Note: Descriptions are shown in the official language in which they were submitted.
CA 02344625 2001-04-02
WO 00/23622 PCTIUS99/2390b
METHODS FOR MANIPULATING COMPLEX NUCLEIC ACID
POPULATIONS USING PEPTIDE-LABELED OLIGONUCLEOTIDES
1. FIELD OF THE INVENTION
The present invention relates generally to methods of labeling, sorting,
comparing and isolating populations of nucleic acids. More particularly, the
present
invention relates to a method for sorting and comparing complex populations of
nucleic
acid, such as cDNA libraries. These complex populations of nucleic acid may be
derived
from cells or tissue types having variations in phenotype of potential
clinical interest. The
methods are referred to generally as ValiGeneS"' Peptide-Labeled
Oligonucleotide methods,
or VG-PLO$"', and involve the use of distinguishable and identifiable peptide
tags linked to
oligonucleotide primers to manipulate nucleic acids.
2. BACKGROUND OF THE INVENTION '
Labeled oligonucleotides have been used for detection of specific sequences.
For example, Burdick and Oakes (Diagnostic Kit and Method Usingva Solid Phase
Capture
Means For Detecting Nucleic Acids, European Patent Publication No. EP 0370 694
A2,
Date of publication May 30, 1990) disclose the use of oligonucleotide primers,
labeled with
a label, with specific nucleic acid sequences which are complementary to a
predetermined
2 0 sequence of interest. This method is limited to the identification of a
known sequence
within a given sample and each pair of primers must correspond to a single
predetermined
PCR product.
Several gene expression assays are now becoming practicable for
quantitating the effect of a drug an expression of a large fraction of the
genes and proteins
25 ~ a cell culture (see, e.g., Schena et al., 1995, Quantitative Monitoring
of Gene Expression
Patterns with a Complimentary DNA Micro-array, Science 270:467-470; Lochort et
al.,
1996, Expression Monitoring by Hybridization to High-density Oligonucleotide
Arrays,
Nature Biotechnology 14:1675-1680; Blanchard et al., 1996, Sequence to array:
Probing
the genome's secrets, Nature Biotechnology 14, 1649; 1996, U.S. Patent
5,569,588, issued
30 October 29, 1996 to Ashby et al. entitled "Methods for Drug Screening").
Raw data from
these gene expression assays are often difficult to coherently interpret. Such
measurement
technologies typically return numerous genes with altered expression in
response to a drug,
typically 50-100, possibly up to 1,000 or as few as 10.
CA 02344625 2001-04-02
WO 00/23622 PCT/US99I23906
Few methods exist to rapidly compare multiple mixtures of nucleic acids to
select genes that are differentially expressed or are shared by some but not
all phenotypes-
of interest. Accordingly, there is a need in the art for methods which rapidly
and eff ciently
allow comparison among nucleic acid populations.
3. SUMMARY OF THE INVENTION
This invention provides methods of labeling, sorting, comparing and
screening multiple, complex populations of nucleic acids. These populations
may be cDNA
Z 0 libraries constructed from phenotypically distinguishable cell or tissue
types of interest.
The method is extremely flexible and is adaptable to perform numerous complex
sorting
and comparing tasks.
The methods of this invention may also be used to increase and supplement
the analytical powers of other techniques of manipulating complex cDNA
population. A
I5 major advantage of the methods of this invention is the ability to screen
multiple
populations of cDNAs derived, for example, from different tissues belonging to
the same
individual or to phenotypically different cell types present concurrently
within a given
tissue sample.
This invention takes advantage of the ability to follow multiple populations
2 0 of nucleic acids through various sorting and molecular comparison
procedures. The
invention employs oligonucleotide primers having distinguishable and
identifiable peptide
tags, which primers can be used to prime PCR reactions from vector sequences.
Using such
oligonucleotide primers, inserts from any given cDNA library can be labeled
with library-
specific peptide tags. The distinguishable tags serve to identify the library-
of origin of any
z 5 given insert. Further, the distinguishable tags can be used to selectively
sort and isolate
inserts based on their library of origin regardless of the complexity of the
mixture of
products. For example, one can use a chromatography matrix having an antibody
specific
to one of the distinguishable peptide tags. Such a matrix can trap or retain
fragments, both
single and double-stranded, which bear the specific peptide tag. Fragments
which do not
3 0 contain this tag will be left free in the flow-through.
The methods of the invention make use of polymerase chain reaction (PCR)
primarily to linearize all inserts within a given cDNA library and to affix a
distinguishable
- 2 -
CA 02344625 2001-04-02
WO 00123622 PCT/(JS99/23906
and identifiable peptide label to all inserts from a particular cDNA library
so as to indicate
their library-of origin. PCR can also be used at various stages to amplify
complex mixtures
of products.
Nucleic acid sample populations may be derived from many different
sources. Such sources may include different phenotypes present concurrently
within a
given tissue sample or different tissues belonging to the same individual. One
phenotype
may, but does not need to be, "healthy" and another typical of a disease
state. The methods
of the invention allow identification of genes that are specifically expressed
in association
with each phenotype as well as a comparison of genes which are expressed
independently of
the phenotype or are shared by some phenotypes but not all.
In a first embodiment this invention provides a method comprising the
following steps: (a) labeling DNA from each of a plurality of cDNA libraries
using PCR
with oligonucleotide primers having a label uniqueto each library; (b)
contacting DNA
labeled in step (a) with a first said label with DNA labeled in step (a) with
a different said
label and (c) sorting DNA contacted in step (b) using one or more molecules,
each molecule
being capable of binding the label unique to each library.
This invention further provides in the first embodiment additional methods
wherein the label unique to each library is a 5'-peptide label.
2 0 This invention further provides in the first embodiment an additional
method
wherein the label unique to each library is biotin.
This invention fiirther provides in the first embodiment additional methods
wherein the one or more molecules is an antibody.
This invention further provides in the first embodiment, methods wherein the
2 5 oligonucleotide primers prime PCR from vector sequences common to the
nucleic acids
within a particular library {thus a different one such primer is used for each
library) or the
oligonucleotide primer primes PCR from vector sequences coinrnon to the
plurality of
cDNA libraries (thus the same oligonucleotide primer is used for priming PCR
for the entire
plurality).
3 0 This invention further provides in the first embodiment a method of
sorting
which comprises (d) denaturing hybrid DNA strands resulting from step (b); (e)
contacting
single strands denatured in (d) with single strand binding protein to prevent
strand
_ 3 _
CA 02344625 2001-04-02
WO 00!23622 PCTIUS99/23906
reannealing; and (f) contacting the single strand binding protein coated
single strands
formed in (e) with one or more molecules, each molecules being capable of
binding one of
the labels unique to each library.
This invention further provides in the first embodiment a method wherein at
least one of the one or more molecules in (e) is an antibody.
This invention provides in a second embodiment a method of cDNA library
comparison comprising: (a) labeling DNA from a first cDNA population by PCR
using
oligonucleotide primers which have a first 5'-peptide label; (b) labeling DNA
from a second
cDNA population by PCR using oligonucleotide primers having a second 5'-
peptide label;
(c) contacting DNA labeled in step (a) with DNA labeled in step (b) under
conditions such
that hybridization can occur and (d) separating DNA having the first and the
second 5'
peptide labels from DNA having only the first or the second 5' peptide label.
This invention.further provides in the second embodiment additional
methods wherein the first cDNA population is from one or more cells or an
organism
subjected to a first condition and the second cDNA population is from one or
more cells or
an organism of the same type not subjected to said first condition.
This invention further provides in the second embodiment additional
methods wherein the first cDNA population is from one or more cells or an
organism
subjected to a first condition and the second cDNA population is from one or
more cells or
an organism of the same type subjected to a second condition.
This invention further provides in the second embodiment additional
methods wherein the first and second cDNA populations axe from cells or
organisms that
differ phenotypically.
This invention further provides in the second embodiment additional
methods wherein the nucleotide sequences of the oligonucleotide primer pair
having the
first 5'-peptide label and the nucleotide sequences of the oligonucleotide
primer pair having
the second 5'-peptide label are the same.
In a third embodiment this invention provides a method of monitoring gene
3 0 expression comprising: (a) contacting mRNA from a cell with an RNA-
dependent DNA
polymerase and a 5'-dephosphorylated target-specific primer (i.e., specific to
the gene of
which it is desired to monitor expression); (b) contacting any DNA:RNA hybrids
- 4 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
synthesized in step (a) with a nuclease to remove single-stranded RNA
extensions; (c) after
step (b) ligating the DNA:RNA hybrids molecules to a partly double-stranded
phasphorylated second primer (e.g., a primer that is not target specific); (d)
labeling
products Iigated in step (c) by PCR with a first primer complementary to the
target-specific
primer used in step (a), said first primer being labeled with a first label
and a second primer
complementary to one strand of the double-stranded phosphorylated second
primer in (e),
said second primer being labeled with a second label that is distinguishable
from the first
label; (e) contacting the PCR products labeled in step (d} with one or more
molecules
Immobilized on a solid support capable of binding the first label; (f) washing
the solid
support; and {g) contacting the support washed in step (fj with one or more
molecules
capable of binding the second label.
This invention further provides in the third embodiment an additional
method wherein the nuclease is mung-bean nuclease.
Z 5 This invention further provides in the third embodiment an additional
method wherein the partly double-stranded phosphorylated second primer is an
M13
forward sequencing primer.
This invention further provides in the third embodiment an additional
method wherein the first label is a peptide label.
2 0 This invention further provides in the third embodiment additional methods
wherein at least one of the one or more molecules in step (e) is an antibody.
This invention further provides in the third embodiment additional methods
wherein at Ieast one of the one or more molecules in step (g) is streptavidin-
linked
horseradish peroxidase.
2 5 A fourth embodiment provides a method of identification of cDNA inserts
represented in a first cDNA library and not represented in a plurality of
other cDNA
libraries comprising: (a) labeling DNA inserts from each cDNA library by
polymerase chain
reaction using oligonucleotide primers having a label unique to each library;
(b) hybridizing
DNA labeled in step (a); (c) contacting DNA hybridized in step (b) with a
plurality of
3 0 immobilized antibodies capable of recognizing the label unique to each of
the plurality of
other cDNA libraries but not the label unique to the first cDNA library; and
(d) recovering
DNA which is not bound by the plurality of immobilized antibodies.
- 5 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99l23906
This invention further provides in the fourth embodiment additional methods
wherein the DNA hybridized from each of the plurality of other cDNA libraries
is in excess
relative to the first cDNA library. Furthermore, additional methods are
provided which
employ from a 2-fold to a 100-fold excess, from a 2.5-fold to a 10-fold excess
and wherein
the excess is a 3-fold excess.
This invention further provides in the fourth embodiment additional methods
wherein the label unique to each library is a peptide label. Furthermore,
methods are
provided wherein the peptide label is 3 to 12 amino acid residues.
Furthermore, methods
~ 0 are provided wherein the label is a thermophilic protein label.
This invention further provides in the fourth embodiment additional methods
wherein one of the plurality of antibodies in step (c) is immobilized on a
separate affinity
column. Furthermore, methods are provided wherein the separate affinity
columns are
physically linked in series in any order.
This invention further provides in the fourth embodiment additional methods
wherein the column flow-through is applied to the separate, physically-linked
affinity
columns one or more times. Furthermore, a method is provided wherein the
column flow-
through is applied to the separate, physically-linked affinity columns three
times.
This invention further provides in the fourth embodiment a method wherein
2 0 the DNA retained by the antibody specific for the label unique to the
first cDNA library is
recovered and cloned.
