close

Вход

Забыли?

вход по аккаунту

?

Biopolymers Origin Chemistry and Biology.

код для вставкиСкачать
of the data given in Figures 12-16 underlines the difficulty
of providing simple rules in this area of structure-property
relationships. Appeal has to be made to the anisotropic responses of both crystalline and amorphous phases to stress
and, additionally, to specific mechanical interactions between
the phases. It is in this latter area, for example, that molecular
weight probably exercises a profound influence on mechanical
behavior.
The work reported here, then, represents just a fraction of
the painstaking task of “mapping” properties in the multidimensional reference frame of structure. It is only as we
continue this mapping that a rational and consistent picture
will emerge.
Acknowledgments are due to seoerai co-workers, namely Drs.
P. J . Owen, P. E. Reed, A . Singh, and Ingrid Voigt-Martin.
Received. June IS, 1973 [A 986 IE]
German version. Angew. Chem. 86, 151 (1974)
[ I ] E. H . Andrcws, P. J . Owcw, and A. Singh. Proc. Royal SOC. (London)
A324, 79(l971).
. J . Chem. I’hys.
121 J . D. Ifofli?ian and J . J .
37. 1723 11962).
[3] J . M . Burruhs-Rimdu and J . !M G. Fufoir. Polymer 13, 407 (1972).
[4] P. J . O*.rn. Ph. D. Thesis. University of London 1970.
[5] E. H . Aiidrcws, Proc. Royal Soc. (London) A 177. 562 (1964).
[6] E . H . Aiidrrws and B. Rc
, .I.
Mater.
Sci. 6, 547 I197 I).
[7] J . J . Klcnic,nf and P H . Gril. J. Macromol. Sci. 8 6 . 3 I (1972).
[8] C . K L. Dories and Ony Eny Long. to be published.
[9] A. Krlltv and M. J . Muchin. J. Macromol. SCI.B I , 41 ( 1967).
[lo] G. S !l fih and S L. Lumhcrr. J. Appl Phys 42, 4614 11971)
[ I I] C. L Grunrr. 5. Wirtidrr/r<h,and R. C . Bopp, J. Polym. Sci. A 2, 7,
2099 ( 1969).
[ 121 D. C . Bassrrr and J . M. Phillips, Polymer 12, 730 ( 197 I ).
[I31 E . H . .4ndrews and P. J . Phillips, J. Polym. Sci. B 1 0 . 321 (1972).
[ 141 J . C . Hulpin and J L. Kardos, J. Appl. Phys. 43, 2235 (1972).
[IS] P. J . I’hrllips and J . Patrl. personal communication.
[I61 P. E. R w d , Ph. D. Thesis, University of London 1970.
[I71 W Yau and R. S. Srein, J. Polym Sci. .4 2, 6, I (19681.
[I81 E. H. A ~ i d r w s ,Purc Appl. (’hem 31. 91 (1972).
[ 191 R . Nufurujun and P. E. Rccd, J. Polym. Sci. A 2, 10, 585 ( 1972).
[20] P Bowdcrr in R . N . Hawurd: The Physics of Glassy Plastics. Applied
Science Publishers. Barking, Essex
[2 I] J . Coo!, and .I. E . Cordmi, Proc. Royal SOC.(London) A 282, 508 11964).
Biopolymers : Origin, Chemistry, and Biology
By Melvin Calvin“]
A brief statement concerning the way in which biopolymers may have originated in the nonbiological world is made, including experimental evidence. This also includes a discussion of such
matters as the way in which the code might have originated, that is, the relationship between
polypeptides and polynucleotides as well as the secondary and tertiary structure resulting
from the primary structure determination. The importance of the interaction of biopolymers
with lipids for the formation of limiting membranes, leading to the formation of cells and other
self-organizing cellular type organelles, is discussed. Thus, the second critical physical-chemical problem for cellular organization, namely, the biopolymer-lipid interaction, is now coming
under scrutiny, both in terms of synthetic systems as well as natural ones.
1. Introduction
The collection of ideas presented here was compiled over
a period of several months into the form of a review article
containing four separate but closely interrelated sections. The
first section discusses the origin of biopolymers, particularly
of proteins, including the chemistry which may be involved.
The next section describes the secondary-, tertiary-, and quaternary-structures of biopolymers and the interaction between
proteins and nucleic acids. This is followed by a section on
the origin of the genetic code. The last section deals with
the interaction between the proteins and lipids.
[*] Prof. Dr. M. Calvin
Laboratory of Chemical Biodynamics
University of California, Berkeley, California
94707 (USA)
[**I This article IS based
on an address presented at the dedication of the
Midland Macromolecular Institute, Midland, Michigan, September 29,
1972. All lectures given a t the dedication will be collected in H.-C. H ias.
Trends in Macromolecular Science (Midland Macromolecular Monographs,
Volume I), Gordon & Breach, New York-London.
Anqew. Chrm iniernaf. Edif. f VoI. 13 (1974) / No. 2
Why has this particular sequence been chosen? Many of the
phenomena (reactions) under discussion are actually extrapolations and modifications of reactions with which polymer
chemists are familiar. However, there seems to be a kind
of curtain separating the polymer chemist, whose background
is oriented to the structure and synthesis of macromolecules
of any kind and the determination of the basic principles
involved, and the group that has grown up in biology and
biochemistry and which haslearned about the physical chemistry and the chemistry of the behavior of the kinds of biopolymers they deal with“]. The biochemists have felt that these
things are almost magic, and they believe new rules, or laws,
or chemistry, may be required to understand in detail how
the naturally-occurring biopolymers (proteins, nucleic acids,
etc.) perform their function. There appears to be some “mystique” about this, so an attempt is made in the following
article to outline some of the reasons for this “mystique”,
which is generally in the minds of those who grew up from
the side of biology. The behavior of biopolymers is interpreted
in terms which are thoroughly familiar and which may even
121
appear somewhat naive to the polymer chemist. It must be
remembered, however, that most of the work discussed here
is being carried out by people who are not polymer chemists,
and that is why some of the material may seem obvious
to those of you who happen to be polymer chemists. Some
of it, however, may not be obvious to you and could possibly
give you fresh ideas for your line of research.
