By
Kevin E. Noonan —

Much
of the gene patenting debate has appeared, to experienced patent practitioners,
to be akin to a conversation with Kevin Kline's character in A Fish Called Wanda, or perhaps a
passage from P.G. Wodehouse's The Luck of
the Bodkins.  In short, it has been
the tale of doctors, scientists, and legal academics interpreting claims to
isolated DNA in ways both bizarre and unlikely to anyone with any experience in
actually claiming DNA.  However, as Dr.
Mason's response illustrates those of us on the patent side of the debate have
contributed to the confusion.

Reviewing
Dr. Mason's assessment of the patents and applications he identified as
encompassing almost the entirety of the human genome reveals two sources of
misunderstanding.  The first is between
the disclosure of patent specifications and the claims:  while sometimes in
their zeal to provide the broadest disclosure possible, patent practitioners
have been overly expansive in defining the scope of the isolated nucleic acids
that are the subject of their clients' inventions, the claims have typically
been more circumspect.  Most claims to
isolated DNA have the canonical structure:

An isolated nucleic acid/human DNA molecule, encoding an amino acid sequence
identified by SEQ ID NO. X.

Occasionally
the word "comprising" is used in such claims:

An isolated nucleic acid/human DNA molecule, comprising a nucleotide sequence
encoding an amino acid sequence identified by SEQ ID NO. X.

And
there are claims to oligonucleotides, usually dependent on the claimed
sequence:

An oligonucleotide comprising about 15 nucleotides of SEQ ID NO: X.

Considered
in a vacuum, in the context of definitions as illustrated in Dr. Mason's
response, it could be thought that such claims encompass a greater scope than
they actually do.  What is missing is
context, both historical and legal, that provides the insights needed to
properly understand the scope of these claims.

The
base canonical form above was developed about thirty years ago, during the age
when "cloning" a DNA molecule that encoded a protein of interest
(erythropoietin, tissue plasminogen activator, blood clotting Factor VIII,
interferon, etc.) was a Herculean feat, and such claims were expressly directed
towards full-length versions of these
genes.  In the days before the Human
Genome Project and the vast amount of genetic information it produced, such
cloned genes had but one utility:  to
permit the production of recombinant cells that expressed the gene product,
typically to be used as one of the first "biologic" drugs.  In order for the claimed DNA molecule to have
that utility, however, what was required almost always was that the full-length
gene be isolated, because truncated proteins could not be expected to have the
desired biological properties (and could have deleterious ones, such as
increased immunogenicity or reduced half-life).  Accordingly, such claims do not have the scope Dr. Mason ascribes to
them, because any such claim would be invalidated for failing to have the
necessary utility.

Moreover,
the scope of such claims are extremely narrow, being limited to
isolated DNA molecules that encode the specifically recited amino acid sequence
identified as SEQ ID NO: X.  An isolated
human DNA comprising any change in this encoded amino acid sequence is not
exclusive to the patentee and can be performed freely without literal
infringement liability by anyone.  (The availability of the doctrine of
equivalents for isolated human DNA claims is severely limited by U.S. Patent
and Trademark Office practice in interpreting the application of 35 U.S.C. §
112(a) to such claims, and by ensuing prosecution history estoppel.)

The significance of this limitation in
scope of isolated DNA claims can be understood by the following example.  If the
single change in the amino acid sequence expressly recited in the claim is from
a valine (Val) residue to an isoleucine (Ile) residue, the number of atoms in
the protein would be increased by a mere 3 atoms (the difference in structure
between these two amino acids is a methylene group,-CH₂-) out a
total of 50-100,000 atoms in the protein.  Yet, even such a molecule
changed to such a small degree would not literally infringe the generic
claim.  Similarly, insertions or
deletions of the sequence, mutations and sequence variants (naturally occurring
or man-made) do not fall within the scope of claims to a specific, particular
isolated human DNA, due to this restricted scope.  Such claims reciting
such full-length "genes" are simply not preclusive of any other gene
in the genome.

Turning to the oligonucleotide claims,
there are two issues.  First, the work of
Bob Cook-Deegan and his collaborators have shown (Kepler et al., "Metastasizing patent claims on BRCA1") that oligonucleotides
on the order of 15-16 nucleotide residues in length are much more ubiquitous in
the genome than was once thought.  This
stems from the simplistic and incorrect assumption that any sequence of n nucleotides will occur once per 4ⁿ times in the genome.  Thus, the random 15-mer was expected to arise
once every 1,073,741,824 basepairs, and the random 16-mer once every
4,294,967,296 basepairs.  The (false)
conclusion was that in a haploid genome comprising 3 x 10⁹
basepairs, the 15-mer and 16-mer (and larger species) would be unique.  The source of the error is that human DNA is
a historical molecule, and that many sequence motifs would be expected to arise
many more times than random statistics would predict; the Human Genome Project
(and basic biology) illustrates the fact that the human genome evolved by gene
duplication and recombination, so that certain motifs (the ATP-binding cassette
sequences, for example) sharing certain amino acid sequence and thus the
nucleotide sequences encoding them, are much more common that was
predicted.  The consequence of this biological fact is not that claims to such
oligonucleotide encompass the entirety of the genome.  Rather, it is that claims of this scope are
invalid, since the prior art is much more replete with these sequences than was
expected.