In a f fth embodiment this invention provides a method of identification of
cDNA inserts represented in a first cDNA library and in a second cDNA library,
and not
represented in a plurality of other cDNA libraries, comprising: (a) labeling
DNA from each
cDNA library by PCR using oligonucleotide primers having a label unique to
each library;
(b} hybridizing DNA labeled in step (a); (c) contacting DNA hybridized in step
(b) with a
plurality of immobilized antibodies capable of recognizing the label unique to
each of the
plurality of other cDNA libraries but not the label unique to the first cDNA
library or the
second cDNA library; and (d) recovering DNA which is not bound by the
plurality of
3 0 immobilized antibodies.
This invention further provides in the f fth embodiment methods wherein
DNA hybridized from each of the plurality of other cDNA libraries is in excess
relative to
- 6 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
the first and second cDNA libraries. Furthermore methods are provided wherein
the excess
is from 2-fold to a 100-fold excess, wherein the excess is from 2.5 fold to a
10-fold excess
and wherein the excess is a 3-fold excess.
This invention further provides in the fifth embodiment methods wherein the
label unique to each library is a peptide label. Furthermore methods are
provided wherein
the peptide label is from 3 to 12 amino acid residues. Furthermore, methods
are provided
wherein the label unique to each library is a thermophilic protein label.
This invention further provides in the fifth embodiment methods wherein
each of the plurality of antibodies in step (c) is immobilized on a separate
affinity column.
Furthermore a method is provided wherein the separate affinity columns are
physically linked in series any in order.
This invention further provides in the fifth embodiment a method wherein
the column flow-through is applied to the separate,physically-linked affinity
columns one
or more times. Further a method is provided wherein the column flow-through is
applied to
the separate, physically-linked affinity columns three times.
This invention further provides in the fifth embodiment methods wherein
DNA recovered in step (d) is further contacted with an antibody specific for
the label unique
to the first cDNA library or the label unique to the second cDNA library so as
to concentrate
2 0 cDNA fragments specific to the first cDNA library and the second cDNA
library.
Furthermore, methods are provided wherein the concentrated cDNA fragments
specific to
the first cDNA library and the second cDNA library are recovered and cloned.
In addition
methods are provided wherein the concentrated cDNA fragments specific to the
first cDNA
library and the second cDNA library are separated. Furthermore, a method is
provided
2 5 whereby the separation is carried out by denaturation, coating with single-
strand binding
protein and contacting with an antibody specific for the label unique to the
first cDNA
library or the second cDNA library.
In a sixth embodiment this invention provides methods for matrix analysis of
a plurality of cDNA libraries comprising: (a) labeling, cDNA inserts from each
of the
3 0 plurality of libraries with a distinguishable label; (b) hybridizing cDNA
inserts labeled in
step (a); (c) contacting cDNA inserts hybridized in step (b) with an affinity
column capable
of binding a distinguishable label; and (d) eluting the affinity column.
CA 02344625 2001-04-02
WO 00123622 PCTIUS99/23906
This invention further provides in the sixth embodiment additional methods
wherein the distinguishable label is a peptide Label, and the step of labeling
comprises
priming PCR from cDNA library vector sequences by use of an oligonucleotide
primer pair
having said peptide label attached to the 5 ends of said primer pair.
This invention further provides in the sixth embodiment a method wherein
the labeled cDNA fragments from each library are hybridized in equal
proportions.
This invention fixrther provides in the sixth embodiment methods wherein
the affinity column capable of binding a distinguishable label is an antibody
affinity
x 0 column. Furthermore a method is provided wherein the antibody-affinity
column is eluted
with a pH gradient.
This invention further provides in the sixth embodiment a method wherein
eluted DNA is denatured to separate strands originating from two different
libraries.
Furthermore, a method is provided wherein the denatured strands are isolated
by: (a)
coating with single-strand binding protein and {b) contacting with an affinity
column
capable of binding a distinguishable label.
The methods of this invention may also be used in another embodiment to
construct subtracted cDNA libraries. The sequences obtained from any of the
above-
described procedures may be used to remove a homologue from libraries known to
share
2 0 such homologue or from any given unknown library. The library to be
subtracted is present
as purified double-stranded clones. This embodiment utilizes the ability of E.
toll RecA
protein to form stable triple-stranded structures between homologous
sequences. Such
triple-stranded structures are present as RecA coated single-stranded
filaments and double-
stranded linear and circular duplexes. The method of this embodiment
comprises:
(a) ~piifying and labeling sequences identified as being shared by different
libraries by
PCR using vector-specific primers with 5'-peptide tag; (b) denaturing the
tagged sequences;
(c) cooling (e.g., flash freezing) the denatured products to prevent
renaturation; (d) adding
an aliquot of RecA protein and non-hydrolyzable ATP (e.g., ATP~yS); (e)
thawing the
mixture; and (f) adding an aliquot of the Library to be subtracted in the form
of closed
3 0 circular clones (e.g. plasmids or phagemids).
_ g _
CA 02344625 2001-04-02
WO 00123622 PCT/US99I23906
4. BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1. Schematic representation of the results obtained from Phase I, Phase
II and Phase III, respectively, of the antibody affinity columns used for
sorting labeled
cDNA fragments.
FIG. 2. Schematic representation of the cDNA fragments comprising the
input and output of Phase IV of the sorting process.
FIG. 3. RNA:DNA hybrid produced by cDNA first-strand synthesis,
including the target-specific primer on the 5' end of the cDNA strand.
FIG. 4. Ligation of the partly double-stranded standard primer of the
RNA:DNA hybrid, including partial ligation to RNA strand only.
5. DETAILED DESCRIPTION OF THE INVENTION
The present invention provides methods referred to generally as ValiGeneS""
Peptide-Labeled Oligonucleotide methods (VG-PLOS"" methods) for manipulating
(e.g.,
labeling, sorting, isolating and/or screening) two or more complex populations
of nucleic
2 0 acids. Such nucleic acids may be derived from a variety of sources,
typically cDNA
libraries representing different phenotypes. For example, the cDNA libraries
used may
represent phenotypes present (i) concurrently within a given tissue (e.g.
normal and
cancerous portions of a biopsy specimen) (ii) within different tissues
belonging to the same
or different individuals, {iii) among different cell Lines, or (iv) within the
same cell line
2 5 subjected to one or more different treatments.
In the methods of this invention generally, polyrnerase chain reaction (PCR)
is employed to Iinearize inserts within a given cDNA library and to affix a
distinguishable
and identifiable label to all inserts from the given cDNA library so as to
indicate the library
(and thus cell type, tissue or organism) of origin. In a preferred embodiment,
3 0 oligonucleotide primers to which peptide labels are linked are used to
prime PCR reactions
from vector sequences of cDNA libraries.
_ g _
CA 02344625 2001-04-02
WO 00/23622 PCT1US99123906
Throughout this application reference is made to peptide labels and
antibodies for binding said labels. In addition to peptide-antibody
combinations, it will be
understood by those skilled in the art that any label can be used, in
combination with a
suitable binding partner. Examples of such labels and binding partners
include, but are not
limited to, digoxigenin-antidigoxigenin, biotin-streptavidin, ligand (e.g.
hormone)-receptor
and carbohydrate-lectin combinations.
As used herein, the complexity of nucleic acid populations or mixtures will
be understood by one of ordinary skill in the art to refer generally to the
number of
distinguishable clones in any given cDNA library or mixture of libraries. The
complexity
of nucleic acids analyzed by the methods of the invention may vary over a very
broad range.
Generally, there is no upper or lower limit on the complexity of a population
or mixture to
be analyzed. For example, in one embodiment the complexity of a population or
mixture
may be from 10 to 10,000,000. Further, the complexity may be from 100 to
1,000,000.
Still further, the complexity may be from 500 to 500,000. In a preferred
embodiment; the
complexity of a mixture analyzed is about (~ 20%) 150,000. In another
preferred
embodiment, the mixture of complexity (~ 20%) 150,000 comprises five
libraries; each
library having a complexity of about 30,000. In another specific embodiment,
the
complexity of the population being analyzed is at least 10~, I04, 10$, or lOb.
2 0 The methodology of the invention utilizes library-specific labels linked
to
primer pairs capable of recognizing vector sequences of the vector used to
construct the
library. In this way, all nucleic acid fragments generated by PCR
amplification with such
primers have identical vector sequences at their 5' and 3' ends, yet also have
a
distinguishable label indicating the library-of origin. Described below
2 5 axe methods for: sorting cDNA fragments to isolate those distinguishable
to a single cDNA
library; sorting cDNA fragments to isolate those common to two libraries;
sorting cDNA
fragments to isolate those common to multiple but not all libraries; and
sorting cDNA
fragments to isolate fragments shared only by two libraries out of all
libraries analyzed. In
addition, set forth below are methods to construct subtraction libraries to
monitor gene
3 0 expression events, and to isolate full-length transcripts of partial
length sequences, such as
expressed sequence tags.
- 10 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
S.1 IDENTIFICATION OF cDNA FRAGMENTS SPECIFIC TO ONE OF
A PLURALITY OF cDNA LIBRARIES
In this embodiment, the inserts from each of a plurality of cDNA libraries
(e.g. A, B, C, D and E) are each linearized and tagged by PCR with a
distinguishable label
(e.g., a peptide tag that comprises an epitope) attached to vector-specific
primers. The
nucleotide sequences of a vector-specific primer pair can be identical among
libraries.
However, each label (tag) is specif c to a library-of origin. In this way, all
fi-agments
produced have (a) identical nucleotide sequences at their S' and 3' ends
corresponding to
the vector-specific primer sequences and {b) a label indicating the library-of
origin. The
labeled inserts are then purified, e.g., by exclusion chromatography, to
remove all reaction
components, including excess peptide-labeled primers. These purified, tagged
PCR
products are then combined, heat-denatured and allowed to reanneal.
The conditions under which renaturation and hybridization are corned out
can vary. In a preferred embodiment of this invention, the combined, heat-
denatured PCR
products are maintained together at 98°C for ten {10) minutes.
The;solution is then allowed
to cool from 98 °C to 85 °C over a period of five (5) minutes.
The temperature of the
mixture is maintained at 85 °C for ten (10) minutes and then cooled to
65 °C over a period of
fifteen (15) minutes. The solution is then maintained at a temperature of 65
°C for a further
2 ~ time period of fifteen (15) minutes. At this point, the reannealing
process is considered
complete.
Here also, the quantity of each PCR reaction product used in the hybidization
reaction can vary. Since isolation of cDNA fragments specific to one library
is desired, the
other libraries are used in excess (i.e., as a "mop"). Specifically, where one
wishes to
isolate fragments present in library A but not in B, C, D and E, one uses
excess B, C, D and
E. The amount of excess B, C, D, and E used will determine the efficiency of
removal. In a
preferred embodiment, 3-fold excess of B, C, D and E is used. In another
embodiment,
from 2-fold to 100-fold excess B, C, D and E is used. In yet another
embodiment, from 2.5
to 10-fold excess B, C, D and E is used. Since the function of using excess B,
C, D and
3 0 E is to act as a "mop" for removal of homologous cDNA fragments from the
library-of
interest (in this example, library A), there is no restriction on the upper
limit of the excess
- 11
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
that can be used. However, a 3-fold excess will efficiently remove the
fragments from
library A that will form hybrids with B, C, D and E without being wasteful.
Aliquots of the reannealed mixture are then contacted with one or more solid
phases, e:g., by passage through separate chromatography columns, having a
binding
partner, preferably antibodies (Abs), specific for tags B, C, D and E,
respectively.
