Enzymes and Molecular Machines Can Be Selected from Random Copolymers

[John Kennard]

Most cell structures are formed or synthesized and most other cell
functions are performed by the molecules called proteins.

Proteins are polymers, complex chain-like molecules each synthesized
by the chemical bonding together or polymerizing of many smaller
molecules called their monomers, and furthermore are copolymers,
polymers consisting of monomers of more than one kind.

The monomers of the proteins are called amino acids, of which twenty
different kinds are commonly incorporated into natural proteins.

The amino acids are small molecules, composed of only ten to twenty-
seven atoms each, depending on kind, averaging about twenty, and
averaging in mass about 2.35 * 10 ^ -22 gram, about one-quarter of one
zeptogram (sextillionth of a gram), or 235 yoctograms (septillionths
of a gram), about eight times as much as the three-atom water
molecule, an oxygen atom single-bonded to each of two hydrogen atoms
(an oxygen atom forms two single bonds or one double bond in a
molecule, while a hydrogen atom forms one single), the total mass of
which is about 2.99 * 10 ^ -23 gram, or thirty yoctograms.

Every amino acid molecule consists of a central carbon atom single-
bonded to a hydrogen atom, an amino group, a carboxylic acid group and
a prosthetic group (a carbon atom forms four single bonds in a
molecule, or one double bond and two singles, or two double bonds, or
one triple bond and one single). An amino group is composed of a
nitrogen atom (which forms three single bonds, or one double and one
single, or one triple) single-bonded to each of two hydrogen atoms.
And a carboxylic acid group is composed of a carbonyl group or moiety,
a carbon double-bonded to an oxygen atom, further single-bonded to a
hydroxy or alcohol group, an oxygen atom single-bonded to a hydrogen
atom. Such bonds and groups are of course the ordinary molecular bonds
and functional groups of carbon chemistry, called "organic chemistry"
due to life's taking such advantage of the ability of carbon to form
large and stable molecules that most carbon on the face of the earth
is or has been part of a living organism. And the amino acid amino and
carboxylic acid groups are of course what give it its name.

Every amino acid is identical to every other in the above, regardless
of kind, in a nine-atom moiety called here its invariant moiety,
composed of its central carbon atom and the hydrogen atom and amino
and carboxylic acid groups bonded to it. And an amino acid of one kind
differs from one of another solely in the elemental composition (the
kinds and number of atoms involved), structure (how those atoms are
bonded together) and consequent properties of its prosthetic group.

The invariant moieties of the amino acids incorporated into a protein,
minus water of polymerization (the equivalent of one molecule of water
is lost for each amino acid incorporated into a protein, the hydroxyl
group from the carboxylic acid group and a hydrogen from the amino
group), comprise what is called its backbone, an elongated and
repetitive structure in which each invariant moiety remnant
incorporated forms a unit identical to every other (although of course
the end-units each bear a free bonding group unused in
polymerization). And the prosthetic groups of those amino acids become
protein side-groups projecting from those protein backbone units and
that backbone.

Proteins of different kinds and functions vary widely in the number of
amino acids of which they are composed. But three hundred amino acids
is a modal protein length and size. And a modal protein mass can be
calculated based on that length and size by multiplying the average
mass of an amino acid by three hundred, and then subtracting the
combined masses of the two hundred and ninety-nine water molecule
equivalents lost, amounting to about 6.16 * 10 ^ -20 gram, about sixty
zeptograms, a little over two thousand times the mass of a water
molecule. And the corresponding modal protein gram number, the number
of such proteins contained in one gram thereof, can be calculated by
dividing that mass into one gram, giving about 1.62 * 10 ^ 19 or about
sixteen quintillion such proteins per gram.

Proteins can rotate around their backbone single bonds, and therefore
all along their backbones, and consequently twist and coil, but the
proteins which perform the functions of the cell generally assume
three-dimensional coiled structures or conformations specific to their
kinds, stabilized in various ways. For example, attractions between
positively- and negatively-charged moieties of molecules form what are
called hydrogen bonds, which separately are weaker but collectively
can be much stronger than any one molecular bond. Such bonds between
nearby backbone units along the protein backbone cause the assumption
of winding or helical conformations along the backbone, as well as
sheet conformations involving multiple turns and side-by-side runs
thereof. Such interactions and resulting conformations are not
specific to any particular protein or moiety thereof, since every
backbone unit is identical to and can form the same such bonds as any
other, and any stretch of backbone could theoretically engage in any
such conformation. Slightly more specifically, the backbone units and
some side-groups hydrogen-bond water molecules, and proteins tend to
coil in such a way as to present such water-soluble or hydrophilic
moieties or groups on their surfaces, and hold those side-groups which
cannot form such bonds inside, in water-insoluble or hydrophobic
cores. But in most of the proteins which perform the functions of the
cell, conformation is specified by side-group interactions, which
depend on the kinds of side-groups available and the order in which
they occur, which depend in turn on which amino acids are incorporated
into the protein and the order in which they are incorporated.