The other distinction that should be
appreciated is that what is disclosed (and even claimed) in patent applications
does not raise infringement issues.  Indeed, filing overbroad claims in International Applications can be a
strategic decision, to produce prior art against competitors (this stratagem
became available in the U.S. with publication of U.S. applications under the
American Inventor Protection Act of 1999).  Mere inspection of the published patent application database can find
patents to a variety of "inventions" unlikely to ever grant with
patentable claims, and their existence is thus not relevant to a serious debate
over the scope of patents claiming
isolated human DNA.

It is difficult to avoid the conclusion
that a failure to appreciate these distinctions informs much of the angst
enunciated by Dr. Mason and other physicians and researchers confronted by the
prospect of infringement liability (or at least the fear that threats from
patent holders will inhibit progress in genetics research).  But the patent community is not entirely blameless
either, as illustrated by Dr. Mason in his Rebuttal.  There exists a certain amount of loose
language in patent specifications that creates understandable confusion as to
the claim scope the patentee believes
she is entitled to, and thus uncertainty as to whether claims will be asserted
as broadly as these specifications can be interpreted to support.  (Of course, some have interpreted these
claims to encompass isolated DNA itself, an interpretation that would render
the claims unpatentable since the time of Friedrich Miescher's work in the 19th
Century.)  Some comfort should be taken
by the fact that no gene patent holder has ever asserted claims to this extent,
and that any attempt to do so would likely result in invalidation of the
claims.  And the patent case law has
provided more and more illustrations of claims (including Myriad's method
claims) being interpreted in view of these overbroad specifications and
invalidated based on broad claim construction.  In short, while there may be specifications suffering from symptoms of
overbroad disclosure, the disease is self-correcting (even more so with the
post-grant review and inter partes
review provisions of the Leahy-Smith America Invents Act).

What is clearly needed is more dialog (and
perhaps fewer assumptions) from both physicians and the patent community.  This debate comes thirty years too late to be
particularly meaningful to claims to isolated nucleic acids.  But as the complexities of "modern"
gene-based inventions increase (and their capacity to be ever more resistant to
reverse engineering), the potential to eschew patent protection will increase,
with the concomitant reduction in disclosure of biologically relevant
inventions.  Such a trend, and Supreme
Court decisions that promote that trend, particularly if based on unrealistic
apprehensions over the (vanishingly small) possibility that claims to isolated
human DNA will negatively impact developing genetic technologies, could
paradoxically promote the outcome that opponents of gene patenting purportedly
are committed to prevent.

By Donald Zuhn —

Yesterday, we posted a
response from Dr. Chris Mason of Cornell University to a recent Patent Docs post and an article
posted by Dr. Chris Holman on Holman's
Biotech IP Blog.  Dr. Mason
co-authored an article in the current issue of Genome Medicine, which contends that due to the non-specificity of
sequence uniqueness across the genome and the broad scope of claims to
nucleotide sequences, the Supreme Court and Congress should limit the patenting
of existing nucleotide sequences.  In his
response, Dr. Mason comments on several aspects of the Holman and Patent Docs posts, which he says suggest
that the analysis from his paper "was mistaken or did not embody an
accurate characterization of the claims' construction."

In response to Dr. Holman's
assertion that Dr. Mason and his co-author Jeffrey Rosenfeld "seem to
assume that every patent with a claim mentioning a gene sequence also claims
every 15mer present in the sequence," Dr. Mason states that the authors
did not make such an assumption, and that their paper instead describes the
results of two distinct analyses.  In the
first analysis, the authors "examined the uniqueness of 15mers in general,
which was shown to be exceedingly non-unique genome-wide."  However, as we noted in our first post on Dr. Mason's paper, the conclusion regarding the broad scope of claims to
15mers — and in particular claims 5 and 6 of Myriad's U.S. Patent No.
5,747,282 — is not particularly ground-breaking given that Kepler et al. similarly concluded in a 2010 Genomics paper that claim 5 of the '282
patent was "exceptionally broad" (see
"Caught in a Time Warp: The (In)validity of BRCA1 Oligonucleotide
Claims").  Kepler et al. also suggested that "if human
genes were random strings of nucleotides, one would expect a human gene to contain
an average of 15 15-mers claimed under the ['282] patent," and in fact
found that 80% of 713 human mRNAs deposited in 1994 (the earliest effective filing
date of the '282 patent is August 12, 1994) contained at least one of the
claimed 15mers.