Alternatively, a single solid phase, e.g., column, having antibodies specific
for all four tags
may be used. Methods of making Ab affinity columns are well known. For
example,
agarose beads (e.g., SepharoseTM or Sepharose CL, Pharmacia) may be activated
by use of
c~'bonyldiimidazole or cyanogen bromide for Ab attachment. The beads are
washed with
dioxane in water and incubated at room temperature with carbonyldiimidazole.
After
incubation the beads are again washed with dioxane. The purified solution of
the desired
antibody may then be added to the activated beads and mixed overnight at room
temperature. The beads are then washed with 1 M NaCl and enthanolamine is
added. The
beads are then ready for binding to the antigen, or may be stored. Many other
methods for
preparing antibody affinity columns are known to those skilled in the art {see
especially,
Antibodies - A Laboratory Manual, Harlow, Ed. Lane, D., Cold Spring Harbor
Laboratory
Press, 198$, pp. S19-540).
Many different antibodies can be used in the single and multiple antibody-
2 0 affinity columns used in the various embodiments of this invention. In a
preferred
embodiment, an antibody is used that specifically binds to a short peptide of
from 6 to 12
amino acids that is used as the label/tag. However, peptides of any length can
be used as
"tags" if the chosen peptide is known to spontaneously renature following heat
denaturation
(e.g. thermophilic proteins}. in a preferred embodiment, the antibody releases
the retained
2 5 peptide label when placed in a weakly acidic solution {e.g., pH S.S).
Elution of antibody
affinity columns on the basis of pH gradient is another preferred embodiment
of this
invention.
The flow-through can be applied to each B, C, D and E column or other solid
phase one or more times. Multiple times is preferred. After several cycles,
all or most of
3 0 the fragments bearing B, C, D or E tags, on either single or double-
strands, will be trapped.
In a most preferred embodiment, the flow-through is applied to the column
three times. The
- 12 -
CA 02344625 2001-04-02
WO OO/Z3622 PCT1US99/23906
flow-through will generally contain single and double-stranded fragments
bearing the A tag
only.
The temperature for running the antibody affinity columns may vary. In a
preferred embodiment the temperature is from 4°C to 50°C. In a
most preferred
embodiment, the antibody-affinity columns are water jacketed and maintained at
a
temperature of 37°C.
The B, C, D and E columns are next washed with a low-salt (e.g., SOmM
NaCI) buffer (e.g., phosphate buffer). The flow-through and washes are pooled.
The
pooled flow-through and washes from the B, C, D, and E columns are then passed
through a
column containing an A-specific antibody only. The trapped A-tagged fragments
can then
be eluted, precipitated and amplif ed using PCR with unlabeled, vector-
specific primers. In
a preferred embodiment, the A-tagged fragments are amplified using 20 cycles
of PCR and
cloned for analysis. These cloned fragments are highly enriched for fragments
specific to
library A (i.e., fragments not found in library B, C, D, or E).
Any method known in the art may be used to elute peptide-labeled fragments
from antibody affinity columns used in the various embodiments of this
invention. In a
preferred embodiment, as mentioned above, columns are eluted by changing pH
(pH
gradient). In a most preferred embodiment, the antibodies, or other solid
phase, chosen will
2 0 release the peptide label in a weakly acidic pH (e.g., 5.5).
The same series of steps can be used to isolate cDNA fragments tagged with
B, C, D or E peptides by changing the "mop". For example, to isolate the
fragments labeled
with B, one would start with an A, C, D and E mufti-antibody affinity column
to retain
everything but B-labeled fragments. The pooled flow-through and washes of this
column
2 5 would then be passed over a column containing only a B-specific antibody.
The same
approach can be used to isolate fragments specific to the C, D and E
libraries.
In the example above, where A-specific fragments are isolated, the material
trapped in the first column (mufti-antibody column) will contain all fragments
tagged with
B, C, D and E labels. This will include hybrids containing one A-tagged
strand. This
3 0 material can also be eluted and further sorted using other embodiments of
the invention, or
as otherwise desired by the practitioner.
- 13 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
5.2 IDENTIFICATION OF cDNA FRAGMENTS SPECIFIC TO TWO OR
MORE OF A PLURALITY OF cDNA LIBRARIES
In this embodiment, the methods of this invention are used to isolate cDNA
fragments that are common to, (i.e., that will form hybrids with} cDNA
fragments from one
or more of the other libraries. For example, if one starts with a plurality of
cDNA libraries
(e.g., A, B, C, D and E), this embodiment allows the isolation of transcripts
common to the
A and C libraries (or to any other two specified libraries). In this example,
by v~iay of
explanation is described the isolation of cDNA fragments from the A library
which form
hybrid duplexes with cDNA fragments from the C library.
In this embodiment, as in Section S.l above, the inserts from each of a
plurality of cDNA libraries (A, B, C, D and E) are linearized and tagged by
PCR with a
distinguishable library-specific label attached to vector-specific primers.
Accordingly, as
above, all fragments produced have identical nucleotide sequences at their 5'
and 3' ends
corresponding to the vector-specific primer sequences, and a distinguishable
label
indicating the library-of origin. The labeled PCR products are then~purified
by exclusion
chromatography to remove all reaction components, including excess labeled
primers. The
purified, labeled PCR products are then combined, heat-denatured and allowed
to reanneal.
In this embodiment, as in the first approach, the quantity of each PCR
2 p reaction product used in the hybridization can vary. In this example,
where one is interested
in isolating A:C hybrids, one would use an excess of libraries B, D and E. The
purpose of
this excess is to act as a "mop" as in the previous approach. There is no
restriction on the
upper limit of the excess of B, D and E. However a 3-fold excess will
efficiently remove
cDNA fragments from both the A and C libraries that will form hybrids with any
fragments
2 5 from B, D and E libraries without being wasteful. Thus, a 3-fold excess is
the preferred
embodiment. In another embodiment, a 2-fold to 100 fold excess of B, D and E
over A and
C is used. In another embodiment, a 2.5 fold to 10 fold excess of B, D and E
over A and C
is used. The degree of excess of B, D and E will determine the efficiency of
the "mop". An
excess of less than 2-fold could be used, but signif cant quantities of A or C
fragment which
~ 0 could hybridize to fragments in B, D or E may remain at the end of the
procedure.
The reannealed mixture is then passed through a chromatography column
containing antibodies specific to the labels on all the libraries except the
two of interest. In
- 14 -
CA 02344625 2001-04-02
WO 00/23622 PCTIUS99/23906
this example, the column would contain antibodies to labels on libraries B, D
and E. The
flow-through of this mufti-antibody column can be applied one or more times.
However,
multiple times is preferred since each pass-through will increase the
percentage of B, D or
E-labeled fragments which will be retained in the column and therefore removed
from the
flow-through. After several cycles, alI or most of the fragments which bear
the B, D or E
label will be removed and the flow-through will contain only fragments bearing
the A or C
label. These fragments will consist of A:A and C:C duplexes and, in addition,
may contain
A:C hybrids (i.e., the fragments-of interest). These A:C hybrids contain cDNA
fragments
~ 0 from the A library which formed hybrid duplexes with fragments from the C
library and
were not removed by the "mop" of fragments from the B, D and E libraries.
The A:C hybrids are the product of interest. The mixture containing A:A
and C:C duplexes and A:C hybrids is next passed through an antibody affinity
column with
immobilized anti-A label antibody. This column will retain all or most
fragments which
~5 bear at least one A-label. This will include A:A duplexes and the A:C
hybrids of interest.
The flow-through may be applied to this single-antibody column one or more
times. The
amount of A-labeled fragments retained will be increased with each pass. In a
preferred
embodiment, the flow-through is passed through the column three times.
The material retained by this column is now eluted and the trapped material
2 p recovered and precipitated. The recovered material consists of A:A
duplexes and A:C
hybrids. These double-stranded fragments are then heat denatured.
Immediately after the denaturation, the resulting single strands are cooled
rapidly to prevent renaturation. For example, this can be accomplished by
rapidly cooling
the heat-denatured material on a bath of dry-ice and methanol. Single-strand
binding
25 protein (SSB) is added to the frozen mixture. This protein will stabilize
single DNA strands
by coating them, thereby preventing renaturation. In this example, the SSB is
then added in
excess to the frozen mixture of denatured single-strands. This frozen mixture
is then
warmed to allow the SSB to enter the solution and contact the single strands.
In a preferred
embodiment, the mixture is heated from the temperature of the dry-ice/methanol
bath to
3 0 37°C and maintained at that temperature for a few minutes (e.g.,
between 5 and 10
minutes). The SSB will coat and stabilize the single DNA strands and will
prevent
reformation of hybrids and the formation of secondary structures.
- 15 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99lZ3906
The single strands of DNA consist of fragments from the A library that
formed hybrids with other fragments from the A library, and fragments from the
C library
that formed hybrids with fragments from the A library. The C library fragments
present at
this stage will be limited to those fragments that were able to hybridize with
A library
fragments and were thus retained on the anti-A antibody column. In addition,
these C
library fragments did not hybridize with and thus were not removed by the
excess of B, D
and E fragments used as the "mop". Accordingly, these C library fragments are
represented
in the A library but not in the B, D, or E library.
In this embodiment, a further step is now employed to separate the SSB-
coated A and C single strands. This step consists of passing the SSB-coated
single-strands
through an antibody-affinity column. This column may contain either
immobilized anti-A
antibody or immobilized anti-C antibody. If the anti-A column is used, then A-
labeled
fragments will be retained by the column and the C tagged fragment will remain
in the
flow-through and washes. If the anti-C antibody column is used, then the C-
labeled
fragments will be retained by the column and may be eluted from this column.
In either
case, the A-labeled fragments and the C-labeled fragments can be recovered.
The recovered
fragments are then extracted to remove SSB, and PCR amplif ed. These C-labeled
fragments may be cloned for further analysis or they can be used as pooled
probes (i.e.,
2 0 "subtraction probes") to remove their homologues from the original A or C
libraries (see
e.g., Section 5.4 below).
In a variation of this embodiment of the invention, more than two libraries-
of interest may be designated. For example, if one is interested in fragments
that may form
hybrids between library A, B and C but not with libraries D and E, then the
first step would
2 5 employ an excess of D and E PCR reaction products over those of A, B and
C. In this
embodiment the multiple antibody column would contain anti-D and anti-E
antibody. The
excess of D and E used would form the "mop" to remove cDNA fragments that
formed
hybrids with any fragments in the A, B, or C libraries. The flow-through and
washes of this
mufti-antibody column would contain A:A, B:B and C:C duplexes but would also
contain
30 A:B, A:C and B:C hybrids if any had formed. These hybrids are the products-
of interest in
this embodiment. If this material is now passed through an anti-A column, then
A:B and
A:C hybrids will be retained. An anti-B antibody column will retain any A:B
and B:C
- 16 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
hybrids, and an antibody column containing anti-C antibody will retain C:B and
C:A
hybrids. The material retained on these three columns may be eluted and
isolated by the
methods used above. This would consist of denaturating the double-stranded
hybrids, rapid
cooling on a dry-ice/ methanol bath followed by warming in contact with an
excess of SSB.
The SSB coated single-strands could now be isolated by passing through an
antibody
column containing a single antibody specific to either A, B, or C label: The
fragments
eluted from these columns would contain the cDNA fragments of interest.