That is, protein amino acid order determines protein conformation.

Protein conformation determines protein shape, whether globular,
elongated or flattened, and whether solid, indented or hollow; protein
mechanical properties, such as whether and how one part of a protein
can bend or rotate with respect to the rest; and protein surface
structure, the protein=92s surface shape and pattern of exposed side-
groups and backbone units.

And protein shape, mechanical properties and surface structure
determine protein function:

Two proteins or other large molecules of complementary shape and
surface charge-patterns upon being brought together will develop
multiple attractions including hydrogen-bonds to one another, such fit
and collective attraction being called an affinity and such collective
bond a complex. Protein complexing is so generally specific in its
requirements of complementary shape and charge-pattern as to be
described as =93lock-and-key=94. And protein complexing is a fundamental
mechanism of protein function: The conformation-determining
attractions and bonds of and within the protein itself can be
considered intramolecular or internal complexing. Cytostructural
proteins complex with one another to form the internal structural
framework of the cell called the cytoskeleton. And every cell contains
protein enzymes catalyzing-- accelerating-- the chemical reactions
used by that cell, which reactions would otherwise run too slowly to
be of use: Body cells typically each synthesize thousands of different
enzymes, and many molecules of each, each more or less specifically
catalyzing its specific reaction operating upon its specific
substrate(s) or reactant(s) (the phrase "lock-and-key" was first
applied to enzyme specificity). And each enzyme catalyzes its reaction
largely through, and its specificity is that of, not so much its
complexing with its substrate(s) as with its reaction's rate-
determining transition state, the highest-energy state through which
that reaction must proceed, stabilizing and therefore lowering the
energy of that state, allowing lower-energy passage through that
state, increasing the probability that a given enzyme-substrate
complex will have the energy needed to pass through that state, and
therefore, in the cell or other reaction mixture where many such
complexes are forming and dissociating, increasing the number of such
able to pass through that state and their reactions proceed to
completion at any given time, and therefore the overall rate of
reaction. In addition, many enzymes catalyze water-sensitive reactions
in their hydrophobic cores. More complicatedly, many if not most
proteins function by virtue of conformation changes, changing back and
forth between two or more conformations in the course and by way of
function, a phenomenon called allostery, and complexing is frequently
combined in protein function with allosteric conformation changes.
Protein complexing of one molecule causing an allosteric conformation
change in that protein enabling or preventing subsequent complexing of
another molecule is a central mechanism of protein function and
control in the cell; for example, some enzymes, including some acting
as cell switches, sensors or governors, are activated or deactivated--
turned on or off-- by conformation changes caused by complexing with
or dissociating from the appropriate molecules, some used specifically
as signals. And other protein enzymes catalyze the degradation of fuel
and use the energy yielded to repetitively alter their conformations
and shapes, acting as motors and machines.

