JP2003510073A - Creating Variable Length and Variable Sequence Linker Regions for Dual or Multi-Domain Molecules - Google Patents

Abstract

  (57) [Summary] Disclosed are methods and compositions for making DNA, RNA or protein molecules having two or more nucleic acid or polypeptide domains, respectively, linked by a linker region. These methods are used to generate a random linker library of nucleic acids encoding a dual domain or multiple domain polypeptide. The linker region is characterized by both length and sequence variability. The above libraries of double or multi domain polypeptides are preferably made in plant cells.

Description

Detailed Description of the Invention

FIELD OF THE INVENTION This invention in the field of molecular biology relates to a library of dual domain nucleic acids and / or proteins in which the domains are linked by a library of linkers that differ in length and sequence.

BACKGROUND OF THE INVENTION Dual domain polypeptides, or dual domain nucleic acids encoding such polypeptides, after patterning, exhibit new and advantageous properties as compared to the original polypeptide or nucleic acid. Can have. Such a polypeptide domain is typically
Linked using a linker region or linker domain. The common nomenclature for such polypeptide constructs is D ₁ -L-D ₂ , where D ₁ and D ₂ are
There are two structural domains that are the same or different, and L is a linker. For example, the two cytoplasmic domains of the transmembrane protein adenyl cyclase linked to a linker domain form a soluble protein (Tang et al.,
Science, 268: 1769-1772 (1995)). The advantage of this soluble form of adenyl cyclase, which has enzymatic activity, is that it can be produced in much higher amounts than the native enzyme (Desuer et al., J.
． Biol. Chem. , 16967-16974 (1996)).

Another type of polypeptide produced by the joining of two domains is a single chain antibody or scFv. These single chain polypeptides contain the variable (V) regions of the heavy (H) and light (L) chains of a selected immunoglobulin (Ig) and are a fraction of that size of native Ig. Recreates the antigen-binding site of (Sker
ra, A.A. Et al., (1988) Science, 240: 1038-1041; P.
luckthun, A .; Et al. (1989) Methods Enzymol. 1
78: 497-515; Winter, G. et al. Et al. (1991) Nature, 34.
9: 293-299: Bird et al. (1988) Science 242: 423.
Huston et al. (1988) Proc. Natl. Acad. Sci. USA
85: 5879; U.S. Pat. Nos. 4,704,692, 4,853,871, 4,946,778, 5,260,203, and 5,455,030). J. A number of US patents and international patent publications by Huston et al. Disclose various two chain or two domain proteins, including single chain antibodies, linked by a linker peptide, and optionally a cleavage site. (US Pat. Nos. 5,888,773, 5,877,305, 5,861,156, 583784).
No. 6, No. 5753204, No. 5534254, No. 5525491, No. 54
Nos. 82858, 5476786, 5330902, 5302526, 5258498, 5132405, 5091513, 5013.
653, WO 9323537A1 (25-NOV-1993)).

The scFv is composed of its N-terminal V _H domain and its C-terminal V _L domain (or vice versa) joined by a peptide linker. Correct folding of the _VH and _VL regions is important for retention of antigen binding capacity by the scFv. The length and sequence of the linker region are important parameters for proper folding and biological function. The scFv chains are easier to express than larger Fv fragments or even larger Ig molecules, which are 4-chain complexes.

Ribozymes are catalytic RNA molecules that cleave other RNA molecules that contain a nucleic acid sequence that is complementary to a particular target sequence in the ribozyme. Two identical or different nucleic acid domains, such as two ribozyme domains, are linked to one or more RN
Bifunctional ribozymes can be created that can act on the A substrate structure. General methods for constructing ribozymes (harpin ribozyme, hammerhead ribozyme and RN
Ase P ribozymes) are known in the art. Castanotto
Et al. (1994) Advances in Pharmacology, 25: 2.
89-317, reviews ribozymes (including group I, hammerhead type, axhead type, harpin and RNAse P). Ribozymes have been described that can successfully target desired specific sequences (eg, HIV sequences) (H
o, A. Et al., WO9426877 (1994); Yu et al. (1993) Proc.
Natl. Acad. USA, 90: 6340-6344, and Dropul.
ic et al. (1992) J. Am. Virol. 66: 1432-1441).

Hammerhead ribozymes and harpin ribozymes are catalytic molecules with antisense and endoribonucleotidease activities. Their intracellular expression may confer, for example, significant resistance to HIV infection. Hammerhead ribozymes are described in Rossie et al. (1991) Pharmac. Ther
． , 50: 245-254; Forster et al. (1987) Cell, 48: 2.
11-220; Uhlenbeck, OC (1987) Nature, 328:
596-600; Haseroff, J. et al. Et al. (1988) Nature, 334.
: 334: 585; Dropulic et al., Supra; and Castanotto et al., Supra, and references cited herein. Harpin ribozymes are described in Hampel et al. (1990) Nucl. Acids Res. , 18:
299-304; Hampel et al., EP 0360257 (1990); Hase.
loff, J.M. P. Et al., U.S. Pat. No. 5,254,678 (1993); Kra.
us, G. U.S. Pat. No. 5,958,768 (1999); Ho, A. et al. ,
WO9426877 (1994); Ojwang et al. (1992) Proc. Na
tl. Acad. USA, 89: 10802-10806; Yamada et al. (1
994) Gene Therapy 1: 39-45; Leavitt et al. (19).
95) Proc. Natl. Acad. USA, 92: 699-703; Lea.
Vitt et al., Human Gene Therapy, 5: 1151-1120.
And Yamada et al. (1994) Virology, 205: 121-1.
26.

For convenience, the conventional single letter nucleotide code is provided in Table 1 to indicate the positions where one or more bases may be present.

[0008]

[Table 1] (When forming an r: y pair, clearly y = c when r = g, etc.) A typical substrate sequence for a harpin ribozyme is nnnng / cn * guc.
nnnnnnnn (where n * g is the cleavage site). The hammerhead ribozyme cleaves at any nux sequence. Thus, the same substrate target, guc, within the harpin leader sequence can be targeted by the hammerhead ribozyme.

The two DNA domains can also be joined to form a dual domain DNA molecule. Specific DNA domains bind proteins such as DNA polymerases, endonucleases, and transcription factors. Therefore, two connected DNs
A domain is a dual domain DNA that binds to one or more DNA binding proteins
Can be linked to form a molecule.

The person skilled in the art is aware of the presence of other nucleic acids or polypeptides which can be advantageously linked to form dual domain nucleic acids or polypeptides having one or more functions. One of ordinary skill in the art will also recognize the general preference of methods for producing such products.

The desired properties of the dual domain DNA, ribozyme or protein molecule are (1
) Constitutes a DNA domain, (2) encodes a ribozyme sequence, or (
3) It can be optimized by modifying the nucleic acid encoding the protein domain. This is accomplished via various conventional techniques. In one approach, the sequence or length of the linker region can be altered to optimize the dual domain molecule. The length and sequence of this linker region are certainly important for the function of the dual domain protein.

Methods for producing scFv double domain proteins with various peptide lengths are known in the art (eg, US Pat. No. 5,837,242). Changes in the sequence or length of the linker can deleteriously affect stability, protease sensitivity, avidity and scFv expression levels. Because the effect of changes in the linker sequence or length on the function of the dual domain polypeptide is generally unpredictable, changing a particular amino acid residue in the linker or changing its overall length. The impact of doing on bioavailability cannot generally be determined a priori.

Therefore, there is a need for a method that allows the generation of nucleic acid libraries encoding D ₁ -L-D ₂ (or higher order) structures, where L has a random length and sequence. There is sex. Dual domain proteins can be expressed from the library and analyzed for properties of interest. Once a protein is identified as having "optimal" properties, its sequence can be determined by resolving the nucleotide sequence of the clone encoding the protein. This approach eliminates the need to make and test individual clones until a clone with the desired properties is found.

The polymerase chain reaction (PCR) is used to create a library of nucleic acid products that have two domains linked by linkers that have different sequences or different lengths. No methods are currently available that allow the simultaneous introduction of both random lengths and sequences into the linker region of a population of nucleic acids.

Expression Systems Many expression systems for heterologous proteins are known in the art. They are,
Includes bacterial systems that have the advantage of rapid and abundant production, but is often limited by the inability to produce properly folded and soluble proteins (protein denaturation and refolding). If not sensitive to the cycle of). The baculovirus system drives expression through the secretory pathway of insect cells, thereby increasing the likelihood of improved protein solubility (Kretz.
schmar, T .; (1996) J. Immunol. Methods 19
5: 93-101; Blocks, B .; Et al. (1997), Immunotech
(Nology 3: 173-184). Since manipulating viruses and propagating insect cells can be time consuming and costly, this system uses expression of certain types of proteins (eg, tumor-specific or idiotype scFv).
Less suitable for individual-specific proteins such as polypeptides). Therefore, a rapid and economical expression system suitable in the art to produce useful bispecific proteins, one example of which is the idiotypic scFv vaccine for treating B-cell lymphomas. There is a need for. The present invention addresses this need.

SUMMARY OF THE INVENTION The inventors have come up with the approach of creating a library of dual domain or multiple (> 2) polypeptides derived from the appropriate encoding nucleic acid, which library comprises Characterized by members with random linkers connecting each pair, where the random linkers have varying lengths and sequences. The nucleotide sequence encoding this linker contains a repeating pattern of degenerate triplet bases. The first and second (and / or higher order) domains can be the same or different from each other. The amino acid composition of the overall linker region can include between 1 and about 20 different amino acids with each repeating pattern of modified triplet bases encoding between 1 and about 12 different amino acids. The preferred linker length varies from 1 to 50 amino acids. In one embodiment, the polynucleotide may be a single chain immunoglobulin or single chain antibody (scFv) molecule, wherein one domain is an immunoglobulin V _H domain and the other domain is an immunoglobulin. It can be a V _L domain.

More specifically, the invention relates to a library of dual domain nucleic acid molecules, each of the nucleic acid molecules comprising: (a) a first and a second domain; and (b) separating and linking the domains. Wherein the linker is (i) a member of a random library of linkers that differs in size and nucleotide sequence and consists of (ii) a repeating pattern of degenerate repeating triplet nucleotides. In the above library, the repeating pattern of degenerate repeating triplet nucleotides of the linker has the following properties: (i) position 1 of each repeating triplet cannot be the same nucleotide as position 2 of the repeating triplet; or ( ii) position 2 of each repeating triplet may not be the same nucleotide as position 3 of the repeating triplet; or (iii) position 1 of each repeating triplet may be the same nucleotide as position 3 of the repeating triplet. Not having a property.

Preferably, the nucleotide at the first and second position of each repeating triplet is selected from any two of deoxyadenosine, deoxyguanosine, deoxycytidine or deoxythymidine. In another embodiment, (i) position 1 of each repeating triplet is deoxyadenosine or deoxyguanosine; (ii) position 2 of each repeating triplet is deoxycytidine or deoxyguanosine; and (iii) Position 3 of each repeating triplet is deoxythymidine.

In another embodiment, two different repeating patterns of denatured triplet bases are combined to create the population of linkers used to create the dual domain molecule. The combination of different repeating patterns of denatured triplet bases is used to increase the complexity of linker sequences obtained from the population. Different repeats can also be used to introduce different structural or biochemical properties into the linker region. For example, denatured triplet vwc and denatured triplet nvt are used as non-template sequences. In this example, the degenerate linker sequence is (vwc) _x (nvt) _y , where x = 1-20 and y = 1-
Twenty. This combination produces linkers containing different combinations of amino acids within each repeat and different linker lengths.

In one embodiment of the above library, at least one domain binds a protein. In another embodiment, both domains bind a protein.

In yet another embodiment, at least one of the domains, preferably both, binds a nucleic acid that is not a member of the library. In any of the above nucleic acid libraries, the first and second domains are preferably coding sequences. As mentioned above, the library is preferably made in plants or plant cells. The invention also provides a double domain or multidomain nucleic acid molecule selected from the above library.

A library of dual domain polypeptide molecules is also provided, wherein each of the dual domain polypeptide molecules is described by the formula D ₁ -LD ₂ (continuing from N-terminal to C-terminal), wherein (A) D ₁ and D ₂ are polypeptide domains, and (b) L is a peptide or peptide linker that is a member of a random library of linkers that differ in size and sequence, wherein the library Is encoded by a nucleic acid consisting of a repeating pattern of degenerate repeating triplet nucleotides.