5.3 MATRIX ANALYSTS OF A PLURALITY OF LIBRARIES
In another embodiment of the invention, an array or matrix comparison is
made among a plurality of cDNA libraries constructed and manipulated, in part,
using
variations of the procedures set forth above. In this embodiment, any cDNA
fragments
present in two of N libraries can be isolated, where >N is the total number of
libraries
~ 5 subjected to the matrix analysis. Further, any cDNA fragments present in X
of N libraries
can be isolated, where X is the number of libraries in which the paxficular
cDNA fragments
isolated are found. Still further, any cDNA fragment present in only two {or
three, or X) of
N libraries can be isolated. Indeed, the cDNA fragments common to any desired
number
(X) of any number (I~ of libraries analyzed can be isolated, and whether or
not these
2 0 fragments are exclusively shared among a subset of N libraries analyzed
can be determined.
A major advantage of this embodiment is the absence of a necessity for
having any knowledge of which genes (i.e. cDNA fragments) may or may not be
represented in a given cDNA library before beginning the comparative analysis.
Further,
the degree or extent of homology (i.e. similarity) among cDNA fragments
obtained from a
2 5 plurality of libraries also need not be known. Still further, one need not
know whether a
specific library-of interest shares any similar cDNA inserts with any other
libraries prior to
beginning the analysis. In summary, the results of a matrix analysis reveals
which cDNA
fragments are common (i.e. similar enough to hybridize) in any two or more of
a plurality of
libraries-of interest.
3 0 To illustrate this embodiment, we again start with five cDNA libraries (A,
B,
C, D and E). First, the cDNA inserts of each library are separately linearized
and labeled by
PCR with a label distinguishable to each library. Again, a 5'-peptide label is
preferred. As
CA 02344625 2001-04-02
WO 00/23622 PCTIUS99/23906
above, the peptide label distinguishable to each library is attached to the S'
ends of an
oligonucleotide primer pair used to prime PCR from library vector sequences.
After the PCR linearizing and labeling procedure, the contents of each
labeled library is separately purified, e.g., by exclusion chromatography.
This step purifies
the linearized, labeled inserts away from unwanted reaction components, such
as excess
peptide-labeled primers. This purification can be performed by any of the
standard methods
well known in the art (e.g., PCR.purification kit from Qiagen, Santa Clarita,
California).
The purified and distinguishably labeled cDNA fragments from each of the
five libraries are then mixed together, heat denatured and allowed to re-
anneal. The relative
proportions of material from each library mixed together in this reaction is
determined by
user discretion. The reaction may or may not employ an excess of material from
one or
more libraries over another one or more libraries. In a preferred embodiment,
the labeled
cDNA fragments from each library are mixed together in equal proportions.
As in the methods of the invention already described, this embodiment
performs a comparative analysis of cDNA libraries based on hybridization and
sorting of
labeled cDNA inserts. Here, however, the analysis employs up to four stages or
Phases of
affinity columns, or any solid phase, capable of binding specific library
labels. As before,
any label known in the art suitable for labeling cDNA strands by PCR may be
used.
2 0 F~er, any affinity column, or any solid phase, known in the art, capable
of binding such
labels may be used. In a preferred embodiment, the affinity columns are
antibody-affinity
columns capable of binding peptide labels.
The precise number of Phases employed in this embodiment, and their order
of use, is determined in part by the result desired by the user. In this
regard, all products of
2 5 the comparative array created need not be analyzed and may be stored. For
example, where
one wishes to isolate inserts shared between any two libraries-of interest and
one is not
concerned with isolating inserts exclusively present in these two libraries,
then the analysis
need only proceed through Phases I and II. However, where one wishes to
isolate fragments
exclusively present between libraries within a library Group, then Phases I,
II and III are
3 0 employed. Further, where one wishes to isolate fragments exclusively
present in libraries
from different library Groups, then Phases I, II and IV are employed. A
library "Group" is
defined as the eluent obtained from a Phase I column, as further set forth in
the sections
- 18 -
CA 02344625 2001-04-02
WO 00/23622 PCTIUS99/23906
below. The following narrative describes the steps employed to achieve the
results just
described. As noted above, while described in terms of antibody affinity
columns and
peptide labels, other types of solid phases with other types of binding
partners to other types
of label, can be used.
5.3.1 PHASE I
In Phase I, the re-annealed mixture of labeled cDNA fragments is applied
sequentially to a series of antibody affinity columns, each column having an
immobilized
antibody capable of recognizing only one of the library labels. Alternatively,
the re-
annealed mixture may be divided into aliquots and applied to each column
individually. In
this five-library example, five columns are used. The order of application of
the re-
annealed mixture to the five individual columns can be any order. For example,
the order
can be A column, B column, C column, D column and E column, respectively. In a
Preferred embodiment, the five columns are physically linked in series. This
arrangement
has the advantages of efficiency in running the columns, of minimising the
volume applied
to the columns, and of reducing losses of column flow-through and washes.
Where the
series of Phase I columns is physically linked, the order of the columns in
the series is again
not important and can be any order.
2 0 Each column in the Phase I series will trap or retain cDNA fragments
having
one specific label. Thus, the A column will retain all A-labeled fragments,
the B column
will retain all B-labeled fragments, etc. Fragments retained, for example, by
the A column
consist of A-labeled hybrids (i.e. A:A, A:B, A:C, A:D and A:E duplexes) and
single-
stranded A-labeled DNA.
2 5 Each of the five Phase I columns is next eluted individually. Where
columns
A through E were physically linked for application of the cDNA mixture and
washes, they
are dis-assembled prior to elution. The material obtained from elution of each
Phase I
column defines a separate library Group, as follows:
3 0 Library Group A
from A column - double-stranded A:A, A:B, A:C, A:D and A:E
duplexes, and single-stranded A-labeled DNA;
_ Zg _
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
Library Group B
from B column - double-stranded B:A, B:B, B:C, B:D and B:E
duplexes, and single-stranded B-labeled DNA;
Library Group C
from C column - double-stranded C:A, C:B, C:C, C:D and C:E
duplexes, and single-stranded C-labeled DNA;
Library Group D
from D column - double-stranded D:A, D:B, D:C, D:D and D:E
duplexes, and single-stranded D-labeled DNA; and
Library Group E
from E column - double-stranded E:A, E:B, E:C, E:D and E:E
duplexes, and single-stranded E-labeled DNA.
~,g In this example; the material-of interest is duplex DNA formed from
hybridization of strands originating in two different libraries. Where five
libraries are used
for the input mixture of Phase I, as in this example, tyventy different
duplexes-of interest
may be formed. For example, in column A, trapped library Group A duplexes-of
interest
consist of A:B, A:C, A:D and A:E duplexes. In column B, trapped library Group
B
2 0 duplexes-of interest consist of B:A, B:C, B:D and B:E duplexes, etc. See
FIG: 1, Phase I,
for a complete listing of the array of products produced. Duplexes having
identical labels
on each strand and any single-stranded DNA trapped in Phase I columns is not
generally of
interest and is therefore not shown in FIG. 1.
2 5 5.3.2 PHASE II
The cDNA fragments (i.e. transcripts) shared between any two libraries-of
interest is isolated in Phase II. Here, the eluent from each of the five Phase
I columns is
rendered single-stranded prior to input over Phase II columns. This may be
performed by
any method known in the art. In a preferred embodiment, single-strand binding
protein
3 0 (SSB) is used. Thus, the eluent from each of the five Phase I columns
(Groups A through E
in FIG. 1) is separately heat-denatured and rapidly cooled (e.g. dry-
ice/methanol bath). An
excess of SSB is added, and each SSB-plus-DNA mixture is then warmed. In a
preferred
- 20 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99123906
embodiment, the temperture is increased to and maintained at 37°C for
10 to 15 minutes.
SSB coats denatured single-strands and prevents renaturation and formation of
secondary
structures.
The output of each of the five Phase I single-antibody columns, now
denatured and stabilized with SSB, is next applied to a series of Phase II
columns. in a
preferred embodiment, each series of Phase II columns is physically Linked.
Each Phase II
column contains a single immobilized antibody specific for one of the
distinuishable .
peptide labels used to label the input cDNA libraries: A Phase II series of
columns may
~ ~ contain as many columns as the number of cDNA libraries N being analyzed.
In an
alternative embodiment, a Phase II series of columns contains N-1 columns,
where the
omitted column corresponds to the Library Graup (e.g. for library Group A, the
A column
may be omitted). The immobilized antibody in each Phase II column captures
single
strands bearing one of the distinguishable peptide libels. Each Phase II
column is then
separately eluted. In this way, pan array of twenty groups of single-stranded
cDNA
fragments is isolated wherein each of the twenty groups contains fragments
shared (i.e.
hybridizable) between two libraries (see FIG. 1, Phase iI).
For example, after elution of each Phase II column in the A library Group
(N-1 approach), the following cDNA fragments are isolated in single-stranded
form:
2 a B column - cDNA fragments originating from the B library also
present in the A Library;
C column - cDNA fragments originating from the C library also
present in the A library;
D column - cDNA fragments originating from the D library also
2 5 present in the A library; and
E column - cDNA fragments originating from the E library also present in
the A library.
As a further example, after elution of each Phase II column in the B library
Group under the N column approach, the following cDNA fragments are isolated
in single-
3 0 stranded form:
B column - B cDNA fragments only;
- 21 -
CA 02344625 2001-04-02
WO 00/23622 PCTlUS99I23906
A column - cDNA fragments originating from the A library also present
in the B library;
C column - cDNA fragments originating from the C library also present
in the B library;
D column - cDNA fragments originating from the D library also present
in the B library; and
E column - cDNA fragments originating from the E library also present in
the B library.
A similar list of isolated cDNA fragments can be constructed for each series
of Phase II columns, thereby completing the array {see FIG. 1, Phase II).
Of course, any fragments in the array created by the output of the Phase II
columns may be cloned for further analysis as desired by the user. Such
fragments may
also be used as "combination probes" for retrieval of corresponding double-
stranded clones
from an existing library using, for example, the RecA method detailed
elsewhere herein.
These fragments are also used as input to Phase III and Phase IV foi'r
isolation of cDNAs
exclusively present in two or more desired libraries, either within or across
library Groups,
respectively, as further set forth below.
2 0 5.3.3 PHASE III
Phase III is used to isolate fragments shared exclusively between two or
more designated members within a library Grvup. For example, Phase III allows
isolation
of fragments shared exclusively between libraries A and B, A and C, A and D,
and A and E,
within library Group A. Further, Phase III allows isolation of fragments
shared exclusively
2 5 between libraries B and A, B and C, B and D, and B and E, within library
Group B. This
pattern is equally applicable to library Groups C, D and E.
Phase III begins with single-stranded fragments eluted separately from each
of a series of Phase II columns in a chosen library Group as described above.
These
fragments are first independently amplified by PCR and labeled with the
relevant peptide
3 0 label. For example, where library Group A is being subjected to a Phase
III analysis,
fragments eluted from the B column of Phase II are amplified using the B-
specific peptide
- 22 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
label, fragments eluted from the C column of Phase II are amplified using the
C-specific
peptide label, etc.
All the independent members belonging to a given library Group are
likewise amplified and labeled by PCR. The amplified products are then mixed,
denatured
and allowed to re-anneal. In this example (library Group A where libraries A
through E are
being analyzed), the reannealed Phase III input mixture is then divided into
four aliquots.
The number of aliquots needed depends on the number of separate labels in the
mixture. In
this example, the A library Group is analyzed and the labels used during the
PCR
~Pliffing process are B, C, D and E. The Phase III columns consist of four
series of
affinity columns, each series consisting of three single-antibody calumns.