With respect to their
second analysis, Dr. Mason states that the authors "used patents that
claimed 15mer sequences in their construction, and we indicated the matches we
could find given their sequence composition."  Unfortunately, given Dr. Mason's views
regarding the subject matter encompassed by the claims of U.S. Patent No.
7,795,422 — one of "58 patents [according to Dr. Mason] whose
claims covered at least 10% of the bases of all human genes," and a patent
Dr. Mason labeled as their "top patent . . . whose claims' sequences
matched 91.5% of human genes" — it is not entirely clear what Dr. Mason
means when he says that the authors "used patents that claimed 15mer sequences in their
construction" (emphasis added). 
An analysis of the claims in the '422 patent, as well as those in U.S.
Patent No. 7,468,248 (a second patent discussed in the Genome Medicine article), clearly
indicates that Dr. Mason's top two candidates do not contain "explicit claims for 15mers that matched 84% of
human genes," as asserted in his paper (assuming that by "explicit
claims," Dr. Mason means that the claims cover or encompass 15mers).  It is also abundantly clear from both the Genome Medicine paper and his subsequent
response that Dr. Mason does not understand the process of claim construction
(which is not surprising given that Dr. Mason is not a patent practitioner).

So, what follows is a short primer on claim construction.  Claim construction is the process of giving proper meaning to the language of the
claims.  Abtox Inc. v. Exitron Corp., 122 F.3d
1019, 1023 (Fed. Cir. 1997).  The
language of the claim frames and ultimately resolves all issues of claim
interpretation, since it is the language of the claim that defines the scope of
the protected invention.  Id. 
However, in determining the scope and meaning of a claim term, three
additional sources may be considered:  the descriptions in the rest of the patent
specification, the prosecution history, and relevant extrinsic evidence.  Phillips
v. AWH Corp., 415 F.3d 1303, 1315, 1317 (Fed. Cir. 2005).

With
regard to the use of the specification and prosecution history to construe
claim terms, the Federal Circuit has stated that:

Claims must be read in view of the specification,
of which they are a part.  The
specification contains a written description of the invention that must enable
one of ordinary skill in the art to make and use the invention.  For claim construction purposes, the
description may act as a sort of dictionary, which explains the invention and
may define terms used in the claims.  . .
.  The written description part of the
specification itself does not delimit the right to exclude.  That is the function and purpose of the
claims.

To construe claim language, the court should
also consider the patent’s prosecution history[.]  . . .  Although
the prosecution history can and should be used to understand the language used
in the claims, it too cannot "enlarge, diminish, or vary" the
limitations in the claims.

Markman v. Westview Instruments, Inc.,
52 F.3d 967, 979-80 (Fed. Cir. 1995) (in
banc), aff’d, 517 U.S. 370 (1996)
(citations omitted).

In
Phillips, the Federal Circuit noted
that where the intrinsic evidence (i.e.,
claim language, specification, and file history) does not resolve ambiguity in
the meaning of a claim term, a court may also consider extrinsic evidence, such
as expert testimony, dictionaries, and treatises.  Phillips,
415 F.3d at 1317-19.  Such
"extrinsic evidence can help educate the court regarding the field of the
invention and can help the court determine what a person of ordinary skill in
the art would understand claim terms to mean."  Id.
at 1319.

With
the above legal principles in mind, we turn back to the claims of the '422
and '248 patents.  The '422 patent
contains five claims, only one of which is an independent claim (i.e., does not depend from, and include
the limitations of, another claim):

1.  A
chemically modified short interfering nucleic acid (siNA) molecule, wherein:
    (a) the siNA molecule comprises a sense strand and an antisense strand, each
strand having one or more pyrimidine nucleotides and one or more purine
nucleotides;
    (b) each strand is independently 18 to 27 nucleotides in length,
and together comprise a duplex having between 17 and 23 base pairs;
    (c) the
antisense strand is complementary to a human Hypoxia Inducible Factor 1 (HIF1) RNA
sequence comprising SEQ ID NO:567;
    (d) a plurality of pyrimidine nucleotides
present in the sense strand are 2'-deoxy-2-fluoro pyrimidine nucleotides and a
plurality of purine nucleotides present in the sense strand are 2'-deoxy purine
nucleotides; and
    (e) a plurality of pyrimidine nucleotides present in the
antisense strand are 2'-deoxy-2'-fluoro pyrimidine nucleotides and a plurality
of purine nucleotides present in the antisense strand are 2'-O-methyl-purine
nucleotides.