[0023]   In a preferred embodiment, the invention provides a library of multidomain polypeptide molecules.
With respect to Bulley, each of the multi-domain polypeptide molecules has a polypeptide domain
D, each pair of D being linked by a peptide or peptide linker L
The numerator is formula D_xL_y, Where x is an integer between 2 and about n
Then, n is preferably given that x and y of arbitrary values are preferably x−1.
Preferably about 20 and y is any number between 1 and (n-1); D ₁ Is attached to a single C-terminal linker; D_n(The "final" C-terminal domain) is a single
Attached to one N-terminal linker; D_n-1For each D₂Is the N-terminal and C-terminal
Randomization of linkers where each L differs in size and sequence.
Which is a member of a library of
Encoded by a nucleic acid sequence consisting of a repeating pattern of ripplet nucleotides
. A preferred library is a library of dual domain polypeptide molecules.
And each of the dual domain polypeptide molecules has the formula D₁-LD₂Described by
here,   (A) D₁And D₂Is a polypeptide domain, and   (B) L is a peptide or the absence of linkers that differ in size and sequence.
This library is a peptide linker that is a member of an artificial library.
Is a nucleic acid sequence consisting of a repeating pattern of degenerate repeating triplet nucleotides.
Is coded.

In the above library of dual domain polypeptide molecules, or library of multidomain polypeptide molecules, each linker of the library preferably comprises (i) a length of about 1 and 50 amino acid residues. And (ii) consists of between 1 and about 20 different amino acids, and (iii) each of the repeating patterns of modified triplet bases encodes between 1 and about 12 different amino acids. In the above library of dual domain or multi-domain polypeptide molecules, a repeating pattern of degenerate repeating triplet nucleotides encoding a linker preferably has the following properties: (i) position 1 of each repeating triplet , (Ii) not be the same nucleotide as position 2 of the repeating triplet; or (ii) position 2 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; or (iii) each Position 1 of the repeating triplet has the property that it cannot be the same nucleotide as position 3 of the repeating triplet.

Preferably, the nucleotides at the first position and the second position of each repeating triplet are selected from any two of deoxyadenosine, deoxyguanosine, deoxycytidine or deoxythymidine. In one embodiment thereof, (i
) Position 1 of each repeating triplet is deoxyadenosine or deoxyguanosine; (ii) position 2 of each repeating triplet is deoxycytidine or deoxyguanosine; and (iii) position 3 of each repeating triplet,
It is deoxythymidine.

The above libraries of dual or multi-domain polypeptides are preferably
Created in plant cells.

Certain embodiments of the invention include any dual domain (or multidomain) polypeptide molecule selected from the libraries described above. One embodiment is
Providing a three domain peptide selected from the above library, this domain is a dual domain scFv polypeptide linked to a third polypeptide domain. The third domain is preferably a toxin polypeptide with therapeutic utility or an enzyme with diagnostic utility, or used as a research tool. The aforementioned polypeptides are preferably produced in plant cells.

The invention further relates to a method of making a library of the above dual domain nucleic acids, the method comprising the steps of: a. Obtaining two template DNA sequences containing the first and second domains, b. Preparing an amplification primer pair for amplifying the first and second domains, each primer pair comprising an upstream primer and a downstream primer, each primer having a 5'end and a 3'end, wherein: The downstream primer of the first domain or the upstream primer of the second domain comprises a non-templated primer, comprising a repeating pattern of repetitive triplet nucleotides in which the non-templated sequence is degenerate, wherein 5 of the repeating pattern of degenerate repeating triplet nucleotides. 'At least two of the terminal triplets have identical degenerate sequences; c. Amplifying the domains using amplification primers to produce at least one population of nucleic acid domains having different lengths and sequences in the non-template sequence; and d. Ligating the nucleic acid domains produced in step (c) to produce a population of dual domain molecules.

In the above method, the repeating pattern of degenerate repeating triplet nucleotides in at least one of the primers preferably has the following properties: (i) a nucleotide where position 1 of each repeating triplet is identical to position 2 of the repeating triplet. Property; or (ii) position 2 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; or (iii) position 1 of each repeating triplet is It has the property that it cannot be the same nucleotide as position 3.

In one embodiment of the above library of dual-domain or multi-domain peptide molecules, the linker of the library of 10 or more residues in length comprises at least 3 residues and 20 or more residues. The linker in the library of lengths contains at least 4 different residues.

In the above method, at least one of the polymers preferably comprises a non-templated endonuclease recognition site.

In the above method, the template DNA sequence is preferably produced by reverse transcription of mRNA.

The method further comprises the step of ligating the group of dual domain nucleic acids into a vector and further the step of introducing the vector into a host. In these methods, the nucleic acid domain generally encodes a polypeptide domain, and the method also preferably comprises the step of expressing the dual domain polypeptide encoded by the dual domain nucleic acid. In a further step, the method may include the step of translating RNA from the vector.

For plant expression, the vector is compatible with the replication and / or expression of nucleic acids in plant cells. The method preferably comprises introducing the transcribed RNA into a plant cell and expressing a dual domain (or multiple domain) polypeptide.

The present invention also provides a group of dual domain polypeptides, or a dual domain polypeptide selected from this group is produced by the above method. Preferably, this group or selected polypeptides are produced in plant cells.

Also provided is a method of producing a dual domain (or suitable variant, multiple domain) polypeptide, which method comprises the steps of: (a) linking the first domain of the polypeptide to a linker. Binding to a nucleic acid encoding a first portion to produce a first nucleic acid construct; (b) binding a nucleic acid encoding a second portion of this linker to a nucleic acid encoding a second domain of the polypeptide, Producing a nucleic acid construct; (c) such that when expressed, the polypeptide has a first domain and a second domain separated by a linker as described by formula D ₁ -L-D ₂ . Incorporating the first construct and the second construct into a transient plant expression vector in frame; (d) transfecting a plant (or plant cell) with the vector, Transiently producing the polypeptide from :; and (e) recovering the polypeptide as a soluble, functionally folded protein.

GENERAL REFERENCES Unless otherwise indicated, the practice of many aspects of the invention will be directed to molecular biology, recombinant DNA.
Conventional techniques and immunology are used, which are within the skill of the art. Such techniques are described in detail in, for example, the following scientific literature: Sambrook, J. et al. Et al., Molecular Cloning: A
Laboratory Manual, 2nd Edition, Cold Spring Ha
rbor Press, ColdSpring Harbor, NY, 1989.
Ausubel, F .; M. Et al., Current Protocols in
Molecular Biology, Wiley-Interscience
, New York, current volume; Albers, B .;
Et al., Molecular Biology of the Cell, Second Edition,
Garland Publishing, Inc. , New York, NY (
1989); Lewin, BM, Genes IV, Oxford Universal.
rsity Press, Oxford (1990); Watson, J. et al. D.
Et al., Recombinant DNA, Second Edition, Scientific Ame.
rican Books, New York, 1992; Darnell, JO.
E et al., Molecular Cell Biology, Scientific.
American Books, Inc. , New York, NY (198
6); Old, R .; W. Et al., Principles of Gene Mani
publication: An Induction to Genetic
Engineering, Second Edition, University of Califo
rnia Press, Berkeley, CA (1981); DNA Clo
niner: A Practical Approach, Volumes 1 and 2 (D. Glover ed.); Oligonucleotide Synthesi
s (N. Gait edition, latest edition); Nucleic Acid Hybrid
ation (edited by B. Haines and S. Higgins, latest edition); Tra
Nscription and Translation (B. Hames Oy
And S. Higgins, latest edition); Methods in Enzymolo
gy: Guide to Molecular Cloning Techni
ques (Edited by Berger and Kimball, 1987); Hartlow
, E. Et al., Antibodies: A Laboratory Manual,
Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, NY, 1988), Collegia.
n, J. E. Et al., Current Protocols in Immuno
LOGY, Wiley-Interscience, New York 199
1. Proteins and functions are discussed below: Schulz, GE et al.
Principles of Protein Structure, Spri
nger-Verlag, New York, 1978, and Creight.
on, TE, Proteins: Structure and Molecule
ar Properties, W.A. H. Freeman & Co. , San
Francisco, 1983.

DEFINITIONS As used herein, the following terms have the meanings ascribed to them unless specified otherwise.

A polypeptide or protein “domain” generally refers to a particular structure and / or
Alternatively, it refers to a region of the polypeptide chain that is folded to confer a biological function (Schullz et al., Supra). Domains can be defined in structural or functional terms. A functional domain can be a single structural domain, but can include more than one structural domain. Such functions include enzymatic catalytic activity, ligand binding,
It may include atomic or intrinsic fluorescent chelation. As mentioned above, within the particular importance of the present invention, each of the V _H and V _L regions of an Ig molecule forms a single structural domain, which domain forms an antigen binding site. Works in unison. The function of the domain is largely defined by the well-defined folded shape. The most commonly used to describe proteins is the "domain"
Is also a region of a nucleic acid (a coding sequence for a polypeptide domain, or a specific function (
For example, any of the nucleic acid structures performing the catalytic function of the ribozyme or the binding of proteins) can be described. The binding domain defined by binding to the binding partner (receptor or ligand) is the V _H of the Ig molecule (see below).
Illustrated by regions and _VL regions, each of these regions forms a single structural domain that operates in concert with the formation of the antigen binding site. Another well known binding domain is the extracellular domain of cell surface receptors that bind each ligand (eg, polypeptide hormone). Furthermore, the part of the polypeptide or peptide ligand that binds to each receptor (eg erythropoietin, GM-CSF or enkephalin) is considered a functional (binding) domain. The portion of the protein that is responsive to capacity for fluorescence (eg, green fluorescent protein-GFP) is also considered a functional domain.

The binding domain of a DNA or RNA molecule is preferably a transcription factor (eg,
cAMP Response Element Binding Protei
n (CREB)), a restriction enzyme (eg EcoRI) or a DNA polymerase (eg Taq DNA polymerase) is the part of the molecule that binds to the protein.

The invention relates in part to methods for making dual domain molecules. In the preferred dual domain molecule, the linker region between the two domains is variable whereas the sequence of the linked domains is held constant.

“Template DNA” refers to DNA that is amplified by an “amplification primer pair” (a group of oligonucleotide primers used in an amplification reaction). This DNA
Can be produced by biological means (recombinant means) or synthetic means (chemical means). In addition, mRNA can be reverse transcribed to form the template DNA used in the amplification reaction.

An “upstream primer” is an oligonucleotide primer or mixture of oligonucleotide primers that anneals to the antisense strand of template DNA.

A “downstream primer” is an oligonucleotide primer or mixture of oligonucleotide primers that anneals to the sense strand of template DNA.

A “non-template sequence” is the part of an amplification primer that contains repetitive nucleotide triplets. The purpose of this sequence is to introduce variability into the linker library and is not complementary to the DNA sequence being amplified (eg, polypeptide domain coding region).

The phrase “repeating pattern of degenerate triplet bases” refers to three bases (in a non-template sequence
Set of triplets) refers to a repeated nucleic acid sequence, where the individual bases in the repeated triplet produce a repeating motif that is independently selected from the defined array.
For example, if the iterative triplet is nws (see Table 1), n is a,
It can be either c, g or t; w can be a or t and s can be g or c, giving repetitive pattern modifications. Here, these iterative triplets are adjacent to each other. These "repeating patterns of denatured triplet bases"
The non-templated sequence of the amplification primer containing is produced in vitro.

“Amplify / amplify” means that at least one round of total template DNA, or a portion thereof,
Preferably, it means a reaction that is replicated multiple times.

“Link / ligate” uses enzymatic and / or chemical methods,
It refers to a covalent bond between two or more DNA strands (3 ′ to 5 ′ ends).

A “non-template endonuclease recognition site” is a sequence within a non-template sequence that is recognized by a restriction endonuclease.

One use of the term “library” herein refers to a group, set or collection of nucleic acid molecules consisting of domains joined by linker sequences, which linkers vary in size and nucleotide sequence. And produced using the methods described. The number of library members included in a library that differ in nucleotide sequence is determined by the number of sequences included in the repeating pattern of degenerate triplet bases. The term "library" also applies to the group of polypeptides encoded by the nucleic acid library.

As used herein, a “linker” at the nucleic acid level is a nucleic acid molecule or sequence that links two nucleic acid sequences that encode two nucleic acid domains or two polypeptide domains. This linker sequence has a pattern of degenerate repeating triplet nucleotides with the following properties: (i) position 1 of each repeating triplet may not have the same nucleotide as position 2 of the repeating triplet; or (ii) each Position 2 of the repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; or (iii) position 1 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet. At the protein level, this linker is the peptide expression product of the linker nucleic acid sequence. In a preferred embodiment, the linker, Gly ₄ Ser or encoding the repeat (a Gly ₄ Ser or repeats) to eliminate sequences.