Each of these
four series of columns contain antibodies specific to three of the four labels
used in the
Phase III PCR amplification step. Where, as here, the A library Group is
subjected to Phase
III analysis, the four series of affinity columns contain:
Series 1 - C, D and E antibodies;
Series 2 - B, D and E antibodies;
Series 3 - B, C and E antibodies; and
Series 4 - B, C and D antibodies.
Phase III Series 1 retains any cDNA fragments labeled with C, D and E,
2 0 allowing B-labeled duplexes and single strands to remain in the flow
through. Within
library Group A, these cDNA fragments are present in libraries A and B, but
not in C, D or
E. Therefore, cDNA fragments exclusively present in libraries A and B have
been isolated.
In the same manner, Phase III Series 2 allows only C-labeled fragments to
pass, Phase III
Series 3 allows only D-labeled fragments to pass, and Phase III Series 4
allows only E-
z 5 labeled fragments to pass. Here, cDNA fragments exclusively present in
libraries A and C,
A and D, and A and E, respectively, have been isolated.
Thus generally, in each series column, one uses one column less than the
number of labels used in the amplifying step. Further, one uses enough series
to cover all
different combinations of columns.
3 0 In an alternative embodiment, the flow-through of each of the four Series
of
multi-antibody columns just described above is next passed through another
antibody
column. These columns each contain a single antibody which is specific for the
labeled
- 23 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99123906
fragments allowed to pass in the Phase III Series columns. This step serves to
concentrate
the fragments, which otherwise might be difficult to recover from a large
volume of flow-
through and washes. The fragments retained by these four single-antibody
columns are
eluted and recovered. This material consists of concentrated cDNA fragments
that are
uniquely shared between two specific libraries. In this example, the fragments
recovered
are uniquely shared between libraries A and B, A and C, A and D, and A and E
within
library Group A.
5.3.4 PHASE IV
The output from Phase II can be further analyzed in Phase IV to determine
whether cDNA fragments shared between any two libraries-of interest in the
array are
distinguishable across library Groups rather than within library Groups. The
Phase IV
analysis thus complements the.Phase III analysis by allowing one to ask
essentially the
same question using different input cDNA fragments. The user thus benefits by
comparing
the results of a Phase III analysis with the results of a Phase IV analysis.
As in Phase III,
the input DNA for Phase IV analysis is obtained from the output of Phase II.
However, the
labels attached in the PCR reactions prior to Phase IV analysis correspond to
the library
Group label and not to the original label of the fragment (see Box in FIG. 2).
2 0 Thus, importantly, the labels attached in the PCR prior to Phase IV
analysis
of Phase II products do not correspond to the library-of origin of a
particular fragment.
Instead, the labels correspond to a library in which a given fragment has
found a homolog
(i.e. the library Group). In this way, an analysis similar to the Phase III
analysis can be
performed across library Groups.
2 5 For example, to perform a Phase IV analysis across library Groups, all
fragments originating from A library and recovered from a B group column in
Phase II are
labeled with B peptide label (see Box in FIG. 2 at "Tag-B"). In a similar
fashion, all
fragments originating from A library which were recovered from a C, D, or E
group column
in Phase II are labeled with C, D and E peptide label, respectively (see Box
in FIG. 2 at
3 0 "Tag-C", "Tag-D" and "Tag-E", respectively).
After PCR amplification and labeling, all such differentially-labeled A-
Library fragments are mixed, denatured and allowed to re-anneal. The re-
annealed mixture
- 24 -
CA 02344625 2001-04-02
WO 08123622 PCT/US99/23906
is then divided into four aliquots, and each aliquot is passed over a mufti-
antibody affinity
column similar to the Series 1-4 columns of Phase III. Thus, each mufti-
antibody column
contains three label-specific antibodies. Where, as here, Phase IV analysis is
performed far
fragments originating in A library, four Series of Phase IV columns contain:
Series 1 - C, D and E antibodies;
Series 2 - B, D and E antibodies;
Series 3 - B, C and D antibodies; and
Series 4 - B, C and E antibodies.
As fax Phase III analysis, the flow-through and washes from each Phase IV
Series column may then be pooled and applied to a column containing a single
antibody
specific for the one label that remained untrapped. For example, the output of
the Phase IV
Series 1 column above would be pooled and passed over a column containing anti-
B.
Representative results for fragments originating in A library and for
fragments originating
in B library are shown in FIG. 4 (see "output of Phase IV"). As was the case
described for
the output of Phase III, the material eluted from each single-antibody column
in Phase IV
consists of concentrated cDNA fragments shared exclusively by two libraries
(i.e., not
found in the other libraries of the analysis).
2 0 5.3.5 FURTHER CONSIDERATIONS
One can also isolate fragments common to three (or more) libraries, to the
exclusion of others, by manipulating the Phase III and Phase IV Series columns
to remove
fewer fragments. For example, in a Phase IV analysis of fragments originating
in library A
directed to isolation of fragments present in A, C and E, but not in B or D,
one could run a
x 5 Phase IV Series column containing just B and D antibodies.
5.4 USE OF METHODS TO CONSTRUCT SUBTRACTED
LIBRARIES
In another embodiment, the methods of this invention can be used to
30 construct subtracted cDNA libraries, i.e., to remove similar clones from
two or more cDNA
libraries. The method described herein takes advantage of the E. coli RecA
protein's ability
to form stable triple-stranded structures as recombination intermediates. RecA
catalyzes a
- 25 -
CA 02344625 2001-04-02
WO 00/23622 PCTIUS99J2390b
homologous pairing and strand exchange reaction during E: coli homologous
recombination. During the first step of this reaction, RecA coats a single
strand of DNA
and initiates an exchange reaction between the single strand and a homologous
region of
double-stranded DNA. A three-stranded nucleoprotein intermediate is formed,
which, in
the absence of ATP, is surprisingly stable (see West, 1992, Annu. Rev.
Biochem., 61:603-
640).
cDNA fragments-of interest, such as those shared by different libraries
identified as described above, are amplified by PCR using peptide-tagged
vector-specific
primers. Thus a distinguishable peptide tag marks a given set of cDNA
sequences. Such
sets of fragments are used concurrently as "subtraction probes". A subtraction
probe set is
purified by exclusion chromatography following PCR and heat-denatured. The
cDNA
mixture is then flash frozen (e.g., dry-iceJmethanol}. An aliquot of RecA
protein is added to
the ice pellet along with non-hydrolyzable ATP (e.g., ATP yS). The ice pellet
is slowly
~a"~'ed. Low temperature and the ATP analog prevent the RecA-bound single-
stranded
DNA from renaturing so that the subtraction probe remains single stranded and
becomes
coated with RecA.
The library to be subtracted, in the form of purified double-stranded,
circular
DNA, is added to the thawed pellet such that the RecA-coated single strands
are present in
2 0 loge excess (20-50 fold). The mixture is heated to 37 °C. The RecA-
coated single strands
scan the double-stranded cDNA library in search of homologous sequences, and
pair with
such sequences. Triple-stranded recombination intermediates are formed,
although strand
exchange will not occur due to the absence of a hydrolyzable form of ATP. The
triple-
stranded structures formed from single-stranded DNA and homologous double-
stranded
2 5 DNA are labeled with the specific peptide tag bound to the single strand.
Such triple-
stranded, labeled structures can now be separated from unlabeled, double-
stranded circular
molecules by passing the solution through a peptide tag-specific antibody
column. Most or
all clones corresponding to the labeled fragments will be removed from the
library if the
RecA-coated single strands are present in large excess over the plasmid
clones.
3 0 The method of this embodiment is independent of the original nature of the
nucleic acid used to construct the library. It can therefore be used with DNA
libraries made
from cDNAs or genomic DNAs.
- 2& -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99123906
In a preferred embodiment, both single-stranded fragments and double-
stranded library plasmids share identical extremities (i:e., 5' and 3' ends)
over at least IO-15
bases, and the homologous fragments are at least 350 by in length. If strong
overall
homology is present, perfect identity between fragments is not required for
RecA to form
stable triple-stranded structures (see, e.g., Rao et al.,1995, Trends In
Biological Science
20:109-113). In another preferred embodiment, the cloned inserts do not exceed
I-2
kilobase (kb) in length so that clones sharing only strong localized
homologies with the
subtraction probes are not selected.
is
5.S USE OF METHODS IN THE MONITORING OF GENE EXPRESSION
In another embodiment of this invention, methods are provided to monitor
gene expression events. Oligonucleotides labeled with specific and
identifiable peptide
labels are used, but in this embodiment the targets (i. e. genes) to be
monitored for
expression are known. These targets may belong to an expression cascade, for
example, if
1' the objective is to define the mechanism of action or physiological effects
of a particular
drug treatment. An alternative use for the methods of this embodiment is to
monitor gene
expression to define a phenotype based on the activation or repression of a
specific
phenotype-associated metabolic pathway. The advantage of this method is to
provide a
simple and rapid means to sort, separate and quantify the product
(representing targets to be
2 0 monitored for expression) based on the peptide label.
In addition, the methods of this embodiment allow a direct quantitative
determination of the amount of target mRNA present. Briefly, a PCR reaction is
carned out
using unamplified cDNA from a first-strand synthesis reaction. For a fixed and
limited
number of PCR cycles (e.g., from about 5 to 20 cycles), the product of the
reaction is
2 5 directly proportional to the initial amount of non-genomic, small-size
(under 2kb) DNA
target present in the reaction. The techniques of quantitative PCR are well
known to those
skilled in the art.
The methods of this embodiment will allow the direct monitoring of gene
expression events, as well as the isolation of partial length transcripts,
without the prior
3 0 construction of the relevant cDNA libraries and starting from very small
biopsy samples
which would be too small to allow construction of new cDNA libraries.
- 27 -
CA 02344625 2001-04-02
w0 00/23622 PCTIUS99123906
The sensitivity and versatility of this embodiment allows its use to analyze
the response of a specific phenotype to a given stimulus or set of conditions.
In addition,
this method provides a rapid and accurate means of directly determining the
physiological
effects of any form of treatment which affects gene expression, such as
treatment with
steroid hormones. Since this is done directly by looking at mRNA production,
it is not
necessary to wait for the overt clinical effects to show themselves or the
production of
serological factors. The method of this embodiment could therefore be used to
make a rapid
assessment of the probable effect of treatment, or to provide rapid and direct
feed-back to
20 allow therapeutic readjustments to be made to optimize outcome.
In this embodiment, total RNA is extracted from the tissue sample using
standard methodologies well known to those skilled in the art. Total RNA is
used for
hybridization to target-specific probes. Each of these probes consists of a
synthetic
oligonucleotide labeled with a. specific peptide epiti3pe or tag at the S' end
and a fluorophore
at the 3' end. The single-stranded probes are mixed with the samples of total
RNA under
conditions allowing hybridization of the probes to their target mRNA
molecules, if present.
Following hybridization, the mix is treated with a single-strand specific
DNase in order to
destroy all non-hybridized excess probes or to effect a separation between the
peptide tag
and the fluorescence label on probes remaining single-stranded. Here, one
skilled in the art
2 0 will recognize that other detectable labels may substitute for the
fluorescence label. The
mixture is then exposed to a solid surface onto Which the tag-specific
antibodies or other
binding partners have been arrayed (e.g. an ELISA plate) hence identifying the
relative
position of each target-specific probe. Only those probes that have hybridized
to their target
will give rise to a fluorescence or other detecable signal at a specific
location on the solid
2 5 surface, the position in the array indicating the presence and identity of
the target and the
signal intensity indicating the relative abundance of each target within the
original RNA
sample.
This embodiment can also be used to isolate full- length forms of only
partial-length transcripts. Total RNA is extracted as previously and aliquots
of the total
3 0 RNA are used for cDNA first-strand synthesis using target-specific, non-
phosphorylated
primers {see, FIG. 3).
