Board of Patent Appeals and Interferences
Patent and Trademark Office (P.T.O.)
*1 EX PARTE KIMISHIGE ISHIZAKA, CHRISTINE L. MARTENS AND KEVIN W. MOORE
Appeal No. 91-2539
April 30, 1992
Application for Patent filed March 31, 1987, Serial No. 032,859; a Continuation-in-Part of Serial No. 789,110 filed October 18, 1985. Glycosylation Inhibiting Factors.
Stephen C. Macevicz et al. for appellants
Primary Examiner--Jacqueline M. Stone
Examiner--Jasemine C. Chambers
Before Goldstein, Pellman and W. Smith
This is an appeal from the decision of the examiner finally rejecting claims 1, 2, 3 and 7, all the claims remaining in the application.
The subject matter on appeal involves a nucleic acid, identified as a sequence of nucleotides and their homologs, designated as glycosylation inhibiting factors (GIF)--natural compounds which are capable of causing T cells to produce a class of IgE binding factors which, in turn, selectively suppress the IgE response. One GIF is related to lipomodulin, a phosopholipase inhibitory protein. The more specific claims further limit the nucleic acid to a sequence that is at least 70% homologous to that defined by the formula in claim 1 (claim 2), to the nucleic acid obtained from mRNA produced from human cell line U-937 (claim 3) and to the vector TGIF-4 deposited with the ATCC under accession number 67311 (claim 7). To illustrate the claims on appeal, a copy of the claims is appended to this decision.
For evidence of obviousness, the references listed below are cited by the examiner:
Yokota et al. (Yokota), "Isolation and characterization of a mouse cDNA clone that expresses mast-cell growth-factor activity in monkey cells," Proc.Natl.Acad.Sci.USA, 81, 1070-1074 (1984).
Wallner et al. (Wallner), "Cloning and expression of human lipocortin, a phospholipase A@2 inhibitor with potential anti-inflammatory activity," Nature, 320, 77-81 (1986).
References to which appellants refer in their brief:
Flint, "Measurement of Messenger RNA Concentration," Genetic Engineering, Setlow et al., Editors, Vol. 2, Dept. of Biochemical Sciences, Princeton University, Princeton, NJ, 47-82 (1980).
Hirata et al. (Hirata), "Presence of autoantibody for phospholipase inhibitory protein, lipomodulin, in patients with rheumatic diseases," Proc.Natl.Acad.Sci.USA, 78(5), 3190-3194 (1981).
Katamura et al. (Katamura), "Biochemical identification of glycosylation inhibiting factor," Ibid, 87, 1903-1907 (1990).
All the claims stand rejected for being unpatentable in view of Wallner taken with Uede and Yokota. It is the examiner's position that Wallner discloses cDNAs encoding a human lipocortin, said cDNAs differing from the claimed invention by encoding a different lipocortin. However, the examiner urges, at the time of the claimed invention, Uede had reported the identification of GIF as a fragment of phosphorylated lipmodulin (same as lipocortin) and had also disclosed various functional assays for GIF. Additionally, the examiner explains, as shown by Yokota, it was well-known in the art that cDNAs were readily isolated from expression libraries using functional assays for screening. The examiner notes that Yokota discloses the isolation of a cDNA clone for mast cell growth factor from a pcD expression library. Therefore, the examiner concludes, the teachings of the three references, taken collectively, would have reasonably suggested to the routineer, the isolation of cDNAs which are similar or identical to those of appellants by employing the well-known pcD expression cloning method and GIF assays.
*2 On the other hand, in the paragraph bridging pages 2 and 3 of the brief, appellants deny that the examiner has established a prima facie case of obviousness. We are told that it is impossible to predict whether a GIF cDNA could be isolated by expression cloning "(1) when the abundance of GIF cDNA in the library to be screened is unknown and (2) when the sensitivity of the biological assay used to detect the expressed GIF is unknown."