As used in the present invention, a “library of linkers” (or “linker library”) at the nucleic acid level is a set or collection or group of nucleic acid molecules or sequences, each of which contains two It is linked to two nucleic acid sequences that encode a nucleic acid domain or two polypeptide domains. Each of this library member has a degenerate repeating triplet nucleotide with the following properties: (i) position 1 of each repeating triplet may not have the same nucleotide as position 2 of the repeating triplet; or (ii) Position 2 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; or (iii) position 1 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet. At the protein level, this linker library is the set of expression products of the group of linker nucleic acid members of this library.

A “single chain antibody” (scFv; also known as “scAb”) is a single chain polypeptide molecule in which an Ig heavy chain variable (V _H ) domain and an Ig light chain variable (V _H ) domain. _{The L} ) domain is artificially linked by a relatively short peptide linker, and the scFv allows the conformation and original antibody (from which the V _H and V _L domains are derived) to retain binding ability and binding properties. It makes it possible to envisage a conformation of the antigen (or epitope) that is specific.

Description of the Preferred Embodiments The present invention employs an expression system (preferably plant-based) to target dual domains, eg, tumor-specific immunity to target individualized B-cell lymphoma. Produces raw. This plant-based transcriptional heterologous expression system described herein produces a correctly folded polypeptide that is surprisingly large and surprisingly protein immunogenic. This system allows the rapid and economical production of useful quantities of such proteins or polypeptides.

Nucleic acids encoding the dual domain products have been introduced into plants using suitable plant viral vectors as described in detail below and are properly folded in plant cells, plant parts and whole plants. Leading to the expression and rapid production of dual domain proteins.

As described herein, the selection of (1) a suitable linker and (2) a transient expression system is such that useful dual domain polypeptide molecules will be for their intended purpose. To be secreted by plant cells in a conformation that allows them to be used in solution (eg, as a tumor-specific immunogen). The scFv produced according to the present invention is advantageously obtained as a highly secreted protein species in these plant cells and is successfully taken up by these cells. For the uses described herein, it allows for simple selection and simple and rapid purification, including as a vaccine composition.

Although plant expression systems are preferred for the reasons listed herein, the present invention provides:
It is not intended to be limited to any particular system. In the production of dual-domain (or multi-domain) nucleic acids and proteins, the inventive approach for the production of random linker libraries of varying complexity has been described in other prokaryotic and eukaryotic hosts (eg, Bacteria, yeast cells or mammalian cells).

In addition to the IgV domain containing scFv vaccines described below, the present invention provides other protein antigens that can be expressed in plants in a similar fashion to achieve proper folding and enhanced immunogens. Can be applied directly to. Examples include antigens common to a particular type of tumor or family of tumors (eg, carcinoembryonic antigen (CEA)), prostate specific antigen (PSA) present in prostate cancer, tyrosinase present in melanoma and many Other known but undiscovered tumor antigens are included. Another type of clonally distributed (self) antigen is the T cell receptor (TCR) domain, which comprises part of the α, β, γ or δ chain V regions (or combinations thereof). Such TCR-based antigens can be labels and thus targets in certain T cell leukemias and lymphomas and autoimmune diseases. Accordingly, autoimmune diseases involving the use of identifiable T cell clones or specific TCR chain V regions are immunized with a polypeptide antigen corresponding to the TCR V region polypeptides produced by the approaches described herein. Regulated / treated by

Other dual domain proteins within the scope of this invention include the viral coat protein domain in combination with another domain of interest. If needed, this molecule
It is purified by utilizing the characteristics of this coat protein.

This protein domain is not limited to a domain expressed on the cell surface; a dual domain protein, where one or both of a polypeptide derived from a cytosolic protein or a protein that functions in solution form is also Is intended. Examples include cytokines such as IL-β and polypeptide hormones.

Other preferred polypeptide domains that bind as dual domain proteins or multidomain proteins using the linker approach of the present invention are transcription factors. They can be constructed such that the active domains of different transcription factors that act in concert or sequentially are combined as a single chain molecule separated by a linker. The size and complexity of this linker is based on the functional requirements for transcription factors (eg, the distance between nucleic acid binding sites for these factors if they must bind and act at about the same time). Selected. Such dual domain or multidomain polypeptides are described as exhibiting advantageous properties in promoting, activating or modulating transcription events. This is 1
The above factors must work, and one is particularly useful where it limits its concentration or utility. This limitation is overcome by producing an artificial dual domain or multi-domain transcription factor, the domain of the other limiting factor is always linked to the domain (s) or one or more non-restricted transcription factors. .

Alternatively, the transcription factor domain can be ligated using the approach of the invention to an inhibitory moiety such as a toxin, so that by binding the transcription factor domain to its target DNA, the toxin can It makes it possible to carry out functions and to inhibit transcription or to block the function of cells. The use of stimulatory or inhibitory transcription factor constructs containing an appropriately flexible linker allows the achievement of new levels of control over cellular function, which heretofore has not been possible to use mixtures of proteins, Allowed by protein domains linked by a limited array of preselected individual linkers. This random linker library approach, which can be selected by appropriate means as described herein, produces a considerable number of array selectivities.

Dual domain or multi-domain polypeptides prepared in accordance with the present invention using a random linker library approach can be delivered exogenously to target cells or inserted into target cells, and It can be combined in expression systems that function autonomously (ie, under the control of cellular factors). This is a conventional method of molecular biology using conventional vectors (eg, conventional vectors such as viral vectors that deliver nucleic acid encoding a polypeptide to appropriate cells by selective or non-selective means). Can be achieved using.

The products of the present invention are intended to be administered to, and expressed in, a dual domain (or multidomain) nucleic acid molecule (eg, a multidomain) nucleic acid molecule in a subject. Two functional DNA vaccines produced).

Unless otherwise intended, the practice of the present invention will employ conventional techniques of molecular biology (recombinant DNA techniques and immunology) that are within the skill of the art. Such techniques are described in more detail in the references listed above.

Focusing on the linker region L between the two polypeptide domains, which amino acid substitutions or adducts will be linked to specific properties of the linker and thus, for example, in biochemical or biological assays: It can be difficult to predict if a particular property of a multidomain polypeptide will be optimized when tested. L
The length and sequence of can affect the activity of the polypeptide product due to the conferring of properties such as solubility, folding, and conformation, sensitivity of proteases or expression levels.

The present invention provides an approach for making a nucleic acid library, which results in a library of polypeptides having linker regions that, when expressed, can vary in both length and sequence. . The present invention allows practitioners to generate and analyze such libraries, which offers the advantage over the prior art that either length or sequence can be varied, but not both.

The present invention is based on the use of known template nucleic acids that encode the protein domain of interest. The nucleic acid encoding the first domain comprises an upstream primer that is complementary to the antisense strand of the template, and a complementary strand to the sense strand of the template DNA, and contains a repeating triplet of nucleotides at the 5'end. Amplify in a PCR reaction using the downstream primer obtained.

The nucleic acid for the second domain is then complementary to the antisense strand of the template DNA and to the upstream primer, which may have a repeating nucleotide triplet sequence at the 5'end, and to the sense strand of the template DNA. It is amplified in a PCR reaction with downstream primers that are complementary.

Either the downstream primer of the first domain and / or the upstream primer of the second domain must contain repetitive triplets of nucleotides to provide the desired variability in length and sequence. Then, the two obtained P
Two DNAs or two Rs, in which the CR products are combined to form a nucleic acid encoding a dual domain protein, or linked by this linker region
Includes NA domain. This resulting molecule (protein, DNA or R
NA) can be analyzed by various means known to those skilled in the art.

The structures of proteins and nucleic acids and their domains are described by well-known biochemical and biophysical methods, especially X-ray crystallography and two-dimensional nuclear magnetic resonance (2D-N).
MR) spectrum). Probing the 3D structure may be sufficient for the depicted polymeric domain. For example, in the 3D structure of the dimeric enzyme glutathione reductase, each subunit has three structural domains (FAD binding domain,
It is exemplified that it is composed of a NADP binding domain and a third domain that forms an interaction between dimers). See Schulz et al., Supra. This Ig
The _VH and _VL domains cooperate to form the antigen binding pocket of an antibody. Thus, these structural domains fold into distinct shapes that are important for molecular function.

Domain Cloning Domains can be isolated by any of several techniques. Generally, the nucleic acid sequence encoding the polypeptide (or RNA) domain of interest is cloned from a suitable cDNA library or a genomic DNA library based on hybridization with an oligonucleotide probe representing this domain.

For the present invention, preferred nucleic acids and proteins are mammalian sequences,
More preferably, it is a human sequence.

Alternatively, the DNA is isolated by amplification techniques using oligonucleotide primers starting with a DNA or RNA template (eg D
Eiffenfach et al., PCR Primer: A Laboratory.
See Manual (1995)). These primers can be used to amplify either partial or full-length coding sequences that may make up the probe (ranging up to about several thousand nucleotides in length). The resulting probe sequence is then used to screen a mammalian library for full length nucleic acid of interest. The use of synthetic oligonucleotide primers and amplification of RNA or DNA templates is described in US Pat. Nos. 4,683,195 and 4,683,202;
PCR Protocols: A Guide to Methods and
Applications (Edited by Innis et al., 1990). PC
From mRNA using methods such as R and ligase chain reaction (LCR),
The nucleic acid sequences of the domains can be amplified from the cDNA or directly from a genomic or cDNA library. Degenerate oligonucleotides can be designed to amplify domain homologs using known sequences encoding the domain. Restriction endonuclease sites can be incorporated into the primer. The gene amplified by the PCR reaction can be purified on an agarose gel and cloned into an appropriate vector.

In expression cloning, antibodies (or other binding pairs) specific for epitopes of the expressed polypeptide are used as probes to isolate nucleic acids from expression libraries. Polyclonal or monoclonal antibodies (mAbs) can be produced by immunizing with one or more peptide fragments of the domain to be cloned.

A library is used to isolate polymorphic variants or alleles of the gene encoding the polypeptide domain of interest using nucleic acid probes, preferably oligonucleotides, under preferred stringent hybridization conditions. To screen. Alternatively, antibody-based expression cloning allows cloning of polymorphic variants or allelic variants or interspecies homologs.

The selection of the source of the cDNA library and its product from the mRNA is performed using conventional methods (Gubler et al., Gene 25: 263-26.
9 (1983); Sambrook et al., Molecular Cloning.
, A Laboratory Manual (2nd Edition 1989); Curr
ent Protocols in Molecular Biology (A
Usubel et al., 1994 or latest version).

The method for preparing the genomic DNA library is conventional in the art. For example, DNA extracted from tissue can be mechanically sheared or enzymatically digested to obtain a fragment of approximately 12-20 kb, which fragments are separated by gradient centrifugation and suitable. Various expression vectors. These vectors are packaged in phage in vitro. Recombinant phage are analyzed by plaque hybridization (Benton
Et al., Science 196: 180-182 (1977)). Colony hybridizations are described, for example, in Grunstein et al., Proc. Natl. A
cad. Sci. USA. , 72: 3961-3965 (1975).

Synthetic oligonucleotides can be used to construct recombinant “genes” for use as probes or for expression of domain polypeptides.

The oligonucleotides were synthesized using an automated synthesizer (Van Devanter et al., Nucl.
eic Acids Res. 12.6159-6168 (1984)) using the solid-state phosphoramidite triester method (Beaucage et al., Tetrah).
edron Letts. 22: 1859-1862 (1981)). Purification of oligonucleotides is typically by natural acrylamide gel electrophoresis or by anion exchange HPLC (Pearson et al., J Chrom. 255.137-149 (1983).
).

The sequences of cloned genes and synthetic oligonucleotides are described in the chain termination method (Wallace et al. , Gene 16: 21-26 (1981)).

Nucleic acid encoding the desired polypeptide is typically cloned into an intermediate vector prior to transfection or transformation of prokaryotic or eukaryotic cells for replication and / or expression of the nucleic acid. To be done. These intermediate vectors (eg, plasmids or shuttle vectors) are typically for use in prokaryotic cells.