- 28 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
The synthesis makes use of an RNA-dependent DNA polymerise (i.e.,
reverse transcriptase) which does not possess RNase-H activity (such as
Moloney marine
leukemia virus reverse transcriptase). Methods for doing this are well known
to those
skilled in the art (Sambroak et al., 1989, Molecular Clonin~~A Laborator~r
Manual, 2nd Ed.,
Cold Spring Harbor Laboratory Press). The result of this synthesis is DNA:RNA
hybrids
with a target-specific primer on the 5' end of the DNA strand. Any RNA
extensions can
now be removed to produce blunt ends by treatment with mung-bean nuclease,
which
cleaves single-strand mRNA extensions.
T~s reaction can be performed under standard conditions for the use of
mung bean nuclease. For example, the DNA:RNA hybrids may be suspended in a
mung
bean nuclease Buffer consisting of 50 mM sodium acetate (pH 5.0 at 25
°C}, 30 mM NaCI,
1 mM ZnS04. Mung bean nuclease in the amount of 1.0 unit per microgram of
DNA:RNA
hybrid is added and the mixture is incubated at 30°C for thirty {30)
minutes. The enzymes
may then be inactivated by phenoUchl~roform extraction or by addition of SDS
to 0.01%.
The blunt-ended hybrids may be recovered by alcohol precipitationF: Kowalski,
D. et al.
{1976) Biochemistry 15, 4457-4463; McCutchan, T.F. et al. (1984} Science 225,
626-628.
The sample is now purified by standard exclusion chromatography. After
purification, the sample consists of the DNA:RNA hybrids together with the
remainder of
2 p the total RNA species initially present. The exclusion chromatography
removes the small
RNA species (such as tRNA) and excess target-specific primer.
At this point, a iigation reaction is corned out using DNA ligase from T4
bacteriophage. The ligase will catalyze the formation of phosphodiester bonds
between
adjacent 3'-hydroxyl and 5'-phosphate termini of DNA or RNA and wilt thus
joiwthe 3' end
2 5 of a double-stranded DNA fragment to the 5' terminus of a double-stranded
DNA:RNA
hybrid molecule. The primer used is a partly double-stranded phosphorylated
second
primer (i.e. a primer that is not target-specific), for example, a M13
"forward" sequencing
primer (see FTG. 4).
Bacteriophage T4 DNA ligase will fully ligate the primer only to the
3 0 phosphorylated end of the DNA:RNA hybrid. However, some of the primer
molecules will
also ligate to the 3' terminus of the RNA strand of the DNA:RNA hybrid. This
will not
affect the result because DNA polymerise enzyme in subsequent steps will not
use RNA as
- 29 -
CA 02344625 2001-04-02
WO 00/23622 PCTIUS99/23906
a template and because no template is available in the 3' direction and so
priming at this site
will not result in elongation by de novo synthesis.
T4 DNA ligase purified from E. toll may be obtained from New England
Biolabs (Waverly, Massachusetts). The reaction may be tamed out in T4 DNA
Ligase
Buffer which contains 50 mM Tris-HCI (pH 7.8), 10 mM MgCl2, 10 mM
dithiothreitol, 1
mM ATP, 25 microgram per milliliter bovine serum albumin. In a preferred
embodiment,
the rection is carried out at 16°C for between four (4) and sixteen
(16) hours. Engier, M.J.
and Richardson, C.C. (1982) in The Enz~n~es (Boyer P.D., ed.) Vol. 5, p. 3,
Academic
Z 0 Press, San Diego, CA.
Following the ligation reaction, the sample is again purified by exclusion
chromatography and amplified by PCR.
The PCR reaction makes use of both a peptide-tagged primer complementary
to the target-specific primer previously used, and a:biotinylated primer
complementary to
the partly double-stranded phosphorylated standard primer used in the ligation
reaction. In
a preferred embodiment, this PCR reaction includes SO nanograms of yeast RNA
per 30
microliters of solution. The number of cycles in this PCR reaction can vary.
In a preferred
embodiment, 20 or fewer cycles is used.
In a variation of this embodiment, the sample can be treated with RNase
2 0 Immediately following the ligation reaction and prior to PCR. This will
destroy all RNA
strands, including single strands of total RNA and the RNA strands of the
DNA:RNA
hybrid molecule. The sample can then be puriEed by exclusion chromatography,
and PCR
amplified and labeled as above. The amplified product is then purified by
exclusion
chromatography to remove all excess primers.
The amount of product produced by the PCR reaction can now be quantified
by a modification of an enzyme-linked immunoassay technique (ELISA). The
purified
reaction mixture may be analyzed in microtiter wells coated with an antibody
specific to the
peptide label that was attached to the target-specific primer. Streptavidin-
linked horseradish
peroxidase can then be added to bind to the biotin moiety attached to the
standard primer of
3 0 the retained PCR products. A horseradish peroxidase substrate can then be
added, and the
reaction product quantified (see e.g. Sambrook et al, 1989, Molecular Cloning
A Laboratory
- 30 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/Z3906
Manual, 2nd Ed., Cold Spring Harbor Laboratory Press at 18.75), indicating the
amount of
target mRNA present in the original sample.
In another embodiment of this method, several different targets can be
simultaneously analyzed. Two or more target-specific primers can be used in
the first-
strand synthesis reaction. Identifiable and distinct peptide-labeled primers
complementary
to the target-specific primers can be used in the PCR reaction. In this
embodiment, the
primers involved are chosen to be compatible in terms of their melting
temperatures (Tm's)
and propensities for secondary stricture formation.
S.6 CHOOSING INPUT PHENOTYPES
The input phenotypes represented by cDNA libraries employed in the
methods of this invention can be chosen as desired by one skilled in the art.
In addition,
one can use methods disclosed in co-pending United States Patent Application
entitled,
"Method For Identifying Genes Underlying Defined Phenotypes" by Iris, F. and
Pourny, J-
L., Serial No. 09/007,905, filing date January 15, 1998 for choosing in
phenotypes. This
co-pending application is incorporated herein by reference in its entirety.
5.7 METHODS AND PRODUCTS OF USE WITH THE INVENTION
2 0 5.7.1 DNA AMPLIFICATION
The polymerise chain reaction (PCR) is used in connection with the
invention to amplify a desired sequence from a source (e.g., a tissue sample,
a genomic or
cDNA library). Oligonucleotide primers representing known sequences can be
used as
primers in PCR. PCR is typically carried out by use of a thermal cycles {e.g.,
from Perkin-
2 5 Elrner Cetus) and a thermostable polymerise (e.g., Gene AmpTM brand of Taq
polymerise).
The nucleic acid template to be amplified may include but is not limited to
mRNA, cDNA
or genomic DNA from any species. The PCR amplification method is well known in
the art
(see, e.g., U.S. Patent Nos. 4,683,202, 4,683,195 and 4,889,818; Gyllenstein
et al., 1988,
Proc. Nat'1. Acid. Sci. U.S.A. 85, 7652-7656; Ochman et al., 1988, Genetics
120, 621-623;
3 0 Loh et al., 1989, Science 243, 21?-220).
Any prokaryotic cell, eukaryotic cell, or virus, can serve as the nucleic acid
source. For example, nucleic acid sequences may be obtained from the following
sources:
- 31 -
CA 02344625 2001-04-02
WO 00/23622 PCT/EJS99/23906
human, porcine, bovine, feline, avian, equine, canine, insect (e.g.,
Drosophila), invertebrate
(e.g., C. elegans), plant, etc. The DNA may be obtained by standard procedures
known in
the art (see, e.g., Sambrook et al., 1989, Molecular Cloning, A Laboratory
Manual, 2d Ed.,
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York; Glover
(ed.), I985,
DNA Cloning: A Practical Approach, MRL Press, Ltd., Oxford, U.K. Vol. I, II).
5.7.2 ADJUSTING STRINGENCY
Other methods available for use in connection with the methods of this
invention include nucleic acid hybridization under low, moderate, or high
stringency
conditions (e.g., Northern and Southern blotting). Methods for adjustment of
hybridization
stringency are well known in the art (see, e.g., Sambrook et al., 1989,
Molecular Cloning, A
Laboratory Manual, 2d Ed., Cold Spring Harbor Laboratory Press, Cold Spring
Harbor,
New York; see, also, Ausubel et al., eds., in the Current Protocols in
Molecular Biology
series of laboratory technique manuals, 1987-1994 Current Protocols, 1994-1997
John
Wiley and Sons, Inc.; see, especially, Dyson, N.J., 1991, Immobilization of
nucleic acids
and hybridization analysis, In: Essential Molecular Biology: A Practical
Approach; Vol. 2,
T.A. Brown, ed., pp. 111-156, IRL Press at Oxford University Press, Oxford,
U.K.; each of
which is incorporated by reference herein in its entirety). Salt
concentration; melting
2 0 temperature, the absence or presence of denaturants, and the type and
length of nucleic acid
to be hybridized (e.g., DNA, RNA, PNA) are some of the variables considered
when
adjusting the stringency of a particular hybridization reaction according to
methods known
in the art.
Conditions of low stringency, by way of example and not limitation, may be
as follows (see, also, Shilo and Weinberg, 1981, Proc. Natl. Acad. Sci. U.S.A.
78,
6789-6792). Filters containing DNA are pretreated for 6 h at 40°C in a
solution containing
35% formamide, SX SSC, SO mM Tris-HCl (pH 7.S), 5 mM EDTA, 0.1% PVP, 0.1%
Ficoll, 1% BSA, and S00 ~.g/ml denatured salmon sperm DNA. Hybridizations are
carried
out in the same solution with the following modifications: 0.02% PVP, 0.02%
Ficoll, 0.2%
3 0 BSA, 100 ug/ml salmon sperm DNA, 10% (wt/vol} dextran sulfate, and S-20 X
106 cpm
szp-labeled probe is used. Filters are incubated in hybridization mixture for
18-20 h at
40°C, and then washed for 1.S h at SS °C in a solution
containing 2X SSC, 25 mM Tris-HCl
- 32 -
CA 02344625 2001-04-02
WO 00123622 PCTIUS99123906
(pH 7.4), 5 mM EDTA, and 0.1% SDS. The wash solution is replaced with fresh
solution
and incubated an additional 1.5 h at 60°C. Filters are blotted dry and
exposed for
autoradiography. If necessary, filters are washed for a third time at 65-68
° C and re-exposed
to film.
Conditions of high stringency, by way of example and not limitation, may be
as follows. Prehybridization of filters containing DNA is carried out for 8 h
to overnight at
65°C in buffer composed of 6X SSC, SO mM Tris-HCi (pH 7.5), 1 mM EDTA,
0.02% PVP,
0.02% Ficoll, 0.02% BSA, and 500 ~.glml denatured salmon sperm DNA. Washing of
filters is done at 37°C for 1 h in a solution containing 2X SSC, 0.01%
PVP, 0.01% Ficoll,
and 0.01% BSA. This is followed by a wash in O.1X SSC at 50°C for 45
min before
autoradiography.
.5.7.3 OLIGONUCLEOTIDE ANALOGS
Nucleic acids used in conjunction with the device of the invention are often
oligonucleotides ranging from 10 to about 50 nucleotides in lengtYi. in
specific aspects, an
oligonucleotide is 10 nucleotides, 15 nucleotides, 20 nucleotides or 50
nucleotides in length.
