Genencor International Inc v Novo Nordisk A/S
[2001] APO 49
•3 October 2001
OFFICIAL NOTICE
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Application : No. 682863 in the name of Genencor International Inc
Title: Oxidatively stable alpha-amylase
Action: Opposition under section 59 by Novo Nordisk A/S
Decision: Issued .
Abstract
Claims 18-20 lack novelty because they inadvertently define the known A4 form of the alpha-amylase rather than the mutated form of the invention. None of the other claims have been found to lack novelty or an inventive step. There is, however, a serious fair basis problem in the specification. The only evidence to suggest that the amino acids, methionine and tryptophan, are important for activity and/or stability in Bacillus alpha-amylase is in relation to positions M15, M197 and W138. There is no basis for extrapolating to the remaining methionine and tryptophan residues which are completely independent of those positions.
PATENTS ACT 1990
DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS
Re:Patent Application No. 682863 by Genencor International Inc and Opposition under section 59 by Novo Nordisk A/S
BACKGROUND
Australian patent application 682863 was filed on 10 February 1994 by Genencor International Inc (Genencor) under the provisions of the PCT. The Australian application claimed priority from an earlier US basic application (08/016395) filed on 11 February 1993.
Application 682863 was advertised accepted on 23 October 1997 and on 23 January 1998, a notice of opposition was filed by Novo Nordisk A/S (Novo). Evidence in the opposition was completed on 16 August 2000 and a hearing was held in Canberra on 1 and 2 February 2001. Amendments to the specification were proposed on 9 November 1999 and these were allowed on 1 August 2000. The hearing was therefore based on the specification as amended after acceptance.
At the hearing, the applicant was represented by Dr Debbie Yin-Foo, patent attorney of Phillips Ormonde & Fitzpatrick, Melbourne and the opponent was represented by Dr John McCann, patent attorney of Spruson & Ferguson, Sydney.
SPECIFICATION
The specification, as amended after acceptance, is directed to mutant alpha-amylases. Alpha-amylases (alpha-1,4-glucanohydrolase, EC 3.2.1.1) are enzymes which hydrolyse internal alpha-1,4-glucosidic linkages in starch largely at random, to produce smaller molecular weight malto-dextrins.
These enzymes are of considerable commercial value, being used in the initial phases (liquefaction) of starch processing, in alcohol production, as cleaning agents in detergent matrices, and in the textile industry for starch de-sizing. They can be isolated from a wide variety of sources, with the most commercially important amylases being produced from Bacillus species. However, the enzymes currently available are not always suitable for the conditions used in such commercial applications.
Therefore, according to the specification, there is a need for an amylase having an altered stability and/or activity profile (oxidative, thermal or pH performance profile) and this can be achieved by selective replacement, substitution or deletion of oxidisable amino acids including methionine (Met or M), and tryptophan (Trp or W).
The specification ends with 33 claims, of which 6 are independent. These are outlined below:
A mutant alpha-amylase that is the expression product of a mutated DNA sequence encoding an alpha-amylase derived from Bacillus, the mutated DNA sequence being derived from a precursor Bacillus alpha-amylase by the deletion or substitution of one or more methionines or a deletion or substitution of one or more tryptophans at a position equivalent to a methionine or a tryptophan in Bacillus licheniformis alpha-amylase with the proviso that the tryptophan deletion or substitution is not W165.
A mutant alpha-amylase having a tryptophan deletion in the precursor alpha-amylase equivalent to any tryptophan in B.licheniformis alpha-amylase as shown in SEQ ID NO:32 with the proviso that the tryptophan deletion is not W165.
A mutant alpha-amylase having a tryptophan substituted or deleted at a position equivalent to W138 in B.licheniformis alpha-amylase.
An alpha-amylase comprising an amino acid sequence corresponding to SEQ ID NO: 37 or a derivative thereof; wherein the derivative hydrolyse (sic) internal alpha-1,4-glucosidic linkages in starch.
A mutant alpha-amylase having enhanced thermal stability, or an enhanced pH performance profile or enhanced oxidative stability, the mutant comprising a substitution of a different amino acid at a position equivalent to M15 in B.licheniformis alpha-amylase.
A mutant alpha-amylase substantially as hereinbefore described with reference to any one of the examples.
DECISION
Section 40 issues
A key factor in this decision is to determine where the inventive idea resides in the current specification. It was well known at the priority date that alpha-amylases were exposed to a wide range of conditions in commercial applications, including conditions where oxidants such as bleach or peracids are present. It was also well-known that these enzymes were not always stable in, or otherwise suitable for, such conditions. The problem being addressed by the specification was how to improve, or alter, the properties of the enzyme.