Proteins can become damaged in use and need replaced, and indeed are
routinely depolymerized and their amino acids reused. Furthermore,
proteins need to be passed down from one generation to the next. There
is therefore a constant need for new copies of proteins. But proteins
are not directly replicated or copied in nature. Instead, the
copolymers called nucleic acids, synthesized from monomers called
nucleotides, are used as templates to synthesize proteins as well as
new copies of themselves. There are two (sub-) classes of nucleic
acids, the deoxyribonucleic acids or DNAs or just DNA, and the
ribonucleic acids or RNAs. Nucleic acids of each class are synthesized
from nucleotides of a corresponding class, the deoxyribotides and the
ribotides, respectively, of which there are only four kinds each.
Nucleotides are larger and more complicated than amino acids, but
every nucleotide molecule is identical to every other of its class in
part, regardless of kind, in an invariant moiety bearing two bonding
groups of different kinds, and a nucleotide of one kind differs from
one of another of its class solely in the composition, structure and
properties of a base or base moiety bonded to its invariant moiety,
all analogous to amino acids. Nucleic acid strands have repetitively-
structured backbones composed of the invariant moieties of the
nucleotides incorporated into them minus water of polymerization, and
side groups projecting from those backbones composed of those
nucleotides=92 base moieties, and free end bonding groups unused in
polymerization, all analogous to proteins. And just as with proteins,
nucleic acid side-group hydrogen-bonding and complexing play central
roles in the form and function of nucleic acids in the cell, both
perfectly analogously to the proteins, and also in a very specific way
used in protein synthesis and nucleic acid replication: The
nucleotides of either class consist of two pairs of kinds in each pair
of which the base moiety become a side-group when a nucleotide of the
one kind is incorporated into a nucleic acid will hydrogen-bond to the
other when close enough and at the proper angle. In a nucleic acid
which is multiply and sequentially complexed with itself or another
nucleic acid the hydrogen-bonded bases or side-groups and the bonds
between them are referred to collectively as base-pairs. And two
nucleic acids of either class, or one of the one class and one of the
other, with complementary nucleotide orders but antiparallel
(opposite) directions, can hydrogen-bond base-pair-by-base-pair all
along both strands, forming a fully-complexed double strand. DNA
generally exists as such double strand, or as a set of such, each of
which when wrapped around and compacted with and by proteins for
transfer and becoming visible under a microscope during cell
reproduction is called a chromosome, and is the very stuff of heredity
or biological inheritance, passed from parents to and conferring their
biological traits upon their offspring. It is and does so by serving
as a permanent record of the amino acid orders of all the proteins
synthesized by every cell of the body, every body cell containing the
identical DNA(s) (but see the description of B cells below), encoding
those orders by means of the genetic code, in which three-
deoxyribotide sequences called codons represent amino acids, and
sequences of codons called genes represent proteins.

In cell reproduction, the cell divides into two daughter cells, and
two DNA double strands or sets thereof, one for each daughter cell,
must be synthesized for each such possessed by the parent cell.
Simplifying greatly, such DNA synthesis or copying or replication is
accomplished by separation of its strands, hydrogen-bonding the
complementary deoxyribotides to the newly-exposed ones, and
polymerizing the new deoxyribotides into complementary new strands
complexed with the old, in what is called, due to each new double
strand produced being composed of one old and one new strand, semi-
conservative replication.

In protein synthesis, and again simplifying greatly, the DNA double
strand is separated in the region of the gene or protein-encoding
sequence being used or expressed and the appropriate strand used as a
template for the bonding of the complementary ribotides in the
transcribing of that gene into in the synthesis of a complementarily-
encoded RNA primary transcript, which after post-transcriptional
processing including the removal of segments and splicing together of
the remainder then serves as a messenger RNA template for protein
synthesis, the codons of that RNA complexing with complementary
anticodons of transfer RNAs, each of which have been previously bonded
to the corresponding amino acids, and which amino acids are then
transferred from those RNAs to the growing protein, in the (usually
repeated) synthesizing of or translating of that RNA into (usually
many copies of) the encoded protein, the process controlled and
catalyzed by the protein-and-RNA complex called the ribosome.

Every living thing of every species on our planet uses nucleic acids
and the genetic code to synthesize its proteins in transcription and
translation, and to inherit and pass on its proteins' amino acid
orders and therefore its biological traits in heredity, proving beyond
a reasonable doubt that all living things of every species on our
planet are related. Since all living things and species are related,
we must share a common ancestry. And such common ancestry and later
divergence of species is called evolution. Evolution proceeds through
mutation or change of trait caused by change of protein caused by
change of gene, which last is the only kind of change which can be
inherited, those changes which are fatal or otherwise disadvantageous
being correspondingly unlikely to be inherited, while those which are
advantageous preferentially so, in both negative and positive natural
selection.

Evolution is therefore an organism-, cell-, trait- and nucleic-acid-
based development of proteins by variation of amino acid order
followed by selection.

The leukocytes or white blood cells (called so because they=92re not
red) called B cells, which synthesize the proteins called antibodies,
comprise one exception to the rule that every body cell contains the
identical DNA: Antibodies play a central role in the body's immunity
to infection, by their complexings with or recognitions of antigen,
mostly proteins upon the surfaces of invading microbes. Free
antibodies secreted by B cells into the lymph and blood complex with
foreign cells and trigger a complex, sequential and cascading
biochemical attack by a group of blood proteins called the complement
system, ending in among other things the construction of fatal protein
pore complexes-- holes-- in the outer membranes of those cells.
Antibodies complex with antigen at a relatively small antigen binding
site on the antibody surface comprised of several adjacent runs of
protein. During B cell development in early childhood the antigen
binding site codon sequences in the antibody gene of each immature B
cell are themselves produced by random and imprecise genetic
recombinations of short codon sequences from a preexisting pool of
such present in DNA. Cells with encoding errors from such
recombination are either re-recombined or destroyed, while those which
continue to mature encode and synthesize antibodies with antigen
binding sites which vary from cell to cell. Such cells undergo
further negative selection re-recombining or destroying those which
recognize the body's own "self" proteins and cells. And the surviving
cells reproduce to form a population of B cells consisting of millions
of different B cell sub-populations or clones in the body each clone
synthesizing its own specific antibody and all therefore collectively
synthesizing millions of different antibodies. Each B cell
synthesizes and exposes on its surface something on the order of a
hundred thousand copies of the antibody specific to its clone. And
such cells upon recognition of antigen are variously stimulated to
reproduce, to secretion of free antibodies into lymph and blood, and
to other immune activities, including further recombinations,
selections and reproductions fine-tuning recognition of antigen.