A cursory analysis of claim 1 (a more thorough analysis being unnecessary) indicates that the claim is
directed to a "chemically modified short interfering nucleic acid (siNA)
molecule."  Therefore, claim 1 does not cover or encompass molecules other than chemically modified short interfering nucleic
acids (siNAs).  More importantly, the siNA
molecules that are covered or encompassed by claim 1 must possess a sense
strand having "a plurality of pyrimidine nucleotides [that] are
2'-deoxy-2-fluoro pyrimidine nucleotides," and an antisense strand having
"a plurality of pyrimidine nucleotides [that] are 2'-deoxy-2'-fluoro
pyrimidine nucleotides."  2'-deoxy-2'-fluoro
pyrimidine nucleotides are not naturally-occurring nucleotides.

In
response to Dr. Holman's statement that "DNA does not contain
2'-deoxy-2-fluoro pyrimidine nucleotides and 2'-deoxy purine nucleotides,"
Dr. Mason notes that "'2'-deoxy purine nucleotides' are simply
the 'A' and 'G' nucleotides in normal DNA."  He does not, however, contend that 2'-deoxy-2'-fluoro
pyrimidine nucleotides are found in "normal DNA."  Dr. Mason also takes Patent Docs to task for stating that "the lone independent
claim of the '422 patent is directed to a 'chemically modified' double-stranded
nucleic acid molecule," and "[i]t is therefore difficult to see how
the claimed sequences of the '422 patent could 'match[] 91.5% of human
genes.'"  Dr. Mason counters that
"[i]t is notable that one of the claimed chemical modifications allowed
from [the '422 patent] claims is '2'-deoxyribonucleotides,' which is the
same thing as dexoyribonucleotides, or DNA."  However, as noted above, the siNA molecules
of claim 1 of the '422 patent must possess a sense strand having "a
plurality of pyrimidine nucleotides [that] are 2'-deoxy-2-fluoro pyrimidine
nucleotides," and an antisense strand having "a plurality of
pyrimidine nucleotides [that] are 2'-deoxy-2'-fluoro pyrimidine nucleotides."  And 2'-deoxy-2'-fluoro pyrimidine nucleotides
are not naturally-occurring nucleotides found in "normal DNA."  Dr. Mason also lists "six other
potential embodiments (of many) from [the '422] patent that could easily cover
unmodified bases or fragments thereof," adding that "[t]he claims
must be interpreted in plain language first, of course, but also in light of
their specifications of the patent." 
Notwithstanding the importance of the specification in claim
construction, absent a clear and specific statement giving a claim term a
special definition, that term should be construed as having the plain and
ordinary meaning given by persons experienced in the field of the
invention.  Renishaw plc v. Marposs Societa' per Azioni, 158 F.3d 1243, 1249
(Fed. Cir. 1998).  Therefore, the
disclosure in the '422 patent specification of unclaimed embodiments such as unmodified
siNA molecules does not trump the express recitation in claim 1 of "[a]
chemically modified short interfering nucleic acid (siNA) molecule"
possessing a sense strand having "a plurality of pyrimidine nucleotides [that]
are 2'-deoxy-2-fluoro pyrimidine nucleotides," and an antisense strand
having "a plurality of pyrimidine nucleotides [that] are
2'-deoxy-2'-fluoro pyrimidine nucleotides."  Thus, contrary to Dr. Mason's assertion, the
claims of the '422 patent do not "claim[] 15mer sequences."

The
same is true for the claims of the '248 patent. 
As we noted in our last post, the claims of the '248 patent are directed
to methods of inferring a
trait of a bovine subject, determining a nucleotide occurrence of a polymorphism
in a bovine sample, and identifying a bovine single nucleotide polymorphism
(SNP) associated with a trait, and not
15mer nucleic acid molecules.  In
response to Dr. Holman's observation that the independent claims of the '248
patent are method claims, Dr. Mason agrees, but suggests that "their
specifications allow for a broad interpretation."  However, the disclosure of "an isolated
polynucleotide that includes at least 20 contiguous nucleotides of any one of
SEQ ID NOS:24493 to 64886" in the specification simply does not convert a
claim directed to a method into a claim directed to a oligonucleotide.  As with the '422 patent, therefore, the
claims of the '248 patent simply do not "claim[] 15mer sequences."