The specification acknowledged that there had been previous studies using recombinant DNA techniques to explore which amino acid residues were important for the catalytic activity of amylases as well as those which important for thermal stability. In the current specification, similar recombinant techniques were used to establish that particular methionine and tryptophan residues were important for activity and/or stability, including oxidative stability.
According to the applicant, having identified that methionine and tryptophan were important for activity and/or stability of alpha amylase, they were then entitled to claim alpha-amylases where any of the methionine or tryptophan residues were substituted or deleted. However, the specification conceded (page 4) that after cysteine, methionine and tryptophan are the most oxidisable naturally occurring amino acids. Given that there are no cysteine residues in the alpha-amylase from B. licheniformis (see Scopes paragraph 25 and figure 1 of the opposed application), methionine and tryptophan are the next most likely amino acid residues to target to improve oxidative stability. This means the idea of selectively targeting methionine or tryptophan over other oxidisable amino acids in proteins was already known at the priority date, and the invention is necessarily more limited.
In the Bacillus alpha-amylase, there are 7 methionine residues (+8, +15, +197, +256, +304, +366, +438) and 16 tryptophan residues (+13, +22, +141, +138, +155, +157, +165, +182, +184, +215, +218, +244, +263, +342, +411, +449). The applicant conceded that not all alterations at methionine or tryptophan residues will affect activity/stability - only those at which the residue occurs near the active site. The applicant also conceded that it was not possible to predict whether an alteration at a particular methionine or tryptophan residue will affect stability of a protein without some knowledge of the 3-D structure of the protein or some other indication of the importance of the particular residue.
The only evidence in the specification that any of these positions is important for stability comes from investigating properties of a range of mutants where substitutions had been made at particular methionine or tryptophan residues, as described in more detail in the table below. A limited number of substitutions were made at the W138 position and the resulting mutants had altered properties from the wild type. No other tryptophan position appears to have been investigated. A number of substitutions were made at each of the M15 and M197 positions with the resulting mutants having altered properties from the wild-type. Of the remaining 5 methionine positions, apart from two amino acid substitutions at position M366, only one amino acid substitution was attempted at each of the other methionine sites. All of these were all vulnerable to H2O2 oxidation at pH 5.0 (see page 22, example 2). No other properties were provided for these other mutants to establish that that their activity and/or stability varied over that of wild-type.
While this suggests that particular positions (M15, M197 and W138) could be important for activity and/or stability, there is no basis to extrapolate from these to the remaining methionine and tryptophan residues which are completely independent of the M15, M197 and W138 positions. Therefore claims to mutant alpha-amylases with substitutions other than M15, M197 and W138 are beyond the scope of the invention and are, consequently, not fairly based.
The opponent argued that not all mutants with substitutions at +15, +197 and +138 have improved oxidative stability. According to the opponent, such substitutions do not achieve the promise of the invention and are not fairly based [as per AMP v Utilux (1971) 45 ALJR 123 (at 131)]. The applicant conceded that not all mutant enzymes had improved performance, but argued that the specification only promised to alter an enzymes stability and/or activity profile.
In my view, while the specification is poorly drafted, there are a number of places that indicate that the aim of the invention was to alter, rather than improve, stability and/or activity. For example, at page 5, the specification noted that enzymes with reduced oxidative stability may be useful in industrial processes that require the rapid and efficient quenching of enzyme activity.
Therefore, I accept that the invention is not necessarily limited to examples of alpha-amylases with improved oxidative stability.
The opponent also argued that there was so much variation in the properties of the mutants, it was not possible to predict, in advance, what properties a particular mutant would have. As a result, the opponent argued that the claims should be restricted to the mutants exemplified in the specification and not to all substitutions at M15, M197 and W138. To demonstrate the point, they provided a summary of the properties of the exemplified range of M15, M197 and W138 mutants which I have presented in the following table.