After referring to a declaration by Dr. Paul Berg discussing the prerequisites for successful cloning of cDNAs and referring to two publications to establish the importance of sensitivity for the screening assay and the abundance of the clones in greater than 0.01%, appellants, at page 4 of the brief, present the following argument:
None of the references (Wallner, Uede, or Yokota) relied on by the Examiner discloses or suggests, either alone or in combination, any source of mRNA containing GIF mRNA, let alone a source of mRNA that contains greater than about 0.1% GIF mRNA. Moreover, none of the references discloses or suggests hybridization probes for detecting GIF cDNA, antibodies capable of detecting an expressed GIF product, or whether the biological assay of Uede is sufficiently sensitive to permit expression cloning of GIF. Thus, one of ordinary skill in the art would not have a reasonable expectation of cloning Appellants' GIF cDNA given Wallner, Uede, and Yokota as a starting point.
After giving careful consideration to the opposing arguments of appellants and of the examiner, as well as to all the evidence of record and the remarks relating thereto, we are unpersuaded of reversible error in the rejection before us, which will be sustained.
Initially, we observe that appellants' parent application, SN 789,110 was filed October 18, 1985 in the name of Kimishige Ishizaka, alone. Nevertheless, since various references of record, published subsequent to the parent application, but prior to the present application, are treated as prior art, we shall do the same. That is, we interpret appellants' action as a tacit agreement that the subject matter of the cited references is not antedated by the disclosure in the parent application.
Turning first to the Wallner publication item, we note the author's assertion that cloned human lipocortin complementary DNA had been cloned and the gene expressed in E. coli. At page 78, right column, the author further teaches that a U-937 cDNA library was constructed. In this connection, we note that appellants' claim 3 characterizes the nucleic acid of claim 2 as being obtained from messenger RNA produced by the human cell line U-937. In light of the teachings in the reference, there would have been a reasonable expectation for the routineer to obtain human lipocortin from a library of the type disclosed by Wallner.
Uede discloses a lymphokine that inhibits glycosylation of IgE-binding factors. The lymphokine is denominated glycosylation-inhibiting factor. In the last line of the abstract, Uede states that the glycosylation-inhibiting factor is a fragment of phosphorylated lipomodulin. At the last page, first line of the last paragraph, the author advises that "[i]t is well known that lipomodulin and macrocortin were formed by stimulation of neutrophils and macrophages with glucocorticoids." It is also indicated that the experiments performed by the author indicate the protein is formed by lymphocytes, after an immunologic stimulus. In the last line of the article, it is stated:
*3 The identification of glycosylation-inhibiting factor as a fragment of lipomodulin indicates that the lymphokine is a phospholipase inhibitory protein, and suggests that some of the other lymphokines may also exert their biologic activities through membrane-associated enzyme(s) on target cells.
The foregoing does not disclose the nucleotide sequence of appellants' glycosolation-inhibiting factor. Likewise, the references do not reveal the availability of a probe that might be employed to recover a nucleic acid having a sequence of nucleotides that would encode a polypeptide exhibiting glycosylation-inhibiting activity. Nevertheless, Yokota describes an alternative method of isolating cDNA clones that encode desired activity. At page 1073, Yokota presents the following explanation:
We describe here the isolation of cDNA clones that encode the MCGF produced by the murine T-cell clone Lyl+2@-/9. When this study was undertaken, there was no nucleotide or amino acid sequence information available for the synthesis of specific nucleic acid probes, nor were there antibodies against MCGF that could be used for immunological detection of the protein. However, a convenient and reliable proliferation assay for MCGF did exist. We showed that X. laevis oocytes injected with mRNA from C1.Lyl+2@-/9 synthesize active MCGF detectable by this proliferation assay.
The author thereafter explains how the initial stages of the screening were conducted and how a specific clone was eventually found that hybridized for the desired mast-cell growth-factor (MCGF). We have no doubt that a person having ordinary skill in the subject art would have used the Yokota method to isolate a sequence of nucleotides capable of encoding a polypeptide exhibiting GIF activity in the reasonable expectation of success.