To
put the above in layman's terms, Dr. Mason could make or use a 15mer derived
from the human Hypoxia Inducible Factor 1 (HIF1) RNA sequence comprising SEQ ID
NO:567 or the nucleotide sequence of SEQ ID NO:21645 and not be found liable
for infringing either the '422 or '248 patents — provided, of course, that he
was not using either sequence to generate the claimed siNA of the '422 patent
or practice the claimed methods of the '248 patent.

In
his response, Dr. Mason argues that even if one disagrees with his analysis of
the '422 patent, related U.S. Patent No. 8,273,866 supports his
assertions.  Claim 1 of the '866 patent
recites:

1.  A
short interfering RNA (siRNA) molecule having a sense strand and an antisense
strand that mediates RNA interference, wherein:
    (a) each strand is between 18 and 24 nucleotides in length;
    (b) the sense strand comprises 10 or more 2'-deoxy, 2'-O-methyl,
2'-deoxy-2'-fluoro, or universal base modified nucleotides, and a terminal cap
molecule at the 3'-end, the 5'-end, or both 3' and 5'-ends of the sense strand;
    (c) the antisense strand comprises 10 or more 2'-deoxy, 2'-O-methyl,
2'-deoxy-2'-fluoro, or universal base modified nucleotides; and
    (d) 10 or more pyrimidine nucleotides of the sense and antisense strand are
2'-deoxy, 2'-O-methyl or 2'-deoxy-2'-fluoro nucleotides.

Dr.
Mason argues that "any sequence that is 18-24 bases can be constructed
from these claims, [and] this may mean that every single 18-24mer is
potentially claimed by this patent." 
Not quite.  Claim 1 is directed to
"[a] short interfering RNA (siRNA) molecule having a sense strand and an
antisense strand that mediates RNA interference."  A "single 18-24mer" is not "[a] short interfering RNA (siRNA) molecule having a sense strand and an
antisense strand that mediates RNA interference."

Turning
back to Dr. Mason's response, he counters Dr. Holman's assertion that the 15mer
claims of Myriad's '282 patent are "extremely rare" by pointing out
that "[a]fter searching for a short time on Google Patents, I was able to
find two potential examples of recently published patent applications
that directly claim a large subset of genes."  As noted in an editorial comment to Dr.
Mason's response, the response was revised per Dr. Mason's request to indicate
that the documents Dr. Mason identified were published patent applications
rather than patents as he stated in the original version of his response.  However, it is unclear whether Dr. Mason
understands that identifying two "published patent applications" as
"issued patents" was not merely an unfortunate typographical
error.  Instead, Dr. Mason's citation of
two published patent applications provides no support for his
argument that 15mer claims can be readily found in recently issued patents
because infringement liability arises only when an individual "without
authority makes, uses, offers to sell, or sells any patented invention"
(35 U.S.C. § 271(a)), and the published patent applications he cites are not
patents.

Finally, when
defending the assertion in paragraph 10 of his Declaration that "Claim #1
and #2 of '282 are so broad that they can include up to 100% of the genes in
the human genome," Dr. Mason counters with more of the same flawed
argument.  He first notes that "55%
homology is a very low threshold," and then points to claim 6, which is
directed to "[a]n isolated DNA having at least 15 nucleotides of the DNA
of claim 2."  But claims 1 and 2 do
not recite 15mer fragments of DNA encoding the amino acid sequence of SEQ ID
NO:2 or the nucleotide sequence of SEQ ID NO:1 that share 55% identity with
those sequences.  Instead, claims 1 and 2 are
directed to a nucleic acid encoding a protein of 1863 amino acids and a nucleic
acid of 5914 nucleotides.

[Ed. Patent Docs recently reported on an
article in the current issue of Genome Medicine which contends that due
to the non-specificity of sequence uniqueness across the genome and the broad
scope of claims to nucleotide sequences, the Supreme Court and Congress should
limit the patenting of existing nucleotide sequences (see "Genome Medicine Article Calls for Limits on Patenting of Existing Nucleotide Sequences" and "Revisiting Genome Medicine Article on 'Pervasive Sequence Patents' That 'Cover the Entire Human Genome'".  The article was authored by Jeffrey Rosenfeld of the University of Medicine & Dentistry of New Jersey
and Christopher Mason of Cornell University.  Dr. Mason (below) has provided Patent Docs with a response to our posts — as well as a response to an article posted by Prof. Chris Holman on Holman's
Biotech IP Blog (see "A Critique of a Recent Article Which Found That Sequence Patents Cover
the Entire Human Genome").  To promote further discussion of the issue of "gene" patenting, we are posting Dr. Mason's response today.  As always, we invite Patent Docs readers to join in the discussion by submitting comments to any of the above posts.  Our reply to Dr. Mason's post will follow tomorrow.]