2. Formation of Biopolymers, Particularly Proteins
Let us begin with a consideration of the general scheme by
which the two major biopolymers are constructed by living
organisms. These are the proteins and the nucleic acids, and
a sequence is shown schematically in Figure 1, beginning
Cel I wall
, ","
~essenger
.. .
Polvsome
is then handed over to a small piece of polymer (RNA) constructed of about 70-100 monomeric units, containing in
it somewhere a certain triplet (three bases in a row) which
is characteristic of a particular amino acid; this distinction
is designated by the different symbols in Figure 1 (circles,
lines, crosses, squares), representing a different sequence of
the three bases. Thus, each activated amino acid is loaded
onto the terminus of a specific small oligomer (small polymer
of the order of 70-100 units). We now have the amino acid
loaded onto a particular transfer RNA (tRNA) with a particular triplet of bases characteristic of that amino acid. These
are now free to attach themselves to the messenger RNA,
the attachment taking place at points in the messenger where
the complementary base triplets are to be found. Thus the
sequence of the amino acids which go into the polymer to
make the polypeptide is established. The polypeptide, is then
"zipped up" to construct the protein. Generally, the protein
is itself a characteristic polypeptide, having a typical shape
and structure, which will be discussed in Section 4.
All of these reactions-the copying reaction (replication), the
transcription into RNA, the activation, the loading, and the
zipping-up on the ribosomes-are dependent for their specificity in part at least on the proteins which are themselves
thus made. This is a "chicken and an egg" problem, and
it sets the stage for the next part of the discussion.
Amino E, Acrlvated
acid
?mipa acid
2
Transler RNA
RNA
[43@
Fig. 1. Protein biosynthesis
with the DNA double helix of the cell which can in some
way (which is still not clarified) reproduce itself. That is,
it can copy a particular sequence of nucleotide monomers
in some particular order, and make a complementary copy,
giving rise to a double set.
In addition, there is another type of reaction which occurs
inside the cell. Instead of copying the DNA into another
DNA strand, part of the DNA (which is an enormously long
polymer, made up of essentially four bases hooked together
by ribose phosphate linkages) can be copied into a ribose
phosphate polymer, with a series of slightly different bases
but related to the initial sequence. That part which is so
copied contains in it the "message" for a particular protein.
That is why this type of material is called "messenger R N A .
The messenger RNA then leaves the cell nucleus and thus
becomes the message for constructing a particular protein. The
message is contained in the sequence of bases in the polymer.
As is known, it takes three of these bases to designate a
particular amino acid. The particular "messages" are then
hooked onto the catalysts, which are themselves made up
partly of protein and partly of nucleic acid and some lipid
substances.
The situation is now ripe to put together the amino acids
into a peptide. In order to do this, the amino acids must
be suitably prepared. Each one of the different length straight
lines in Figure 1 represents a different amino acid. The zigzag
line represents the amino acid, acylated with a phosphate
residue (aminoacyl phosphate or aminoacyl adenylate). This
122
How did this reflexive system, which is represented schematically in Figure 1, arise? As chemists, we would like to see
if chemistry could have designed such a system and put it
together in just this way. This is, of course, what should
be understood as the nonbiological origin of the polymers
and of the genetic code. How could this relationship between
particular triplets and particular amino acids come into being?
I believe this relationship is the result of chemistry and not,
as some of the biologists believe, a "frozen accident". However,
experimental evidence for this is, as yet, forthcoming. Most
polymer chemists will find this a rather amenable point of
view, and it is to be hoped that many will turn their attention
to devising experiments which will show that there is some
reason for a triplet corresponding to a particular amino acid
and not to some other one. This remains to be proved in
the laboratory and is still a subject of considerable controversy.
Let us now return to the possible ways in which the proteins
and nucleic acids might have come into being in a nonliving
world: The world with which we are dealing is a world in
water; it is not a world in methylene chloride, dioxane, DMF,
or other solvents. And all of the reactions which are described
in Figure 1 occur in water. Whatever reactions we use, whatever
reactions we call upon to do the originating of the protein
and the nucleic acid, should be essentially in an aqueous
medium to begin with. This can be modified somewhat by
the resulting polymer itself in terms of the protein-lipid interaction. The reactions do, however, occur primarily in an aqueous
environment.
We set out to try and generate polypeptides and other polymers in an aqueous environment[']. Chemically and thermodynamically this is rather a difficult operation. The formation
of the biopolymer from the monomer is, in the case of the
proteins, nucleic acids, carbohydrates, and lipids, the result
of a dehydration condensation. To make a dehydration condensation occur in water is a tricky operation. Living
organisms have long ago learned how to do this, but the
Angew. Chem. mternat. Edit.
1 Vol. 13 ( 1 9 7 4 ) 1 No. 2
question now is whether we, as chemists, can accomplish
something similar. The reactions which must be accomplished
in this aqueous environment are shown in Figure 2. The
H
,
HOCH,CHOHCH7-O~-~-ICH,I,
-
+.H,N-CH-C-NH-CH-CO,H
H
-
Dipeptide
“
Polymer
,
HOCHzCHOHCH,-O-!-ICH,I,H
u
l s t e r bond
Fig. 2. Dehydration condensation in the formation of polypeptides. carbohydrates, fats.
proteins and peptides are formed from a bifunctional molecule,
from which a water molecule can be eliminated, a bifunctional
dimer remaining which, in turn, can grow at either end. Formally, the same kind of system occurs in each of the four
cases-proteins, polysaccharides, lipids, nucleic acids. In each
case, the water molecule is eliminated and a new bifunctional
molecule is formed. This type of operation can go on continuously, but, in !he case of the lipids, not to the same extent
as with the proteins or polysaccharides; elimination of water
gives rise to the ester bond which can react further. This
is, however, not quite in the same category but it is possible
to make fairly large molecules. This case has been mentioned
at this point, since the lipids play an important role in the
whole evolutionary scheme.
