use a gaseous phase or an aqueous phase, whether you use electrical
discharges, ultraviolate light, ionizing radiation, optical radiation, or
thermal energy, you get exactly the kind of mixture of amino acids,
hydrocarbons, saccharides, nucleic acids, phosphated nucleotides, porphyrins
and other biologically significant compounds predicted by the "primordial
soup" hypothesis and a number of other scenarios. And proteinoids have been
formed from mixtures of amino acids containing nucleic acids, fatty acids and
> between .0001 and .000001 g/l -about the range of concentrations in the
> mid-atlantic today).
The implication here is that the "soup" would have been too dilute to form
macromolecules, but this is also incorrect. According to Fox and Dose (pg.
154): "Also, since amino acids pass from a solid state [after dehydration]
to a molten mixture [after heating] of polymer with oligomers and
monomers...no question of an act of concentrating amino acids from a solvent
need arise. Also, as stated earlier, an aqueous solution may be heated until
the dry amino acids left by evaporation condense. The amino acids are at
infinite concentration in the dry state. An infinitesimal amount of amino
acid can polymerize on heating as indicated, providing the small but
sufficient proportion of nonneutral amino acid is present." And Fox and Dose
also report on condensation experiments with hydrogen cyanide, nucleotides
and saccharides (pg. 182-93).
> If one were to try the experiment with condensed
> "prebiological soup," tar would be the only product.
Again, incorrect, first of all because condensation experiments require that
the "soup" be dehydrated, and secondly because experiments reported by Fox
and Dose clearly demonstrate that you get macromolecules and not tar.
> 4) The ratio of 50% Glu and Asp necessary for success in these experiments
> bears no resemblance to the vastly higher ratio of Gly and Ala found in
> nearly all primitive earth synthesis experiments.
Again, incorrect. First of all, as I pointed out above, Art's 50% ratio is
unnecessary. Secondly, glycine and alanine are not the only amino acids
produced in large amounts. As Fox and Dose explain (pg. 10): "When we
examine all of the relevant experiments comparatively (page 130), we find
that the compounds most common to organisms, i.e., adenine, alanine, aspartic
acid, glycine, etc., are those that appear mosty frequently and in largest
proportion in 'origin' experiments." Note that they included a nonneutral
amino acid. How much you get of any amino acid does depend upon the method
you use, though. Electrical discharge experiments produce only 1% the amount
of aspartate as compared to glycine, but ionizing radiation like x-rays can
produce 50% the amount of aspartate as compared to glycine and thermal energy
can produce 13% to 63% the amount of aspartate as compared to glycine (Fox
and Dose pg. 89). Thirdly, glycine and alanine can be easily modifed into
glutamate and aspartate by reaction with formate or acetate. Fourthly, the
very nature of the thermal condensation process means that even small amounts
of nonneutral amino acids can catalyze large amounts of proteinoids.
In conclusiion it would seem that, despite his claims to the contrary, Art's
knowledge of the relevant research is in fact quite limited and full of
In this next part, Art switches gears suddenly and starts discussing whether
life can be synthesized in the lab. I should point out, however, that I do
not agree with the way these characteristics are worded, so I am placed in
the awkward position of defending statements I would myself critique.
However, Art's evaluation of them is so ludicrous that I cannot ignore it.
For a general discussion on the properties of proteinoid microspheres, see
Fox and Dose pg. 203-20.
> Now lets examine the claims of those who feel they are on the track of
> creating a living cell
Art is using an old debators trick here: divide and conquer. Rather than
treat all four points as the connected argument they are, he treats them
separately, under the assumption that if he can refute each one individually
he can refute the whole argument. Such an assumption is fallacious, but
considering that Art cannot dispute the whole argument he has no choice but
to use this tactic.
> >> 1. "Delineate itself from its environment through the production and
> >> maintenance of a membrane equivalent, most probably a rudimentary or
> >> quasi-active-transport membrane necessary for selective absorption of
> >> nutrients, excretion of wastes, and overcoming osmotic and toxic
> >> gradients,
> This is something a simple membrane of cellophane can do.
