вход по аккаунту


Chemical Storage and Processing of Information in Biological Systems.

код для вставкиСкачать
(2) can be isolated on a preparative scale, the sulfur
generally occurs in subsequent products.
in 70 % yield at only 70 ‘C; since (2e) is stabilized by resonance, it only starts to decompose to benzonitrile at 190 ‘C.
After 30 min at 100 OC, benzenesulfenyl chloride ( I n ) affords
chlorobenzene (2n) in 60 yield, together with diphenyl disulfide and SzC12. The p-methyl and p-methoxy groups o n
the benzene ring of (Ib) and ( I c ) , respectively, accelerate
the decomposition which is complete after only 10 min at
90 “C; nitro groups inhibit cleavage.
Aliphatic sulfenyl chlorides ( I d ) , especially those which
branch at the cc-carbon atom, also decompose at about
100°C into alkyl chlorides, although here the yields are
lower than in the aromatic series.
The requisite thermolysis temperature depends on the
stability of the sulfenyl derivative in question. Thus, benzenesulfenyl thiocyanate ( l e ) forms phenyl isothiocyanate (2e)
Thermolytic cleavage of monosulfides into disulfides is
already known [2] but proceeds at higher temperatures; it
deserves a similar interpretation. For example, diphenyl
disulfide (If) gives an almost quantitative yield of diphenyl
sulfide (2f) and sulfur within about an hour at 200 “C.
Received, September 14th, 1964 [Z 824/644 IE]
German version: Angew. Chem. 76, 861 (1964)
[I] Cf. H . Lecher and F. Holschneider, Ber. dtsch. chem. Ges. 57,
755 (1924); H . Lecher, F. Holschneider, K.Koberle, W. Speer, and
P . Stocklin, ibid. 58,409 (1925); H . Rheinboldt and F. Motr, ibid.
72, 668 (1939); E. Schneider, Chem. Ber. 84, 913 (1951).
[2] Cf. C. Craebe, Liebigs Ann. Chem. 174, 189 (1874).
Chemical Storage and Processing of Information in Biological Systems
The Gesellschaft fur Physikalische Biologie and the Neuroscience Research Foundation organized a conference o n
May 5th and 6th in Berlepsch Castle near Gottingen (Germany). The subject of the conference was “Chemical Means
of Storage and Readout of Information in Biological Systems’’. Some 90 participants, about one-third from abroad,
discussed the four topics “Molecular Interactions”, “Antigen-Antibody Reactions”, “RNA Replication” and “Possible Relations to Psychic Memory”. The very lively discussions were interrupted by a n evening of chamber music,
with works by Baroque composers, improvised by the participants themselves.
Molecular Interactions
M . Eigen, Gottingen (Germany), opened the conference with
an introduction to the problem of molecular interactions. Although our knowledge of the interactions occurring among
simple inorganic ions is relatively good, the interactions of
large organic ions and molecules are as yet largely unexplored.
However, it is these very interactions that are important in
biological systems.
Ionic interactions are normally non-specific, i. e. a free ion can
be situated with almost equal probability at a point x or
x + Ax. Although hydrogen bridges are spatially fixed, they
are structurally non-specific, and hydrophobic interactions,
as such, are also without appreciable specificity. In biological
systems, however, interplay of these three types of interactions
leads to considerable specificity, owing to the spatial fixation
of the mutally interacting partners. Examples of this are seen
in the tertiary structure of pro?eins, in antigen-antibody reactions, and in enzymatic processes.
Hydrophobic interactions, e. g. between uncharged sidechains of a protein or between the various heterocyclic bases
of nucleic acids, are of importance to the molecular structure
primarily in polar solvents, and hence particularly in aqueous
solution. They cause the disappearance of the structure of the
solvent between hydrophobic surfaces. The resulting gain in
free energy, together with other effects, particularly x- x interactions, leads to stabilization.