I
Adenine
is available and at the other end the 3’-alcohol hydroxy group.
The nucleic acids are built on the same principles as the
two other major biopolymers shown in Figure 2.
We asked the question: Is there any way to generate the
polypeptides? Many reactions are known which might be
used for the linking of polypeptide bonds in an aqueous
environment. Only a slight modification of the existing knowledge of reactions was required. One of the important reagents
that has been used for the dehydration condensation
of a variety of functional groups is the carbodiimide
R-N=C=N-Rf*l.
The carbodiimide structure has the
ability to remove water, in stages, by more or less specific
reactions with acid groups such as carboxyl, phosphate, and
certain alcohols. This leads to condensation reaction^'^].
Carbodiimide has been used widely for nucleotide and
polypeptide synthesis, but usually not in aqueous environments. It was necessary to change the R groups, to make
them useful in aqueous environments. This can be achieved
by making the R groups hydrophilic so that the carbodiimide
can be dissolved in water. This is useful since the carbodiimide
does not hydrolyze very rapidly directly but does so by virtue
of its reaction with the carboxyl group and then with the
phosphate.
’
‘?HZ
t [mini-
Fig. 4. Polypeptide formation by dicyandiamide in homogeneous solution.
Adenyltc acid
I
i
Oinucleofide IApApl
,m
Fig. 3. Dehydration condensation in the formation of polynucleotides.
Figure 3 shows the methods for the formation of the nucleic
acids by dehydration condensation reactions. Here, actually,
there are three points of dehydration involved. The first cleavage of water effects formation of the amino glycoside on
a base, whereby the N H group of the base reacts with the
glycosidic semiacetal hydroxy of the sugar. The second takes
place between the phosphate ester and the primary hydroxy
group with formation of the terminal 5’-phosphate. In this
case the ribose sugar is shown which forms RNA. (The DNA,
of course, is the material in which the 2’-carbon lacks the
hydroxyl, i. e., carries two hydrogen atoms). The third dehydration condensation occurs between the monophosphate ester
and the secondary hydroxy group on the 3’-carbon atom
of another nucleotide. The result is a bifunctional molecule,
with both functions available; at one end the phosphate ester
Angrw. Chrm. inrrrnar. Edir. 1 Val. 13 ( 1 9 7 4 ) f No. 2
Now arose the next question: How could this type of reaction
occur in a natural environment? It turned out that there
was a very easy way in which this particular structure could
have evolved in a natural environment. As we know, ammonia
and HCN were two of the primary molecules on the earth’s
surface. Reactions between these two materials can lead to
cyanamide, probably through cyanic acid or hydroxylamine.
Cyanamide is a tautomeric form of carbodiimide which, however, is not stable; it tends to dimerize to the dicyandiamide
(DCDA) H2N-C(=NH)-NH-CN,
which is the common
form of cyanamide. It also contains the same type of n-bonds
as carbodiimide. We used DCDA as one of the starting materials to demonstrate this type of dehydration condensation
in the formation of biopolymeric materials. We took an amino
acid, adjusted the pH to slightly acid conditions, and added
the water-soluble carbodiimide material (dicyandiamide) to
[*] R
= e . g . cyclohexyl
123
see if we could produce polypeptides in the homogeneous
aqueous environment. The results of some of these experiments
are shown in Figure 414’. We started with glycine and carried
out the reaction at pH 3.5; the glycine disappears and the
dimer (diglycine), trimer (triglycine),and tetramer (tetraglycine)
are formed. The reaction is very slow and does not proceed
very far. One can hardly call the tetraglycine a “polymer”,
but at least the reaction does occur in aqueous solution.
The final study is, perhaps, the most elegant, and one doesn’t
know whether it will continue. This is the work of Aharon
Katchalsky, who used aminoacyl adenylates as starting mater-
The next question was: Is the reaction in any way selective?
Do the amino acids self-select each other when they condense?
This is a natural question to ask in the ordinary course of
evolutionary studies. About five or six years ago, Steinman,
in trying to answer this question, attached an amino acid
B on polymer beads via the carboxyl group and used an
amino protected amino acid A to couple to it. He measured
the relative rates of coupling of amino acid A to the polymerbound amino acid B in an homogeneous environment; this
is a model, of sorts, for polypeptide formation. Table 1 shows
the relative rates of coupling. It can be seen that the rates
differ by a factor of about ten[51.The is the measured coupling
efficiency, and the calculation is based upon the frequency
of a particular pair (phenylalanylglycine, glycylphenylalanine,
or isoleucylglycine, etc.) in today’s existing proteins with a
statistical analysis of the situation, normalizing the frequencies
to glycylglycine as unity. There seems to be slight relationship
between the efficiency of coupling in the heterogeneous system
and the occurrence in nature of these particular pairs of peptides. This idea has been developed still further, and there
are now better ways of studying the possible self-selection
of amino acids in peptide formation.
Amino acid
Table 1 Comparison of experimentally derermined dipeptide yields and
frequencies calculated from known protein sequences relative to Gly-Gly = 1.0.
Values
Dipeptide [a]
Gly-Gly
Gly-Ala
Ala-Gly
Ala-Ala
Gly-Val
Val-Gly
Gly-Leu
Leu-Gly
Gly-Ile
Ile-Gly
Gly-Phe
Phe-Gly
Obs.
Calc.
1 .o
1.o
0.7
0.6
0.6
0.2
0.3
--_________
0.8
0.8
0.7
0.5
0.5
0.5
05
0.3
0.3
0.1
0.1
03
0.2
0.1
0.1
0.1
0.1
[a] The dipeptides are listed in terms of increasing volume of the side chains
of the constituent residues. Gly =glycine, Ala =alanine, Val = valine. Leu
= leucine, Ile = isoleucine, Phe = Phenylalanine. Example: Gly-Ala = glycylalanine [5].