To begin with, this is an example of a logical fallacy called a synecdoche.
This is where one tries to dismiss a general notion by citing a single poor
or ridiculous example. As I will soon explain, cellophane membranes are
little like biological membranes, either in structure or function, so using
the former to refute the claim that the latter is a characteristic of life is
Up to a certain point, yes, cellophane membranes can do what biological
membranes can. Cellophane bags and cells both allow small molecules like
water, oxygen, carbon dioxide, salts and even glucose to diffuse in and out
more or less freely, while severly restricting the ability of larger
molecules like polypeptides, polysaccharides and polynucleotides to do the
same. The difference lies not in what they do, but how they do it.
Cellophane is made up of a crosslinked polymer whose structure leaves gaps
that serve as holes through the membrane. The tighter the crosslinkage, the
smaller the holes. The polymer weave itself is impermiable; nothing can
cross the membrane except by going through a hole. It is the size of the
holes that determine the size of molecules allowed to diffuse freely. Any
molecule small enough to pass through the holes can do so freely based solely
on osmotic pressure; any molecule too large to pass through the holes can
never do so regardless of osmotic pressure. This is the basis for dialysis.
A cell, however, has a biological membrane. Biological membranes are made of
amphiphilic molecules that are held togther by a combination of hydrophobic
cohesion and hydrophilic interaction, not crosslinkages. This means there
are no holes; this also means that the membrane is semi-permiable. Small
molecules that interact weakly with the membrane can pass more or less
freely; larger molecules that would interact more strongly pass through with
greater difficulty. The diffusion of molecules across the membrane is still
driven by osmotic pressure as it is with cellophane bags. However, there is
no maximum limit to the size of molecule that can pass a biological membrane;
any molecule, no matter how big, can conceivably pass through. That is one
significant difference between a biological membrane and a cellophane
Another is that cells often need to control the amount of small molecules
they contain. A cell that absorbs too much sodium for example can die, so
cell membranes have to contain portals that can actively pump out small
molecules that otherwise get in freely. A cellophane membrane could never so
this. Similarly, there are times when cells have to let in large molecules
that would otherwise have great difficulty crossing the membrane, so again
membranes have special portals that allow these large molecules to enter
while preventing the cell's contents from escaping. This is something else a
cellophane membrane could never do. This what the people who created these
criteria meant when they said "quasi-active-transport membrane necessary for
selective absorption of nutrients, excretion of wastes, and overcoming
osmotic and toxic gradients." A cellophane membrane could never defy osmotic
pressure or physical size to selectively transport specific molecules, but
biological membranes can, even the biological membranes of a proteinoid
Proteinoid microspheres respond to osmosis (Fox and Dose pg. 213), but they
also demonstrate selectivity (pg. 213-4). As such, microspheres are more
like cells than they are like cellophane bags.
> That does not make cellophane alive.
Neither does it make a biological membrane alive, yet a biological membrane
is an important characteristic of something that is alive: a cell. For a
cell to be alive it must have a membrane, even though the membrane itself is
not alive. Similarly, for protocells to be alive they must also have a
membrane, even though that membrane is also not alive. In deliberately
trying to confuse the characteristic itself with the whole structure, you are
presenting a fallacious point.
> >> 2. Capture, transduce, store, and call up energy for utilization
> Capturing energy is something any soap bubble or cellophane membrane can
> do. Transducing that energy is what happens when it is carried across the
> membrane to the inside, where it can be stored, and used for work
> (expansion of the bubble).
This is another synecdoche. In this case energy (heat actually) crosses the
boundary due to a thermaldynamic gradient. If it is hotter outside the
bubble than in, the heat flows inside; if it is hotter inside the bubble than
out, heat flows outside. When heat flows in, it increases the Brownian
motion of the ait trapped inside, which increases the pressure the air can
exert on the boundary, which increases the volume of the bubble; hence the
bubble expands. When heat flows out, the exact opposite happens; hence the
bubble contracts. This is a purely physical reaction and the bubble has no
influence or control over it.
