 |
Description  |
|
|
The present invention relates to molecular transformation procedures useful
in the synthetic preparation of organic and biochemical materials and,
more particularly, to the preparation of biologically active polymers such
as peptides and the like.
The area of peptide synthesis has received considerable attention in recent
years. A significant problem has existed in synthetically achieving a high
molecular weight, pure polypeptide wherein the amino acid sequence of the
peptide actually prepared corresponds to that sought. To approach
realization of the synthesis with the desired purity has heretofore been
quite laborious.
The synthetic preparation of a polypeptide is a multi-stage molecular
transformation procedure whereby a desired product is constructed by
sequential chemical reactions of a precursor and an added compound, with
the precursor at any given stage being the chemically reacted, reactable
precursor from the preceding stage. Thus, the procedure is reiterative.
In this process a first amino acid is reacted with a second to form a
dipeptide, schematically represented by formula I
##STR1##
The peptide so formed at this stage is separated from the unreacted acids
and a third amino acid is then reacted with the dipeptide to form a
tripeptide. The procedure is reiterated until a polypeptide having the
desired amino acid sequence, customarily termed the target peptide, is
obtained. Sequence failure, whereby a portion of the elongated polypeptide
chains have an improper amino acid sequence, can result from several
causes. One can be the failure to remove residual free acid from the
reaction mixture prior to reaction with a subsequent amino acid. The
presence of such unreacted acid presents the possibility that a portion of
the chains will be improperly elongated with the residual acid rather than
with the acid desired at that stage of the sequence.
Yet a further and perhaps more significant cause of sequence failure is the
incomplete reaction of all of the chains present with the amino acid added
at each stage of the synthesis. In the preparation of low molecular weight
polypeptides, the presence of chains containing different numbers of acids
can be analytically ascertained and the desired peptide chains isolated.
Conventional analytical techniques do not permit this to be done with
respect to the higher molecular weight varieties, however, because the
difference in molecular weight between properly and improperly synthesized
chains is simply too small to be detectable.
The peptide synthesis procedure described above has been represented as
involving the sequential reaction of amino acids with a polypeptide chain.
In this respect, there are two approaches; one being growth of the peptide
from the C-terminal end (the end of the chain with the
##STR2##
group) and the other being growth from the N-terminal end (the end with
the --NH.sub.2 group). It is well recognized that during the synthesis
either the alpha amino group (N-terminal route) or carboxyl group
(C-terminal route) of the added acid must be blocked so that the added
acid reacts with the polypeptide chain and reaction between the molecules
of added acid cannot occur. It is necessary, therefore, that the blocked
group of the added acid, after reaction with the chain, be deblocked for
the subsequent acid addition step. Failure to achieve deblocking at any
stage of the synthesis can introduce a sequence failure. Moreover,
deblocking must be accomplished in a fashion whereby the peptide being
synthesized is not adversely affected.
In order to achieve solubility, conventional methods of peptide synthesis
customarily are carried out in a non-aqueous medium, particularly for high
molecular weight peptides containing protected amino acid side chains.
This has necessitated the use of harsh coupling reactions to effect
peptide bond formation and the accompanying likelihood of chain disruption
such as racemization or fragmentation. Moreover, especially with respect
to the higher molecular weight polypeptides, non-aqueous reaction
solutions may not permit the peptide to assume its naturally occurring
configuration. In nature, of course, the peptides are manufactured in an
aqueous environment.
Accordingly, the present invention provides an improved method for
preparing a substantially pure polypeptide of predictable amino acid
sequence, which method is susceptible to automation and which can reliably
be used to prepare pure, high molecular weight target peptides.
The present invention also provides a method for isolating an elongated
polypeptide from its reaction environment so that the proper sequence can
result on further reaction, which method is easy to accomplish with a
minimum expenditure of time and minimum peptide loss.
The present invention further provides a facile method for minimizing the
difficulties attendant on the reiterative preparation of polymers which
result from sequence failure due to incomplete reaction by efficiently
removing failed sequences from the growing chain population so that
recovery of a desired pure polypeptide can be accomplished more
conveniently.
This invention further provides a novel method for deblocking a peptide
which is easy to accomplish and does not result in destruction of the
chain being fashioned. In addition, the deblocking is accomplished in a
manner which ensures the optical purity of the peptide being formed.
The present invention further provides a method for peptide synthesis
wherein the reiterative elongation of the growing chain can be rapidly
accomplished in an aqueous medium without disruption of the polymer chain.