Several
recent posts and commentaries have suggested that the analysis from our recent
paper, "Pervasive Sequence
Patents Cover the Entire Human Genome," was mistaken or did not embody
an accurate characterization of the claims' construction.  I would like to comment on several aspects of
the posts from both Holman's
Biotech IP Blog and the Patent Docs blog, but foremost I would like
to thank both sites for their dialogue and debate on this important issue.

Part One: The Holman Critique:  "A Critique of a Recent Article Which Found That Sequence Patents Cover
the Entire Human Genome"

1)  In his post, Prof. Holman
states "the authors seem to assume that every patent with a claim
mentioning a gene sequence also claims every 15mer present in the sequence,
i.e., every contiguous 15 nucleotide sequence appearing in the gene."

We
do not.  We performed two distinct analyses
that were described in the main text.  First, we examined the uniqueness of 15mers in general, which was shown
to be exceedingly non-unique genome-wide (no gene is unique at the 15mer level).  Secondly, we used patents that claimed 15mer
sequences in their construction, and we indicated the matches we could find
given their sequence composition.  From
these, we found many exact matches, ranging from 4% (for BRCA1) to potentially
as high 91%.

2)  Prof. Holman
continues, "In my experience, claims of this type are extremely rare.  I
looked at [a] hundred patents identified as gene patents in the Jensen Murray study
and found that most only claim the full-length gene sequence.  . . .  I looked
through hundreds of gene patents trying to find another 15mer claim analogous
to those in the Myriad patents and could not find one.  The patent claims at
issue in the Myriad case will be expiring within the next few years.  . . .  I doubt
that this sort of broad 15mer claim has been issued by the patent office in recent
years, or if it has it seems to be extremely rare."

After
searching for a short time on Google Patents, I was able to find two potential examples
of recently published patent applications that directly claim a large subset of genes.

US 20130030040 A1

Methods and compositions
for increasing sialic acid production and treating sialic related disease
conditions

14. An isolated nucleic acid
molecule the sequence set forth in SEQ ID NO:1 or SEQ ID NO:2.

"In
some embodiments, the therapeutic product is a polynucleotide, while in other
embodiments, the therapeutic product is a polypeptide.  In some embodiments, the
polynucleotide is a DNA molecule, which can comprise the full-length coding
region for a protein, the coding region for a domain of a protein, or a coding
region for a protein fragment, which is shorter than a recognized and
identified domain of a protein.  Thus, the polynucleotides disclosed herein can
range from oligomers of at least 15 base pairs in length to DNA molecule
comprising the full-length coding region for a protein."

This patent application refers to the
gene "GNE," which allows it to also claim 15mers.  Given the thousands of nucleotides in this
patent application, it turns out that this patent application can cover 1,306 other genes, using the
same 15-mer matching algorithm.

US 20130041209 A1

Methods and compositions
for improved fertilization and embryonic survival

"An
isolated nucleic acid molecule comprising a nucleotide corresponding to a
nucleotide at a first polymorphic position selected from the group consisting
of positions 85146, 85161, 85216, 85292, and 85300 of the nucleic acid sequence
shown in FIG. 1 (SEQ ID NO: 1), and at least 10 contiguous nucleotides of SEQ
ID NO: 1 adjacent to the first polymorphic position, wherein position 85146 is
guanine, position 85161 is guanine, position 85216 is adenosine, position 85292
is cytosine, or position 85300 is guanine; or an isolated nucleic acid molecule
comprising a nucleotide corresponding to a nucleotide at a second polymorphic
position selected from the group consisting of positions 35728, 36016, and
38867 of the nucleic acid sequence shown in FIG. 2 (SEQ ID NO: 2), and at least
10 contiguous nucleotides of SEQ ID NO: 2 adjacent to the second polymorphic
position, wherein position 35728 is guanine, position 36016 is guanine, or
position 38867 is guanine.

2. A nucleic acid molecule
according to claim 1, which comprises at least 10, at least 11, at least 12, at
least 13, at least 14, or at least 15 contiguous bases of SEQ ID NO: 1 adjacent
to the first polymorphic position, or of SEQ ID NO: 2 adjacent to second
polymorphic position.

7. A nucleic acid molecule
according to claim 1, wherein the first or second polymorphic site is within 4
nucleotides of the center of the nucleic acid molecule.

8. A nucleic acid molecule
according to claim 7, wherein the first or second polymorphic site is at the
center of the nucleic acid molecule."

Notably, these are highly
polymorphic sites in the human genome, so if you look at any piece of DNA
anywhere near this gene, you will easily match the specified variants from this
gene, and then be entitled to any genes that are on the same fragment.  From simple pipetting, this can be anywhere
from 10,000-100,000 bases.