In many biologically important molecular structures, the
interactions described are cooperative, i. e. one interaction
facilitates the occurrence of another of the same type. Consequently, after a critical value has been reached, a relatively
small change in the external conditions can lead to a radical
change in structure. Thus, the p H or temperature range within
which a nucleic-acid double helix changes into two coiled
strands is much narrower than would be expected if the known
interactions were simply additive in their effects. Systems with
cooperative interactions can therefore respond t o external influences by all-or-none reactions.
Owing to their short half-lives, all molecular interactions,
whether ionic, hydrophilic, or hydrophobic in nature, are
suitable onIy for t r a n s i e n t assimilation of information. Perm a n e n t storage of information, i . 2 . for about the duration
of a human life, probably requires reactions with an activation energy of at least 25 kcal/mole, i. e. reactions which lead
to the formation of covalent bonds.
Incidentally, this lecture was based o n a manuscript which
had been sent in advance to the conference participants [*I
and in which an attempt was made to redefine a number of
concepts borrowed from psychology, but nevertheless applicable to molecular processes. These concepts range from
simple “interaction” on the basis of a force law, through ,,distinction” and “recognition” to “memory”, and include functions such as “storage”, “transfer”, and “readout” of information, and go on to “adaptation” and “learning”. We are
quite familiar with molecular processes of distinction and of
storage, transfer, and recognition of information in enzymology and molecular genetics. Learning and memory functions,
o n the other hand, pass beyond a simple adaptation and are
known only on the cellular level.
In the discussion, which was led by L. Onsager, New Haven,
Conn. (U.S.A.), J. Kendrew, Cambridge (England), reported
o n the structure of myoglobin. Only 70 % of the molecule has
ahelical structure. Between the regions which are arranged in
this manner, there are seven relatively short sections which
are non-helical. The longest of these sections consists of
eight amino acids. The amino-acid sequence (the myoglobin
molecule contains a total of 153 amino-acid units) gives no
answer, in the light of present-day knowledge, to the question
why a helical region begins and why it ends. The four proline
units in the molecule, which are sterically incompatible with
an a-helix, evidently play a part, and so d o the six serine and
five threonine residues, whose side-chains interfere with the
hydrogen-bonded structure of the =-helix. The end of a
helical region is almost certainly also influenced by hydrogen
bridges between the free carboxyl groups of aspartic acid units
and amide NH groups in the main chain, since such interactions are possible only where a helical region ends.
[*I To be published.
Angew. Chem. internal. Edit. / Vvl. 3 (1964)
/ No. 10
Myoglobin has to a large extent the same tertiary structure as
the a- and p-chains of hemoglobin. The homologies in the
amino-acid sequences, i.e. in the primary structure of the
three protein chains, are astonishingly slight in comparison with the similarities in tertiary structure: only 23 out
of the 153 amino acids present, i . e . only about 15 %, are
homologous. However, if we compare the amino acids whose
side-chains are located in the interior of the spherical molecules, the number of homologies increases to about 33 %. It
appears, therefore, that hydrophobic interactions are of major
importance for the tertiary structure of myoglobin and hemoglobin, and hence also, indirectly, for the function of these
proteins. This would agree with the fact that abnormal hemoglobins which are formed by mutations are almost always
modified in the amino acids whose side-chains point outwards, as long as the mutations are not lethal.
0. Sinanoglu, New Haven, Conn. (U.S.A.), pointed out that
deoxyribonucleic acid (DNA) can form a double helix in
water, but is denaturated in other similar solvents (formamide or alcohols). In the search for a parallelism between
this phenomenon and other physical properties of the solvent,
a relationship was found to exist with the interfacial energy:
the higher the interfacial energy, the smaller the denaturing
influence of the solvent on the D N A structure.
D . Crothers, Gottingen (Germany), found that the relaxation
period for the helix-coil transition in the D N A of TZ phages
depends o n the square of the molecular weight. This dependence disappears almost completely above a molecular weight
of about 2x 107. To explain this result, it was assumed that
there are at least six single-strand bridges per TZ-DNA molecule of molecular weight >2xlO7, i.e. that the strands of
the D N A double helix are broken at several points, and that
the molecule is held together at these points only by the
strand which is not broken. These measurements also
permitted the calculation of a coefficient of friction for
the unwinding of the double helix into two separate strands
by rotary diffusion. This coefficient is about 103 times
greater than would be expected in water, probably owing to
the greater microscopic viscosity of a n ordered water structure
in the immediate vicinity of the D N A molecule. The fact that
the viscosity of the aqueous solution can be increased one hundredfold by addition of glycerol, without appreciably changing
the coefficient of friction, lends support to this hypothesis.