[0083]   (Linker area)   The function of the linker L is that the first and second peptides (or cores)
Acid) domain, allowing the two domains to fold correctly,
Thereby assembling into a functional molecule. Amino acid linker L is V _H And V_LIn scFv embodiments linking domains, L is 1 and about 50
The length of residues between can vary. Each L is preferably between 1 and about 20
It consists of different amino acids, and repeats bases of each degenerate triplet base.
The turns encode between 1 and about 12 different amino acids. The optimal linker is
, V_HAnd V_LContributed significantly to the correct folding of the domain, resulting in
The scFv is (a) soluble and (b) bound to the antigen, or (c)
And) act as an antigen to elicit a relevant immune response.

In one embodiment, the linker is resistant to cleavage by the protease that the end product would be expected to encounter when used.

In contrast, the linker also contains amino acids, or short sequences that function as cleavable sites for proteases that may be used to separate one or several domains from each other at the appropriate time. Can be designed to be incorporated.

In addition, the linker can be designed to confer an affinity to another molecule or matrix that facilitates subsequent purification of the expression of the fusion domain based on the properties of the linker. One example includes the incorporation of a histidine (His) tag that allows purification on metal (eg, nickel) affinity columns. Other affinity tags are well known in the art and need not be described in the specification.

Depending on the two domains that are linked, the sequence and length of L can vary widely.

Linkers may be selected based on their ability to fuse two polypeptide domains at the same time and facilitate property-based purification and characterization of one (or both) domains. Examples include the fusion of a selected protein domain with glutathione S-transferase (GST), which can then be purified on a glutathione-agarose affinity matrix (Smith.
(1988) Gene, 67: 31-40). The linker used by Smith et al. Was then modified by Guan et al. (Anal. Biochem. 192: 26.
2-267 (1991)), the amino acid sequence PGISGGGG.
A glycine-rich stretch known as "glycine flexor" having G [SEQ ID NO: 1] was introduced. Such linkers, which are within the scope of the present invention, facilitate the cleavage of GST from its fusion partner, in this example the protein tyrosine phosphatase.

Vectors for producing these types of fusion proteins are well known in the art and many are commercially available. For example, New England Biola
bs provides pMAL-p2 and this vector encodes a maltose binding protein that can be fused to a domain sequence cloned into this vector.
In pMAL-p2, the amino acid sequence of the linker between the maltose binding protein and the added domain is NNNNNNNNNNNLGIEGR [SEQ ID NO: 2]. The stretch of asparagine facilitates purification of the fusion protein on an amylose affinity column.

The linker used to link the Ig V _H and V _L domains to the scFv is the 15 amino acid sequence GGGGGSGGGGSGGGGS (SEQ ID NO: 3).
And is usually written as (Gly _4- Ser) ₃ . Many other linkers for scFv generation are described by Lawrence et al., FEBS Letters, 425: 4.
76-484 (1998), Solar et al., Protein Engineer.
ing, 8: 717-723 (1995), Alfthan et al., Protein.
Engineering, 8: 725-731 (1995), Newton et al., Biochemistry, 35: 545-553 (1996), Ager et al., Human Gene Therapy, 7: 2157-2164 (1996).
) And Koo et al., Applied and Environmental M.
Microbiology, 64: 2490-2496 (1998). The library approach of the present invention produces many useful linkers beyond those described above.

Making Variable Lengths and Sequences in the Linker Region A preferred approach is to make a library of _two domain polypeptides (D ₁ -LD ₂ ), where each live Rally members differ from everything else in L. That is, randomness between domains is found in the linkers that connect them. This is particularly allows the creation of an array of D ₁ -L-D ₂ product in plant expression systems from which folded optimally, one or an array of products which function optimally may be selected.

In this approach, two cloned domains are amplified and a linker of variable length and variable sequence is introduced between them using an amplification method such as PCR. To achieve this, a portion of the 3'end of the downstream primer for the upstream domain and the 3'end of the upstream primer for the downstream domain are complementary to each domain sequence to be amplified. (
"Downstream" and "upstream" are with respect to the linker). However, a portion of the 5'end of the downstream primer for the upstream domain and / or the 5'end of the upstream primer for the downstream domain is not complementary to each domain being amplified. The non-complementary segment of this primer (called the "non-template sequence") comprises a repeating pattern of degenerate triplet bases connecting the upstream to the downstream domain at the nucleic acid level.

The upstream and downstream primers for amplifying D ₁ and D ₂ are
Mixed with NA primer and other necessary reactions. For more information, see Inn
See is et al., supra. The reaction mixture is subjected to multiple temperature cycles, D
The NA duplex is melted allowing annealing of the primer to the template and polymerization of the PCR product. During the first cycle, the DNA polymerase performs the "first strand" synthesis until the temperature rises sufficiently to melt the duplex. Then, when the temperature drops below the annealing temperature, the primer anneals to the first strand DNA. The DNA polymerase then creates a "second strand" when the cycle polymerization temperature is reached. This results in an exponential accumulation of amplified domains. Due to the non-template sequences, the amplified domain-encoding DNA forms a population of molecules (libraries) with a degenerate base repeat pattern at the 3'end of the upstream product and at the 5'end of the downstream product.

Due to the nature of the repeated pattern of degenerate triplets in the non-template sequence of the amplification pair, PCR products are diverse in sequence and length in the L region. The length variability is likely due to duplex formation of the L region of the primer with a bubble or loop in the middle due to base pair mismatches. 3'-
The 5'exonuclease and 5'-3 'polymerase activities act to delete or extend the length of the primer sequence.

To shorten the L sequence, a primer containing a repeating triplet is annealed to the complementary strand, which is already incorporated into the L sequence. The degenerate primer can then anneal to form a duplex with bubbles at the site of the unpaired base and release the unpaired 3'extension (overhang), as shown below. Get (underlined).

[0096]   Duplex with bubble and 3'overhang

[0097]

[Chemical 1] (Upper sequence is SEQ ID NO: 50; lower sequence is SEQ ID NO: 51) Enzymes, such as PFU or Vent, that have 3'-5 'exonuclease activity, have their duplex annealed. Cleavage the 3'extension of the complementary strand in the 5'direction until reaching the portion. In this manner, one or more triplet repeats
It can be removed from the PCR product, thereby shortening the peptide linker L by one (or more) amino acid.

For the extension of linker L, the “upper” strand can anneal to the complementary strand, resulting in
Duplexes with 5'extensions are formed as follows: Duplexes with bubbles and 5'overhangs.

[0099]

[Chemical 2] (The upper sequence is SEQ ID NO: 50; the lower sequence is SEQ ID NO: 51). A polymerase (eg, Taq polymerase) present in the amplification reaction can extend the PCR product by one or more triplet repeat codons. Due to its 5'-3 'polymerase activity, the enzyme can satisfy a 5'extension, thereby extending the linker region by one or more repeating triplets. This extends the peptide linker by one or more amino acids. The polymerase in PCR is 3'-5
In the absence of'exonuclease activity, and in the absence of an enzyme with 3'-5 'exonuclease activity, extension of triplet nucleotides should occur.

In order to promote bubble formation, at least one 5'end must contain the same degenerate base in at least two stop codons to prevent slippage. That is, two triplet repeats having the same sequence at the 5'end of at least one of the primers used to amplify the domain (eg, 5'rst-rst3 ', or 5'ysa-ysa3', etc.) Must exist.

To retain the proper reading frame, which is important when the fusion nucleic acid expresses the protein (when using scFv), some rules are for the L region of the primer, It should be observed when designing the degeneracy of the non-template region. Degenerate triplet iterations should follow one of the following rules: (a) position 1 of the triplet may not contain the same base as position 2; or (b) position 2 of the triplet is position It cannot contain the same base as 3; or (c) position 1 of the triplet cannot contain the same base as position 3. For example, the repeating triplets rst and ysa follow these rules. The following combination of bases meets these rules: rst = agt, act, ggt
, Gct and ysa = tca, tga, cca, cga. Other degenerate sequences also meet these rules. For example, str (which is gta, gtg, cta,
Or ctg) or ayr (which can be aca, acg, ata, or atg) can function as a repeating triplet.

Another degenerate triplet sequence useful in the present invention is nvt, which can be any of 12 different codons encoding 11 different amino acids. The degenerate triplet nws can be any of 16 different codons encoding 12 different amino acids. The degenerate triplet csy does not follow this rule. Because it can be ccc (which does not follow). Similarly, any other degenerate sequences that may be triplets of the same base (ie ccc, aaa, ggg, or ttt) do not follow these rules and are therefore excluded as repeating triplets.

Restriction enzyme recognition sequences can be incorporated into the primers to facilitate, for example, cloning and orientation of the IgV region domains (or any other polypeptide domain) relative to each other. For example, restriction endonuclease sites, 'be incorporated into end, this 5 upstream domain 5' upstream amplification primer for D ₁ domain end, the 5 'ends of the restriction vector fragment is subcloned Facilitate the connection of. Similarly, the same or different restriction sites can be incorporated at the 5'end of the downstream amplification primer for the downstream domain. The resulting PCR product can then be restricted with each endonuclease for subsequent ligation into a vector having a sequence complementary to the PCR product. Alternatively, the same restriction sites can be used and subclones screened by DNA sequencing, PCR, restriction enzyme digests, etc. to determine if the correct orientation was achieved.

Ligation of PCR Product The 3 ′ end of the upstream PCR product and the 5 ′ end of the downstream PCR product can be ligated to each other (Methods in Enzymology: Guide to M).
molecular Cloning Techniques, Berger et al.,
Ed., 1987). If both ends of these products are blunt, the 5'phosphate can be phosphorylated with T4 polynucleotide kinase and the reaction products can be ligated with T4 DNA ligase. If the ends of the PCR product can be complementary or can be made complementary through restriction endonuclease digestion, cohesive end ligations are performed where the complementary ends are ligated with T4 DNA ligase. . Similarly, the 5'end of the upstream PCR product and / or the 3'of the downstream PCR product.
The 'end can be ligated to the restriction vector with a blunt end or a sticky end ligation.

Multiple PCR reaction products of D ₁ and D ₂ can be combined to increase the sequence and length complexity of the linker region of a population of dual domain molecules (eg scFv). For example, a PCR reaction of D ₁ and / or D ₂ (where the degenerate triplet repeats 6 times) is a PCR reaction of D ₁ and / or D ₂ (where the degenerate triplet repeats 9 times). ) And can be ligated to an appropriate vector. The combination of PCR products increases the length and sequence complexity observed in the L region.

The complexity of the resulting linker sequences in a population or "library" can be predetermined by the number of different amino acids designed into the non-template sequences of the PCR amplification primers used to amplify the domain. The number of amino acids encoded by the non-templated sequence is determined by the nucleotide degeneracy designed for the triplet of each codon.

In one example, the desired complexity of linker sequences present in the library is
Limited to two amino acids (Ala and Gly). A preferred non-templated sequence for this linker combination is the codon triplet gst (= gct and gg
t), where gct encodes Ala and ggt is Gly.
Code

In the second example, the desired complexity of the linker sequences present in the library is
Increased to 6 amino acids (Ala, Gly, Ser, Thr, Lys and Asp). A preferred non-templated sequence for this linker combination is the codon triplet rvt (= gct, ggt, agt, act, aat and gat).
, Where the following amino acids are encoded:

[0109]

[Chemical 3] Using the same approach, higher order multidomain polypeptides (eg,
3 domain polypeptide or 4 domain polypeptide). These can include all different domains, or one or more domains can be repeated. The general structure for such a molecule is (where:
D is a polypeptide domain and L is a linker):

[0110]

[Chemical 4] Different linkers between different domains can vary in complexity. This depends on the structural relationships required for the proper functioning of each domain for its intended purpose. Thus, in the example of an scFv molecule with a single idiotype or a single ligand binding specificity, these two domains must function in agreement for proper binding. ScFv with desired binding specificity
A three-domain polypeptide (where the third domain D ₃ is the toxin)
"Interaction" between the toxin domain and either of the two binding domains
There are some restrictions on. In this case, linker L ₂ between one toxin domain of the scFv domain may be different with the linker L ₁ between two domains containing scFv polypeptide (D ₁ and D _2), from a linker L ₁ It doesn't have to be complicated.