| Property | Substitution at M15 | Substitution at M197 |
| Specific activity | Specific activities of M15X mutants varied from greater to 200% to less than 25% of the wild type activity (figure 14). M15L did not show increased specific activity (table III). | Specific activities of the limited number of M197X mutants assayed also varied. None were better than the wild type, and M197L was significantly worse (table III) |
| Thermal stability | None of the M15 mutants had significantly better thermal stability, some were worse and some significantly worse (figure 15). M15L was equivalent to wild type. | Only two M197 substitutions had slightly better thermal stability (perhaps not significant) – M197C and M197T (figure 10) and many were worse (including M197L) –see also page 32). |
| H2O2 oxidative stability | Only M15L data provided (figure 7)– the specification conceded that M15L was still inactivated by H2O2 (page26). The opponent argued that no significant improvement of oxidative stability over wild type. | H2O2 oxidative inactivation data only provided for 2 mutants - M197T and M197L which were significantly resistant to H2 O2 oxidation (figure 7) although the specification indicated on page 32 that M197A and M197T were also resistant to peroxide inactivation. M197C was said to be inactivated readily by air oxidation (see page 32). |
| Chloramine T oxidative stability | No data provided | Data only provided for mutants M197T, M197L, M197A, M197T/W138F and M197T/W138Y. All performed better than wild type (example 10 and figures 17-19). |
| Inactivation in a dishwasher detergent | No data provided | Data only provided for M197T (including the A4 form) and M197A (A4 form only). Only M197T provided acceptable results (figure 11), M197A was inactivated before the first assay could be performed (page 33). |
| Starch liquefaction | Only provided for certain M15X mutants (those that had at least equivalent specific activity to the wild type) (table IV and figure 16). M15L, M15T, M15N, M15D all appear to have greater jet performance results compared to wild type, M15S and M15V had less, with M15V having only around 15% of wild type. |
I agree with the opponent that there is considerable variation in the properties of the exemplified mutants. I also note that prior art cited by the opponent [EP 285123 by Suomen Sokeri OY – (JFM-5)] acknowledges that the choice of a suitable amino acid for substitution often involves much trial and error because there is little knowledge concerning the natural laws governing structure and function of proteins (see page 2, lines 41-45).
Having said that, differences between members of a class of proteins may not be critical if there is also at least one beneficial property common to the whole class and provided there is a reasonable basis in the specification to extrapolate to the whole class. In such cases, the fact that some trial and error is needed to generate mutant proteins is not fatal as long as the specification enabled the other products of the class to be produced without further invention. As concluded by Heerey J in Genetics Institute, Inc v Kirin-Amgen, Inc (No. 3) [1998] 740 FCA (25 June 1998), claims need not be limited to specific protein sequences if:
there was a beneficial property common to the class of proteins;
the patent application disclosed a principle of general application; and
the exemplified coding sequence enabled other products of the class to be produced.
In the table above, the properties of four M197 mutants were described (M197T, M197L, M197A and M197C). Although their overall properties varied considerably, one property, in particular, (oxidative stability) was consistent. All of the mutants had improved oxidative stability except M197C which was said to be inactivated readily by air oxidation. From these results, the skilled worker can reasonably predict that methionine at the M197 position is important for the oxidative stability of alpha-amylase. Even the results with M197C are consistent with this principle because cysteine is known to be more oxidisable than methionine.
Applying the Genetics Institute, Inc v Kirin-Amgen [supra] test above, the mutants have a "common beneficial property" which is an altered oxidative stability compared with the wild-type enzyme. There is a general principle of application because the specification teaches that alterations at M197 affect oxidative stability. Further, while some trial and error may be required, there has been no suggestion that the skilled worker would not be able to generate the full range of mutants at the M197 position and investigate their properties from their common general knowledge. In other words, the specification enabled other products of the class to be produced.
As a result, the applicant is entitled claim all substitutions at M197, as perOlin Mathieson Chemical Corporation v Biorex Laboratories [1970] RPC 157 at page 193.:
"If it is possible for the patentee to make a sound prediction and to frame a claim which does not go beyond the limits within which the prediction remains sound, then he is entitled to do so."
Of course, Olin v Biorex [supra] also noted that the applicant in making a prediction is taking the risk that "a defendant may be able to show that his prediction is unsound or that some bodies falling within the words he has used have no utility or are old or obvious or that some promise he has made in the specification is false in a material respect". However, this suggests that once an applicant has made a reasonable prediction, the onus is on the opponent to prove the applicant's prediction is incorrect rather than on the applicant to demonstrate otherwise. The opponent has failed to do this and consequently has not established that the claims which encompass all substitutions at M197 lack fair basis.
The opponent further argued that the invention did not extend to deletions at M197 because no deletions had been exemplified. However, in my view, the applicant has established the broad principle that methionine at the M197 position is important for the oxidative stability of alpha-amylase. It seems reasonable, based on this, to predict that removal of methionine at that site by either substitution or deletion will alter the stability of alpha-amylase. As a result, the applicant is entitled to claim both deletions and substitutions at that position as the opponent has failed to establish that the applicant's prediction is unsound.