Additionally, the present invention provides a peptide synthesis procedure
which does not necessitate elaborate protection of amino acid side chains
which customarily decrease the aqueous solubility of the polymer being
prepared.
The present invention provides a process for synthesizing a peptide chain
having a distinct sequence of amino acid segments which comprises
(1) reacting a pure precursor complex, wherein a first amino acid segment
of the peptide chain to be prepared is covalently bonded to a handle, and
wherein said segment contains a free terminal carboxyl group or a free
terminal amino group, with a second amino acid segment containing a free
N.sup..alpha. -amino group and a blocked carboxyl group when the precursor
has a free terminal carboxyl group or a free carboxyl group and a blocked
N.sup..alpha. -amino group when the precursor has a free terminal amino
group, in an aqueous medium;
(2) optionally removing the unreacted precursor complex;
(3) reversibly coupling the handle of the reacted complex to an adsorbent
immobilized on an insoluble support;
(4) separating the reacted complex from unreacted second amino acid
segment;
(5) releasing the handle from the adsorbent;
(6) optionally deblocking the carboxyl group or amino group on the reacted
complex;
(7) optionally reiterating steps (1) to (6) until the desired number of
amino acid segments are added to the precursor;
(8) optionally releasing the peptide chain from the handle and recovering
the product.
The present invention also provides a process for synthesizing a peptide
chain having a distinct sequence of amino acid segments which comprises
reacting a pure precursor containing a first amino acid segment of the
peptide chain to be prepared having a free terminal carboxyl group or a
free terminal amino group, with a second amino acid segment containing a
free N.sup..alpha. -amino group and a blocked carboxyl group susceptible
to enzymatic hydrolysis when the precursor has a free terminal carboxyl
group or a free carboxyl group and a blocked N.sup..alpha. -amino group
susceptible to enzymatic hydrolysis when the precursor has a free terminal
amino group, in an aqueous medium; and deblocking the product peptide
enzymatically.
The present invention further provides a process for preparing a peptide
chain having a distinct sequence of amino acid segments which comprises
reacting a pure precursor containing a first amino acid segment of the
peptide chain to be prepared having a free terminal carboxyl group or a
free terminal amino group, with a second amino acid segment containing a
free N.sup..alpha. -amino group and a blocked carboxyl group when the
precursor has a free terminal carboxyl group or a free carboxyl group and
a blocked N.sup..alpha. -amino group when the precursor has a free
terminal amino group, in an aqueous medium; and removing unreacted
precursor by enzymatic degradation or by scavenging.
According to the present process, chain elongation in solution can be
accomplished by reacting a precursor complex with the selected sequencing
segment to be added at a particular stage of the polymer synthesis. The
precursor complex can, of course, contain the initial segment of the chain
or an existing chain of segments onto which additional segments are to be
attached. For the purposes of this invention, the use of the term
"segment" or "amino acid segment" includes, where applicable, derivatives
of the segment which actually exist in the ultimate chain being fashioned.
The term "segment" or "amino acid segment" can refer to a single amino
acid or a series of amino acids.
In polypeptide synthesis the added segment is an amino acid residue and the
precursor is an elongatable peptide chain having either a free terminal
amino or carboxyl group. Peptide bond formation and chain elongation thus
is accomplished through either acylation of the amino group on the chain
by the carboxyl moiety of the acid being added (N-terminal route) or
acylation of the amino group of the added acid by the carboxyl group on
the growing chain (C-terminal route).
Preferably, the precursor is part of a larger, water soluble complex which
contains a water soluble handle attached to the precursor through the
non-elongatable end thereof. That is, for C-terminal chain growth the
chain is anchored to the handle at the N.sup..alpha. -group of the first
amino acid residue of the sequence. For N-terminal growth the chain is
anchored to the handle through the carboxyl group of the first amino acid
residue of the sequence. Chain elongation, therefore, is effected while
the precursor is a part of the complex. So that the complex is stable in
aqueous medium, attachment between the handle and precursor is preferably
covalent and effected in a manner which permits subsequent release so that
eventual recovery of synthesized, pure target fragment can be effected.
Since the growing chain is covalently complexed to the water soluble
handle, aqueous solubility of the chain during the addition reaction is
markedly enhanced, even when the chain is quite large. Also, as later
described, the use of a handle, such as a polynucleotide, can facilitate
separation of the complex from its reaction environment.