J. Lelzninger, Baltimore, Md. (U.S.A.), reported on higher
forms of organization at the molecular level. Every enzyme
has a catalytically active site a t which transformation of the
substrate into the product takes place. However, there are evidently also other centers which are independent of the catalytically active site and which interact with substances referred
to as allosteric effectors. These alter the tertiary structure
(and hence the activity) of the enzyme. This gives rise to a new
possibility for the control and regulation of the activity of enzyme chains. The product of the enzyme last in a chain may
be an allosteric effector which modifies a preceding enzyme of
the chain in such a way that its activity is decreased. In contrast to enzyme inhibition by feedback, where the inhibitor
must be similar to the substrate of the inhibited enzyme (since
it interacts with the catalytically active site), the effector in
allosteric inhibition does not need to resemble any substrate
whatsoever of the enzyme chain (since its point of interaction
is different from that of the substrates).
The enzymes of a n enzyme chain probably always form complexes during the catalysis of successive steps, even when they
are found to be soluble o n isolation. The complex formation
is probably due t o the presence in the enzymes of bonding
sites which may have the character of allosteric centers and
which appear to be specific for one another. The enzyme complexes which have been most thoroughly studied to date, and
which can be isolated as such, are the fatty-acid synthetase
and the a-keto-acid dehydrogenase systems. In both systems,
the substrate is bonded covalently to the enzyme complex
while it passes through the reaction chain.
Membrane-bound enzyme systems exhibit another feature,
namely directionality. For example, when a n ATP-ase reaction (ATP + H20 + A D P + POj-) occurs in a membrane,
Angew. Chem. internat. Edit. f Yol. 3 (1964) / N o . 10
it can give rise to a p H gradient across the membrane when
the ions of water participating in the reaction come from different sides of the membrane. The same is true of the respiratory chain (XH2 -+ 1/202 + X + HzO), if the ions of the water
formed are given up o n different sides of the membrane.
Membranes themselves are forms of higher organization a t
the molecular level. They consist, among other things, of
double layers of phospholipids separated by layers of protein.
Membranes constitute about 60 % of the total cell material
and reach large surface areas (for example, the total surface
of the plasma membrane of a liver cell is about 8000 p2and
the mitochondria1 membranes of the same type of cell have a
total surface area of 29000 p2). Bearing in mind that the phospholipids of a membrane, like proteins, have many sidechains of different types, it becomes clear that the membranes
of a cell offer enormous possibilities for coding, i.e. for the
storage of information. It would thus be feasible, for example,
that proteins could be bonded to a membrane only according
t o the underlying phospholipid pattern.
Antigen-Antibody Reactions
G. Edelman, New York (U.S.A.), gave an excellent introduction to the present-day knowledge of the structure of antibodies. Antigen-antibody reactions were discussed in this symposium because they are examples of “memory” at the cellular level.
If an antigen is injected into a mammal, antibodies can be detected in the serum from about the fifth day onwards and sediment out in the ultracentrifuge with a sedimentation constant
of 19 S. Their production reaches a maximum after a further
four days, and then decreases rapidly, provided the antigen
dose does not exceed a threshold value. On injection of a quantity of antigen above this threshold value, the conditions are
initially the same: 19.9 antibodies can be detected after about
five days. Their production then declines, and 7s antibodies
appear, whose quantity reaches a maximum which is not exceeded as long as n o further antibody is injected.
The dimensions of a 7s antibody are about 240x 57x 19 A3
(determined from small-angle X-ray scattering). 7s Antibodies consist of four proteins which are similar in pairs
and which are referred to as the L and H components.