With regard to the W138 mutants, the two examples shown with substitutions at W138 have improved oxidative stability over the wild type. This suggests that the W138 site is also important for oxidative stability and the applicant can make a reasonable prediction based on this. For the same reasons as the M197 mutants, the applicant is therefore entitled to claim all substitutions or deletions at W138 until the opponent can prove the prediction is unsound.
The opponent argued that the only stability data provided for W138 mutants was in relation to double mutants with M197T rather than single mutants and that, as a consequence, the invention should be restricted to double mutants involving W138. However, in my view, the applicant has demonstrated that the W138 site is important for oxidative stability. There is no reason why this importance is less so because it was identified in an analysis of a double mutant. The applicant is therefore entitled to claim the single mutants unless the opponent can demonstrate the applicant's prediction is unsound. In any case, the opponent’s argument appears to be one of inutility (that single W138 mutants will not work) which is not a ground of opposition.
With regard to the M15 mutants, the specification conceded that M15L was still inactivated by H2O2 and no other oxidative stability data was provided. As a result, the applicant has not demonstrated that M15 is important for oxidative stability. However, the M15 mutants all varied from the wild type with regard to starch liquefaction activity. M15L, M15T, M15N, M15D all appear to have greater jet performance results compared to wild type while M15S and M15V had less (with M15V having only around 15% of wild type).
The M15 mutants therefore show a common beneficial property of having a variation in starch liquefaction activity from the wild-type and, consequently, there is reasonable support for the general principle that the M15 site is important for starch liquefaction activity. Further, there is sufficient teaching to enable the skilled worker to generate the claimed range of useful M15 mutants. Therefore, for the same reasons as M197 and W138, the applicant is entitled to predict that substitutions at M15 will affect starch liquefaction and (until the opponent can establish the prediction is unsound) can claim any substitution or deletion at that position.
In conclusion, the invention does not lie in the choice of any methionine or tryptophan residues as the applicant has argued but in the identification of the particular methionine or tryptophan residues which the applicant has shown to affect stability (ie: M15 and M197 and W138). Therefore, claims 1, 2, 9, 12-21, 25, 27- 29, 31, 32 which encompass sequences containing substitutions other than these are not fairly based. The applicant has demonstrated that the M15 and M197 and W138 sites are important for stability and/or activity and so, is entitled to predict all substitutions/deletions at those positions until an opponent can establish the prediction is unsound. The opponent did not discharge this onus and hence, has therefore not established that any of the other claims lack fair basis.
Other Fair Basis Issues
The opponent made two further arguments in relation to fair basis.
The opponent noted that the claims cover all alpha-amylases derived from Bacillus although only Bacillus licheniformis alpha-amylase was exemplified. Professor Rogers, in evidence in reply, argued that without any detailed knowledge of the relationship between homology and 3D structure between Bacillus alpha-amylases generally, there was no basis at the priority date or the filing date of the opposed application for extending coverage of the invention to Bacillus alpha-amylases other than Bacillus licheniformis. Professor Rogers also noted that Scopes, in evidence in answer at paragraph 29.4 acknowledged there is very little homology amongst the known Bacillus alpha-amylases and that, in any case the regions of high homology between the Bacillus alpha-amylases as marked in figure 3 do not correspond to the M15, M197 or W138 positions.
In my view, Professor Scopes comments in paragraph 29.4 are ambiguous. It is not clear to me whether he is acknowledging that there is little homology among the known Bacillus alpha-amylases, as the opponent suggested, or whether he saying that there is little homology between the known Bacillus alpha-amylases and known alpha-amylases from other genera. I am therefore not willing to accept his concession at face value. Quantitative evidence in the statutory declarations comparing Bacillus alpha-amylase sequences show an 81% homology between B. licheniformis and B. amyloliquefaciens alpha-amylase and a 66% homology between B. stearothermophilis and B. licheniformis alpha-amylase (see Rogers, evidence in reply, paragraph 12.1). These reasonably high levels of identity are consistent with the principle that there is a related family of enzymes within the Bacillus genus.
This relationship is confirmed in one of the citations raised by the opponent [Gray et al J.Bacteriol. (1986) 166(2): 635-643 (JFM-58)] which looked at the structural genes encoding the thermophilic alpha-amylases in Bacillus species. However, while the citation indicated that the alpha-amylases from B. licheniformis, B. amyloliquefaciens, B coagulans and B. stearothermophilis were related and form part of the one enzyme family, this relationship did not extend to the B. subtilis alpha-amylase (see the paragraph 1 under "Discussion", page 642).