> Anyone who has carried out experiments on
> osmosis in high school biology class has done all of these with a passive
> cellophane membrane.
Still another synecdoche. As with the soap bubble, the inflow of energy (in
this case the mechanical energy of moving water molecules) is caused by
osmotic pressure created by a solute concentration gradient. If the
cellophane bag contains a large molecular solute that cannot cross the
membrane, then when the bag is placed in pure water an osmotic pressure is
created that causes water molecules to flow into the bag in an attempt to
dilute out the solute. The increase in the number of water molecules
increases the pressure on the inside of the bag, which increases its volume;
hence it expands. If, however, a bag containing pure water is placed in
water saturated with the large molecular solute, the osmotic pressure created
causes water molecules to flow out of the bag, creating the opposite effect;
hence the bag contracts. This is also a purely physical phenomenon, but the
bag has no more control or influence over it than the soap bubble had over
But this is not what the people who selected these criteria meant by the
capture, transduction, storage, and utilization of energy. If all that cells
did was rely on purely physical or chemical phenomena to acquire, transduce,
store and utilize energy, with no control or influence over the process, they
would never survive. Cells don't just expand or contract, they grow. That
is, they create additional structures that they use to increase their size by
expanding and strengthening their membranes, thus giving themselves greater
volume; they then fill that extra room with other structures. The synthesis
of specific structures requires more energy than simple physical expansion
does. This energy cannot come from simple gradients; it has to come from
either light or chemicals. The light must be specifically captured and
converted into chemical energy, stored in the form of molecules that the cell
can ready access and use to create new structures. Chemicals must be
specifically ingested and broken down to release their energy, which is then
used to create energy storage molecules that the cell can access and utilize
to create new structures. Each step is controled by the cell, using
tranducers, specific molecular machines that do the actual conversion. This
is fundamental different from the purely physical processes you described.
Proteinoid microsphere protocells also specifically capture light (Fox and
Dose pg. 253) or ingest chemicals, use molecular transducers to convert the
energy into storage molecules, then use these molecules to create new
structures (Fox and Dose pg.171-6, 217, 253-4). This makes them
fundamentally much more like modern cells than soap bubbles or cellophane
> >> 3. Actively replicate, not just passively polymerize or crystallize,
> If you can explain what you mean by "active" and "passive" process is, I
> will understand what the difference between the growth of a crystal and the
> growth of a proteinoid microsphere is, in physical terms.
Polymerization and crystallization are passive because they utilize no
specific template and simply rely on physiochemical forces and ambient energy
to form structures that convey random information, or simple repetitive
information. Replication is active because a specific template is used, the
process is directed by specific molecular machines and stored chemical energy
is utilized to create a specific nonrandom non-repeating sequence that
conveys complex ordered information. Cells replicate structures, as do
protocells (Fox and Dose pg. 232-7).
> The crystal is
> supplied with elements externally that then accrue to the crystal and
> increase its size. It may also bud off into new crystals as well. The
> proteinoid microspheres are supplied with elements externally that accrue
> to the microcell and increase its size. It may also bud off into new
> microspheres as well. The difference is.....?
This is an example of the strawman argument. By the description given above
there is no difference, but the description of how microspheres behave is
incorrect. Your description implies that microspheres are solid, but they
are not (see Fox and Dose pg. 208-13). They consist of a membrane enclosing
a hollow space filled with liquid. This space can contain proteinoids or
even smaller microspheres. The outer membrane consists of no more than two
layers. Microspheres appear to grow by accreting proteinoids onto their
outer surface (Fox and Dose pg. 218), but this is not simply layering on
proteinoids the way a crystal layers on atoms or molecules. The proteinoids
actually form a structured membrane. The old inner membrane then dissociates
into individual proteinoids that then reorganize into ultrastructures within
the hollow space. Those microspheres that contain proteinoids capable of
synthesizing peptides can add them to the interior ultrastructure, forcing
the microsphere to expand and thus create new membranes. The result is that
microspheres grow the way cells do, by creating (or incorporating) new
structures that they use to expand and strengthen their membranes, then
filling the increased space with other structures that perform specific
functions. This is fundamentally different from the way a crystal grows.