Antibody development and function are therefore cell- and nucleic-acid-
based developments of proteins by variation of amino acid order
followed by selection.

Such functional copolymers as the proteins are called "mechanomers"
here (distinguishing naturally-occurring mechanomers as "biopolymers"
where convenient); the science of mechanomers is called
"mechanomerics"; and applied mechanomerics is called "mechanomeric
engineering", including "mechanomeric medicine".

Mechanomeric engineering requires some technique for developing
mechanomers to perform desired functions, such as enzymes and
molecular machines.

And just as proteins are developed in nature by variation of amino
acid order-- and therefore of protein conformation (coiled structure),
shape, mechanical properties and surface structure, and therefore of
function-- followed by selection, so too artificial mechanomers can be
developed by selecting those performing desired functions from among
random mechanomers, mechanomers with random monomer orders,
conformations, shapes, mechanical properties, surface structures and
functions, in what is called here "mechanomeric selection".

Evolution's scale, in numbers of mechanomers/biopolymers and
selections and across time, keeps it from being proof that
mechanomeric selection is a practicable technology.

However, the development of each individual's antibody complement in
childhood, and still more such development in the even smaller and
faster system of the duck embryo in the egg, and the function of those
antibodies, and most of all the medical exploitation of such
development and function in vaccination and the medical and industrial
development and use of antisera and monoclonal antibodies, all furnish
so many everyday small-scale proofs that proteins performing such more
or less simple complexings or molecular recognitions as antibodies can
be developed by random synthesis followed by selection.

Enzyme function furnishing so many more examples of such complexings,
enzymes can therefore be developed likewise.

Proteins performing any equally simple functions or small combinations
thereof likewise.

Biopolymers of the other important class of such, the nucleic acids,
likewise, within the limitations imposed by their smaller numbers of
kind of monomers (and smaller variations in structure between those).

And mechanomers of other classes likewise.

Nucleic acids should in fact not be mechanomerically selected, to
prevent unwanted genomic introductions. Mechanomer classes mechanomers
of which are to be selected should not use naturally-existing
monomers, to prevent naturalizations of replicative systems (see
below). And mechanomer classes mechanomers of which are to be selected
should be non-toxic and biodegradable.

Random polymerization of a mixture of monomers of the different kinds
of the appropriate class will produce a mixture of different random
mechanomers of the desired class, a random mechanomer stock, and
replication or molecular copying of the mechanomers in such stock a
mixture of many replicands of each of those mechanomers, a mixture of
many mechanomeric clones, a replicated random mechanomer stock. Such
stocks will be the fundamental tools of mechanomeric selection. And
such polymerization and replication will be catalyzed by enzymes, at
least one polymerase and replicase respectively, both of another class
of mechanomer than that of those being synthesized to avoid unwanted
operations upon those enzymes themselves, both operating in the same
direction along and continuously upon the growing mechanomer during
its synthesis to insure that the replicands have the same
conformations as the original, and both themselves mechanomerically
selected in early mechanomeric selection (see below).

Mass properties of mechanomers significant to mechanomeric selection
include (1) incidence of well-conformedness and (2) coincidence and
incidence of function among random mechanomers or in random mechanomer
stocks; (3) gestation (-time) of mechanomeric function (time for
detectable effect to accumulate, taken here to be that of an enzyme);
and (4) incidence of degradative enzymes among random mechanomers or
in random mechanomer stocks:

(1) Incidence of well-conformedness: Proteins generally each assume a
single stable conformation, or change between two or three
conformations by way and in course of function, but perhaps fewer than
one in one billion random amino acid orders will specify such well-
conformed proteins, and such incidence is taken here to be that of
well-conformed mechanomers in random mechanomer stocks. Many
mechanomeric functions might be performed by mechanomers which do not
assume such conformations (if only because conformed by complexing by
way and in course of function), and such incidence will be adequate
for the selection of mechanomers performing the simplest functions
anyway (see below), but selection of mechanomers performing more
complex functions will require use of some technique(s) for increasing
such incidence. Three such techniques, in order of increasing
complexity and decrease in synthesis of poorly-conformed mechanomers,
are what are called here "diagonalization", "fuzzy replication" and
"splicing":

Diagonalization: Chromatographing random mechanomers along one side of
a square medium or matrix and then at a right angle to the original
direction until the spectrum lies largely along and is enriched in
well-conformed mechanomers along the diagonal, well-conformed
mechanomers being more sharply localized in chromatography, such
mechanomers extracted and the procedure repeated, using different
media.

Fuzzy replication: Using an inaccurate or fuzzy replicase-- see
below-- to replicate an original well-conformed mechanomer, perhaps
with a function similar or even identical in part to that desired, and
synthesize a random mechanomer stock, analogous to evolution.

Splicing: Of random or fuzzily-replicated segments into the
appropriate areas of otherwise well-conformed mechanomers, analogous
to antibody antigen binding site development, followed by replication
to insure that those mechanomers assume the conformations they would
have assumed upon continuous polymerization and will upon replication
for production.

(2) Coincidence and incidence of function: Proteins vary widely in the
numbers and orders of the amino acids of which they are composed, but
three hundred amino acids is a typical natural protein length and
size, and if all proteins with all possible amino acid orders of that
length were synthesized, the total mass of protein synthesized would
be several hundred powers of ten times the mass of our galaxy.
Plainly, if each and every protein function could be performed by only
one specific protein with one specific amino acid order, no biological
process or artificial procedure could ever develop such. But the
evolution of proteins and other mechanomers, and the development and
function of antibodies in the body, and vaccination and the
development and use of antisera and monoclonal antibodies, all prove
not only that mechanomers with different monomer orders can share a
given function but that there must be a fantastically high degree of
coincidence of function among them. Hundreds out of the millions of
different antibodies in the body typically complex with a given
antigen, which incidence of one in ten thousand is taken here to be
that of such simplest function among well-conformed random mechanomers
(taking the restriction of antibody complexing to its antigen binding
site alone to cancel out multiple antibody complexing of different
parts of antigen). And the greater the number of functions performed
by a mechanomer, and the greater their complexities, the lower will be
such incidence of such mechanomer, the incidence with two sites
performing such functions taken here to be about one in ten thousand
squared or one in one hundred million, and the incidence with three
one in ten thousand cubed or one in one trillion.

(3) Gestation time: Ten thousand random proteins three hundred amino
acids in length will collectively mass a little over six hundred
attograms, which stock if replicated to one gram will average about
1.6 quadrillion replicands and one hundred micrograms of each protein,
which replicands if of an enzyme each molecule of which produces ten
product molecules per second each with the mass of an amino acid will
take about four and a half minutes to produce one milligram of such,
while one trillion such proteins will mass a little over sixty
nanograms, which replicated to one gram will average about sixteen
million replicands and a picogram of each, which as such enzyme will
take about ten months to produce one microgram of product.

(4) Incidence of degradative enzymes: Depolymerases depolymerizing and
other enzymes degrading mechanomers of their own class will occur in
every random mechanomer stock and make it unstable. Such reactions and
enzymes for the most part will be simple ones, the collective
incidence of such enzymes in such stocks will be correspondingly high,
such stocks will be correspondingly unstable, and such problems will
be exacerbated by replication. Random mechanomer stocks should
therefore be freshly prepared for mechanomeric selection. If such
stock must be stored it should be kept cold, decreasing reaction rates
in general, and dry, if depolymerization incorporates solvent into the
free monomers, as with proteins, amino acids and water. Such stock
might also be matriciated (see below), separating most mechanomers in
the stock and causing degradative enzymes to preferentially degrade
their own replicands, and even bound after matriciation to some
matrix, fixing the mechanomers in place and completely halting such
degradation.

What is called here "matricial mechanomeric selection", analogous to
antibiotic sensitivity testing, will be the simplest and most common
form of such selection, at its own simplest matriciating (spreading
and arraying) a sample of a replicated random mechanomer stock across
or through or into a thin layer; overlaying that replicated random
mechanomer matrix with any materials and subjecting it to any other
conditions needed for the desired function; analyzing that matrix
identifying locations in which the desired function is being
performed; extracting the mechanomers from those locations for further
replication and testing, perhaps by another round of such selection
(using a different matriciation to redistribute the mechanomers in the
sample-- see below); and replicating the mechanomer finally selected
for its performance of the desired function for production.

