BPAI Board of Patent Appeals and Interferences Patent and Trademark Office (P.T.O.) *1 EX PARTE LARS THIM, MOGENS T. HANSEN AND KJELD NORRIS Appeal No. 90-2870

Board of Patent Appeals and Interferences

Patent and Trademark Office (P.T.O.)



Appeal No. 90-2870

August 22, 1991

HEARD: May 7, 1991



 Application for Patent filed March 19, 1986, Serial No. 841,327. Insulin Precursors, Process For Their Preparation And A Process For The Preparation Of Human Insulin.



Morris Fidelman et al. for Appellants



Primary Examiner--Richard A. Schwartz



Examiner--Richard C. Peet



Before Winters, Goolkasian and Kimlin












 This is an appeal from the examiner's final rejection of claims 1-4, 8, 9 and 14, which are all the claims of the application.



 Claim 1 is illustrative and reads as follows:

   1. Human insulin precursors of the formula:




B-X-Y-A  (I)


wherein B and A are the B- and A-chain of human insulin cross-linked as in human insulin and wherein X and Y are each lysine or arginine selected from the group consisting of B-Lys-Arg-A; B-Lys-Lys-A; and B-Arg-Lys-A.



 The references relied on by the examiner are:




Goeddel et al. (Goeddel) EPO  0,055,945  Jul. 14, 1982

Brake et al. (Brake) EPO      0,121,884  Oct. 17, 1984


Barman, Enzyme Handbook, Vol. II, 1969, pages 608-609.



 Kemmler et al. (Kemmler), The Journal of Biological Chemistry, "Studies on the Conversion of Proinsulin to Insulin," 1971, pages 6786-6790.



 Appellants' invention concerns the production of human insulin using recombinant genetic engineering technology. More particularly, the invention concerns the preparation of certain precursors of human insulin known as "proinsulins."



 Human insulin is composed of two polypeptide chains termed A and B chains, held together by sulfide linkages between cysteine moieties of each polypeptide. Insulin is not produced as such by the human cell but, rather, is produced as a single polypeptide chain containing a sequence of about 30 residues known as a "connecting peptide" (C-peptide) which joins the carboxyl end of the B-chain and the amino terminus of the A-chain of the insulin molecule. The connecting peptide (C-chain) is removed by proteolytic enzymes upon secretion. The connecting peptide in human proinsulin contains the amino acid couplet Arg-Arg at the amino end and the amino acid couplet Lys-Arg at the carboxyl end. The proinsulins claimed by appellants differ from human proinsulin in that the connecting peptides of appellants' proinsulin molecule are not the 30 residue polypeptide chain of natural proinsulin, but are couplets of only 2 peptides; the peptide couplets of the claims being Lys-Arg (claim 2), Lys-Lys (claim 3) and Arg-Lys (claim 4). Claims 8, 9 and 14 also include Arg-Arg as an additional couplet within their scope.



 Appellants' claims stand rejected as follows.



  *2 Claims 1-4 stand rejected under 35 U.S.C. § 103 over Goeddel.



 Claims 8 and 14 stand rejected under 35 U.S.C. § 103 over Goeddel in view of Brake.



 Claim 9 stands rejected under 35 U.S.C. § 103 over Goeddel in view of Brake and further in view of Kemmler and Barman.



 Claims 8, 9 and 14 stand rejected under 35 U.S.C. § 112, first paragraph, as being broader than the enabling disclosure. [FN1]



 We have carefully considered all the arguments set forth in appellants' Brief and Reply Brief as well as the arguments set forth at the oral hearing. Nevertheless, we are unpersuaded of error in the examiner's rejection. Accordingly, we affirm all rejections for the reasons set forth by the examiner, adopting his reasoning as our own.



 There is no question that the concept of reducing the 30 amino acid polypeptide chain of the connecting peptide (C-peptide) to only 2 amino acids is taught by the prior art Goeddel '945 reference. Pages 8 and 9 of the '945 Patent teach that proinsulins may be made wherein the "C" chain of amino acid units is an analog comprising as few as 2 amino acid units. Goeddel teaches that preferably the bridging "C" chain should have sites which permit its excision by enzymatic means and suggests that preferred sites are the amino acid couplets found on the end units of normal "C" chains including the Arg-Arg and Lys-Arg units. Preferred "C" chain analogs described by Goeddel contain either the two acid couplet Arg-Arg or a six amino acid polypeptide chain composed of the sequence Arg-Arg-Gly-Ser-Lys-Arg. Goeddel indicates that the purpose of using the peptide couplets which terminate the C-chain of human insulin as components of Goeddel's "C" chain analogs is to ensure ready cleavage of the "C" chain because these particular sequences are very susceptible to proteolytic cleavage using standard enzymes such as trypsin and carboxypeptidase B, the very same enzymes used by appellants to remove the C-peptide. We are in complete agreement with the examiner that one skilled in the art having the teachings of Goeddel before him would have considered it obvious to use the Lys-Arg couplet and, indeed, any couplet capable of being readily cleaved using proteolytic enzymes such as trypsin and carboxypeptidase B.



 Claims 8, 9 and 14 differ from claims 1-4 insofar as they concern the production of the proinsulin analogs in a yeast expression vehicle. The examiner has rejected these claims based on the combined teachings of Brake and Goeddel.