In a library of multi-domain polypeptides, every pair of domains is
It need not be linked by a linker according to the invention. Thus, two or more adjacent domains may be directly linked (1) so that they may occur in their native state (if they are derived from a natural dual domain or multidomain protein). ), Or (2) may be joined by “conventional” linkers well known in the art. In yet another embodiment, a particular linker identified using the present invention and derived as a member of a random linker library is a random linker between two given domains in a multidomain polypeptide. It may preferably be chosen for use as a non-linker. These various embodiments may be presented in the following (non-limiting) manner:

[0112]

[Chemical 5] In the four formulas shown above, L ₁ and L ₂ represent random linker members of the library of the invention. All multiple domains shown linked to adjacent domains without the use of a link L may (1) be linked directly to each other as described above; (2) by conventional linkers known in the art. Or (3) may be linked by a fixed linker found in a random linker library according to the present invention, but inserted at a particular position as a given non-random, unaltered linker. May be. As described in the Summary section above, the multi-domain polypeptides herein can consist of up to about 20 domains. For example, a 10 domain polypeptide can have anywhere between 1 and 9 linkers according to the invention. When a 10 domain polypeptide has one such linker L ₁ linking two domains, the other eight domains are either directly linked to each other or linked by conventional or other predetermined linker groups. It is either

Expression Systems for the Production of Dual Domain Polypeptides Many well known heterologous expression systems in bacteria, insects, mammals and plants have been discussed above, each with its advantages and disadvantages. . The invention is particularly suitable for plant expression.

Many transformation methods allow the expression of heterologous proteins in plants. Some relate to the construction of transgenic plants by incorporating into the plant genome a DNA sequence encoding the protein of interest. The time required to obtain transgenic plants would be too long for certain embodiments of rapid production such as tumor vaccine polypeptides. An attractive solution, an alternative to such stable transformation, is the transient transfection of plants with expression vectors. Both viral and non-viral vectors that enable such transient expression are available (Kumagai, MH et al. (1993).
Proc. Nat. Acad. Sci. USA 90: 427-430; Shi.
vprasad, S .; Et al. (1999) Virology 255: 312-32.
3; Turpen, T .; H. Et al. (1995) BioTechnology 13
: 53-57; Pietrzak, M .; Et al. (1986) Nucleic Aci
d Re. 14: 5857-5868; Hooykaas, P .; J. J. And Schillperort, R .; A. (1992) Plant Mol. Bio
l. 19: 15-38), but viral vectors are easier to introduce into host cells, spread by infection and amplify expression, and are therefore preferred.

The chimeric genes, vectors and recombinant viral nucleic acids of the invention are constructed using conventional techniques of molecular biology. Viral vectors expressing heterologous proteins in plants are preferably (1) a native viral subgenomic promoter (Dawson, WO et al. (1988) Phytopathology).
78: 783-789 and French, R. et al. Et al. (1986) Science
231: 1294-1297), (2) preferably one or more non-native viral subgenomic promoters (Donson, J. et al. (1991) Pro.
c. Nat. Acad. Sci. USA 88: 7204-7208 and Ku.
magai, M .; H. (1993) Proc. Nat. Acad. Sci. U
SA 90: 427-430), (3) sequences encoding viral coat proteins (whether native or not), and (4) nucleic acids encoding the desired heterologous protein. Vectors containing only non-native subgenomic promoters can also be used. The minimal requirement for the vector of the present invention is the combination of the replicase gene and the coding sequence to be expressed, driven by a native or non-native subgenomic promoter. This viral replicase is expressed from the viral genome and is required to replicate extrachromosomally. Subgenomic promoters express foreign or heterologous coding sequences and useful genes for any protein (eg, those that encode viral proteins that facilitate viral replication, proteins required for translocation, capsid proteins, etc.). To enable. The viral vector is encapsidated with the encoded viral coat protein, yielding a recombinant plant virus. This recombinant virus is used to infect suitable host plants. Thus, the recombinant viral nucleic acid can replicate in host plants, become globally prevalent, and direct the synthesis of RNA and proteins to produce the desired heterologous protein in the plant. In addition, the recombinant vector maintains the non-viral heterologous coding sequences and regulatory elements for a sufficient time for the desired expression of the coding sequence.

The recombinant viral nucleic acid is prepared from any suitable plant viral nucleic acid, although members of the tabamovirus family are preferred. The native viral nucleotide sequence can be modified by known techniques, provided that the required biological function of the viral nucleic acid (replication, transcription, etc.) is preserved. As mentioned above, one or more subgenomic promoters may be inserted. They may regulate the expression of flanking heterologous coding sequences in infected or transfected plant hosts. The native viral coat protein may be encoded by this RNA,
Alternatively, this coat protein sequence can be deleted and a different plant virus ("
It may be replaced by a sequence encoding a coat protein of "non-native" or "foreign virus"). The foreign viral coat protein gene can be placed under the control of either a native or non-native subgenomic promoter.
The foreign viral coat protein should be capable of encapsidating recombinant viral nucleic acids to produce functional, infectious virions. In a preferred embodiment, the coat protein is a foreign viral coat protein encoded by a nucleic acid sequence, which nucleic acid sequence is flanked by either the native viral promoter or a non-native subgenomic promoter. Preferably, the nucleic acid encoding the heterologous protein, exemplified as an immunogenic polypeptide expressed in plants, is under the control of the native subgenomic promoter.

A key element of the invention, which is partly responsible for the proper folding and abundant production of heterologous proteins (eg immunogenic scFv polypeptides), is the ability to engineer newly synthesized proteins. The presence of a signal peptide sequence directed to the plant secretory pathway. The sequence encoding the signal peptide is fused in frame with the DNA encoding the expressed polypeptide. The preferred signal peptide is α
-Amylase signal peptide.

In another embodiment, sequences encoding transfer proteins are also incorporated into the viral vector. This is because transfer proteins facilitate the rapid cell-to-cell transfer of virus in plants and facilitate the overall infection of the entire plant.

Either RNA or DNA plant viruses are suitable for use as expression vectors. DNA or RNA can be single-stranded or double-stranded. Single-stranded RN
A virus may preferably have a + strand, but-strand RNA viruses are also contemplated.

Recombinant viral nucleic acid is prepared by cloning in a suitable producer cell. Conventional cloning techniques (for both DNA and RNA) are well known. For example, using a DNA virus, an origin of replication compatible with the producer cell is
It can be spliced into viral DNA.

With RNA viruses, a copy of the full-length DNA of the viral genome is first prepared by conventional procedures: eg, viral RNA is reverse transcribed to produce DNA.
A polymerase is used to form subgenomic fragments of double-stranded DNA. This DNA is cloned into a suitable vector and inserted into producer cells. This DNA fragment is mapped and combined with the appropriate sequences to produce a full-length DNA copy of the viral genome. A subgenomic promoter sequence (DNA), with or without the coat protein gene, is inserted at a non-essential site in the viral nucleic acid, as described herein. Non-essential sites are those that do not affect the biological properties of the viral nucleic acid or the assembled plant virions. The cDNA complementary to the viral RNA is placed under the control of a suitable promoter so that (recombinant) viral RNA is produced in the producing cell. If RNA must be capped for infectivity,
This is done by conventional techniques.

Examples of suitable promoters include the lac promoter, lacuv5 promoter, trp promoter, tac promoter, lp1 promoter and ompF promoter. Preferred promoter is phage SP6
It is a promoter or T ₇ RNA polymerase promoter.

The producing cell may be prokaryotic or eukaryotic, and may be Escherichi.
a coli, yeast, plant cells and mammalian cells.

Many plant viral vectors are available and are well known in the art (Grierson, D et al. (1984) Plant Molecular Bi.
LOGY, Blackie, London 126-146; Gluzma.
n, Y et al. (1988) Communications in Molecular
r Biology: Virtual Vectors, Cold Spring
(Harbor Laboratory, New York, pp. 172-189)
. The viral vector and its control elements must obviously be compatible with the plant host to be infected. Suitable viruses are: (a) viruses from the group of tobacco mosaic viruses (TMV) (eg TMV,
Tobacco light green mosaic virus (TMGMV), cowpea mosaic virus (C
MV), alfalfa mosaic virus (AMV), cucumber green spot mosaic virus-watermelon strain (CGMMV-W), oat mosaic virus (O)
MV)), (b) viruses from the group Brome mosaic virus (BMV), such as BMV, broad bean mosaic virus and cowpea yellow mosaic virus, (c) other viruses (eg rice necrosis). Virus (RNV), twin virus (g
eminivirus) (eg, tomato golden mosaic virus (TGMV))
, Cassava latent virus (CLV) and maize streak virus (MSV)
).

A preferred host is Nicotiana benthamiana. A host plant, as the term is used herein, can be a whole plant, plant cells, leaves, rhizomes, flowers or any other plant part. The plant or plant cell is grown using conventional methods.

N. A preferred viral vector for use with Benthamiana is the expression vector pBSG1250 (pTTOSA derivative) containing a hybrid fusion of TMV and tomato mosaic virus (ToMV) (Kumaga).
i, MH et al., (1995) Proc. Natl. Acad. Sci. USA92
: 1679-1683). The inserted subgenomic promoter must be compatible with the TMV nucleic acid and be correctly located in the infected plant (eg,
It must be able to direct the transcription of (adjacent) nucleic acid sequences. Coat protein is
Allows the virus to infect the entire plant host. TMV coat protein is described in N. promotes systemic infection of benthamiana.

Infection of plants with recombinant viral vectors is accomplished using a number of conventional techniques known to promote infection. These include, but are not limited to, leaf abrasion, abrasion with solutions and high velocity water sprays. Viral vectors can be delivered by single leaf manual, mechanical or high pressure spray.

Purification of Protein / Polypeptide Products Dual domain polypeptides produced in plants are preferably recovered and purified using standard techniques. Suitable methods include homogenization or grinding of plants, or production of plant parts in liquid nitrogen, followed by protein extraction. If it is not desirable to homogenize the plant material for some reason, the polypeptide can be removed by vacuum infiltration and centrifugation followed by sterile filtration. Protein yield can be estimated by any acceptable technique. Polypeptides are purified by size, isoelectric point, or other physical property. Further purification steps may be performed following isolation of the total secreted protein from plant material. Immunological methods using antibodies specific for the epitope of the desired polypeptide (eg,
Immunoprecipitation) or, preferably, affinity chromatography may be used.

To facilitate purification, viral vectors can be engineered so that the protein is produced with an affinity tag that can be leveraged during the purification step. An example of such a tag is the histidine (His) tag, which allows purification on metal (eg nickel) affinity columns. Other affinity tags are well known in the art and need not be described herein.

A variety of solid supports can be used in the method of the invention: agarose (
®), Sephadex ®, derivatives of cellulose or other polymers. For example, using Staphylococcal protein A (or protein L) immobilized on Sepharose®, first incubate the protein with a specific antibody in solution and then mix the mixture with immobilized protein A
The target protein may be isolated by contacting with it, which binds and retains the antibody-target protein complex.

Using any of the aforementioned methods or other well known methods, the polypeptide is purified from plant material to a purity greater than about 50%, more preferably greater than about 75%, and even more preferably about 95%. Purified to a purity higher than%.

Determination of Correct Folding The conformation of proteins in solution is important for certain properties such as immunogenicity. The conformation of related epitopes of the dual domain polypeptide in solution is preferably similar to or mimics the same epitope of the native protein. By producing polypeptides in plants and targeting them to the secretory pathway of plants, the present invention ensures that the polypeptides are secreted in a soluble, optimally folded form.

A preferred reagent used in determining correct folding is a particular antibody, preferably a mAb, which binds to an epitope of the polypeptide if the (1) chain is correctly folded. , (2) Does not bind when the epitope is denatured. Antibodies are used in any of a number of immunological assays, including dot blots, western blots, immunoprecipitations, radioimmunoassays (RIA), and enzyme immunoassays (EIA) (eg enzyme linked immunosorbents). Assay (ELISA)). In a preferred embodiment, Western blots and ELISAs when such antibodies are available.
Is used to assess the correct folding of the relevant part of the dual domain (or multidomain) polypeptide produced in plants.

Further Analysis of Dual Domain Molecules The DNA encoding the dual domain polypeptides can be sequenced to give the deduced amino acid sequence of the encoded product. If the DNA molecule has been subcloned, it can be excised from the vector using restriction enzymes, and the resulting fragments analyzed on an agarose gel to determine the size of the fragments.

If the DNA molecule itself has a binding domain of interest, the subcloned DNA molecule (or excised fragment) can be assayed for binding to a related ligand.

If the DNA molecule encodes a dual domain ribozyme, the ribozyme RNA can be transcribed from the vector. The coding sequence can be excised with restriction enzymes and contacted with RNA polymerase (along with ribonucleotides and other necessary factors) to transcribe the dual domain RNA. The ribozyme can then be quantified and its enzymatic activity measured in an appropriate assay.