An oligonucleotide can be DNA or RNA or chimeric mixtures or derivatives or
modified
versions thereof, or single-stranded or double-stranded, or partially double-
stranded. An
2 0 oiigonucleotide can be modified at the base moiety, sugar moiety, or
phosphate backbone,
or a combination thereof. An oligonucleotide may include other appending
groups, such as
biotin, fluorophores, or peptides.
An oligonucleotide may comprise at least one modified base moiety which is
selected from the group including but not limited to 5-fluorouracil, 5-
bromouracil,
5-chlorouracil, 5-iodouracil, hypoxanthine, xanthine, 4-acetylcytosine,
5-(carboxyhydroxylmethyl} uracil, 5-carboxymethylaminomethyl-2-thiouridine,
S-carboxymethylaminomethyluracil, dihydrouracil, beta-D-galactosylqueosine,
inosine,
N6-isopentenyladenine, 1-methylguanine, 1-methylinosine, 2,2-dimethylguanine,
2-methyladenine, 2-methylguanine, 3-methylcytosine, 5-methylcytosine, N6-
adenine,
3 0 7-methylguanine, 5-methylaminomethyluracil, S-methoxyaminomethyl-2-
thiouracil, beta-
D-mannosylqueosine, 5'-methoxycarboxymethyluracil, S-methoxyuracil, 2-
methylthio-N6-
isopentenyladenine, uracil-5-oxyacetic acid {v), pseudouracil, queosine, 2-
thiocytosine,
- 33 -
CA 02344625 2001-04-02
WO 00/23622 PGT/US99/23906
5-methyl-2-thiouracil, 2-thiouracil, 4-thiouracil, 5-methyluracil, uracil-5-
oxyacetic acid
methylester, uracil-5-oxyacetic acid (v), 5-methyl-2-thiouracil, 3-(3-amino-3-
N-2-
carboxypropyl) uracil, and 2,6-diaminopurine.
An aligonucleotide may comprise at least one modified phosphate backbone
selected from the group including but not limited to a phosphorothioate, a
phosphorodithioate, a phosphoramidothioate, a phosphoramidate, a
phosphordiamidate, a
methylphosphonate, an alkyl phosphotriester, and a formacetal or analog
thereof.
An oligonucleotide or derivative thereof used in conjunction with the
methods of this invention may be synthesized using any method known in the
art, e.g., by
use of an automated DNA synthesizer (such as are commercially available from
Biosearch,
Applied Biosystems, etc.}. As examples, phosphorothioate oligonucleotides may
be
synthesized by the method of Stein et al. (1988, Nucl. Acids Res. 16, 3249),
methylphosphonate oligonucleotides can be prepared by use of controlled pore
glass
polymer supports (Sarin et al., 1988, Proc. Nat'1 Acad. Sci. U.S.A: 85, 7448-
7451), etc. An
oligonucleotide may be an a-anameric oligonucleotide. An a-anomeric
oligonucleotide
forms specific double-stranded hybrids with complementary RNA in which,
contrary to the
usual ~3-units, the strands run parallel to each other (see Gautier et al.,
1987, Nucl. Acids
Res. 15, 6625-6641).
2 0 Oligoriucleatides may be synthesized using any method known in the art
(e.g., standard phosphoramidite chemistry on an Applied Biosystems 392/394 DNA
synthesizer). Further, reagents for synthesis may be obtained from any one of
many
commercial suppliers.
Spacer phosphoramidite molecules may be used during oligonucleotide
2 5 synthesis, e.g., to bridge sections of oligonucleotides where base pairing
is undesired or to
position labels or tags away from an oligonucleotide portion undergoing base
pairing. The
spacer length can be varied by consecutive additions of spacer
phosphoramidites. Spacer
phosphoramidite molecules may be used as 5'- or 3'- oligonucleotide modifiers.
Such
spacers include Spacer Phosphoramidite 9 (i.e., 9-O-Dimethoxytrityl-
triethylenegiycol, 1-
3 0 [(2-cyanoethyl)-(N,N-diisopropyl)]-phosphoramidite, and Spacer
Phosphoramidite 18 (i.e.,
18-O-Dimethoxytrityl-hexaethyleneglycol, 1-[(2-cyanoethyl)-(N;N-diisopropyl)]-
phosphoramidite), both available from Glen Research (Sterling, Virginia).
- 34 -
CA 02344625 2001-04-02
WO OOI23622 PCTIUS99/23906
Other spacers are available for use in standard oligonucleotide synthesis. For
example, Spacer Phosphoramidite C3 and dSpacer Phosphoramidite can be used to
destabilize undesirable self hybridization events within capture
oligonucleotides or to
destabilize false hybridization events between incorrectly-matched
template/probe
complexes. Such spacers, when positioned at the 3' end of an oligonucleotide,
will also
prevent incorrect extension products from being generated when included in a
PCR reaction
mixture.
One spacer available from Glen Research, Spacer Phosphoramidite C3 (i.e.,
3-O-Dimethoxytrityl-propyl-1-[(2-cyanoethyl)-(N,N-diisopropyl)]-
phospharamidite), can
be added to substitute for an unknown base within an oligonucleotide sequence.
A branching spacer may be used as one method to increase label
incorporation into an oligonucleotide. Such a branching spacer may also be
used to
increase a detectable signal by..hybridization through multiply branched
capture probes or
PCR primers. Branching spacers are available commercially, e.g., from Glen
Research.
Biotinylated oliganucleotides are well known in the-art. An oligonucleotide
may be biotinylated using a biotin-NHS ester procedure. Alternatively, biotin
may be
attached during oligonucleotide synthesis using a biotin phosphoramidite
(Cocuzza, 1989,
Tetrahed. Left. 30, 6287-6290). One such biotin phosphoramidite available from
Glen
2 0 Research is 1-Dimethoxytrityloxy-2-(N-biotinyl-4-aminobutyl)-propyl-3-O-{2-
cyanoethyl)-
(N,N-diisopropyl)-phosphoramidite. This compound also has a branch point to
allow
further additions. The branched spacer used in this biotin phosphoramidite has
been
described by Nelson et al. (1992, Nucl. Acids Res. 20, 6253-6259).
Another 5'-biotin phosphoramidite, namely [1-N-(4,4'-Dimethoxytrityl)-
~ 5 biotinyl-6-aminohexyl]-2-cyanoethyl-(N,N-diisopropyl)-phosphoramidite, may
be used to
biotinylate an oligonucleotide. This compound is sold by Glen Research under
license from
Zeneca PLC.
Fluorescent dyes may also be incorporated into an oligonucleotide using
dye-labeled phosphoramidites. Two such labels are 5'-Hexachloro-Fluorescein
3 0 Phosphoramidite (HEX), and 5'-Tetrachioro-Fluorescein Phosphoramidite
(TET), both
available from Glen Research.
- 35 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
5.7.4 PRODUCTION OF LABELED OLIGONUCLEOTIDES
Oligonucleotides may be labeled with a wide variety of lables for use in the
various embodiments of the invention. For example, European Patent Publication
No. EP
0370 694 A2, entitled, "Diagnostic Kit and Method Using a Solid Phase Capture
Means For
Detecting Nucleic Acid", by Burdick and Oakes, publication date May 30, 1990,
discloses
methods of linking labels to oligonucleotides.
Methods of attaching peptides to oligonucleotides are well known to those
with ordinary skill in the art, e.g., see, 1.) Soukchareun S. et aL,
Preparation and
ch~acterization of antisense oligonucleotide-peptide hybrids containing viral
fusion
peptides. Bioconjug. Chem.,1995, 6(1):43-53; 2) Tung CH, et al., Preparation
of
oligonucleotide-peptide conjugates. Bioconjug. Chem.,1991, 2{6):464-465; 3)
Bruick RK,
et al., Template-directed ligation of peptides to oligonucleotides. Chem.
Biol.,1996,
3(1):49-56; 4) Tung CH, et aL, Dual-specificity interaction of HIV-1 TAR RNA
with Tat
peptide-oligonucleotide conjugates. Bioconjug. Chem.,1995, 6(3):292-295; 5)
Robles J., et
al., Synthesis and Enzymatic Stability of Phosphodiester-Linked Peptide-
Olignonucleotide
Hybrids, Bioconjug. Chem., 1997, 8(6):785-788 ; and 6) Rajur S.B., et al.,
Covalent
Protein-Oligonucleotide Conjugates for Efficient Delivery of Antisense
Molecules,
Bioconjug. Chern.,1997, 8(6}:935-940.
2 0 Oligonucleotides linked to various peptides for use in the methods of this
invention may be obtained for example, from Cybergene S.A. (11 rue Claude
Bernard, zl
nord, 35400, Saint Mallo, France) and Glen Research (22825 Davis Drive,
Sterling,
Virginia 20164). Further information from Glen Research can be obtained
through their
web site (www.glenres.com).
2 5 One specific method for linking a peptide to an oligonucleotide
recommended by Glen Research is as follows {see also, www.glenres.com}. A
heterobifunctionai crosslinking reagent is used to link a synthetic peptide
having an N-
terminal lysine residue to a 5'-thiol-modified oligonucleotide. Such a
crosslinking reagent
is N-maleimido-6-aminocaproyl-(2'-vitro, 4'-sulfonic acid} phenyl ester (mal-
sac-HNSA).
3 0 The sodium salt of mal-sac-HNSA is available from Bachem Bioscience.
Conveniently,
reaction of the mal-sac-HNSA crosslinker with an amino group releases a
dianion phenolate
(i.e: 1-hydroxy-2-vitro-4-benzene sulfonic acid). This dianion phenolate is
also a yellow
- 36 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
chromophore. The chromophore feature provides (i) a means for quantifying the
extent of
completion of the coupling reaction (where greater yellow color intensity
corresponds to a
more complete coupling reaction), and (ii) an aid in monitoring the extent of
separation of
an activated peptide (i.e. a peptide crosslinked to mal-sac-HNSA and ready for
contacting
with a 5'-thiol-modified oligonucleotide) from free crosslinking reagent
during gel filtration.
The specific steps employed when using a mal-sac-HNSA crosslinker may
be as follows. First, a peptide is synthesized having an N-terminal lysine.
Alternatively, a
peptide having an internal lysine may be used since the lysine epsilon amino
group is
actually more reactive than the lysine alpha amino group. Second, an
oligonucleotide is
synthesized having a 5'-thiol group using methods known in the art. Third, the
peptide is
reacted with an excess of mal-sac-HNSA in a sodium phosphate buffer {pH 7.1).
Fourth,
the peptide-mal-sac conjugate is separated from free crosslinker and the
buffer is exchanged
to sodium phosphate (pH 6) using a gel filtration column (e.g. NAP-5,
Pharmacia, Uppsala,
Sweden). Fifth, a thiol-modified oligonucleotide is activated, desalted and
buffer-
exchanged to sodium phosphate (pH 6) on a gel filtration column. "'Sixth, the
activated
peptide is reacted with the thiol-modified oligonucleotide. Finally, the
peptide-
oligonucleotide conjugate is purified by ion exchange chromatography (e.g.
Nucleogen
DEAF-500-10 or equivalent). The elution order from the ion exchange column is
as
2 0 follows: free peptide first, peptide-labeled oligonucleotide next, and
free oligonucleotide
last.
5.7.5 ANTIBODIES AND PEPTIDES
Antibodies of use with the methods of this invention include any antibodies
2 5 lrnown in the art. Such antibodies may be used, for example, to manipulate
the nucleic
acids of interest. In this regard, a nucleic acid may be manipluated by
antibody binding to
the nucleic acid itself ar to an antigen (e.g., a protein, peptide or hapten)
which is bound
(either covalently or non-covalently) to the nucleic acid. In a preferred
embodiment,
nucleic acids are manipulated using peptide antigens covalently attached to
PCR primers.