 There is no question that the examiner is correct in taking the position that Brake provides the necessary impetus for preparing the proinsulin analogs of Goeddel in a yeast expression vehicle. As noted in the "Background" section of appellants' specification, the prior art was aware that the use of E. coli as a transformant organism has the drawback that expressed proinsulin products are not secreted from the cells but accumulate intracellularly in the E. coli host organism. The Brake reference notes that an advantage of yeast as an expression vehicle is that human proinsulin is secreted outside the cell into the nutrient medium where it may be isolated in high yield. In this regard page 2, lines 30- 35 and page 15, lines 16-19 of Brake are most pertinent.



  *3 Appellants urge, however, that the results achieved upon using the yeast expression vehicle are greater than would have been expected considering the rather poor performance of proinsulins having more than 2 amino acids in the C-chain when expressed by yeast. Appellants' results are set forth at Table 1 at page 13 of the specification and reiterated at page 5 of the Brief. The results in Table 1 indicate that insulin precursors having more than 6 amino acids in the C-peptide chain yield a fermentation broth having dissolved therein a lesser amount of proinsulin than precursors having C-chains of six or less amino acids, the results, expressed in ??mo1/1 of amino reactive insulin being 0.14 for Construction No. 2, a ten acid chain; 0.31 for Construction No. 6, a six acid chain; 1.0 for Construction No. 7, a two acid (Arg-Arg) chain and from 1.6-2.0 for other two acid chains composed of Arg and/or Lys combinations. Though appellants urge strongly that the superior results of the two acid chain C linkages in yeast mandate patentability, the examiner has taken the position that the evidence is insufficient to overcome the prima facie case established.



 It is by now well settled that obviousness is a question of patent law and not of chemistry. In re Papesch, 315 F.2d 381, 386, 137 USPQ 43, 47 (CCPA 1963). We are mindful of the decisions in In re Murch, 464 F.2d 1051, 175 USPQ 89 (CCPA 1972) and In re Chupp, 816 F.2d 643, 2 USPQ2d 1437 (Fed.Cir.1987), which indicate that an unexpected improvement in properties or results achieved may, in some cases, be sufficient to impart patentability to compositions or processes which are otherwise prima facie obvious. Nevertheless, in this instance, where the art provides ample reason to modify the prior art products and processes to obtain at least one important property (secretion into the fermentation broth) the examiner is required to weigh the expected results against the unexpected results along with the scope of the subject matter claimed. In re May, 574 F.2d 1082, 197 USPQ 601, 609 (CCPA 1978).



 There must be "clear and convincing evidence of unobvious results in order to overcome a prima facie case of obviousness." In re Lohr, 317 F.2d 388, 137 USPQ 548 (CCPA 1963). The presentation of objective evidence of nonobviousness does not, in and of itself, mandate a conclusion of nonobviousness. In re Chupp, supra. The fact finder, in this case the examiner, is entitled to his own ideas, within reason, as to what evidentiary facts will persuade him of unexpected results. Thus, whether the rebuttal evidence is sufficient to persuade the examiner is an evidentiary matter left, within reason, to the trier of fact. In re Johnson, 747 F.2d 1456, 223 USPQ 1260, 1263 (Fed.Cir.1984).



  *4 It is evident that in the case before us the examiner has weighed heavily the very strong incentive to use yeast as the vehicle for expression of the analog proinsulins of Goeddel in order to have the proinsulin expressed into the fermentation broth. From a production viewpoint this advantage would greatly enhance the yields based on a given cell population since the broth may be removed without destroying the cell population as would be the case were E. coli used. The examiner has also evaluated the yields proposed by appellants as basis for the grant of a patent and has noted that the yield obtained using Goeddel's preferred Arg-Arg bridging C-chain falls within the superior yields set forth in claim 14. The examiner has concluded that the yield achieved using yeast as an expression vehicle is not so significantly superior that it overcomes the prima facie case of obviousness. We cannot state that the examiner's position is unreasonable and, accordingly, we affirm the examiner's rejections under 35 U.S.C. § 103.



 Turning next to the rejection of claims 8, 9 and 14 under 35 U.S.C. § 112, first paragraph, we are in complete agreement with the examiner that the specification is not enabling with regard to strains of yeast other than Saccharomyces cerevisiae strain MT105 and the resulting fermentation broth. The background portion of appellants' specification indicates that there are enzymes in yeast which tend to degrade proinsulin type precursors. Accordingly, the examiner's reasoning regarding unpredictability is not unreasonable and may be said to have a sound scientific basis.



 Claim 14 represents a further problem from a patentability viewpoint insofar as appellants have acknowledged that the fermentation broth claimed therein is not limited to one derived from culturing yeast and contains no upper limit on the amount of human insulin precursor contained in the broth. With regard to the lack of upper limit, we are of the opinion that the decision of In re Fisher, 427 F.2d 833, 166 USPQ 18 (CCPA 1970) is controlling. With regard to the scope of enablement, we remind appellants that the term "fermentation" is not limited to fermentations involving yeast but, rather, is used to describe reactions in which hydrogen atoms are taken from one organic substance and given to another. This reduction reaction can be effected by numerous microorganisms other than yeast including Clostridium, E. coli, Lactobacillus and Enterobacter. [FN2] Appellants' disclosure is simply not enabling for the different types of broths prepared by the different types of microorganisms.



 The examiner's rejections of claims 1-4, 8, 9 and 14 under 35 U.S.C. § 103 and 35 U.S.C. § 112 are affirmed.



 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).









Sherman D. Winters






John T. Goolkasian






Edward C. Kimlin






FN1. This rejection under 35 U.S.C. § 112, first paragraph, was added by the examiner in his Answer.



FN2. See the attached pages 150 and 151 of the text "Microbiology" by Nester et al.


<< Return to Board of Patent Appeals and Interferences Index