Water soluble synthetic polymers are a class of substances which can be
employed as handles. Representative examples of this class of substances
are polyvinyl alcohol, polyvinylpyrrolidone, poly(acrylamide-acrylic acid)
or polyethylenemaleic anhydride. Polyamides such as those amino acid
polymers containing a glutamic acid or aspartic acid segment also are
suitable substances for use as handles. Another useful substance which can
be used as a handle is polyethyleneglycol.
Water soluble polynucleotides also constitute a useful class of substances
which can be employed as handles. Representative examples of useful
polynucleotides, named as acids, include polyadenylic acid, polyuridylic
acid, polythymidylic acid, polycytidylic acid and polyguanylic acid.
Preferably the acids have at least ten repeating ribosephosphate moieties
and are commercially available.
The manner of achieving attachment of the first amino acid, or a short
chain peptide, to a handle should be such that the target peptide can be
thereafter removed under mild conditions. To this end, a further aspect of
the present invention resides in including, as a part of the handle, an
endopeptidase-specific spacer arm onto which the first protected amino
acid segment of the target peptide is added. The configuration of this
spacer varies with respect to the synthetic route, that is C- or
N-terminal, to be employed. The use of an endopeptidase specific spacer
arm has the advantage of mild pH and temperature conditions for removal of
the target peptide. While, for example, saponification can be used in an
N-terminal route if the first amino acid of the peptide is attached
directly to a polynucleotide handle, saponification is harsh and can lead
to racemization.
Thus, considering attachment in more detail, for synthesis via the
C-terminal route with a polynucleotide containing handle, the ribose of
the 3' end of a polynucleotide is oxidized to a dialdehyde with subsequent
coupling to an endopeptidase specific spacer arm by reductive alkylation
which involves formation of an amine, dialdehyde adduct followed by
reduction of that adduct in aqueous solution with, for example, sodium
borohydride. Therefore, the arm has, on one end, a primary amino group
reactable with the oxidized ribose moiety of the nucleotide. The other end
thereof contains a carboxyl group which, after having acylated a
N.sup..alpha. -amino group of an added acid, can be hydrolyzed by the
action of an endopeptidase. Polypeptides themselves containing a carboxyl
terminated arginine or lysine residue constitute a useful class of such
spacers, especially where the N-terminal residue, or other residues, are
from hydrophobic acids requiring no protection such as glycine, alanine,
or valine. The dipeptide, glycyl-L-arginine, is a useful spacer arm which
can be coupled, by reductive alkylation, to a polynucleotide according to
the method of Royer, et al., BBRC, 64, 478 (1975). The dipeptide spacer,
attached via a tertiary amine linkage to the nucleotide, thus has a free
carboxyl group available for addition of a first carboxyl protected amino
acid or polyacid segment for the preparation of a target peptide by the
C-terminal route.
When the use of polyethyleneglycol or polyvinyl alcohol as a handle is
desired, the alcohol group is converted to an alkoxide derivative, for
example by reaction with potassium tert-butoxide, and the alkoxide
derivative is reacted with ethyl bromoacetate to provide the carboxymethyl
derivative. This derivative is hydrolyzed to give the free acid which is
coupled to the endopeptidase specific spacer arm using a water soluble
carbodiimide activator.
Polyvinylpyrrolidone, poly(acrylamide-acrylic acid) and polyethylenemaleic
anhydride are subjected to basic hydrolysis to form derivatives having
free carboxyl groups. The polyamides are selected from those having a free
carboxyl group. These substances also are coupled with the spacer arm as
stated above.
After preparation, removal of the target peptide from its complex with the
handle can be effected by use of a highly specific endopeptidase, for
example, trypsin, which cleaves only those peptide bonds whose carbonyl is
that of arginine (or lysine). The enzyme for this cleavage can be used
either bound to a support or free in solution at mild alkaline pH.
For N-terminal peptide synthesis, preparation of the polynucleotide handle
again requires oxidation of the ribose moiety, with a spacer arm being
attached thereto through a secondary amine linkage via reductive
alkylation. However, in this case, the end of the spacer arm disposed for
covalent coupling to the first amino acid to be added must contain a free
amino group so that N-terminal chain growth can occur. Thus, the spacer
arm contains a primary amine group in both terminal positions, one to
react with the nucleotide and the other to complex with the first acid
residue of the sequence. Moreover, in order to also have the required
endopeptidase activity necessary for eventual removal of the target
peptide from the handle, the amine group used for covalent coupling to the
first acid residue of the desired sequence can be provided by the
alpha-amino group of arginine or lysine or derivatives thereof.