There are two other methods by which polypeptides have
been made, one of which was carried out in a nonaqueous
environment. Table 2 shows the results of experiments in
which Sidney Fox used molten glutamic acid containing a
small amount of pyrophosphoric acid as a solvent (hardly
an aqueous environment) and to which an equimolar mixture
of all the other amino acids, other than aspartic and glutamicI6],
was added. Here, again, it can be seen that the incorporation
into the resulting polypeptides is not statistical. The main
point about these data is that they show the variable frequency
of incorporation, ranging from 0.5% to 5 % , and that selectivity is evident.
124
NH2
Total
%
%
Amino acid
5.02
27.6
ial and montmorillonite clay as the catalyst, both of which
are readily available in nature. The aminoacyl adenylate is
available in nature because the adenylic acid can be generated
by ordinary chemical reactions and the coupling of the amino
acid to the adenylic acid (linking a phosphate and a carboxyl
anhydride) can be easily achieved using the carbodiimide
dehydration condensation reaction in an aqueous environment. So the system with which Katchalsky began is achievable
biologically. Katchalsky’s discovery ought to be described in
detail because it is the starting point of a very interesting
idea. The aminoacyl adenylate used is the activated amino
acid in the first stage of normal biological amino acid activation; although it is usually a part of an enzyme complex
it, nevertheless, is a suitable starting material for this type
of study. At first, Katchalsky tried to polymerize this material
in a homogeneous aqueous solution. The aminoacyl adenylate
can be regarded as a mixed anhydride of a carboxylic acid
group and phosphoric acid which also contains the free amino
group of the amino acid. Thus, the free amino group can
act as a nucleophile toward the carboxy group of the anhydride
grouping.
Polypeptide adenylale
I alternative
LJ
Fig. 5. Polymerization of aminoacyl adenylate: alternative reactions ( K a r chalskr).
Angew. Chem. internat. Edit.
/ Val. 13
(1974)
/ No. 2
Figure 5 outlines Katckalskj’s basic idea; but it shows more
as well. If one considers the aminoacyl adenylate skeleton
attached to the polynucleotide chain one can see that the
amino group can act as a nucleophile toward the acyl group
of the anhydride (reaction a), a reaction that leads to the
formation of a polypeptide (po1y)adenylate and free adenylic
acid. The alternative reaction, which Katchalsky had considered, but had not published any experimental information
on, is that in which the secondary hydroxy of the sugar
will attack-once again as a nucleophile-the phosphate in
the mixed anhydride (reaction b), resulting in the reverse
type of polymerization, the products being a free amino acid
and the (po1y)aminoacyl polyadenylate. There are, of course,
alternative reactions which can occur (rearrangements, etc..)
to stop or slow down the polymerization reactions.
As Katchalsky reported in his publication of 1970[’”] and
in more detail in May 1972[7h1in Gottingen, shortly before
his death, the reaction is slow in homogeneous solution and
the polymers which are formed are not very large (8-10
units). He found, however, that if he performed this reaction
in the presence of a properly prepared montmorillonite, the
whole system changed. The reaction became extremely rapid
and the polypeptides reached fairly high degrees of polymerization-i. e. high for this kind of a reaction[’].
We have both attempted this type of reaction and have, indeed,
been able to prepare polypeptides and polynucleotides from
the same starting material. We can now begin to see the
real relationship between the polynucleotides and polypeptides-or the nucleic acids and proteins-and how they can
arise as a result of the stereochemistry of the reaction.
Table 3. Composition of the main fraction obtained from 1 g alanyl adenylate
on montmorillonite.
Fraction
Component
MW
Degree of
polymerization
Wt.
[mg]
I
Peptide
Peptide
Peptide
Adenylic acid
Peptide adenylate
Peptide adenylate
Peptide adenylate
Peptide adenylate
Adenylic acid
640
1120
1900
9
16
27
2130
2310
3020
4000
30
32
42
56
10 3
35.2
20.0
470.5
8
5.2
10.8
I1
1
No. in
Fig. 6
4
2
I
3
6
17.1
260.1
1
unidentified inorganic salts and two unidentified organic compounds. When I spoke to Katchalsky he said he thought
they might be polynucleotides; and I think so too.
Perhaps someone else will pursue this work further in the
future: In my opinion it represents an extremely important
step towards understanding the nature of polymerization reactions. Why is it that you d o not get a random distribution
but discrete groups of polymers? I think these discrete groups
occur because no simple addition reaction takes place. Amino
acids are added and polynucleotides are constructed and subsequently form a polypeptide-polyadenylate, and these materials
regulate the size of the polypeptides which are found. It will
probably turn out that the polynucleotide is also discrete
in some way.
3. The Origin of the Genetic Code
I
I
I
It is known that in today’s living organisms the specification
of the linear array of amino acids in a polypeptide is contained
in a corresponding linear array of bases in a polynucleotide.
A series of three bases in the polynucleotide (DNA) specifies
a specific amino acid in a polypeptide, or protein. This correspondence of a triplet of bases with a particular amino acid
is universal in all living organisms. One form of expressing
I
I
I
I
I
I
I
I
I
I
I
Table 4. The genetic code.
jI 6
first letter
I
A
,
6000
AAA
AAG
7000
V [rnll-
G
U
AAC
AAU
Asn
ACA
ACG
C
Katckalsky found that a number of discrete polypeptides were
formed from alanyladenylate and montmorillonite. Figure 6
shows a chromatogram of the second fraction. Table 3 gives
more information about the two principal components contained in the initial extract of this reaction. Fraction I consists
mostly of adenylic acid and three types of polypeptides. Fraction 11, however, contains mostly polypeptides with some
adenylic acid. In the gel permeation chromatogram of fraction
I1 (Figure 6) one can see the presence of some nine components;
the largest component is adenylic acid, and the components
1, 2, 3 and 4 are polypeptides. In addition, there are two
Lys
CAA
CAG
GAA
GAG
CAC
CAU
GAC
GAU
CCA
CCG
GCA
GCG
UAA
UAG
terminate
UAC
UAU
Tyr
A
Fig. 6. Polypeptides from alanyl adenylate on montmorillonite (Karckalskyl.