The DNA molecule encoding the dual domain polypeptide is first expressed. If desired, the DNA further comprises sequences that have been modified to allow or optimize expression in a suitable host, or in vitro transcription / translation system. Once expressed,
The polypeptide is subjected to a suitable functional assay (eg, measuring enzymatic activity (of either domain)). Also, the quantitative physical properties of this dual domain polypeptide can be determined, for example, by SDS-PAGE. Electrophoretic separation can be followed by direct staining of the protein or probing with Western blots and appropriate antibodies that recognize epitopes in either domain. If the domain has binding activity, or has other functions as described above, this can be measured by conventional means.

Having generally described the invention so far, the invention will be more readily understood with reference to the following examples, which are provided by way of illustration and unless otherwise specified, the present invention. It is not intended to limit the invention.

The following examples are provided by way of illustration only and not as a limitation. Those of ordinary skill in the art will readily recognize a variety of non-critical parameters, which can be varied or modified to obtain essentially similar results.

Example 1 Generation of Autologous / Tumor Antigens from a Single Patient (CJ) Including the Idiotype of CJ B Cell Lymphoma An immunogenic scFv protein designated "CJ" was added to human lymphoma patients Derived from (with initial CJ), and this protein had (Gly ₄ Ser) ₃ as its linker. Patient CJ had been treated with an early passive immunotherapy trial. CJ molecule (specifically, its V region epitope (s))
) Is recognized by an anti-Id mAb designated 7D11. McCormick
, AA et al., Proc Natl Acad Sci USA (1999) 96:
See also 703-708.

In an early attempt to generate a human scFv polypeptide, the CJ V region gene was sequenced and cloned into a bacterial expression system using the (Gly ₄ Ser) ₃ linker. Targeting to the periplasm using the PEL-b reader, but C
J scFv proteins were separated by insoluble inclusion bodies. No anti-CJ anti-idiotypic antibody response was detected when mice were immunized with CJ scFv generated in bacteria.

Derivatives of CJ were generated by producing a linker with random length and sequence that is part of the general PCR-based cloning strategy described herein.

Four reactions were run. In the first and second, the following synthetic oligonucleotides:

[0144]

[Chemical 6] Was used to amplify the V _H domain-encoding sequence from a cDNA clone of lymphoma cells from patient CJ. The SphI restriction site is underlined. X of the first reaction
Was 6:

[0145]

[Chemical 7] . In the second reaction, x is 9 and SEQ ID NO: 7:

[0146]

[Chemical 8] (Generally, the number of triplets (x) can be from 1 to about 50).

In the third and fourth PCR reactions, the sequence encoding the V _L domain was replaced with the following synthetic oligonucleotide:

[0148]

[Chemical 9] Was used to amplify from a CJ cDNA clone. The AvrII site is underlined. In the third reaction, z was 6:

[0149]

[Chemical 10] In the fourth reaction, z was 9 (SEQ ID NO: 11):

[0150]

[Chemical 11] (In general, the number of triplets (z) can be from 1 to about 50).

Following amplification, the four PCR products were purified and digested with SphI for the V _H chain PCR product and AvrII for the V _L chain PCR product. The digest was electrophoresed on an agarose gel and the four digested PCR fragments were purified, combined and the Geneware® expression vector pBSG125.
0 (pTTOSA derivative). This expression vector is TMV
And a ToMV hybrid fusion (Kumagai et al., Supra) and digested with the restriction enzymes SphI and AvrII. In certain Geneware® vectors, the SphI site is downstream of the TMV U1 CP subgenomic promoter and α-amylase signal peptide sequence. Primer V _H F definitive SphI site, Sp in α-amylase signal peptide sequence
In-frame with hI site. Ge for both V _H and V _L PCR fragments
After ligation to the neware® vector, the DNA was
First PCR of phosphate by treatment with polynucleotide kinase and ATP
It was incorporated into the blunt 5'end of the product.

After the kinase reaction, this DNA was ligated back to itself to create a circular plasmid. The ligated DNA was transformed into E. coli (using electroporation). E. coli and transform the transformed cells with 50 μg / ml
Of the ampicillin were plated on selective medium. Plasmid DNA was isolated from individual ampicillin resistant E. coli. E. coli colonies were purified and transcribed with T7 RNA polymerase to produce infectious transcripts of individual clones.

Transcripts were generated essentially as described by Lindbo et al., Plant Cell 5: 1749-.
1759 (1993), using a PEG-based transfection protocol. Tobacum plant protoplasts were transfected and the transfected protoplasts were incubated in protoplast culture medium for several days. The latter medium was 265 mM mannitol, 1X Murashige minimal organic medium (Gibco / BRL), 1.5 mM KH ₂ PO ₄ , 0.2 μg / ml 2, 2.
4-dichlorophenoxyacetic acid, 0.1 μg / ml kinetin
, And 5% coconut water (Sigma). Protoplasts were cultured at a density of approximately 10 ⁶ cells / ml. Plasmid DNA was purified from at least 10-50 individual colonies from each cloning experiment.

About 1 to 4 days after transfection, protein samples were collected from individual protoplast samples. Culture medium (200-500 μl) was concentrated approximately 10-fold by high speed vacuum evaporation or Microcon sample concentrator.

This cloning strategy contained a signal peptide sequence designed to facilitate secretion of the protein product by plant cells into the culture medium,
Media samples were also analyzed by SDS-PAGE followed by Coomassie blue staining and / or Western blot.

Standard (Gly _4- Ser) ₃ linker sequence incorporating the starting scFv; other sc
Fv chains were randomly selected from transformants obtained from linker library cloning experiments. This experiment utilized a cloned PCR product generated from four primers (SEQ ID NOS: 4-11, above). Culture supernatants from equivalent numbers of cells were electrophoresed (SDS-PAGE) and the gels were run on mAb 7D1.
Transferred to nitrocellulose membrane for Western analysis with 1 (see above).

Some of the selected linker library members screened randomly expressed and accumulated more CJ protein than the CJ scFv with the conventional linker (Gly ₄ -Ser) ₃ expressed and accumulated. It seemed to do.

DNA in which library members express particularly high amounts of CJ scFv were sequenced. The results are shown in Table 2. Plasmid DNA for selected clones was prepared and sequenced by standard methods. From the nucleotide sequences of various CJ-derived constructs, the linker sequence of individual clones was deduced. Table 2 resulting enumerating several nucleotides and amino acid linker sequences farmland, and "relative expression" (which is, of expression compared to the same protein but with the (Gly ₄ -Ser) ₃ linker Means quantity).

DNA sequencing revealed that these clones did not have the same nucleotide or amino acid sequences, but rather exhibited amino acid length and nucleotide length diversity. Table 2 shows sampling of clones with L ranging from 13-20 amino acids. This range was clearly the result of mispriming during PCR amplification of the _VH and _VL coding sequences. Since the linker coding sequence of the oligonucleotides used in this experiment contains stretches of low complexity nucleotide sequences (ie asy _x or rst _z and), multiple mispriming events are possible. In relation to the DNA polymerase / exonuclease activity present during PCR, this can lead to an increase or decrease in the number of codons containing the L sequence.

[0160]

[Table 2] The amount of CJ scFv protein produced was also altered (relative to CJ scFv with the (Gly ₄ -Ser) ₃ linker). This indicates that both the length and the sequence of the linker region influence the amount of protein produced by plant cells or plants.

Example 2: Expression of scFv Product in Whole Plant The process described in Example 1 was repeated except that whole plant was used with an appropriate expression system to produce the scFv product.

Expressed products are screened by SDS-PAGE / Coomassie blue staining and / or Western blotting. These results show that various amounts of s
It shows that the cFv product was produced. The highest yield clones were used as vaccine scF
Selected for the production of v.

Expression System A DNA fragment encoding a CJ human lymphoma V region-containing dual domain scFv fragment was generated as in Example 1 and vector pBS.
It was cloned into G1250. In this vector, the TMV coat protein subgenomic promoter is located upstream of the CJ sequence insertion site. Following infection, the TMV coat protein subgenomic promoter
"Tsp") directs initiation of CJ RNA synthesis in plant cells. Rice α-amylase signal peptide (O'Neil) fused in-frame to CJ sequence
1, SD et al. (1990) Mol. Gen. Genet. 221: 235-244
) Encodes a 31-residue polypeptide that targets the protein to the secretory pathway (
Firel, S, et al., (1994) Transgenic Res. 3: 326-
331)) and subsequently CJ s expressed with the C-terminal Gly of the signal peptide.
It is excised from the N-terminal Met of the cFv protein. The sequence encoding the CJ scFv was introduced between the gene for the protein migrating at 30K and the gene for ToMV coat protein (Tcp). The T7 phage promoter was introduced upstream of the viral cDNA allowing transcription of infectious genomic + strand RNA.

The capped infectious RNA was transferred to the T7 message from Ambion (me
Sage) kit was used to make in vitro from 1 μg of plasmid. The synthesis of this messenger RNA was quantified by gel electrophoresis and approximately 2 μg of in vitro transcribed viral RNA was isolated from N. be
nthamiana (Dawson, WO et al., (1986) Proc. Natl.
． Acad. Sci. USA 83: 1832-1836) lower leaves (about 1-
2 cm size) with abrasive. Transcription of the subgenomic RNA encoding the CJ scFV protein was initiated post infection at the indicated transcription start points. High levels of subgenomic RNA species were synthesized in virus-infected plant cells (Kuma
gai, MH et al., (1993) Proc. Natl. Acad. Sci. USA
90: 427-430), and serve as a template for transcription and subsequent accumulation of the CJ scFv protein.

Clonal Characterization Indications of infection could be discerned after 5-6 days as a lobe leaf malformation with some discernible leaf plaques and growth retardation. Secreted proteins were isolated 11-14 days after seeding. The leaf and stem material was collected, weighed, and then osmotic buffer (10
Subjected to a vacuum of 700 mmHg for 2 minutes in 0 mM Tris HCl (pH 7.5) and 2 mM EDTA). The secreted protein (hereinafter referred to as the "interstitial fraction" or "IF") was supported from the infiltrated leaves by gentle centrifugation at 2000 g (Beckman JA-14). It was collected on a nylon mesh disk and concentrated about 10 times with a Centricon-10 (Amicon) concentrator. Total protein was analyzed by the Bradford method (Bradf
ord, M.D. (1976) Anal. Biochem. 72: 248-254)
And stored at −80 ° C. until used.

Secreted material was analyzed for the presence of soluble CJ scFv protein by SDS-PAGE followed by Western blotting with CJ mAb 7D11. About 3 μg of IF protein was separated by SDS-PAGE and 15 in standard Tris-glycine buffer containing 20% methanol.
Transfer to a nitrocellulose membrane at 0V for 1 hour. After transfer, the blot was blotted to a blotting buffer (50 mM Tris (pH 8), 150 mM NaCl, 1 mM).
EDTA, 2.5% nonfat dry milk, 2.5% BSA and 0.05% Twee
n 20) at room temperature for 20 minutes, then 1 μg / ml of purified 7D
Incubated in blotting buffer containing 11 antibody for 16 hours at 4 ° C. Wash 3 times for 15 minutes (100 mM Tris (pH 8), 150 mM
After NaCl, 1 mM EDTA and 0.1% Tween 20), the membrane was
μg / ml goat anti-mouse IgG-HRP (Southern Biotech
Incubation was carried out for 1 hour in blotting buffer containing After washing 3 times for 15 minutes, the Western blot was analyzed with Enhanced Chemiluminescence (ECL) (Am) according to the manufacturer's instructions.
developed by ersham). The exposure time was in the range of 1-5 seconds. No cross-reactivity with plant proteins was observed (I from control infected plants).
Test F extract).

Individual clones were sequenced, reading frame and original CJ Ig
The amino acid identities to the sequences were analyzed and then screened for protein expression in infected plants. Figure 1 shows the results of 9 individual CJ scFv expressing clones that showed various levels of protein accumulation. Clones 20 and 30 showed high levels of expression and accumulation of protein dimers. Clone C contained a modification of the (Gly ₃ Ser) ₄ linker.

From the sequence data, the linker sequence for each clone was deduced. The clone numbers in Table 3 are the same as those listed in Table 2. As mentioned above, relative expression refers to scFv proteins with a (Gly ₃ Ser) ₄ linker.