One convenient manner of preparing the handle for N-terminal chain growth
is to first reductively couple a short chain diamine to the aldehyde
containing nucleotide, Royer et al., supra. Thereafter, N.sup..alpha.
-amino protected arginine or lysine, eg, PC-L-ArgOH, is attached through
the acid carboxyl group to the available amine of the diamine, for example
by using a water soluble carbodiimide activator, and the protecting group
removed, for example by using the enzyme L-pyrrolidone-carboxylpeptidase,
to yield the desired handle having a free amino group available for
peptide synthesis by the N-terminal route.
When use of a vinyl polymer or polyamine handle is desired, derivatives
containing a free carboxyl group are prepared as above. The carboxyl
derivative is coupled with a diamine, such as ethylenediamine, using a
carbodiimide. The N.sup..alpha. -amine protected arginine or lysine spacer
arm is attached to the diamine as indicated above for polynucleotide
handles.
Subsequent recovery of the target peptide from the handle can be effected
by the two stage use of an arginine or lysine specific endopeptidase as
previously discussed followed by an arginine or lysine specific
exopeptidase, for example carboxypeptidase B. The first enzyme releases
the arginine or lysine terminating target peptide from the remainder of
the handle while the second removes the C-terminal arginine or lysine
residue and simultaneously liberates the target peptide. As stated above,
both enzymes may be used bound or in solution at mild alkaline pH.
If lysine or arginine is to be present in the target peptide, the side
chains of these amino acid segments must be protected. If such side chains
are not protected, the target peptide itself would be fragmented by the
endopeptidase used for separation from the handle. Typical protecting
groups are trifluoroacetyl for the epsilon amino group of lysine and nitro
group protection for the guanidinium side chain of arginine. Deprotection
of these residues can be accomplished by routine procedures well known
for this purpose after separation of the target peptide from the handle.
In building a spacer arm having arginine as the terminal acid for reaction
with the first amino acid of the target peptide, it has been found that
enhanced coupling yield is obtained if nitro protected arginine is used.
To achieve subsequent enzymatic release of the target peptide at the
arginine linkage, the nitro protecting group must first be removed. Where
the target peptide itself contains arginine, added in protected form as
above described, deprotection of arginine in the handle is accomplished
before addition of further arginine, eg, after several non-arginine amino
acid residues have been added.
A further useful type of spacer arm is one which can be chemically released
from the target peptide under mild conditions. Cyanogen bromide cleavage
at the carboxyl end of methionine (Met) is an embodiment of this aspect,
eg, with about a 100 fold excess of CnBr, cleavage is achieved in water in
about 1 hour at 30.degree. C. Thus, a spacer arm joined to the target
peptide through the Met carboxyl group is useful for C-terminal peptide
growth. For N-terminal growth, Met in the spacer arm is separated from the
first amino acid of the target peptide by a basic amino acid such as
arginine. Met cleavage with cyanogen bromide then yields the target
peptide with the Arg-homoserine lactone dipeptide terminus. The lactone is
first removed with the enzyme CPA followed by removal of Arg with CPB to
yield the target peptide. Again, if Met occurs in the sequence of the
target peptide, protection, eg, formation of the sulfoxide, is necessary.
After release from the handle, methionine sulfoxide is reduced to
methionine with a thiol such as mercapto ethanol.
In order to prevent the added amino acid from reacting with itself during
chain elongation, the primary alpha amino group or, as the case may be,
the carboxyl group thereof, as well as other reactive groups except for
the intended reactive moiety, must be appropriately blocked or protected.
As hereinafter discussed, a preferred blocking or protecting group for an
alpha amino group or carboxyl group is one which can be enzymatically
removed. Hereinafter, the symbol, .alpha., refers to the term alpha.
A preferred aspect of the present invention, particularly with respect to
peptide synthesis by the C-terminal route, resides in using non-activated
amino acid ester derivatives containing a free N.sup..alpha. -amino group
to effect reaction with the precursor. Compounds within this class include
those prepared from single amino acids as well as other compounds such as,
for example, those containing one or more peptide bonds prepared from the
same or different amino acids. These amino acid derivatives containing an
ester blocked carboxyl group can be represented as follows:
##STR3##
wherein n is an integer of zero or more; A is an amino acid side chain
which can be different in each repeating unit when n is greater than zero;
and R is a blocking group which prevents the derivative from acylating a
molecule containing a free amino group. Preferably, R is a short chain,
less than about 10 carbon atoms, straight or branched alkyl group, which
as hereinafter discussed can be removed enzymatically. Other suitable
ester groups are the benzyl and nitrobenzyl groups.