The molecular weight of the L component is 20 000-24 000,
that of the H component 55000-60000. Two H components are linked together by disulfide bridges, and one L component is bonded to each H component. H and L components
can be separated and recombined, and it is possible in this way
to obtain H/L hybrids from different antibodies of the same
type of animal, and recently even of different types of animals.
It has not yet been settled whether the information required
for the formation of each specific antibody is already present
in the cell and the antigen simply triggers off the corresponding
reaction chain, or whether the cell “learns” the formation of
the antibody from the antigen.
In the discussion, 0. Westphal, Freiburg/Brsg. (Germany),
reported o n attempts to immunize cattle with pure Pneumococcus polysaccharides as antigens. There is an optimum dose,
viz.200-4OOy of polysaccharide per injection. If more antigen
is given, the quantity of antibody no longer increases, but
decreases again on further increase of the antigen dose. The
production of antibodies against various antigens is additive,
but some cows form no antibodies at all against these antigens;
this suggests that competent cells (competent for the recognition of antigens and production of antibodies) must be present. The larger the antigen, the better is the production of the
antibody. It is increased by the use of cross-linked polysaccharide antigens.
It is an interesting fact that y-globulin components which
form precipitates with erythrocytes are found in the hemolymph of the horseshoe crab. This appears to indicate the
presence of a primitive antigen-antibody system. So far antigen-antibody reactions have been detected only in higher
.A. Rubin, Brookline, Mass. (U.S.A.), reported on the forniation of antibodies after surgical transplantations. If 6-mercaptopurine is given to a patient on the same day as or on the
day after a kidney graft,the transplanted organ is not rejected
by the body. Thus, 6-mercaptopurine interferes with the formation of antibodies against the foreign kidney, probably at
the stage of macrophage-RNA.
In experiments which have not yet been completed it is being
checked whether the organs of an immunized animal which
synthetize antibodies contain R N A types characteristic of the
antibodies. To this end, rabbits were injected with human albumin and ferritin. At maximum antibody production the
animals were slaughtered and the ribosomes in their spleens
centrifuged in a density gradient.
Ribonucleic Acid Replication
The introductory lecture was delivered by R . SmeNie, Glasgow (Scotland). RNA-polymerase synthesizes ribonucleic acid
from the four ribonucleoside triphosphates in the presence of
DNA and Mn2+. Native and single-stranded DNA (DNA
from QX-174 phages or DNA preparations denatured by
heating) may serve as matrices. The nucleotide sequence of
the synthetic R N A is complementary to the base sequence of
the DNA; this indicates pairing of complementary bases
during the R N A synthesis. The complementary nature of the
matrix and the product is shown by analysis of immediately
adjoining bases (nearest-neighbor analysis), by the formation
of DNA/RNA hybrids, and by the base ratios in the matrix
and in the product. Both strands of native D N A are used in
vitro as matrices for the formation of RNA, since singlestrandedCDX-174DNA gives an RNA with a c o m p l e m e l i t a r y base composition, whereas double-stranded QX-174
D N A gives a product with the i d e n t i c a I base composition. On the other hand, only o n e of the two strands of
native DNA is evidently copied in vivo. This is indicated by
the following experiment: the two strands of a native phage
DNA could be separated by centrifuging in a density gradient.
Each of these strands gave RNA i n vitro, whereas m-RNA
synthetized by bacteria which had been infected with the
phages forms a hybrid with only one of the two D N A strands.