Chromatography of fraction I 1 on Sephadex G-25 with gel bed 3 x 4 x 150cm
1-9. see text and Table 3.
Anyrw. Ckrm. i n f r r n a f . Edit. f Vol. 13 (1974) / NO. 2
C
I
g
5
Pro
Thr
ACC
ACU
CCC
CCU
Asp
UCA
UCG
UCC
UCU
GGA
GGG
UGA
UGG
-- -
--___-.-____
~
2
AGA
AGG
2
G
Ax
AGC
AG U
Ser
AUA
AUG
Ile
Met
Glv
CGC
CGU
GGC
GGU
CUA
CUG
GUA
GUG
U
Leu
AUC
AUU
,le
CUC
CUU
Ser
Ala
GCC
GCU
UGC
UGU
UUG
- -
terminate
Trn
Cys
Leu
Val
GUC
GUU
UUU
Phe
125
this is shown in Table 4. Here one can see there is a certain
amount of redundancy in the code, but the universality of
it throughout the living world is now fairly well established.
Thus, in addition to the question of how these biopolymers
may actually have been formed abiogenically we must turn
to the question of how the linear relationship between the
two major biopolymers, that is, the polypeptides (proteins)
and the polynucleotides (DNA), evolved. That is: What is
the possible origin of the genetic code?
There are those who believe that the particular specificity we
now find in living organisms between a particular triplet and
a particular amino acid, as assembled in Table 4, is the result
of an accident in some early catalytic reaction which, by
virtue of a selective advantage in producing autocatalytic
systems, has now been frozen into all of biology['! Selection
followed because a single autocatalytic system eventually
dominated all the others. Another way of expressing this
"frozen accident" notion is to state that a renewed chemical
evolution under exactly the same starting conditions could
very well lead to a completely different codeal relationship[''. ''I. It is my personal belief, however, that this would
not be the case. I believe that the coda1 relationship reflects
some characteristic molecular interactions between amino
acids (polypeptides) and nucleotides (polynucleotides).
A modicum of evidence for this already exists in the form
of some experiments carried out on the rate of coupling of
amino acids to nucleotides. Some years ago['21we attached
a nucleotide (adenylic acid) to a synthetic polymer (polystyrene); the polystyrene was in the form of microspheres. The
next step was to measure the rate of coupling of two different
amino acids to each of these two polystyrene-nucleotide preparations. The coupling was performed using an N-protected
amino acid adenylate. The reaction is shown in Figure 7
G-O-[PO-C@de
tl
FHZC6"5
+
zNH-cH-c"' 0'
ZNH-CH-C'
HO
P
I
OH
i
*o
CHZC6H5
[y)
N",
Ade=
I
Z = C,H5CH,-O-C0
c=o
CH-NHZ
I
CHZC6H5
Fig. 7. The coupling of a polymer-AMP complex with the anhydride of
an N-protected amino acid (24h at room temperature in pyridine) to an
N-protected phenylalanyl-AMP polymer.
and the results, as well as those of analogous reactions, are
given in Table 5f1j1.It can be seen that glycine reacts more
rapidly with adenine and cytosine than does phenylalanine,
and adenine reacts more rapidly with both amino acids than
does cytosine. The overall range of reactivity is roughly a
factor of three.
Here the beginning of a kind of selectivity in a reaction rate
is
However, I believe that this kind of selectivity
will be very much enhanced if both the amino acid and the
126
nucleotides are each, respectively, part of a polymer. For
example, one would expect the alternative reactions (a)
and (b) indicated in Figure 6 to be very dependent upon
the nature of the amino acid and the nature of the nucleotide
Table 5. Coupling of a polymer-nucleotide complex with the anhydride
of an N-protected amino acid. The proportion [%]of bound nucleotides
reacted are quoted.
Adenine
Cytosine
6.7
10.0
2.9
6.5
acid
Phenylalanine
Glycine
involved. Furthermore, one would expect that selectivity would
be increased with the length and character of the polynucleotide, on the one hand, and possibly even with the polypeptide
on the other. This last experiment is yet to be done, but
is underway.
4. Secondary and Tertiary Structure of Protein and
Interaction with Nucleic Acids
This part of our discussion will be concerned with what is
known about the secondary, tertiary and quaternary structures, above all, of proteins. There are several principles which
should be emphasized here: One is the way in which proteins
fold up, in their secondary and tertiary structures, another
is the way enzymes interact with substrates, and a third is
the way in which the proteins interact with the nucleic acids.
(The protein-lipid interaction is discussed in Section 5.)
Myoglobin, with its tertiary structure, is shown in Figure
8. The left-hand side of the figure shows the alpha helical
structure and the drawing on the right-hand gives a more
three-dimensional view of the same material, to show that
the myoglobin is a folded protein, with the heme stuck in
the center. This secondary and tertiary structure is built into
the molecule as a result of the amino acid sequence in the
polypeptide which takes up this structure, given the opportunity, in water"']. In the structure of cytochrome c, the
hydrophobic sidechains of the amino acids are more or less
in the middle of the molecule and the hydrophilic parts are
on the outside of the molecule. This is the only characteristic
which has so far approached a generalization on the structure
of soluble proteins. In general, the structure of soluble proteins
is such that when they fold up in their secondary. and tertiary
structures they d o so in a way which places the hydrophobic
chains on the inside (away from the water environment) and
the hydrophilic chains on the outside, towards the water environment.
It is for this reason that I had the reservation, mentioned
earlier, about the generation of polypeptide and polynucleotide
structures in a purely aqueous environment. Modifications
can occur during the course of formation of these biopolymers
which will remove their functions from the water environment
into a nonaqueous environment, by virtue of their own structure, so to speak.
Angew. Chem. internat. Edit.
1 Vol. 13
(1974)
1 No. 2
Fig. 8. Tertiary structure of myoglobin.
The lysozyme molecule has a cleft, into which the substrate
is incorporated. Figure 9 shows the mechanism of the polysaccharide hydrolysis, and the actual function of the active site.
thus letting the other part go free. The Asp glycosidic ester,
in turn, is very rapidly hydrolyzed. This is the way in which
the hydrolytic enzymes and many of the transferases function.