The derivatives represented by formula II above are prepared by known
esterification techniques such as the acid-catalyzed reaction of an amino
acid with an alcohol. Using these derivatives, reaction with a precursor
containing a free carboxyl group can be accomplished at ambient
temperature in water at acid pH utilizing a water soluble carbodiimide as
a coupling reagent. Because of coupling at an acid pH value, racemization
is minimized.
When an N-terminal route is selected, again the conventional means of
coupling an N.sup..alpha. -blocked acid to the free amino precursor
involving use of a water soluble carbodiimide is an attractive and
practical approach. Preferably, as will be discussed, the N.sup..alpha.
-blocking group is enzymatically removable. Another coupling means is the
use of active N.sup..alpha. -blocked amino acid derivatives to effect
reaction with the precursor. Active amino acid esters are one example of
such derivatives. As is recognized (Bodansky and Klausner, The Chemistry
of Polypeptides, ed. Katsoyannis, p. 21, Plenum, 1973), these active
esters spontaneously form peptide bonds in solution at room temperature
with minimal adverse racemization.
The active esters can be prepared by reacting the acid moiety of a
N.sup..alpha. -protected amino acid with an alcohol having substituents
which make it readily displaceable by an attacking amino group on the
precursor chain. The preparatory reaction can be accomplished in an
organic solvent in the presence of a carbodiimide. Aliphatic alcohols
containing one or more electron withdrawing groups, phenol (and
thiophenol) derivatives and hydroxylamine derivatives are useful alcohols.
Particular examples of useful active esters are those containing the
following displaceable leaving groups: cyanomethyl, carboethoxymethyl,
propargyl, N-hydroxysuccinimide, N-hydroxylphthalimide, p-nitrophenyl,
2,4,5-trichlorophenyl, as well as others given in the foregoing reference.
While the active esters are preferred, amino acid derivatives prepared
with other readily displaceable groups on the carboxyl moiety are also
useful. These groups include, for example, those such as azido, imidazole,
halo, acyl and phosphoryl.
With enzymatically deblockable N.sup..alpha. -groups, the active amino acid
derivatives, and especially the esters, constitute a useful class of
compounds for peptide synthesis using the N-terminal route. Both the
acylation reaction with the chain and the deblocking procedure for
subsequent elongation can be accomplished in solution under very mild
conditions, thus minimizing any adverse effects on the polymer being
synthesized. Also, as will be hereinafter discussed, the use of the active
esters can obviate blocking other side chains on certain amino acids which
ordinarily need appropriate protection.
In essential aspects, the compounds constituting the above class of active
esters are those which contain an amino acid derivative having an
activated terminal carboxyl group and an N-blocking group susceptible to
removal by a corresponding and specific enzyme. In one embodiment, these
compounds can be represented as follows:
##STR4##
wherein B.sub.ez is an enzymatically removable N.sup..alpha. -blocking
group; X is a group readily displaceable by an amino group; and n and A
are as identified with respect to formula II. The L-pyrrolidonecarboxyl
(pyroglutamyl) group is a useful N.sup..alpha. -acyl blocking group.
Kurath and Thomas, Helv. Chim. Acta., 56, 1658 (1973) and Doolittle,
Methods in Enzymol, 19, 558 (1970) illustrate the manner in which
L-pyrrolidonecarboxylic acid can be used to prepare the N.sup..alpha.
-L-pyrrolidonecarboxyl derivatives of amino acids.
The first of these methods involves preparing a N- protected
pyrrolidonecarboxyl N-hydroxysuccinimide ester (Z-PC-NHS) by the
dicyclohexyl carbodiimide mediated coupling of benzyloxycarbonyl (Z)
protected pyrrolidone carboxylic acid (PC) to N-hydroxysuccinimide (NHS).
The resulting Z-PC-NHS dissolved in a solvent is then coupled in aqueous
solution in the presence of a base with an amino acid (AAOH) to form the
Z-PC protected acid (Z-PC-AA-OH). Removal of the Z group is then
accomplished by catalytic hydrogenation yielding the pyrrolidonecarboxyl
protected amino acid. Alternatively, a trifluoroacetyl protecting group
can be used in place of Z which can be removed by a pH adjustment to 10.