3 0 Such antibodies include but are not limited to polyclonal, monoclonal,
chimeric and
humanized antibodies, as described below. Further, single chain antibodies,
Fab fragments
- 37 -
CA 02344625 2001-04-02
WO 00/23622 PCTlUS99/23906
and F(ab')2 fragments, fragments produced by a Fab expression library, anti-
idiotypic (anti-
Id) antibodies, and epitope-binding fragments of any of the above may also be
used.
Polyclonai antibodies which may be used with the invention are
heterogeneous populations of antibody molecules derived from the sera of
immunized
animals. Various procedures well known in the art may be used for the
production of
polyclonal antibodies to an antigen-of interest. For example, the production
of polyclonal
antibodies, various host animals can be immunized by injection with an antigen
of interest
or derivative thereof, including but not limited to rabbits, mice, rats, etc.
Various adjuvants
may be used to increase the immunological response, depending on the host
species, and
including but not limited to Freund's (complete and incomplete), mineral gels
such as
aluminum hydroxide, surface active substances such as lysolecithin,
polyanions, peptides,
oil emulsions, keyhole limpet hemocyanins, dinitrophenol, and potentially
useful human
adjuvants such as BCG {bacille Calmette-Guerin) and corynebacterium parvum.
Such
Z5 adjuvants are also well known in the art.
Monoclonal antibodies which may be used with the invention are
homogeneous populations of antibodies to a particular antigen. A monoclonal
antibody
(mAb) to an antigen-of interest can be prepaxed by using any technique known
in the art
which provides for the production of antibody molecules by continuous cell
lines in culture.
2 0 These include but are not limited to the hybridoma technique originally
described by Kohler
and Milstein (1975, Nature 256, 495-497), and the more recent human B cell
hybridoma
technique (Kozbor et al., 1983, Immunology Today 4, 72), and the EBV-hybridoma
technique (Cole et al., 1985, Monoclonal Antibodies and Cancer Theraby, Alan
R. Liss,
Inc., pp. 77-96). Such antibodies may be of any immunoglobulin class including
IgG, IgM,
2 5 IgE, IgA, IgD and any subclass thereof. The hybridoma producing the mAbs
of use in this
invention may be cultivated in vitro or in vivo.
Monoclonal antibodies which may be used with the methods of the invention
include but are not limited to human monoclonal antibodies. Human monoclonal
antibodies
may be made by any of numerous techniques known in the art (e.g., Teng et al.,
1983, Proc.
30 Nat'1 Acad. Sci. U.S.A. 80, 7308-7312; Kozbor et aL, 1983, Immunology Todav
4, 72-79;
Olsson et al., 1982, Meth. Enz~rr~ol. 92, 3-16).
- 38 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99IZ3906
A chimeric antibody may be used with the methods of the invention. A
chimeric antibody is a molecule in which different portions are derived from
different
animal species, such as those having a variable region derived from a marine
mAb and a
human immunoglobulin constant region. Various techniques are available for the
production of such chimeric antibodies {see, e.g., Morrison et al., 1984,
Proc. Nat'1 Acad.
Sci. U.S.A. 81, 6851-6855; Neuberger et al., 1984, Nature, 312, 604-608;
Takeda et al.,
1985, Nature, 314, 452-454) by splicing the genes from a mouse antibody
molecule of
appropriate antigen specificity together with genes from a human antibody
molecule of
appropriate biological activity.
A humanized monoclonal antibody may be used with the methods of the
invention. Briefly, humanized antibodies are antibody molecules from non-human
species
having one or more complementarily determining regions (CDRs} from the non-
human
species and a framework region from a human imnlunoglobulin molecule. Various
l~ techniques have been developed fox the production of humanized antibodies
{see, e.g.,
Queen, U.S. Patent No. 5,585,089, which is incorporated herein byreference in
its entirety).
An immunoglobuiin light or heavy chain variable region consists of a
"framework" region
interrupted by three hypervariable regions, referred to as complementarily
determining
regions (CDRs). The extent of the framework region and CDRs have been
precisely
2 0 defined (see, Kabat et aL, 1983, Sequences of proteins of immunological
interest, U.S:
Department of Health and Human Services).
Alternatively, techniques described for the production of single chain
antibodies (U.S. Patent No. 4,946,778; Bird, 1988, Science 242, 423-426;
Huston et al.,
1988, Proc. Nat'1. Acad. Sci. U.S.A. 85, 5879-5883; and Ward et al., 1989,
Nature 334, 544-
2 5 546} can be adapted to produce single chain antibodies useful in the
device of the invention.
Single chain antibodies are formed by linking the heavy and light chain
fragments of the Fv
region together via an amino acid bridge, resulting in a single chain
polypeptide.
Antibody fragments which recognize specific epitopes may be generated by
known techniques. For example, such fragments include but are not limited to:
the F(ab')Z
3 0 fragments which can be produced by pepsin digestion of the antibody
molecule and the Fab
fragments which can be generated by reducing the disulf de bridges of the
F(ab')2 fragments.
Alternatively, Fab expression libraries may be constructed (Huse et al., 1989,
Science, 246,
- 39 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
1275-1281) to allow rapid and easy identification of monoclonal Fab fragments
with the
desired specificity.
Further general methods of antibody production and use are suitable for use
in connection with the methods of the invention. For example see Harlow and
Lane, 1988,
Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold
Spring
Harbor, New York, which is incorporated herein by reference in its entirety.
The single-lettter amino acid code corresponds to the three-letter amino acid
code of the Sequence Listing set forth hereinbelow, as follows: A, Ala; R,
Arg; N, Asn; D,
Asp; B, Asx; C, Cys; Q, GIn; E, Glu; Z, Glx; G, Gly; H, His; I, Ile; L, Leu;
K, Lys; M, Met;
F, Phe; P, Pro; S, Ser; T, Thr; W, Trp; .Y, Tyr; and V, Val.
Suitable antibodies for use with the methods of this invention include the
following, available from Affinity Bioreagents, Inc., 79, rue des Morillons,
75015, Paris,
France.
1) Catalog No. PA 1-047 (affinity-purified rabbit IgG). The
corresponding peptide recognized by this Ab is KFSREKKAAKT
(SEQ ID NO:1).
2) Catalog No. PA 1-039 (affinity-purified rabbit immunogobins). The
corresponding peptide recognized by this Ab is DQKRYHEDIFG
(SEQ ID N0:2}.
3) Catalog No. PA 1-036 (purified rabbit IgG). The corresponding
2 5 peptide recognized by the Ab is DLKEEKDINNNVKKT (SEQ ID
N0:3).
4) Catalog No. PA 1-014 (purified rabbit antibody). The corresponding
peptide recognized by this Ab is CTGEEDTSE (SEQ ID NO: 4).
5) Catalog No. PA 3-013 (affinity purified IgG). The corresponding
peptide recognized by this Ab is PEETQTQDQPM (SEQ ID NO:S).
-
CA 02344625 2001-04-02
WO 00/23622 PCT/US99l23906
6) Catalog No. PA 1-815 (rabbit anti-serum}. The corresponding peptide
recognized by this Ab is QKSDQGVEGPGAT (SEQ ID N0:6).
7) Catalog No. PA 3-034 (rabbit polyclonal serum IgG). The
corresponding peptide recognized by this Ab is DIGQSIKKFSKV
(SEQ ID N0:7). This polyclonal antibody will also recognize
QRADSLSSHL {SEQ ID No:B).
In addition, antibodies for use with the methods of this invention may be
obtained from Medical & Biological Laboratories Co., Ltd., 440 Arsenal Street,
Watertown,
Massachusetts 02171, U.S.A.
These include the following:
1 ) Code No. 561 {Rabbit IgG from anti-serum). The corresponding
peptide recognized is YPYDVPDYA (SEQ ID N0:9).
2) Code No. 562 (Rabbit IgG from anti-serum). The corresponding
peptide recognized is EQKLISEEDL (SEQ 117 NO:10).
3) Code No. 563 (Rabbit IgG from anti-serum). The corresponding
peptide recognized is YTDIEMNKLGK {SEQ ID NO:11).
The invention described and claimed herein is not to be limited in scope by
2 5 the specific embodiments herein disclosed since these embodiments are
intended as
illustration of several aspects of the invention. Any equivalent embodiments
are intended to
be within the scope of this invention. Indeed, various modifications of the
invention in
addition to those shown and described herein will become apparent to those
skilled in the
art from the foregoing description. Such modifications are also intended to
fall within the
3 0 scope of the appended claims. Throughout this application various
references are cited, the
contents of each of which is hereby incorporated by reference into the present
application in
its entirety.
- 41 -
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
SEQUENCE LISTING
<110> VALIGENE CORPORATION
<120> METHODS FOR MANIPULATING COMPLEX NUCLEIC ACID
POPULATIONS USTNG PEPTIDE-LABELED OLIGONUCLEOTIDES
<130> 9408-025-228
<140> PCT/US99/23906
<141> 1999-10-15
<160> 11
<170> PatentIn Ver. 2.0
<210> 1
<211> 11
<212> PRT
<213> Oryctolagus cuniculus
<400> 1
Lys Phe Ser Arg Glu Lys Lys Ala Ala Lys Thx
1 5 . 10
<210> 2
<211> 11
<212> PRT
<213> Oryctolagus cuniculus
<400> 2
Asp G1n Lys Arg Tyr His Glu Asp Ile Phe Gly
1 5 10
<210> 3
<211> 15
<212> PRT
<213> Oryctolagus cuniculus
<400> 3
Asp Leu Lys Glu Glu Lys Asp Ile Asn Asn~ Asn Val Lys Lys Thr
1 5 10 15
<210> 4
<211> 9
<212> PRT
<213> Oryctolagus cuniculus
<400> 4
Cys Thr Gly Glu Glu Asp Thr Ser Glu
1 5
1
SUBSTITUTE SHEET (RULE 26)
CA 02344625 2001-04-02
WO 00/23622 PCT/US99I23906
<210> 5
<211> 11
<212> PRT
<213> Oryctolagus cuniculus
<400> 5
Pro Glu Glu Thr Gln Thr Gln Asp Gln Pro Met
1 5 10
<210> 6
<211> 13
<212> PRT
<213> Oryctolagus cuniculus
<400> 6
Gln Lys Ser Asp Gln Gly Val Glu Gly Pro Gly Ala Thr
1 5 10
c210> 7
<211> 12
<212> PRT
<213> Oryctolagus cuniculus
<400> 7
Asp Ile Gly Gln Ser Ile Lys Lys Phe Ser Lys Val
1 5 to
<210> a
<211> to
<212> PRT
<213> Oryctolagus cuniculus
<400> 8
Gln Arg Ala Asp Ser Leu Ser Ser His Leu
1 5 10
<210> 9
<211> 9
<212> PRT
<213> Oryctolagus cuniculus
<400> 9
Tyr Pro Tyr Asp Val Pro Asp Tyr Ala
1 5
<210> 10
<211> 10
<212> PRT
<213> Oryctolagus cuniculus
<400> 10
Glu Gln Lys Leu Ile Ser Glu Glu Asp Leu
1 5 10
2
SUBSTITUTE SHEET (RULE 26)
CA 02344625 2001-04-02
WO 00/23622 PCT/US99/23906
<210> 11
<2I1> 11
<2I2> PRT
<213> Oryctolagus cuniculus
<400> 11
Tyr Thr Asp Ile Glu Met Asn Lys Leu Gly Lys
1 5 10
3
SUBSTITUTE SHEET (RULE 26)