Thus, the tertiary structure is evolved to give the particular
tertiary architecture which will take hold of the proper substrate.
Fig. 10. Structure of collagen. For details see text
tysozyme main chain
Lysozyme
main chain
jbsst'il
Fig. 9. Hydrolysis of a polysaccharlde (rings A-E)
on lysozyme.
One can see how the poiysaccharide lies in the active site
of the lysozyme. The glycosidic linkage is hydrolyzed: one
part of the substrate is transferred t o the polymer (Asp 52)
Angrw. Chem. inturnat. Edit.
1 Vol. 13
(1974)
1 No. 2
The next figure (which is an old one) reaches into the area
visible under the electron microscope. In collagen (Fig. lo),
we can see that molecules can be refolded into their tertiary
structure and, even further, that these molecules can actually
reassemble into more complex quaternary structures which
are dependent upon their primary sequence" 61. The upper
part of Figure 10 shows the separated collagen fibrils, and
the bottom part the reaggregated fibrils, which strongly resemble the natural ones. The fibrils are highly ordered arrays
as a result of protein-protein aggregation.
121
The next few figures show the dissociation and re-aggregation
of the tobacco mosaic virus (a relatively simple virus particle)
which consists of a set of identical protein molecules and
with its uniform particles, and Figure 12 shows the reassembly
of the TMV protein without the nucleic acid. The protein
is separated from the nucleic acid, redissolved, and, by changing the salt concentration, reaggregated with the result shown
here. One can see that the material has reaggregated with
the correct dimensions in one way, but not in the other:
in contrast to the native TMV, all the particles have different
lengths because there is no nucleic acid “information” to
give the correct dimensions. In Figure 13 you can see the
reconstituted TMV virus particle, which occurs when the nucleic acid component is put back into the protein solution.
This same type of experiment has been done with much more
complex particles such as T4 bacteriophages, MS2 phages,
eft. Progress is also being made in the reassembly of very
complex structures[’*].The TMV particle itself is shown diagramatically in Figure 14. The tertiary structure of the subunit
protein material in TMV has not yet been established, but
we know that there is a sequence of 120 amino acids.
Fig. I I. Tobacco mosaic virus (TMV), native.
one nucleic acid molecule. These two can be separated and
then reassembled again into a complete virus particle“ ’I. More
maximum radius
mean radius
Fig. 12. Reconstituted TMV protein.
[Aj&G
4
20
t
Radius ot hole
Fig. 14. Diagrammatic structure of TMV.
5. Interaction of Protein and Lipids
Fig 13. Reconstituted TMV
complex virus particles have now been reconstituted as well.
Figure 1 1 shows the native tobacco mosaic virus (TMV),
128
Protein-lipid interaction is one of the areas of very high scientific activity, in the technological polymer area and more selectively than hithertofore in the biochemical laboratory. Lipids
have been used for handling biochemical proteins for some
time. The present concept of membrane structure-protein
embedded in the bilipid membrane-is shown in Figure 15.
A phospholipid consists of the hydrophilic ends (the phosAngrw. Chrm. internat. Edir. J Vol. 13 ( 1 9 7 4 ) J
No. 2
structed themL2’].The spheres are generally called liposomes;
their osmotic properties, transport properties, and a variety
of other properties resemble those of living cells. How d o
materials get in and out through that membrane? What is
the mode of their transport? Can specific proteins be associated
with the membranes to produce these specific results? These
are questions still unanswered. Studies are now in progress;
this field is really just blossoming out as a new area and
marks the advent of a new era. Only now have the polymer
physical chemists become active in this field.
Fig. IS. Present concept of membrane structure showing protein embedded
in bilipid membrane.
phoethanolamine end represented by spheres) and the lipid
chains. The globules which are membrane proteins in various
stages of insertion into the bilipid membrane make the total
actual biological membrane itself. These proteins may be either
structural elements or transport proteins, which carry material
into the cell through the biological membrane. The fact that
many of the proteins are directly associated with lipids implies
that there must be hydrophobic connections on the outside
of the proteins in order to be able to perform this type of
insertion. This has not yet been established in the laboratory.
Clearly, there are hydrophobic parts on the proteins, but
whether they are on the outside or the inside, or both, is
a subject of some concern.
As is known, it is quite common to use detergents, anionic,
cationic and non-ionic, for the isolation of a particular enzyme
from a living cell. The procedures for the soluble proteins
(soluble enzymes) are now well established. The insoluble
enzymes, however, are a different story. The isolation of membrane-attached enzymes from the membrane of the cell, the
membrane of the mitochondrion, or the membrane of other
internal cellular structures present more difficulties ;it is necessary to use detergents of various types for enzyme extraction.
A few years ago, we naively (and this quality has some merits
as well as demerits) began to isolate the enzymes in which
we were interested (the RNA-instructed DNA polymerase,
RIDP). We had to learn that detergents are necessary. We
also learned that when the detergents were eliminated, the
enzyme itself seemed to be eliminated, or at least the activity
of the enzyme was eliminated. This gave rise to the idea
that perhaps the activity of the enzyme was dependent upon
the presence of detergent molecules, i. e., the lipid component
in some way was changing the conformation of the enzyme
and inducing its activity, or at least raising it. Therefore,
the residual enzyme activities which are observed when synthetic detergents are not used for extraction are simply due to
the natural phospholipid, which was released when the solution
was sonicated.
We found that when we added small amounts of certain
types of non-ionic detergents it was possible to recover the
enzyme activity. This constitutes a new development. N o one
really believed that the intrinsic activity of the enzyme was
dependent upon its association with a detergent molecule.
Previously it has been assumed that the detergent molecules
are solubilizing entities to dissolve the enzyme out of the
lipid membrane. The detergent does, in fact, take the enzyme
out of the lipid membrane, but large amounts of the detergent
are not needed to make the enzyme active.
Fig. 16. Liposomes of phospholipid and cytochrome‘ (Home and Watkins).