The Doolittle method involves reaction of the t-butyl amino acid esters
with PC in the presence of a carbodiimide in an organic solvent followed
by removal of the butyl group and acid regeneration. However, low yields
are likely.
To obtain high conversion, the elongation reaction can be accomplished with
a large excess of the added sequencing segment, amounting to at least a
2:1 equivalent ratio, and preferably at least 5:1. However, the solution
after reaction may nevertheless contain unreacted transformable precursor.
In a reiterative procedure, the presence of such unreacted precursor can
produce a sequence failure. While blocking of the unreacted precursor
could be accomplished prior to the next reaction, blocking does not alter
the molecular structure of the precursor and the possibility of sequence
failure continues to exist if deblocking was to occur subsequently.
Therefore, a further aspect of this invention centers on pruning. Pruning
is selectively and chemically removing unreacted precursor from reacted
compound after the formation of the latter, thereby leaving a properly
reacted compound which is free of material containing the base molecular
structure of the unreacted precursor. Pruning is important in achieving
ultimate separation and recovery of a pure target peptide.
With particular respect to the synthesis of polypeptides, which are
elongated through acylation of a terminal amino on the precursor using
N.sup..alpha. -blocked, free carboxyl segments or through acylation of a
free N.sup..alpha. -amino group on a C-terminal carboxyl blocked segment
by a free carboxyl group on the precursor, pruning can be effected by
enzymatically hydrolyzing those precursor chains which failed to elongate
and thereby degrading such chains. The unreacted chains, of course, still
contain either a free amino group in the case of N-terminal growth, or a
carboxyl group in the C-terminal case, and thus can be enzymatically
attacked using an appropriate enzyme. On the other hand, those chains
which did properly elongate will have their terminal reactive group
(either amino or carboxyl) protected by a blocking group and will not
undergo hydrolysis.
A preferred method of enzymatic pruning is to pass the reaction solution
through a column which contains a water insoluble support material having
immobilized on its surface an enzyme which selectively hydrolyzes
substances either from the N-terminus or C-terminus. An aminopeptidase,
such as aminopeptidase M or leucine aminopeptidase, is suitable for
hydrolysis at the N-terminus (Royer and Andrews, 1973, J. Biol. Chem, 248,
1807). The hydrolysis is carried out at a temperature between 0.degree.
and 50.degree. C. and at a pH of 6.5 to 7.5. For hydrolysis directed at
the C-terminus, a carboxypeptidase such as carboxypeptidase A, B, C or Y
is useful. These Y and C enzymes, at pH 4-6, have been demonstrated as
having non-specific, C-terminal exopeptidase activity. Hayashi et al., J.
Biol Chem., 248, 2296 (1973) and Kuhn et al., Biochemistry, 13, 3871
(1974). A temperature between 0.degree. and 60.degree. C. is employed. All
of these enzymes are specific for L- amino acid residues and, as will be
hereinafter discussed, unreacted precursor will only be present in the
L-isomer form.
An alternative method of pruning involves scavenging the unreacted
precursor from the reaction solution, such as by attaching it to a water
insoluble support, and thereafter separating the solution from the
support. With particular respect to a precursor having a free amino
terminus, a manner of accomplishing this is to immobilize onto a support
an electrophilic reagent which has specific covalent reactivity for the
free terminal amino group of the unreacted precursor and, thereafter, pass
the reaction solution into intimate contact with the support in order to
bond the unreacted precursor thereto. A suitable electrophilic reagent is
the mixed disulfide formed by reaction of a thiol derivative and
mercaptosuccinic anhydride. For C-terminal scavenging, a support
containing free primary amino groups can be used in conjunction with water
soluble carbodiimides.
Subsequent to pruning of unreacted precursor, the properly elongated chains
are separated and recovered from excess unreacted amino acid. This
separation preferably is effected while the elongated complex is
reversibly coupled to an insoluble support. Reversible coupling, for the
instant purposes, is to be considered as attachment by means of a
non-covalent and non-ionic association between two substances which have a
specific affinity for each other in an aqueous medium, which affinity can
be dissipated without chemical reaction. Reversible coupling thus permits
attachment to and release from the support without the use of harsh
conditions which might adversely affect the transformed compound.
To achieve reversible coupling to the support, the precursor can be one
part of a larger, water soluble complex which contains a polynucleotide
handle attached to the precursor through the non-elongatable end thereof.