The replication of virus R N A was the subject of the lecture by
S . Ochoa, New York (U.S.A.). If Escherichia coli is infected
with MS2 phages, which contain single-stranded RNA, the
cells form a n R N A synthetase, which can be isolated. The enzyme contains ribonucleic acid in a form which is partly
stable towards ribonuclease. This is evidently doublestranded RNA. On annealing with other types of RNA, this
RNA, which is referred to as the replicative form, gives a
hybrid only with MS2-RNA. It must therefore be assumed
that the cells infected with t h e MS2 phages first synthesize a
complementary R N A strand whicb forms a double helix with
the original bacteriophage RNA. This replicative form then
serves as a matrix for the synthesis of new phage RNA. Destruction of the replicative R N A bound to the enzyme leads
to irreversible inhibition of the enzyme. The formation of new
bacteriophages in infected cells probably follows the scheme:
phage R N A + cell
polyribosomes + phage protein
H. H y d h , Goteborg (Sweden), gave an introductory report
on experiments with rats which had to learn a new mode of
behavior and whose brains were then analysed for their RNA
content i n the appropriate sections (analysis of the cell nuclei
of individual neurons; sensitivity of detection: ppg). “Righthanded” rats learned to take their food from a tube with the
right fore-foot. The tube was then moved so that the food
could only be reached with the left fore-foot. The part of the
somatosensory cortex which determines whether the animal
is right- or left-handed was then analysed. A distinct increase
was observed in the RNA content in the cell layers of the right
half ofthe brain. The corresponding cells of the left half of the
brain were used as controls. The RNA isolated from the cells
of the right (learning) half of the brain also exhibited a higher
ratio of purine to pyrimidine bases.
In a second experiment, the animals had to crawl up a thin
steel wire 1 m long and inclined at an angle of 45 in order to
reach their food. With a daily learning period of 45 min, the
ratstookonly4--5 days to master the new mode of behavior.
The large nerve cells and the neuroglia of the lateral vestibular nucleus were analysed, using neurons and neuroglia
from other parts of the brain of the same animal and from
other animals as controls. An increase in the R N A content of
the nerve cells and in the adenine: uracil ratio of the nuclear
RNA in the nerve cells and that of the neuroglia R N A was
again found.
The interpretation of these results may be that repressed
sections of the chromosomal DNA in the brain cells become
active during learning and produce a corresponding R N A
which in turn leads to the synthesis of specific proteins. The
presence of these proteins or the rate of their production
might then affect the transmission of nervous impulses. The
neuroglia ma) also participate in this process, possibly by controlling the induction of the RNA synthesis in the neuron.
Finally, F. 0. Schmitt, Cambridge, Mass. (U.S.A.), showed
some possibilities for the storage of information in the central
nervous system. The system consisting of neurons, nerve
fibers, and synapses in the brain can be regarded as a threedimensional network, resembling the tracks of a marshalling
yard. The fate of an impulse then essentially depends on the
part ofthe network through which it is transmitted. This may
be decided by the neurons. One way in which they could carry
out this decision is by sending specific proteins through their
nerve fibers to the synapses; the proteins would then permit
the transmission only of certain impulses. It is certain that
protein is being constantly synthetized in the neurons and
migrates through the nerve fibers to the synapses. However, it
is not yet known whether this protein is used only for the
metabolism of the nerve cells,or as a structural material for the
production of specific routes of communication, or even as a
molecular information store.
[VB 824/151 IE]
German version: Arigew. Chem. 76. 718 (1964)
Translated by Express Translation Service, London
Cycloadditions of Alkenylamines and
G. Opitz, Tubingen (Germany)
R N A synthetase
complementary R N A
replicative R N A + sinxle stranded R N A
phage protein
However, it is not yet known whether the R N A synthetase
which has been isolated forms also the complementary RNA,
or whether another enzyme is responsible for this reaction.
Possible Relations to Psychic Memory
Cycloaddition of alkenylamines onto ketenes, isocyanates, or
sulfenes affords P-dialkylaminocyclobutanones( I ) , P-dialkylamino-P-lactams (Z), and P-dialkylaminotrimethylene sulfones ( 3 ) , respectively. Whereas the enolizable P-aminocyclobutanones rearrange on heating to alkyl P-aminovinyl ketones, compounds ( I ) , which are not capable of enolization,
are thermally stable; so are p-aminotrimethylene sulfones ( 3 ) .
At higher temperatures, 9-amino-P-lactams (2), which are
also formed from amidines and ketenes, are in equilibrium
with the corresponding alkyleneamines and isocyanates.
Angew. Chem. internal. Ediz. ] Vol. 3 (1961) No. I 0
Без категории
Размер файла
427 Кб
informatika, chemical, biological, system, storage, processing
Пожаловаться на содержимое документа