As described above, differences were observed in the expression of various CJ scFv-based clones in whole plants. Interestingly, some clones expressed in plant protoplasts did not express in whole plants. For example, clone # 16, which is strongly expressed in plant protoplasts, was clearly not expressed in the whole plant. Nevertheless, the disclosed methods for making linker regions with varying lengths and sequences allow screening of many clones for expression in either plant protoplasts or whole plants.

The quality of the CJ protein optimized by the random linker library was verified by two methods. First, the CJ protein was immobilized on 7D1.
Purified by affinity chromatography using 1 anti-idiotype mAb. This method requires the CJ protein to bind to the anti-Id column under physiological conditions. Such binding does not occur if the protein is not properly folded. Bind the protein under normal pH, and 5
Eluted with 0 mM dimethylamine (pH 11.5), then immediately dialyzed against normal saline. Material was quantified by ELISA using 7D11 and using standard protein determination methods.

Second, and more stringently, the assay for CJ protein quality was a functional assay in animals. Clone CJLL20 (for linker library pick # 20) was purified by 7D11 affinity chromatography and administered to 5 mice three times with weekly immunizations at 30 μg each. Serum was sampled 10 days after the third injection. Sera were tested for specific response to the CJ idiotype in a sandwich ELISA using native idiotype (1D12) or isotype-matched unrelated human antibodies. The results are shown in Figure 2.

Non-specific antibody responses to xenogeneic human Ig determinants were detected in only 3 out of 5 animals and at very low doses (detected as minimal cross-reactivity of mouse sera to unrelated human antibodies). Were present.

The sera of all 5 mice had high titers of anti-CJ antibody (Figure 2). Therefore,
The immune response elicited by the dual domain scFv polypeptide was highly specific for the original _VH and _VL domains of the original Ig, as expected and desired. These results suggested that the proteins produced in plants were correctly folded, so that an appropriate immune response could be induced when administered to a subject.

Table 3: Analysis of selected members of CJ linker library experiments in whole plants.

[0175]

[Table 3] * RE = Relative expression for (Gly ₃ Ser) ₄ clones.

Example 3: Expression of scFv Products in Whole Plants The process described in Example 2 was followed, except that different human scFvs with unknown expression characteristics were used with an appropriate expression system to produce scFv products. Repeated.

The expressed products were screened by SDS-PAGE / Coomassie blue staining. These results indicated that varying amounts of scFv product were produced based on the linker composition. The highest yield clones are selected for production of vaccine scFv.

Expression System A DNA fragment encoding a double domain scFv fragment containing the V region of Go19 human lymphoma was generated as in Example 1 and insert cloning of TMV and TMGMV-U5 and SphI and AvrII. Vector p1324-MBP containing a hybrid fusion of rice α-amylase signal peptide having a site (modified 30B vector (Shivprased,
S., et al., (1999) Virology 255: 312-323)).

In this vector, the TMV coat protein subgenomic promoter is
It is located upstream of the insertion site of the Go19 sequence. Following infection, this TMV coat protein subgenomic promoter directs the initiation of Go19 RNA synthesis in plant cells at the transcription start point ("tsp"). Rice α-amylase signal peptide fused in frame to the Go19 sequence (O'Neill, SD et al. (19
90) Mol. Gen. Genet. 221: 235-244) encodes a 31-residue polypeptide that targets proteins to the secretory pathway (Firel, S, et al., (1994) Transgenic Res. 3: 326-331)), and subsequently a signal. It is cleaved between the C-terminal Gly of the peptide and the N-terminal Met of the expressed Go19 scFv protein. The sequence encoding the Go19 scFv was introduced between the gene of the protein migrating at 30K and the gene of TMGMV-U5 coat protein (Tcp). The T7 phage RNA polymerase promoter enables transcription of the infectious genomic + strand RNA into the viral cDNA
It was introduced upstream of NA.

The Go19 V region was amplified in four separate PCR reactions. In the first and second reactions, the V _H domain-encoding sequence was amplified from a cDNA clone derived from the lymphoma cells of patient Go19 using the following synthetic oligonucleotides:

[0181]

[Chemical 12] The SphI restriction site is underlined above. In the first reaction, x is 4:

[0182]

[Chemical 13] In the second reaction, x is 9 (SEQ ID NO :) 29):

[0183]

[Chemical 14] (In general, the number of triplets (x) can be from 1 to about 50).

In the third and fourth PCR reactions, the sequence encoding the V _L domain was amplified from Go19 cDNA cloning using the following synthetic oligonucleotides:

[0185]

[Chemical 15] The AvrII restriction site is underlined above. In the first reaction, z is 6:

[0186]

[Chemical 16] In the second reaction, z was 9 and gave SEQ ID NO: 33:

[0187]

[Chemical 17] (In general, the number of triplets (z) can be 1 to about 50).

Prior to PCR amplification, the V _H R and V _L R oligonucleotides were treated with polynucleotide kinase and ATP to remove 5 of the oligonucleotides.
'Phosphate was added to the end. 4 PCRs following amplification
The product was purified and the _VH and _VL products were ligated together to create scFv. The scFv ligation product was purified again, restriction digested with SphI and AvrII,
The digested scFv is then gel isolated and ligated into the Geneware® vector. The ligated DNA was transformed into E. E. coli was transduced (using electroporation) and the transformed cells were plated onto selective medium containing 50 μg / ml ampicillin. Plasmid DNA was isolated from individual ampicillin resistant E. It was purified from E. coli colonies.

The capped infectious RNA was transferred to the T7 message from Ambion (me
ssage) kit was used to make in vitro from approximately 0.5 μg of plasmid. The synthesis of this messenger RNA was assessed by gel electrophoresis and approximately 2 μg of in vitro transcribed viral RNA was
Encapsulated with purified TMV-U1 coat protein in 00 mM sodium phosphate, pH 7.0 for a minimum of 6 hours at room temperature. The encapsulated transcript was transformed into N. benthamiana (WO Dawson et al., (19
86) Proc. Natl. Acad. Sci. USA 83: 1832-18
36) Lower leaves (about 1-2 cm in size) were applied with an abrasive. Go1
Transcription of the subgenomic RNA encoding the 9 scFV protein was initiated post-infection at the indicated transcription start points. High levels of subgenomic RNA species were transferred to virus-infected plant cells (MH Kumagai, et al., (1993) Proc. Natl. Aca.
d. Sci. USA 90: 427-430) and serve as a template for transcription and subsequent accumulation of the Go19 scFv protein.

Clonal Characterization Indications of infection could be discerned after 5-6 days as longitude leaf malformations with some discernible leaf plaques and growth delays. Secreted proteins were isolated 11-14 days after seeding. About 0.1 g of infected leaf material was collected and placed in a 96-well glass fiber filtration block (Whatman / Polyfiltronics) and placed in permeation buffer (20 mM Tris HCl (pH 7.0), 10 mM 2-mercaptoethanol). Soak. The tissue is subjected to vacuum at 700 mmHg for 30 seconds, the vacuum is released, and the vacuum process is repeated at least one more time.
Residual buffer is removed by slow spin at 30 xg in a plate centrifuge. Secreted proteins (hereinafter "interstitial fraction" or "I")
Designated as "F") is recovered from the infiltrated leaves by gentle centrifugation at 1700 xg in a plate centrifuge and collected in 96-well polypropylene plates.

Secreted material was analyzed by SDS-PAGE for the presence of soluble Go19 scFv protein. IF (27 μl containing approximately 5 μg protein) was separated by SDS-PAGE. The linkers from individual clones were sequenced, analyzed for reading frames and amino acid content, then screened for protein expression in infected plants. FIG. 3 shows the results of 22 individual Go19 scFv expressing clones that showed various levels of protein accumulation. Clones C5 and E1 and E9 showed high levels of expression with minimal protease degradation.

From the sequence data, the linker sequence for each clone was deduced as shown in Table 4.

Table 4: Analysis of selected members of Go19 linker library expression experiments in whole plants.

[0194]

[Table 4] * RE = Relative expression against Go19scFv library clone. As noted above, differences were observed in the expression of various Go19 scFv-based clones throughout the plant and some degradation indicated by the presence of protein accumulation between the 6.5 and 21 kDa marker bands. did. The disclosed methods for making linker regions with varying lengths and sequences allow the screening of many clones for expression in either plant protoplasts or whole plants.

Example 4: scFv-Detectably Labeled Conjugate The mAb to HER-2 / neu was a 6-day-old breast cancer cell line SK-Br.
-3 (ATCC HTB 30). Such treatment sensitizes these cells to chemotherapeutic agents (US Pat. No. 5,677,171).
issue).

The process of Example 1 was repeated using the _VH and _VL regions of the scFv that specifically bind to the HER-2 / neu (erbB-2) protein. The scFv gene encoding such a polypeptide is described in Wels et al., Biotechn.
LOGY 10: 1128-1132 (1992). Using the repeated triplet nucleotide sequence as in Example 1, erbB-2s
The 3'end of the cFv DNA construct is ligated to the 5'end of the horseradish peroxidase gene using the appropriate PCR primers modeling the method described in Example 1.

High yield clones are identified by measuring for peroxidase activity in the supernatant. In a control sample of a breast cancer cell line highly expressing HER-2 / neu, high affinity and high affinity were again determined by immunohistochemical detection with substrate and chromophore. Conventionally labeled mAbs to HER-2 / neu (eg DAKO HerceptTest, Dako Corp.
． , Carpinteria, CA) to determine which clone produces an acceptable scFv protein.

Example 5 scFv-Toxin Conjugate Production The process of Example 4 was repeated with the following modifications. Ricin A
The strand is linked to the 3'end of the scFv DNA construct via a linker region of the invention (creation of repeated triplet nucleotides).

Plant cell clones are grown in 24-well plates and screened first by measuring secreted protein (PAGE followed by Coomassie blue staining). Two day culture supernatants from these wells in which each clone was grown were tested for cytotoxic activity on target cells in SK-B in 6 well plates (Costar).
Test by incubation of r-3 with active culture. Cytotoxicity to these targets is determined after 48 hours by microscopy.

High-producing clones that produce strong cytotoxicity are selected. Callus is formed from these cultures to regenerate plants for growth and large-scale production on farmland.

Humanized mAbs against HER-2 / neu are FDA-approved therapeutics for breast cancer (HERCEPTIN, Genentech, Inc., S.).
out San Francisco, CA). It is expected that toxin-conjugated scFv specific for the same antigen will be at least comparable and preferably more cytotoxic to human breast cancer cells.

Example 6: Production of Dual Domain Ribozymes The process of Example 1 is repeated except that DNA encoding two different ribozyme domains is used. RNA having characteristics of each ribozyme domain is obtained by transcribing a vector containing subcloned double ribozyme domains.
To produce.

The amount of transcribed RNA product can be determined by hybridization with oligonucleotide probes, spectrophotometric measurements and the like. The amount of activity of either ribozyme domain can be measured using a suitable assay.

Example 7: Generation of Dual DNA Domains The process of Example 1 is repeated except that two different DNAs, one of which binds the protein, are used. Plasmid DNA can be produced in quantity and double DNA domain molecules can be excised with restriction endonucleases. The resulting fragment has two linked DNA domains and can be assayed for its activity to bind a DNA binding protein (eg, transcription factor, restriction endonuclease, polymerase, etc.).

All of the above-referenced references, whether specifically incorporated or not, are incorporated herein by reference.

[Sequence list]

[Brief description of drawings]

FIG. 1 shows a Western blot analysis of the scFv protein produced in Example 1 in plant cytoplasm. CJ is a sc with a (Gly ₄ Ser) ₃ linker
It is Fv. Lane numbers refer to clone numbers. Kilodalton (kD
) Size is shown on the left.

FIG. 2 shows Western blot analysis of scFv proteins produced in Example 2 in whole plants. CJ is an scFv with a (Gly ₄ Ser) ₃ linker
Is. Lane numbers refer to clone numbers. The kilodalton (kD) size is shown on the left.

FIG. 3 is a ComF of the scFv protein produced in Example 3 in all plants.
3 shows an SDS-PAGE analysis stained with assie. Numbers in lanes refer to clone numbers and arrows indicate scFv proteins. Kilodalton (k
The size of D) is shown on the left.