Matriciation must be ordered-- for example by affine chromatography,
chromatographing a sample of a replicated random mechanomer stock
using one chromatographic medium and blotting the resulting linear
chromatogram into one side of a different medium and chromatographing
that a right angle to the first, forming a square or two-dimensional
matrix, perhaps itself blotted into a final test matrix and medium--
to localize the replicands and effects of each different mechanomer in
its characteristic location on the matrix, maximizing concentration of
effect and minimizing gestation (time for effect to accumulate to
detectability) and analytical sensitivity needed; to perform parallel
testing of mechanomers under different or incompatible conditions,
using identical matriciations of multiple samples of a replicated
random mechanomer stock and comparing mechanomer behaviors at their
identical locations from matrix to matrix; to perform parallel
recovery of mechanomers from a matrix parallel to a test matrix from
which it would be difficult or impossible to recover the tested
mechanomers; and as noted above to separate mechanomers and therefore
decrease mechanomeric interactions on and in the matrix, causing
enzymes degrading mechanomers of their own class to preferentially
degrade their own replicands.

Matricial analysis will of course use infrared spectroscopy and
nuclear magnetic resonance imaging where appropriate.

It will also use "orthogonal analysis", by "orthogonalization" or
third-dimensional separation of the matrix, for example by blotting
the matrix into one end of and separating its components using a very
wide chromatographic column (and orthogonal standards inoculated into
the margin of the original square matrix marking in the orthogonal
matrix or column planes or bands of interest).

But matricial analysis will above all use what is called here
"mechanomeric indication", overlaying the matrix with a previously-
selected enzyme, called here an "indicase", which under some condition
resulting from the performance of the desired mechanomeric function
catalyzes a reaction converting a substrate, called here an
"indicator", causing a color-change on the matrix. Such indicase will
be readily developed by overlaying a replicated random mechanomer
matrix with indicator, then subjecting that matrix to the desired
condition, and noting where indicator was not converted and no color-
change took place until that condition was applied. Mechanomeric
indication by its analysis at the molecular level, analysis by
complexing, cumulative indication as colored indicator accumulates,
and ability to use the product of one indicase to trigger another to
amplify indication, will render most mechanomeric selection amenable
to being performed as matricial mechanomeric selection. Even though in
such selection of any mechanomer the function of which is more complex
than that of a simple enzyme catalyzing the indicating reaction, false
indications by any given indicase will outnumber true, making
advisable the development and simultaneous use of multiple indicases
in such selection, and selecting only from regions on the matrix where
multiple indication is taking place.

What is called here "mechanomeric evolution", freely mixing
unreplicated random mechanomers and monomers with a replicase
complexed with a previously-selected what is called here "conditional
replicase inhibitor" which inhibits replication except under some
condition resulting from the performance of the desired mechanomeric
function, will test the greatest possible number of random mechanomers
at a time for a desired function and therefore facilitate the
selection of mechanomers performing more complex functions occurring
more infrequently in random mechanomer stocks. Such conditional
replicase inhibitor will be readily developed by first developing an
indicase which converts colored indicator to colorless, then
overlaying two parallel random mechanomer matrices with monomers,
replicase and that indicase, subjecting one such matrix to some
condition resulting from the desired function, and observing where in
the latter indicator color persists. Mechanomeric evolutionary system
sizes will be limited by same-class replicase-pair takeovers (see
below), and in such selection of any mechanomer the function of which
is more complex than that of a mechanomer disinhibiting the replicase,
false evolutions will outnumber true.

The enzymes and other mechanomers needed for mechanomeric selection
will themselves be mechanomerically selected, in what is called here
"early mechanomeric selection":

Replication being a more complex function than and indeed including
polymerization, replicases must be more complex and therefore occur
more rarely in random mechanomer stocks than polymerases, but cross-
class replicase pairs, one from each of two mechanomer classes
replicating mechanomers of the other, will evolve in what is called
here =93mechanomerogenesis=94, in which random mechanomers and monomers of
both classes are mixed, and such replicases upon encountering one
another engage in a more or less exponential course of mutual
replication (with, note, a more or less exponentially-increasing heat
of replication), with the first such event and pair likely taking over
the system. Many such events will produce same-class pairs, which
pairs will also limit mechanomeric evolutionary system sizes (see
above), and many cross-class replicases produced will be incapable of
replicating mechanomers incorporating monomers of all the kinds of the
appropriate class supplied, and more or less inaccurate or fuzzy in
their replication of the mechanomers they can replicate (such reduced-
monomer-set replicases will often be workable until better ones are
developed, and such fuzzy replicases will be useful for increasing the
incidence of well-conformed mechanomers in random mechanomer stocks--
see above).