3)  Prof. Holman
then says, "DNA does not contain 2'-deoxy-2-fluoro pyrimidine nucleotides and
2'-deoxy purine nucleotides, these are synthetic analogues to the nucleotides
that appear in DNA."

This
is completely false.  "2'-deoxy
purine nucleotides" are simply the "A" and "G" nucleotides
in normal DNA, and they indeed appear in normal DNA.  Modifications of these bases are allowed in this
patent for "chemical variations" of RNA, even though (in this case)
they just turn RNA bases into DNA bases.

4)  Prof. Holman
continues, "the authors reported that US7468248 contains 'explicit
claims for 15mers that matched 84% of human genes.'  In fact, the '248
patent has only two independent claims, both of them method claims."

We
agree that these are method claims, but their specifications allow for a broad
interpretation.  In particular, from the
patent:

"In
one embodiment, the present invention provides an isolated polynucleotide that
includes at least 20 contiguous nucleotides of any one of SEQ ID NOS:24493 to
64886, a polynucleotide at least 90% identical to the 20 contiguous nucleotide
fragment, or a complement thereof, wherein the isolated polynucleotide includes
a nucleotide occurrence of a single nucleotide polymorphism (SNP) associated
with a trait, wherein the SNP corresponds to position 300 of SEQ ID NOS:19473
to 21982."

5)  Prof. Holman
then says, "the publication of the article highlights the limitations of
peer review (assuming Genome Medicine engages in peer review)."

The
publication of contentious research or discordant viewpoints does not
demonstrate a failure of the entire peer-review system.  Rather, discussion and debate are key drivers
of scientific progress through these peer-reviewed publications.  Five scientific reviewers and two patent
attorneys reviewed and approved our article before its publication, and the site
of the Journal clearly indicates that it conducts a thorough peer review.  We also note that the journal is open-access,
allowing for easier dissemination of data and results, and that the Journal
hosts a large set of leading researchers on its Editorial Board.

Part
Two: The Patent Docs blog post:  "Revisiting Genome Medicine Article on 'Pervasive Sequence Patents' That 'Cover the Entire Human Genome'"

1)  Patent Docs
states that "the lone independent claim of the '422 patent is directed to
a "chemically modified" double-stranded nucleic acid molecule. 
It is therefore difficult to see how the claimed sequences of the '422 patent
could "match[] 91.5% of human genes."

It
is notable that one of the claimed chemical modifications allowed from their
claims is "2′-deoxyribonucleotides," which is the same thing as
dexoyribonucleotides, or DNA ("Ribonucleotides versus Deoxyribonucleotides").

Also,
here is a listing of six other potential embodiments (of many) from this patent
that could easily cover unmodified bases or fragments thereof.  The claims must be interpreted in plain
language first, of course, but also in light of their specifications of the
patent, which include:

1.1)  In one embodiment, the invention features one or more chemically modified siNA
constructs having specificity for HIF1 expressing nucleic acid molecules, such
as RNA encoding a HIF1 protein.  In one embodiment, the invention features a RNA
based siNA molecule (e.g., an siNA comprising 2′-OH nucleotides) having
specificity for HIF1 expressing nucleic acid molecules that includes one or more
chemical modifications described herein.

1.2)  Non-limiting examples of such chemical modifications include without limitation
phosphorothioate internucleotide linkages, 2′-deoxyribonucleotides, 2′-O-methyl
ribonucleotides, 2′-deoxy-2′-fluoro ribonucleotides, "universal base"
nucleotides, "acyclic" nucleotides, 5-C-methyl nucleotides, and
terminal glyceryl and/or inverted deoxy abasic residue incorporation.

1.3)  In another embodiment, the invention features a double-stranded short
interfering nucleic acid (siNA) molecule that down-regulates expression of a
HIF1 gene comprising an antisense region, wherein the antisense region
comprises a nucleotide sequence that is complementary to a nucleotide sequence
of the HIF1 gene or a portion thereof, and a sense region, wherein the sense
region comprises a nucleotide sequence substantially similar to the nucleotide
sequence of the HIF1 gene or a portion thereof.  In one embodiment, the
antisense region and the sense region independently comprise about 15 to about 30
(e.g. about 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30)
nucleotides, wherein the antisense region comprises about 15 to about 30 (e.g.
about 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30)
nucleotides that are complementary to nucleotides of the sense region.

1.4)  In one embodiment, an siNA molecule of the invention comprises no
ribonucleotides.  In another embodiment, an siNA molecule of the invention
comprises ribonucleotides.