At page 4 of the brief, appellants contend that the references do not suggest a source of mRNA containing GIF mRNA in a concentration greater than about 0.1% GIF mRNA. Nevertheless, Yokota teaches an enrichment technique at page 1071, left column, and in the declaration by Dr. Paul Berg, beginning at page 2, the declarant acknowledges that the "relative abundance of a mRNA can be increased by a variety of means: (1) fractionation by sucrose gradient centrifugation or gel electrophoresis provided that there is a suitable assay available; (2) subtractive hybridization if an inducible or tissue-specific source of mRNA is available; or (3) selective extraction of membrane- or polysome-associated mRNAs for enrichment of mRNAs encoding secretory proteins." Declarant also explains that successful cloning of cDNAs corresponding to low abundance mRNAs depends on the availability of (i) a sensitive and selective biological assay for the encoded gene product, (ii) antibodies to the protein encoded by the desired mRNA, or (iii) a hybridization probe specific for the desired clone. Declarant, we are aware, cautions that the approaches are not available in every case and are often "difficult to apply technically." However, appellants have not informed us of any objective evidence showing that a person having ordinary skill in the subject art could not have used available enrichment techniques to increase the abundance of GIF mRNA in a U937 cDNA library to a point wherein the desired material would be present in an operable concentration. On the basis of the above, we find that the examiner has established a prima facie case of obviousness. Accordingly, following the mandate of the court in In re Johnson, 747 F.2d 1456, 223 USPQ 1260 (Fed.Cir.1984), we reevaluate all the evidence anew and weigh the evidence of obviousness against that of non-obviousness.
*4 Insofar as the examiner has relied upon Wallner, Uede and Yokota, we believe these references have been adequately discussed. Thus, we focus on appellants' evidence of non-obviousness. In their brief, appellants refer to declarations by Dr. Paul Berg, by coappellant Dr. Kimishige Ishizaka and by coappellant Dr. Christen Martens. Thus, at page 4 of the brief, the following sentence appears:
Otherwise, the two cDNAs and their respective products are completely different structurally, as shown by the Declarations by Dr. Martens (no homology at nucleic acid level) and Dr. Ishizaka (monoclonal antibodies specific for Wallner's lipocortin do not cross-react with Appellants' GIF).
Likewise, at page 5, appellants, after observing that the Katamura reference is discussed in the declaration by Dr. Ishizaka, further state:
Preliminary to our discussion of the declaration evidence, we stress that appellants have the burden of explaining the data in any declaration they proffer as evidence of non-obviousness. Compare In re Payne, 606 F.2d 303, 203 USPQ 245 (CCPA 1979); In re Borkowski, 505 F.2d 713, 184 USPQ 29 (CCPA 1974) and In re Goldberg, 35 CCPA 1225, 168 F.2d 527, 78 USPQ 69 (1948). We suggest that a mere reference to the declarations does not adequately discuss the teachings of the declarants.
In the declaration by Dr. Martens, declarant states that she compared the sequence shown in the reference with that inserted in the clone TGIF4 and found no significant homology. Therefore, we are told, it would "be virtually impossible to use one as a probe to screen for the other in a cDNA library by cross-hybridization." However, as shown by our earlier discussion, the examiner does not rely upon Wallner to disclose a probe. Rather, reliance was placed upon Yokota for the teaching of expression cloning as a technique that might be employed to isolate the desired GIF mRNA. Therefore, this declaration has little probative value.
The declaration by Dr. Paul Berg, as observed by the examiner, is from a non-related application and discusses references upon which the examiner does not rely. Hence, to the extent that Dr. Berg discusses the two previously cited references, the declaration has no value as evidence herein. On the other hand, Dr. Berg's general discussion of the problems accompanying low concentration of total mRNA is relevant to a consideration of obviousness with respect to the cited references. However, appellants have not established by competent evidence that an unworkable concentration of mRNA would be present. Additionally, after explaining why very low concentrations may result in experimentally unreasonable hybridization times, Dr. Berg acknowledges the steps that may be taken to increase the relative concentration of the desired material and mentions the availability of at least one approach that is necessary for successful cloning of cDNAs corresponding to low abundance mRNAs. Since we have not been informed how Dr. Berg's discussion relates to the use of expression cloning, as described by Yokota, we find the noted declaration inadequate to outweigh the evidence of unobviousness.