A synthetic “biological cell” is shown in Figure 16. These
liposomes (soft spheres) are formed by shaking the solution
of a phospholipid (lecithin) with a protein cytochrome. The
multiple and single bilipid layers are visible in the electron
micrograph; these layers are filled with protein[”! This work
was done about eight years ago in Great Britain. I have
called the spheres “Bangasomes” after the man who first conAnyew. Chem. internat. Edrt. / Vol. 13 ( 1 9 7 4 ) 1 No. 2
E 20l
t
Detergent [rnmolillFig. 17. Activation of RNA-instructed DNA polymerase by detergent.
100%=maximum activity.
129
The activation of the RNA-instructed DNA-polymerase by
detergents is shown in Figure 17. RIDP is the enzyme which
copies RNA into DNA, which is popularly known as “reverse
transcriptase activity”. Research on this enzyme resulted in
one of the major breakthroughs of the last two or three
years in this field, particularly in connection with tumor viruses. The activity of the enzyme is increased by addition
of suitable amounts of various types of non-ionic detergents121’22!On a molecular basis, this is a positive molecular
interaction. Observation of interaction happened to be made
during the development of techniques for the study of enzymes.
Not uncommon to this type of work, this turned out to
be the central problem.
;HI
p 3
~O-ICH,-CH,-O
H,C-F-CH,-F
CH,
I~.,~H
CH,
p3
!HI
H,c-F-cH,-;~oCH,
CH,
n-C,,H,,-0-
ICH,-CH~-OI,,.,,H
The structure of the non-ionic detergents which were used in
this study is shown in Figure 18. Some of the detergents
(the Tritons) are aromatic and others are not. This fact is
important. The effect of three different detergents on the
ability of a certain drug, which has a hydrophobic tail and a
hydrophilic end, to inhibit the enzyme is shown in Figure 19.
An attempt was made to plot the critical micelle concentration
for each of the detergents. The ability to deter the drug from inhibiting the enzyme is dependent upon micelle formation1221.
We interpret this finding as follows: if there is too much
detergent present there is micelle formation ; the micelles
dissolve the drug, and thus remove it from the hydrophobic
portion of the enzyme. The drug no longer can act on the
enzyme, so the enzyme returns to its full activity.
Figure 20 shows two magnifications of a tissue culture which
has been transformed into malignancy. As soon as the cells
have been transformed into malignancy they overgrow each
other, and they make little “piles’’ called foci. We have found
that by suitable adjustment of the drug concentration it is
possible to prevent the virus from transforming cells as determined by this focus formation inhibiti~n”~!We have even
gone a stage further and been able to show that the formation
ICH,-CH,-0l2,H
H - I 0 -CH,-CH,I,.,OH
Fig. 18. Structure of detergents used in RNA-instructed DNA polymerase
activation. From top t o bottom: Triton X-100, Triton X-1017, Triton D N
65, Brlj 35, polyethylene glycol-400.
We were actually studying the inhibition of the RIDP enzyme
activity by certain drugs with a view toward the possible
use of such drugs to inhibit the transformation of normal
cells into cancer cells by the viruses which carry this enzyme.
The RIDP enzyme copies the virus-RNA in DNA which is
then inserted into the DNA of the cell, thus making a transformed cell. The rationale was to find a drug which would
inhibit the RIDP enzyme and thus block the transformation
into cancer cells. This meant that we had to study inhibition
of the enzyme activity under various conditions. It also gave
rise to our studies on the effect of detergents on enzyme
activity. We found that we had some excellent enzyme inhibitors in the drug itself. Rifampicin and its derivatives are
lipophilic substances. Here, it turns out that if there is too much
4
I
1
Fig. 20. MSV focus formation on Balb/3T3 cells (dark field illumination)
of tumors in whole animals can be prevented‘241.Again, this
required a suitable combination of detergent and drug to
prevent the detergent from spoiling the activity of the drug.
I hope that some of the things described above will give the
reader some concept of the kind of a china shop one can
get into, if one has the will to do it. I have tried to give those
of you who are not in the biochemistry-molecular biology
business some idea of the way in which a polymer chemist
could influence the development of our fundamental concepts
of the nature of life, its origin, and the application of our
knowledge to the problems of the day.
The work described in this paper, which will also be published
in the “International Journal of Polymeric Materials”, was
sponsored by the U . S . Atomic Energy Commission and by the
Elsa U . Pardee Foundation for Cancer Research.
Received: June 15, 1973 [A 984 IE]
German version: Angew. Chem. 86. 111 (1974)
Fig. 19. Suppression of drug effect on RNA-instructed DNA polymerase
by detergents ( 0. 0 ,A ) (see text).
detergent the drug is ineffective in its inhibition of RIDP
enzyme activity.
130
[ I ] P. J . Flory, Angew. Chem. 86, 109 (1974): Angew. Chem. internat. Edit.
13, 97 (1974).
[2] M . Caluin - Chemical Evolution. Oxford University Press, Oxford, England 1969, p. 152ff.
[3] See Ref. [2]. p. 161ff.: D H . Kenyon and G . Stemman; Biochemical
Predestination. McGraw-Hill, New York 1969, p. I82ff.
Angrw. Chem. inrumof. Edit. 1 Vol. 13 ( 1 9 7 4 ) 1 No 2
[4] D. H . K m y o n , G. Stemman, and M . Calvin, Biochim Biophys. Acta
124, 339 (3966).
[ 151 For a discussion of the principles involved in three-dimensional structure
and self-assembly, see Ref. [2], pp. l87---193.
[ 5 ] G . Strinman and M . N . Cole, Proc. Nat. Acad. Sci. USA 58, 735 (1967).
[ 161 For a discussion of the principles involved in quaternary structure.
see Ref. [2], p. 196.
1171 For a discussion of reassembly, see Ref. 123, pp. 216-218.
[6] K . Harada and S . W Fox in S. W Fox: Origins of Prebiological Systems
and Their Molecular Matrices. Academic Press, New York 1965, p. 296.