─────────────────────────────────────────────────── ─── Continued front page    (81) Designated countries EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, I T, LU, MC, NL, PT, SE), OA (BF, BJ , CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, K E, LS, MW, MZ, SD, SL, SZ, TZ, UG , ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, C R, CU, CZ, DE, DK, DM, EE, ES, FI , GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE, KG, KP, KR, K Z, LC, LK, LR, LS, LT, LU, LV, MA , MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, S K, SL, TJ, TM, TR, TT, UA, UG, US , UZ, VN, YU, ZA, ZW (72) Lindau, John A. Inventor.             United States California 95688,               Bakerville, Sundance Drive             143 (72) Inventor Tarpen, Thomas             United States California 95688,               Bakerville, Santa Fe Court             160 F-term (reference) 4B024 AA01 BA43 BA61 CA04 CA07                       DA01 EA01 EA04 FA02 GA11                       HA01 HA20                 4B064 AG01 AG26 CA11 CA19 CC24                       DA01                 4H045 AA10 AA11 AA20 AA30 BA10                       BA41 CA40 DA00 DA75 EA20                       EA31 FA72 FA74

Claims (51)

[Claims] 1. A library of dual domain nucleic acid molecules, each of the nucleic acid molecules comprising: (a) a first and a second domain; (b) a step of separating and linking the domains. And wherein the linkers are (i) different in size and nucleotide sequence, and (ii) consisting of a repeating pattern of degenerate repeating triplet nucleotides, wherein the linkers are members of a random library of linkers. 2. A library of molecules according to claim 1, wherein said repeating pattern of degenerate repeating triplet nucleotides of said linker has the following properties: (i) position 1 of each repeating triplet is said repeating triplet. (Ii) position 2 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; or (iii) each repeating triplet A position 1 of which is not the same nucleotide as position 3 of the repeating triplet. 3. A library of molecules according to claim 2, wherein the nucleotides at the first and second positions of each repeating triplet are deoxyadenosine, deoxyguanosine, deoxycytidine or deoxythymidine. A library selected from any two. 4. The library of claim 3, wherein (i) position 1 of each repeating triplet is deoxyadenosine or deoxyguanosine; (ii) position 2 of each repeating triplet is deoxycytidine or Deoxyguanosine; and (iii) Deoxythymidine at position 3 of each repeating triplet is a library. 5. A library of molecules according to claim 1, wherein at least one of said domains binds a protein. 6. A library of molecules according to claim 5, wherein both said domains bind a protein. 7. A library of molecules according to claim 1, wherein at least one of said domains binds a nucleic acid which is not a member of said library. 8. A library of molecules according to claim 7, wherein both of said domains bind nucleic acids that are not members of said library. 9. A library of molecules according to any one of claims 1 to 4, wherein the first and second domains are coding sequences. 10. A library of molecules according to any one of claims 1 to 8 produced in plant cells. 11. A library of molecules according to claim 9 prepared in plant cells. 12. A dual domain nucleic acid molecule selected from the library of any one of claims 1-8. 13. A dual domain nucleic acid molecule selected from the library of claim 9. 14. A dual domain nucleic acid molecule selected from the library of claim 10. 15. A dual domain nucleic acid molecule selected from the library of claim 11. 16. A library of dual domain polypeptide molecules comprising:
Each of the dual domain polypeptide molecules is described by the formula D ₁ -L-D ₂ , wherein (a) D ₁ and D ₂ are polypeptide domains, and (b) L is a peptide, Or a peptide linker that is a member of a random library of linkers that differ in size and sequence, wherein the library is encoded by a nucleic acid sequence consisting of a repeating pattern of degenerate repeating triplet nucleotides. 17. A library of multi-domain polypeptide molecules, which comprises:
Each of said multi-domain polypeptide molecules comprises a polypeptide domain D, each pair of said polypeptide domain D being linked by a peptide or peptide linker L, each molecule being described by the formula D _x L _y , wherein x is an integer between 2 and 20; y is an integer between 1 and 19 with the proviso that any value of x, y = x−1; D ₁ is a single Attached to one C-terminal linker; most C-terminal D attached to a single N-terminal linker; each of D ₂ -D ₁₉ attached to N-terminal and C-terminal linker; And a member of a random library of linkers that differ in sequence, the linker library being encoded by a nucleic acid sequence consisting of a repeating pattern of degenerate repeating triplet nucleotides. Over. 18. A library of double domain polypeptide molecules according to claim 16, or a library of multidomain polypeptide molecules according to claim 17, wherein each linker in said library comprises: ) Having a length of amino acid residues between about 1 and 50, and (ii) different amino acids between 1 and about 20, each repeating pattern of degenerate triplet bases being between 1 and about Encodes different amino acids between 12 and
Library. 19. The library of polypeptide molecules of claim 18, wherein the repeating pattern of degenerate repeating triplet nucleotides encoding the linker has the following properties: (i) position 1 of each repeating triplet. Is not the same nucleotide as position 2 of the repeating triplet; or (ii) position 2 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; or ( iii) A library having the property that position 1 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet. 20. A library of polypeptide molecules according to claim 19, wherein the nucleotides at the first and second positions of each repeating triplet are deoxyadenosine, deoxyguanosine, deoxycytidine or deoxythymidine. A library selected from any two of 21. A library of polypeptide molecules according to claim 20, wherein (i) position 1 of each repeating triplet is deoxyadenosine or deoxyguanosine; (ii) position 2 of each repeating triplet. , Deoxycytidine or deoxyguanosine; and (iii) deoxythymidine at position 3 of each repeating triplet. 22. The library of dual domain polypeptide molecules of claim 16, or the library of multidomain polypeptide molecules of claim 17, produced in plant cells. 23. A library of polypeptide molecules according to claim 18, produced in plant cells. 24. A library of polypeptide molecules according to claim 19 produced in plant cells. 25. A library of polypeptide molecules according to claim 20, produced in plant cells. 26. A library of polypeptide molecules according to claim 21, produced in plant cells. 27. A dual domain polypeptide molecule selected from the library of claim 16. 28. A multi-domain polypeptide molecule selected from the library of claim 17. 29. A dual domain or multi-domain polypeptide molecule selected from the library of claim 18. 30. A dual domain or multi-domain polypeptide molecule selected from the library of claim 19. 31. A dual domain or multi-domain polypeptide molecule selected from the library of claim 20. 32. A dual domain or multi-domain polypeptide molecule selected from the library of claim 21. 33. A 3-domain peptide selected from the library of claim 17, wherein the dual domain scFv polypeptide is linked to a third polypeptide domain. 34. The three domain polypeptide of claim 33, wherein the third domain is a toxin polypeptide or enzyme. 35. A method of making a library of double domain nucleic acids according to claim 1, wherein the method comprises the following steps: a. Obtaining two template DNA sequences comprising said first and said second domains; b. Preparing an amplification primer pair for amplifying said first and second domains, each primer pair comprising an upstream primer and a downstream primer, each primer having a 5'end and a 3'end, wherein Wherein the downstream primer of the first domain or the upstream primer of the second domain comprises a non-templated primer, the non-templated sequence comprises a repetitive pattern of degenerate repeating triplet nucleotides, wherein a degenerate repeating triplet nucleotide At least two of the 5'-terminal triplets of the repeating pattern of have the same degenerate sequence; c. Amplifying the domains using the amplification primers to generate at least one population of nucleic acid domains having different lengths and sequences in non-template sequences, d. Ligating the nucleic acid domains produced in step (c) to produce said population of dual domain molecules. 36. The method of claim 35, wherein the repeating pattern of degenerate repeating triplet nucleotides in at least one of the primers comprises:
The following properties: (i) position 1 of each repeating triplet cannot be the same nucleotide as position 2 of the repeating triplet; or (ii) position 2 of each repeating triplet is the same as position 3 of the repeating triplet. Or (iii) position 1 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet. 37. The method of claim 35, wherein at least one of said primers comprises a non-templated endonuclease recognition site. 38. The method of claim 35, wherein the template DNA sequence is produced by reverse transcription of mRNA. 39. The method of claim 35, further comprising ligating the population of dual domain nucleic acids to a vector. 40. The method of claim 39, further comprising introducing the vector into a host. 41. The method of claim 40, wherein the nucleic acid domain encodes a polynucleotide domain and the method expresses a dual domain polypeptide encoded by the dual domain nucleic acid. The method further comprising: 42. The method of claim 39, further comprising the step of transcribing RNA from the vector. 43. The method of claim 42, wherein the vector is compatible with replication and / or expression of the nucleic acid in plant cells.
The method further comprising introducing the transcribed RNA into a plant cell and expressing a dual domain polypeptide. 44. A population of dual domain polypeptides, or a dual domain polypeptide selected therefrom, produced by the method of claim 41. 45. A population of dual domain polypeptides, or a dual domain polypeptide selected therefrom, produced by a method according to claim 43 in plant cells. 46. A method of producing the polypeptide of claim 27, the method comprising the steps of: (a) a first domain of the polypeptide for making the first nucleic acid construct. A nucleic acid encoding the first portion of the linker is ligated to the nucleic acid encoding the second portion of the linker to produce a second nucleic acid construct. Linking to a nucleic acid encoding a second domain of (c) said first and said, wherein said polypeptide, when expressed, is separated by said linker as described by formula D ₁ -L-D ₂ . Incorporating said first and second constructs in-frame into a transient plant expression vector so as to carry a second domain, (d) said plant transiently producing said polypeptide To do And (e) recovering the polypeptide as a soluble and functionally folded protein, the method comprising the steps of: (i) transfecting a plant with the vector; 47. The method of claim 46, wherein the early stage plant is a plant cell. 48. A linker nucleic acid molecule or linker nucleic acid sequence linking two nucleic acid sequences encoding two nucleic acid domains or two polypeptide domains, the linker nucleic acid molecule or linker nucleic acid sequence having the following properties: (I) position 1 of each repeating triplet cannot be the same nucleotide as position 2 of the repeating triplet; or (ii) position 2 of each repeating triplet is the same as position 3 of the repeating triplet. A property that cannot be a nucleotide; or (iii) position 1 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; and (iv) a molecule or sequence linking the domains is , does not encode Gly ₄ Ser or repeated, have the property, the modified iterative Toripure Having a repeating pattern of nucleotide in the linker nucleic acid molecule or linker nucleic acid sequence. 49. A linker nucleic acid molecule or a library of linker nucleic acid sequences, which links two nucleic acid sequences encoding two nucleic acid domains or two polypeptide domains, each of said linker nucleic acid molecule or linker nucleic acid sequence. With the following properties: (i) position 1 of each repeating triplet cannot be the same nucleotide as position 2 of the repeating triplet; or (ii) position 2 of each repeating triplet has position 2 of the repeating triplet. A property that cannot be the same nucleotide as position 3; or (iii) a position 1 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; and (iv) bind the domain the molecule or sequence that does not encode an Gly ₄ Ser or repetitive nature, the To have a repeating pattern of modified repetitive triplet nucleotide library. 50. A method for producing a library of linker nucleic acid molecules or linker nucleic acid molecule sequences according to claim 49, which method comprises the steps of: (a) the first and second domains; And (b) preparing an amplification primer pair for amplifying the first and second domains, wherein each primer pair is an upstream primer and a downstream primer. Wherein each primer has a 5 ′ end and a 3 ′ end, wherein the downstream primer of the first domain or the upstream primer of the second domain comprises a non-template sequence, Contains a repeating pattern of degenerate repeating triplet nucleotides, wherein a small number of triplets at the 5'end of the repeating pattern of degenerate repeating triplet nucleotides. At least two having the same degenerate sequence; (c) using an amplification primer to generate at least one population of nucleic acid domains having different lengths and sequences in the non-template sequence, Amplifying, (d) linking the nucleic acid domains produced in step (c) to produce said population of double domain molecules, (e) said linker from said population of dual domain molecules Cleaving or amplifying the nucleic acid molecule or the linker nucleic acid sequence. 51. A method for making a linker nucleic acid molecule or linker nucleic acid sequence that links two nucleic acid sequences encoding two nucleic acid domains or two polypeptide domains, the method comprising the following properties: (I) position 1 of each repeating triplet cannot be the same nucleotide as position 2 of the repeating triplet; or (ii) position 2 of each repeating triplet is the same nucleotide as position 3 of the repeating triplet. (Iii) position 1 of each repeating triplet cannot be the same nucleotide as position 3 of the repeating triplet; and (iv) said molecule or sequence that binds said domain is It does not encode Gly ₄ Ser or repeated, have the property, the modified repeated triplet j Kure A tide pattern, the method comprising the steps of: (a) creating a library of said linker nucleic acid molecule or linker nucleic acid sequence according to the method of claim 49, (b) said linker molecule or Selecting and isolating the linker sequence from the library.