The insoluble support conveniently is contained in a column and has
covalently affixed to its surface a polynucleotide adsorbent which has
specific affinity for the polynucleotide handle complexed to the reacted
precursor. As the solution containing the complex is passed through the
column, the elongated precursor is reversibly coupled to the insoluble
support by affinitive interaction between the handle and the adsorbent.
Separation of the elongated chains, in complexed form, from chemically
unrelated substances, such as the unreacted amino acid reactant which does
not contain the covalently bonded handle, is thereby effected. The
coupling can be simply reversed by heat, the institution of a competing
association, or a change of pH. In order to achieve reversible coupling,
the polynucleotide selected as the handle should have a base which is
complementary, as to spatial arrangement and affinitive interaction, with
the base of the polynucleotide adsorbent. Examples of useful complementary
base pairs are adenine with either uracil or thymine and cytosine with
guanine. It should be appreciated that polynucleotides of the "copolymer"
type also can be used, especially when they are of the "block" form
containing alternating and repeating segments of complementary base pairs.
In this instance, of course, the same polynucleotide can be used as both
the handle and adsorbent.
When the elongation reaction is carried out using a precursor which
contains a polyethylene glycol, vinyl polymer or polyamide handle,
separation of the unreacted added amine acid segment is carried out by
conventional methods, such as dialysis, ultrafiltration, extraction, etc.
A further preferred feature of the present invention provides a method for
enzymatically deblocking the elongated complex before it is used as a
further precursor in subsequent stages. Here it is, of course, necessary
that the blocking group on the elongated complex be selectively degradable
by enzymatic action.
Turning first to that aspect of the present invention wherein chain
elongation is accomplished through the C-terminal end of a growing chain
by reaction with an amino acid segment of Formula II, the blocking group
on the acid segment is a short chain alkyl group or benzyl group coupled
to the acid through an ester linkage. One reason for this is that
deblocking after reaction with the precursor can be accomplished
enzymatically using an esterase, thereby hydrolyzing off the ester group
to yield the free C-terminal carboxyl group for subsequent elongation of
the chain. A carboxypeptidase such as carboxypeptidase Y is useful for
this purpose so long as the pH is maintained in the range of pH 8-9,
preferably at pH 8.5. At a pH of 8.5 this enzyme exhibits optimum esterase
activity to the exclusion of peptidase activity, while, as previously
discussed, at a lower pH it is exclusively an exopeptidase. The hydrolysis
reaction is carried out at a temperature between 0.degree. and 60.degree.
C.
A further significant advantage accompanying the use of this enzyme for
deblocking is that hydrolysis is only effected with respect to esters of
L-amino acids. Thus, those chains containing blocked esters of D-amino
acids are not hydrolyzed by the enzyme and are not available for
subsequent growth. As a result, a high degree of optical purity with
respect to the target peptide can be achieved.
When growth from the N-terminus is desired, the L-pyrrolidonecarboxy group
is a useful blocking agent for the .alpha.-amino group on the added acid.
The elongated complex containing this blocking group then is exposed to an
enzyme, such as L-pyrrolidonecarboxylpeptidase, which has the necessary
specificity at a temperature between 0.degree. and 60.degree. C. and a pH
of 7 to 8. This enzyme is effective in deblocking only derivatives of
L-amino acids. Thus, any D-isomer terminating chains remain blocked and
are effectively no longer available for subsequent growth.
In either of the foregoing cases, intimate contact should be achieved
between the blocking group on the chains and the enzyme in order to effect
substantially complete deblocking of the L-terminated chains. Accordingly,
it is preferred that contact be achieved while the elongated precursor is
dissolved in an aqueous medium. Moreover, in order to easily separate the
deblocked compound and the enzyme and to minimize enzyme loss, the enzyme
preferably is immobilized on a water insoluble support. Therefore, a
preferred manner of accomplishing the deblocking is to pass the aqueous
solution of the blocked elongated complex through a column which contains
an insoluble support having the enzyme immobilized thereon. As should be
apparent, with respect to C-terminal synthesis, a column containing
carboxypeptidase Y immobilized on a water insoluble support may be used
both for pruning unreacted precursor and for deblocking the terminal
carboxyl group of the blocked elongated complex merely by adjusting the pH
to achieve the desired exopeptidase or esterase activity respectively.