Once a workable replicase pair is selected, at least one polymerase of
each class polymerizing monomers of the other will be selected by
matricial mechanomeric selection, using random mechanomers synthesized
by purely-chemical (non-enzyme-catalyzed) polymerization and then
replicated. Such polymerase pairs must operate in the same directions
as their classmate replicases, although replicases and polymerases
operating in both directions will be selected to select mechanomers
which in the course of synthesis coil in such ways as to bury the ends
first synthesized and prevent replication, as well as those which vary
in their conformations depending on direction of synthesis.

Finally, sets of mechanomers-- growth hormones stimulating cell
reproduction and cytodifferentiators converting cells of a sample type
to those of others-- will be selected and refined and expanded which
allow construction of cytopalettes, sets of cultures of cells of
different types, for use in matricial empirical mechanomeric
developmental matricial overlays in parallel testing of mechanomers
for toxicity (including environmental safety. Cytopalettes will
include multicytotypic such as neuromuscular junctional cultures.
Cytopalettes cultured from cell samples from individual patients will
allow the custom selection of mechanomeric pharmaceuticals for use in
idiotherapies, individual or customized therapies of refractory
infections and idiopathic diseases, including cancers (see below). And
such cells and tissues will also be used for replenishment and
replacement, and the engineering of organs for (more or less)
autotransplantation.

The utility of mechanomeric selection is highlighted by its
applications to itself above in mechanomerogenesis and other early
mechanomeric selection (including the selection of cytopalette
mechanomers), mechanomeric indication, and mechanomeric evolution.

There are, of course, many other relatively easily foreseeable
applications for such technology, in particular with regard to
medicine and industry.

Mechanomers specifically toxic to microbes of a given species but not
to those of other species or to humans will be mechanomerically
selected. At its simplest, such selection will take the form of
parallel matricial mechanomeric selection: overlaying a set of
parallel identical replicated random mechanomer matrices with,
respectively, a culture of the microbes in question, cultures of
microbes of as many other species as practicable, both human
commensals (to avoid for example disturbing normal human
gastrointestinal microflora and possibly facilitating fulminating
toxic overgrowths) and a selection of environmentally-significant
species, and cytopalette cultures of as many different human cell
types from as many different humans as practicable; observing for
regions where on the first matrix the microbial cells die but on the
others no cells do, deaths signaled by mechanomeric indication for
greatest sensitivity; extracting the random mechanomers from those
regions from yet another parallel matrix; re-matriciating the
mechanomers from each such region using different media to separate
them; and repeating, until a specific and effective antimicrobial is
selected. Multiple agents should be selected to overcome microbial
resistance to any one such and to reduce the amount of each such agent
needed and therefore any side-effects therefrom. And note that such
selection of antimicrobials will enjoy even greater flexibility than
microbes themselves do in their developments of resistances to
antimicrobials, since it will be limited neither to starting from
naturally-existing mechanomers nor to mechanomers of naturally-
existing classes. Furthermore, it will be much faster than the
microbial development of antimicrobial resistance.

Mechanomers preventing viral destruction of cells will be
mechanomerically selected. At its simplest, such selection will take
the form of parallel matricial mechanomeric selection: overlaying a
set of parallel identical replicated random mechanomer matrices with,
respectively, cytopalette cultures of as many different human cell
types from as many different humans as practicable, and cultures of
human microbial commensals (to avoid for example disturbing normal
human gastrointestinal microflora and possibly facilitating
fulminating toxic overgrowths); overlaying the human cell matrices
with a solution of the virus in question; observing for regions where
on the human cell matrices the cells do not die but on the microbial
matrices no cells do, deaths signaled by mechanomeric indication for
greatest sensitivity; extracting the random mechanomers from those
regions from yet another parallel matrix; re-matriciating the
mechanomers from each such region using different media to separate
them; and repeating, until a specific and effective antiviral is
selected. And just as with the mechanomeric selection of
antimicrobials, multiple agents should be selected to overcome viral
resistance to any one such and to reduce the amount of each such agent
needed and therefore any side-effects therefrom, and such selection of
antivirals will enjoy even greater flexibility and speed than viruses
themselves do in their developments of resistances to antivirals.