*5 The remaining declaration, that by Dr. Ishizaka, after presenting the declarant's credentials, describes the characterization of the glycosylation inhibiting factor in the Uede article. Dr. Ishizaka follows his description of Uede with the statement:
From these observations we concluded that lipomodulin and GIF were probably versions of the same protein. It was also believed that both belonged to the lipocortin "family" of proteins because of earlier observations that lipomodulin could inhibit phospholipase activity, although the structural relationship between GIF/lipomodulin and the other lipocortins reported in the literature remained unknown.
Declarant refers to the 1990 Katamura publication to support his statement that purification of GIF to the extent it could be identified as a single band on a gel by sodium dodecyl sulfate polyacrylamide gel electrophoresis has only recently been accomplished. Additionally, appellants' statements in the declaration seem to conflict with those appearing at pages 174 and 175 of the Ishizaka article "Regulation of IgE synthesis" in Ann.Rev.Immunol., 2, 159-182 (1984), cited at page 3 of the present specification to support the statement "recently, Ishizaka and his co-workers have discovered and characterized a class of compounds which they have designated as glycosylation inhibiting factors (GIFs)." Thus, it does not appear that the declaration is entitled to much probative weight relative to the evidence of obviousness. Moreover, the intertwining of appellants' invention with that of later developments by other workers would seem to raise a question as to when and who invented the subject matter now claimed.
Regarding the Hirata publication item, appellants, at page 5 of their brief, refer to the observations in the publication item, wherein it was concluded that lipomodulin and GIF were probably versions of the same protein. However, on the basis of the Katamura publication, appellants state: "[i]t is now known that the lipocortins of Wallner are structurally distinct from Appellants' GIF." We have already adverted to the Katamura publication item. Moreover, insofar as they are attempting to employ the publication as a substitute for a declaration, with its concomitant safeguards, appellants should be advised that 37 C.F.R. 1.132 does not recognize this method of overcoming a reference. Furthermore, the only specific mention of Wallner's work is in the discussion beginning with the sentence bridging pages 1906 and 1907 of the article. There, the author states that "GIF does not appear identical to lipocortin 1 or lipocortin 2." Also, "[n]either the monoclonal anti-GIF (31C3) nor antilipomodulin (141-B9) bound recombinant human lipocortin 1, which has 90% identity with rat lipocortin in amino acid sequences and reacted with anti-rat lipocortin." It will be seen that end note 23 refers to the Wallner article.
Although all the evidence has been given due consideration, we find the evidence of obviousness outweighs the evidence of non-obviousness. Consequently, the rejection of record logically must be sustained.
*6 Under the provisions of 37 C.F.R. 1.196(b), we exercise our discretion and set forth a new ground of rejection of claims 1, 2 and 3 for failing to comply with the requirements of 35 U.S.C. 112, first and second paragraphs. As clearly explained by the court in In re Moore, 439 F.2d 1232, 169 USPQ 236 (CCPA 1971), claims must be analyzed in light of the specification, and not in a vacuum. Thus, although appellants' claims superficially appear definite, in light of the disclosure at pages 5 and 6 of the specification, we find the claims to be indefinite. At page 5 of the disclosure, the invention is described as including fragments of the nucleotide sequence disclosed in Formula 1 which are capable of being used as hybridization probes for screening cDNA or genomic libraries to obtain additional nucleic acids encoding GIFs. The nucleotide sequences of Formula 1 may range in size from about 18 to 20 bases to the full length of the disclosed sequence. Furthermore, claim 1 recites "a sequence of nucleotides effectively homologous to the nucleotide sequence." The term "effectively homologus" is defined at page 6 of the specification and in the last three lines on page 6 is stated to be "at least 50% homologous to the sequence to be isolated." Therefore, it would appear that only 9 or 10 bases in formula I are necessary to meet the terms of the claim. However, we find no disclosure in the specification showing that a fragment of this size has been effectively employed by appellants or could be employed by others without undue experimentation.