[7] a) M . Parcht-Horowitz, J . D. Breger, and A. Katchalsky, Nature 228,
636 (1970); b) M . Paecht-Horowitz, Angew. Chem. 85, 422 (1973); Angew.
Chem. internat. Edit. 12, 349 (1973).
[ 8 ] M . Parcht-Horowitz and A. Katchalsky, J. Mol. Evol. 2, 91 (1973).
[9] M . Eigm, Naturwissenschaften 58,465 (1971); cf. H . Kuhn. Angew. Chem.
84, 838 (1972): Angew. Chem. internat. Edit. 11, 798 (1972).
[lo] F . H . C. Crick, J . Mol. Biol. 38. 267 (1968).
[ I I]
[I21
[I31
[I41
L. E. Orgel, J. Mol. Biol. 38, 3x1 (1968).
M . A. Harpold and M. Calvin, Nature 219, 486 (1968).
M. A . Harpold and M . Calvin, Biochim. Biophys Acta 308, 117 (1973).
M . Calvin, Proc. Roy. SOC. Edinburgh, Section 70, 273 (1969).
[ 1 X] D. J . Kushner. Bacteriol. Rev. 33, 302 ( 1969).
[19] D. Papahadjopoulos and N. Miller, Biochim. Biophys. Acta 135, 624
(1967); D. Papahadjopoulos and W Watkms, ihid. 135, 639 (1967).
[20] A D. Bunghum, J . DeGier, and G . D. Grruitch, Chem. Phys. Lipids
I , 225 (1967).
[21] F . M . Thompson, L. J. Llberttni, U . R. Joss, and M (hloin, Science
178, 505 ( 1972).
1221 F . M . Thompson, L . J . Liberrini, U . R. Joss, and M Calvin. to be published.
[23] M . Calvin, Radiation Res. 50, 105 (1972).
1241 U . R . Joss, A . M . Hughes, and M. Calvin, Nature, New Biol. 242,88 (1973).
C 0 M M U N I CAT10N S
cis-trans-Isomerism of Thyrotropin Releasing
Hormone (TRH) in Aqueous Solution"*]
By Wolfgang Voelter, Oskar Oster, and Karl Zechl'l
A prerequisite for explaining the mode of action of peptide
hormones is an accurate knowledge of their structure. Since
nuclear magnetic resonance measurements can yield a wealth
of information about the structure of dissolved peptide molecules we have investigated synthetic''] TRH by pulse Fourier
transform 3C-NMR spectr~scopy[~l[**'].
CONHz
Figure 1 shows the broad-band proton decoupled spectrum
of TRH. Almost all the signals can be assigned by comparison
with the spectra of simple amino acids, TRH derivatives,
and derivatives of TRH components obtained as intermediates
in the synthesis of the hormone. An additional aid for assignment of the signals is provided by the proton off-resonance
spectrum[*J.
Remarkably, as can be seen in Figure 1, the intense signals
of the C-atoms 3, 4, 5, and 6 are each accompanied within
a distance of 1-2 ppm by a further signal (3', 4 , 5 ' , and 6', resp.)
whose intensity is only 1@-20% of the remaining signals.
To explain this phenomenon we have synthesized the proline
derivatives and proline-containing dipeptides and recorded
their I 3C-NMR spectra.
[*] Prof. Dr. W. Voelter, Dr. 0. Oster, and Dr. K. Zech
Chemisches lnstitut der Universitat
74 Tubingen, Auf der Morgenstelle (Germany)
[**I This work was supported by the Deutsche Forschungsgemeinschaft.
[***I All the compounds investigated have rotations, melting points, and
analytical data a s given in the literature: for the preparation of the substances
cf. Refs. [ I , 51. The PFT "C-NMR spectra were recorded on a Bruker
HFX-90 multinuclear NMR spectrometer (22.628 M H z for 'H: pulse width
4- --5 ws; pulse interval 0.41 s)
Angew. Chrm. internat. Edit.
/ Vol. 13 (1974) / No. 2
The carbon atoms of the proline ring of Boc-Pro-NH2 give
rise to two signals (Fig. 2). The spectrum can be interpreted
only in terms of the presence of a cis-trans isomeric pair.
13C-NMR spectroscopy thus proves to be an excellent tool
for the detection of cis- and trans-isomers of proline compounds; previously, these were mainly differentiated by proton
resonance s p e c t r o s ~ o p y ~ ~ l .
Although the intensities of PFT I 3C-NMR resonances are
not suitable for quantitative determinations they can provide
a rough estimation of the equilibria of chemically similar
carbon atoms. Thus, a comparison of the intensities of the
I3C-NMR signals of proline carbon atoms would indicate
that 20-30% of the cis and 80-70% of the trans isomers
of Boc-Pro-NHz are present in equilibrium in aqueous solution
at room temperature.
The assignment of signals to the cis or truns isomers by
3C-NMR spectroscopy is also possible in the case of Boc-GlyPro. 'H-NMR investigation^'^^ have already shown that solutions of this compound contain up to 70--80% of the trans
isomer. The more intense resonances of the proline carbon
atoms in the PFT 13C-NMR spectrum of Boc-Gly-Pro can
therefore be assigned to the trans isomer.
Furthermore, the positions of the proline carbon-atom signals
in the spectrum of glycylproline are characteristic for the
cis isomers of prolines.
cram
c h
On the basis of these findings the resonances of the C', Cp,
CY, and C8 atoms of the proline rings of the compounds
listed in Figure 2 can be accurately assigned to the cis or
trans isomers.
Figure 2 shows, moreover, that Boc-Ala-Pro must be present
almost exclusively as the trans isomer in aqueous solution.
From the intensities of the acetylproline resonances it follows
that an approximately 1 : 1 cis-trans isomeric equilibrium is
set up in aqueous solutions ofthis compound at room temperature.
131
Документ
Категория
Без категории
Просмотров
1
Размер файла
5 176 Кб
Теги
chemistry, origin, biologya, biopolymers
1/--страниц
Пожаловаться на содержимое документа