It should also be pointed out that like cells microspheres will only grow so
large, then no larger, whereas there is no limit to how big a crystal can
get. As for budding (Fox and Dose pg. 217, 218), this also appears to occur
as a result of accretion on the outer membrane, but as explained above, this
is not simply a new crystal being formed but a kind of membrane within a
membrane. When it grows large enough it detaches and begins life as a new
microsphere. Microspheres can in fact proliferate in this fashion, because
each generation can grow into a new parent that creates more buds.
Microspheres can also fission like cells (Fox and Dose pg. 214-5). This
doesn't necessarily involve replication (though it can), but the contents do
have to reorganize themselves to form the doublet that will then cleave.
> >> 4. Write, store, and pass along seemingly conceptual information that
> >> 'gives orders' for what is to be manufactured in the future, and to
> >> actually bring to pass those processes and "factory products" out of
> >> linguistic-like coded (codon) messages ('recipes') into physical
> >> biochemical, biological, and thermodynamic reality."
> Here absurdity reaches new heights. Certainly in every sense that
> proteinoid microspheres are capable of these things, inorganic crystals are
> also. They are capable of giving 'orders' about what is being added in the
> future, and actually bringing to pass those processes and products out of
> linguistic-like coded messages (ordered arrays of molecules that are part
> of the crystal surface) into reality.
And again, still another synecdoche. This process for crystals is extremely
simplified and is capable of producing only a limited repetitious sequence
that can do nothing except continue to produce the same limited repetitious
sequence. In other words, crystals don't do anything except increase in size
as a result of simple physiochemical process they can neither control nor
influence. Those microsphere capable of synthesizing peptides or
polynucleotides do so by using existing proteinoids as templates (Kox and
Dose pg. 254). The newly created peptides and polynucleotides can then serve
as templates for producing more molecules, but they also preserve and pass
along whatever function the proteinoid was able to perform. These structures
can form what are called endomicroparticles that can then be passed onto
other microspheres, thus giving them new functions they never had before, or
they can be released to serve as nuclei for the growth of new microspheres
(Fox and Dose pg. 218, 220). Crystals cannot write conceptual information,
but microspheres can. Crystals store information only in the sense that
their lattice structure contains the information needed to extend the lattice
further, but microspheres can actually create replicates they can store away
until needed. Crystals can pass along the information in their lattice
structure only if they break up into smaller pieces and dispurse, but
microspheres can actually conjugate and form junctions between themselves
that can then allow stored information to be passed on. Once again, the
behavior of microspheres more closely mimics the behavior of cells than the
behavior of crystals.
> Are the cellulose membranes or soap bubbles or crystals alive according to
> the above definition?
No, because you neglected to tell everyone that that the people who proposed
these characteristics also stated that they were defining life as "any system
which can independently do **all four** of the following...[emphasis added]."
In other words, for a candidate to be alive, it must have all four of those
characteristsics; not one, not two, but all four. Cellophane membranes have
the first two, soap bubbles have the second and crystals have the last two,
but neither have all four, so they are not alive. Proteinoid microsphere
protocells, on the other hand, have all four, so they are alive.
Again, I can only conclude that Art's knowledge of basic biology and
protocells is much less than he believes it to be, or would like other people
to believe it to be. Which makes some sense, because he is a geologist, not
a biologist. But what I fail to understand is why he chooses to make such a
fool of himself when he could easily read the scientific literature on
protocells and get the real picture.
Kevin L. O'Brien