Cytopalettes cultured from cell samples from individual patients will
allow the custom selection of mechanomeric pharmaceuticals for use in
idiotherapies, individual or customized therapies of refractory
infections and idiopathic diseases, including cancers.

Mechanomers toxic to cancer cells but not to cells of their parent (or
other normal) cell types will be mechanomerically selected, for the
individualized mechanomeric oncotherapies necessitated by the
individualized nature of cancers. Cancer cells are mutated cells, with
at least one and usually multiple DNA abnormalities and consequently
abnormal messenger RNAs transcribed from those DNAs and abnormal
proteins translated from those RNAs, and the differences between such
DNAs, RNAs and proteins and their normal versions will allow the
mechanomeric selection of mechanomers which are toxic to such cells
only (e.g., lethal enzymes activated by complexing with such abnormal
proteins only). At its simplest, such selection will take the form of
parallel matricial mechanomeric selection: overlaying a set of
parallel identical replicated random mechanomer matrices with,
respectively, a culture of the cancer cells in question, perhaps a
culture of the parent cell-type of that cancer, and cytopalette
cultures of as many other (normal) cell types generated and cultured
from the patient as practicable; observing for regions where on the
first matrix the cancer cells die but on the others none do, deaths
signaled by mechanomeric indication for greatest sensitivity;
extracting the random mechanomers from those regions from yet another
parallel matrix; re-matriciating the mechanomers from each such region
using different media to separate them; and repeating, until a set of
individualized antineoplastic or oncotherapeutic mechanomers is
selected. It is unlikely that even with the use of multiple
oncotherapeutic mechanomers that all the cells of a cancer will be
killed, and indeed some if not most cancers more or less steadily and
rapidly mutate, so the mechanomeric oncotherapies of such cancers will
therefore involve not only multiple mechanomers but also one or more
extra passes or rounds of such multiplex therapy, as the cancer
reoccurs to be re-sampled and a new set of oncotheraputic mechanomers
selected to deal with it, with the majority of such cells destroyed by
each pass, and a genetically-different minority left (if any) to begin
again (note that even if not directly destroyed, the cells of such
cancer held in check long enough should become ever more mutated,
function ever more poorly, with ever greater numbers of its cells
being fatally mutated, or exhibiting mutated proteins detectable and
those cells therefore destroyed by the immune system, in what might be
called "neoplastic burnout").

Cytopalette cells and tissues will be used for replenishment and
replacement of tissues lost to injury or illness, including
senescence, as well as the construction of organs for (more or less)
autotransplantation.

As far as industrial applications of mechanomeric selection go, many
expensive and/or toxic and/or otherwise hazardous industrial chemical
catalysts will of course be replaced by mechanomerically selected
enzymes. Such enzymes will catalyze many reactions at lower
temperatures and pressures than at present, lowering energy use and
cost, and many others in water rather than more expensive and/or toxic
and/or otherwise hazardous solvents. Such catalysis due to enzymatic
specificity will furthermore reduce side-reactions and increase the
efficiency and economy of such reactions. And it will also allow new
reactions to be developed and run as well, such as higher-order ones
involving greater numbers of reactants at a time.

Most importantly, mechanomerically selected enzymes will afford
artificial photosynthesis of fuels, organic chemical industrial
feedstocks, and bulk nutrients such as cooking-oils and sugar,
reversing the dumping of crustal carbon into the atmosphere over the
last several centuries.

Thus the first glimpses of the new medicine and photosynthetic economy
that will be afforded to us by mechanomeric selection.


The above, the fourth version of an analysis begun in 1992 as a
private study of senescence and gerontotherapy, was based principally
on two secondary sources:

Hendrickson, James; Cram, Donald; and Hammond, George. Organic
Chemistry. Third Edition. New York: McGraw-Hill Book Company, 1970.
The classic introduction to the classic science.

Alberts, Bruce; Johnson, Alexander; Lewis, Julian; Raff, Martin;
Roberts, Keith; and Walter, Peter. Molecular Biology of the Cell.
Third/Fourth Editions. Garland Science, 1994/2002. A tidal-wave of
information on the core functions of the biopolymers.

Any errors in fact or analysis are of course the present author=92s.


Keywords: antibiotic, antiviral, enzyme, mechanomeric selection,
mechanomers, medicine, MeSe, molecular machine, oncotherapy,
artificial photosynthesis, virotherapy

http://mechanomers.blogspot.com/2012/03/enzymes-and-molecular-machines-can-=
be.html

--

A structure this pretty just had to exist.

< James Watson

[Moderator's note: Originally submitted March 16, 2012]