1.5)  In one embodiment, the invention features a chemically synthesized
double-stranded RNA molecule that directs cleavage of a HIF1 RNA via RNA
interference, wherein each strand of said RNA molecule is about 15 to about 30
nucleotides in length; one strand of the RNA molecule comprises nucleotide
sequence having sufficient complementarity to the HIF1 RNA for the RNA molecule
to direct cleavage of the HIF1 RNA via RNA interference; and wherein at least
one strand of the RNA molecule optionally comprises one or more chemically
modified nucleotides described herein, such as without limitation
deoxynucleotides, 2′-O-methyl nucleotides, 2′-deoxy-2′-fluoro nucleotides,
2′-O-methoxyethyl nucleotides etc.

1.6)  In any of the above-described embodiments of a double-stranded short
interfering nucleic acid (siNA) molecule that inhibits expression of a HIF1
gene, wherein a majority of the pyrimidine nucleotides present in the
double-stranded siNA molecule comprises a sugar modification, each of the two
strands of the siNA molecule can comprise about 15 to about 30 or more (e.g.,
about 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 or
more) nucleotides.

2)  Even
if you disagree with our analysis of the '422 patent, you can look at patent
8,273,866, which is in the
same family as 7,795,422.

Notably,
Claim 1 says:

"1.  A short interfering RNA (siRNA) molecule having a sense strand and an antisense
strand that mediates RNA interference, wherein:
    (a)
each strand is between 18 and 24 nucleotides in length;
    (b)
the sense strand comprises 10 or more 2′-deoxy, 2′-O-methyl,
2′-deoxy-2′-fluoro, or universal base modified nucleotides, and a terminal cap
molecule at the 3′-end, the 5′-end, or both 3′ and 5′-ends of the sense strand;
    (c)
the antisense strand comprises 10 or more 2′-deoxy, 2′-O-methyl,
2′-deoxy-2′-fluoro, or universal base modified nucleotides; and
    (d)
10 or more pyrimidine nucleotides of the sense and antisense strand are
2′-deoxy, 2′-O-methyl or 2′-deoxy-2′-fluoro nucleotides."

The
patent also specifies considerable flexibility for the terminal cap molecule,
stating "[t]he cap moiety can be an inverted deoxy abasic moiety, an
inverted deoxy thymidine moiety, or a thymidine moiety."

Again,
given that "2'deoxy nucleotides" and "a thymidine moiety"
are simply normal DNA bases, and that any sequence that is 18-24 bases can be
constructed from these claims, this may mean that every single 18-24mer is
potentially claimed by this patent.

3)  Patent Docs
then says, "while the '248 patent recites oligonucleotides — which
the specification states "[i]n certain aspects" can be "at least
15 nucleotides in length" — it is difficult to see how the authors can
conclude that the '248 patent has "explicit claims for 15mers that matched
84% of human genes."

See
above for our further analysis on this point.  Our main point is that 15mers from these patents match both human and
bovine genomes, and that claim construction becomes non-specific (even across
species) at these low k-mer sizes.

4)  "In paragraph 10 of his
Declaration, Dr. Mason surprisingly states that "Claim #I and #2 of '282
are so broad that they can include up to 100% of the genes in the human genome."

I
made this point for two reasons:

4.1)  The 55% homology is a very low threshold, and it allows matches to many other
genes, because "homology" and "identity" are not the same
thing in biology; homology allows for far more flexibility in the mismatches.

4.2)  The Myriad patents claim any "isolated DNA having at least 15 nucleotides
of the DNA of claim 2," which was my focus.  Notably, this is not the same as claiming 15 contiguous nucleotides.  Most other gene patents I have found will
describe contiguous nucleotides,
which obviously limits their scope.  But,
if we allow any 15 nucleotides, in any order, it is easy to match these
sequences to the entire genome and indeed every gene.

5)  "As
with the '422 and '248 patents, Dr. Mason appears to be having some difficulty
ascertaining the subject matter that is actually encompassed by claims 1 and 2
of the '282 patent."

I
would agree, insofar as I am worried about the likely overly broad scope of
these claims.  Since I risk liability
whenever I perform genetic testing on my own DNA, or the DNA from any of my
patients, I welcome the Supreme Court or the legal scholars to clarify the
issue.  In the absence of a statutory
research exemption for infringement liability or some other guarantee, I am
restricted from researching thousands of genes for many years to come.

But,
if any lawyer is confident enough about the irrelevance of these patents that
he or she would be willing to state, in writing, the willingness to defend me
in Court and pay all legal fees or damages if I get sued, then I will happily
join you and start working again to develop new tests, tools, and algorithms to
ameliorate and eliminate human diseases.

Thank
you,

Christopher
Mason, Ph.D.

[Ed. At Dr. Mason's request, his response has been revised to indicate that US 20130030040 A1 and US 20130041209 A1 are published patent applications rather than patents as Dr. Mason stated in the original version of his response.]