Consequently, we can neither determine the intended scope of claims 1, 2 and 3 nor find adequate enabling support for the breadth that would seem to be attributable to these claims. In this connection, attention is invited to Amgen Inc. v. Chugai Pharmaceutical Co., 927 F.2d 1200, 18 USPQ2d 1016 (Fed.Cir.1991), wherein, the court, in dealing with the question of enablement for claims directed to erythropoietin (EPO) analogs, agreed with the district court's finding of non-enablement. At 18 USPQ2d 1027, the court explained:
We believe the trial court arrived at the correct decision, although for the wrong reason. By focusing on the biological properties of the EPO analogs, it failed to consider the enablement of the DNA sequence analogs, which are the subject of claim 7. Moreover, it is not necessary that a patent applicant test all the embodiments of his invention, In re Angstadt, 537 F.2d 498, 502, 190 USPQ 214, 218 (CCPA 1976); what is necessary is that he provide a disclosure sufficient to enable one skilled in the art to carry out the invention commensurate with the scope of his claims. For DNA sequences, that means disclosing how to make and use enough sequences to justify grant of the claims sought. Amgen has not done that here. In addition, it is not necessary that a court review all the Wands [FN1] factors to find a disclosure enabling. They are illustrative, not mandatory. What is relevant depends on the facts, and the facts here are that Amgen has not enabled preparation of DNA sequences sufficient to support its all-encompassing claims.
*7 We are convinced that the foregoing is applicable to claims 1, 2 and 3 on appeal.
With respect to claim 7, this is directed to vector TGIF-4, which contains an insert corresponding to the nucleotide sequence of Formula I. Since the scope of this claim is clear and support for said claim is provided in the specification, the new rejections are not applicable to claim 7.
For the reasons expressed in the answer and those discussed above, the examiner's decision rejecting claims 1, 2, 3 and 7 is affirmed. Additionally, a new ground of rejection has been set forth under the provisions of 37 C.F.R. 1.196(b) as to claims 1, 2 and 3.
Any request for reconsideration or modification of this decision by the Board of Patent Appeals and Interferences based upon the same record must be filed within one month from the date hereof (37 CFR 1.197).
With respect to the new rejection under 37 CFR 1.196(b), should appellants elect the alternate option under that rule to prosecute further before the Primary Examiner by way of amendment or showing of facts, or both, not previously of record, a shortened statutory period for making such response is hereby set to expire two months from the date of this decision. In the event appellants elect this alternate option, to preserve the right to seek review under 35 U.S.C. 141 or 145 with respect to the affirmed rejection, the effective date of the affirmance is deferred until conclusion of the prosecution before the examiner unless, as a mere incident to the limited prosecution, the affirmed rejection is overcome.
If appellants elect prosecution before the examiner and this does not result in allowance of the application, abandonment or a second appeal, this case should be returned to us for final action on the affirmed rejection, including any timely request for reconsideration thereof.
No time period for taking any subsequent action in connection with this appeal may be extended under 37 CFR 1.136(a). See the final rule notice, 54 F.R. 29548 (July 13, 1989), 1105 O.G. 5 (August 1, 1989).
AFFIRMED 37 C.F.R. 1.196(b)
BOARD OF PATENT APPEALS AND INTERFERENCES
Irving R. Pellman
William F. Smith
FN1. In re Wands, 858 F.2d 731, 8 USPQ2d 1400 (Fed.Cir.1988).
1. A nucleic acid capable of encoding a polypeptide exhibiting glycosylation inhibiting factor activity, the nucleic acid having a sequence of nucleotides effectively homologous to the nucleotide sequence defined by the formula:
2. The nucleic acid of claim 1 wherein said polypeptide exhibits human glycosylation inhibiting factor activity, and wherein said sequence of nucleotides is at least seventy percent (70%) homologous to said nucleotide sequence defined by said formula.
3. The nucleic acid of claim 2 obtained from messenger RNA produced by the human cell line U-937.
7. The vector TGIF4 deposited with the American Type Culture Collection under accession number 67311.