Turning now to the combined use of the above features in a multi-stage
polypeptide synthesis and with reference to the drawing, the initial step
is the reaction in the vessel 10 of a first amino acid derivative with a
handle to form a water soluble covalent complex containing the first amino
acid residue of the intended sequence. The added acid contains an
enzymatically removable N.sup..alpha. -amino or C-carboxy protecting group
depending on the route selected. Unreacted amino acid derivative is
removed from the reaction solution containing the initial complex by
passing the solution through a column 18 containing a water insoluble
support 16 which has immobilized on its surface an adsorbent which can
affinitively interact with the handle. Preferably, when the handle and
adsorbent are polynucleotides, the solution is maintained at about
4.degree. C. The support is then washed several times with 7.5 pH
phosphate buffer at this temperature. Thereafter, the complex is eluted
from the support as an aqueous solution free of the added acid derivative
by simply drawing buffer through the column at an elevated temperature,
preferably from 40.degree. C.-60.degree. C.
The solution so obtained then is passed through another column 20 in order
to remove the blocking group on the terminal acid segment of the complex.
Accordingly, this column contains an insoluble support 22 having
immobilized on its surface an enzyme having specificity for the protecting
group. Then the solution is introduced back into a clean reaction vessel
and, since the blocking group has been removed, chain elongation can be
effected with the second amino acid of the intended sequence. As with the
first acid, the second acid is derivatized so as to be appropriately
N.sup..alpha. - or C-blocked.
The foregoing procedure (involving steps I, III and IV) is then reiterated
to successively add the desired acids on to the complex containing the
growing polypeptide chain until a short polypeptide, for example a
hexapeptide, has been prepared. It will be noted that, up to this point,
pruning of chains which failed to react with added acid has not been
employed. As will become apparent, there is no particular advantage to be
derived from including this step (Step II) in the early stages of the
synthesis, although it can be used without any adverse consequences if
desired.
At this point, the solution recovered after separation of unreacted acid
from the short chain peptide is enzymatically treated to release the
elongated chains from the handle. The solution is passed back over the
support containing the immobilized adsorbent to remove the separated
handle and the terminal amino or carboxyl blocked short chain target
peptide then is isolated from the solution. Since the occurrence of
sequence failure in any of the foregoing steps results in the presence of
chains having less than, for example, six amino acid residues, separation
and isolation of the desired short chain peptide easily can be
accomplished by conventional techniques, such as ion exchange or gel
filtration chromatography.
The preparation of long chain polypeptides by the process of the present
invention may utilize a short chain polypeptide as a precursor. The short
chain polypeptide precursor may be prepared by the present process as
illustrated above or synthetically prepared by other methods. In addition,
naturally occuring short chain polypeptides may be used as the precursor.
In any event, the pure short chain polypeptide is attached to the handle,
enzymatic deblocking is effected, and the short chain peptide complex then
is used as the precursor for chain elongation with the next amino acid
derivative. It is at this point that the above described pruning of
unreacted chains preferably is initiated (Step II). To this end, the Step
I reaction solution, which contains the complex of the elongated
polypeptide, unreacted excess blocked acid, and unreacted complex of the
short chain polypeptide precursor from the vessel 10 is passed through
another column 12 containing an insoluble support 14 having an alpha-amino
group or terminal carboxyl-specific exopeptidase immobilized on its
surface.
On passing through this column, the unreacted precursor chains, which
contain an unblocked terminal amino or carboxyl group, are enzymatically
degraded and, therefore, pruned from the desired chain population.
Thereafter, this step (Step II) is incorporated into the above described
reiterative sequencing procedure as the chain is elongated with additional
amino acid derivatives.
Finally, after the desired target polypeptide has been prepared, the
polypeptide chains are released from the handle and the target peptide
separated and purified. It will be appreciated that, due to the
incorporation of the enzymatic degradation step (Step II) for each
sequence after the preparation of the short chain polypeptide precursor,
the final reaction solution contains very few and, preferably,
substantially no polypeptide chains which differ from the target peptide
by less than the number of amino acid residues in the precursor. Thus,
conventional separation techniques can be used.
Furthermore, it will be appreciated that the foregoing, generally described
reiterative procedure, is useful with respect to both the C-terminus and
N-terminus routes to peptide synthesis. The principal differences between
the two routes reside in the manner in which the growing chain is attached
to the handle and in the selection of blocking groups and enzymes. Also,
there can be a difference in the manner in which activation for chain
elongation is accomplished, for example the use of active esters for
N-terminal growth versus carbodiimide mediated coupling. The latter, which
is useful with respect to both C- and N-terminal growth, is preferred.
Most preferred is the C | | |