In my view, it is reasonable for the applicant to predict that similar substitutions in different Bacillus alpha-amylases might have the same effect as observed in B. licheniformis, provided the enzymes were related. Professor Rogers argued that it is only when sequence identity exceeds 70% that approaches using comparative model predictions of the 3D structures can be considered successful (evidence in reply paragraph 12.4). He further argued that the level of sequence homology between B. stearothermophilis and B. licheniformis is in a region of uncertainty and must raise questions therefore about the extension of the claim to B. stearothermophilis.
As noted above, the onus is on the opponent to prove the applicant’s prediction is unsound as per Olin Mathieson Chemical Corporation v Biorex Laboratories (supra) In this case, even if Professor Rogers is later proven right with regard to B. stearothermophilis, his arguments at this stage are mere conjecture. Although there is inherent unpredictability in the field, there is a known principle that protein sequences are often conserved between related species and the applicant is entitled to make a reasonable prediction based on that principle (see, for example, Genetics Institute, Inc v Kirin-Amgen, Inc (No. 3) supra and note page 13 of the specification). The evidence demonstrates that the B. licheniformis and B. stearothermophilis are part of the one enzyme family. It is not necessary for the applicant to provide data to show the extension of the invention to B. stearothermophilis as the opponent contends (see Rogers, evidence in reply paragraph 12.3). Rather, the opponent has to show that it does not extend that far. While the opponent argued that the homologous regions were not around the M15 or M197 sites, this does not mean that the prediction is necessarily unsound. Folding in more homologous regions could still place the +15 and +197 amino acids into key positions where substitution could affect stability.
The opponent did not provide any evidence showing that the same amino acid substitutions would not work in other related Bacillus species alpha-amylases nor did their arguments suggest that it was not possible for the same substitutions to work. In fact, much of the opponent’s argument was based on uncertainty in the exact levels of homology between the Bacillus species alpha-amylases and uncertainty as to whether the results would extend to other alpha-amylases. This is not sufficient for the opponent to discharge their onus. The benefit of doubt goes to the applicant and the opponent has not established that claims which encompass related alpha-amylases sequences from Bacillus species other than B. licheniformis lack fair basis.
With regard to the B subtilis alpha-amylase, the evidence shows that this enzyme is unrelated to the B licheniformis enzyme. Without a relationship, and given the inherent uncertainty in the field, I accept that there is no reasonable basis to extrapolate from B licheniformis to the B subtilis alpha-amylase. The same problem will occur if there are other unrelated alpha-amylases from Bacillus. As a consequence, claims which encompass the B subtilis alpha-amylase (or other Bacillus alpha-amylases unrelated to B licheniformis) go beyond the scope of the invention and are not fairly based (ie: claims 1-12, 15-17, 22-32).
The opponent also submitted that claim 18, and dependent claims 19-20, were not fairly based. These claims referred to a mutated alpha-amylase having an amino acid sequence corresponding to SEQ ID NO. 37. However, the alpha-amylase shown in SEQ ID NO 37 is the known A4 form of the alpha-amylase (having an additional 4 alanines at the N-terminus) and is not mutated at any of the methionine or tryptophan residues. The claim therefore has none of the features provided in the specification said to overcome any of the difficulties of the prior art. The applicant acknowledged that they did not seek to encompass the known A4 form of the alpha-amylase and that there could therefore be an unintended problem with claim 18 and its dependent claims. I accept that there is a problem with the claims as the opponent suggests and that these claims are not fairly based and further, do not define the invention.
Full Description
In written submissions provided at the hearing, the opponent raised two issues under full description. Firstly, they argued that there was insufficient description to support all aspects of the claimed invention. In my view, this argument really concerns whether the scope of the claims is commensurate with the description. As such, it needs to be (and has been) considered as part of fair basis rather than full description. I therefore do not propose to consider the argument further under this ground.
Secondly, they referred to the decision of Sami S. Svendsen v Independent Products Canada (1968) 119 CLR 156 at 165 which held that a sufficient description was one that enabled “the reader to discern what was the invention in the total thing that it [the specification] describes.” They argued that inconsistencies between the claims, the title of the specification and the specification itself regarding whether the invention concerned an improved (or merely an altered) oxidative stability resulted in there being a problem in ascertaining the nature of the invention.
I note that this argument is similar to that successfully used in the full bench Federal Court decision in Arico Trading International Pty Ltd vKimberly-Clark Australia Pty Ltd [1999] FCA 1191 (30 August 1999). However, this decision was overturned in the High Court which handed down their decision shortly after the current hearing on the opposed application [see Kimberly-Clark Australia Pty Ltd v Arico Trading International Pty Ltd [2001] HCA 8 (15 February 2001)]. The opponents arguments are made moot by the High Court decision. The High Court found that the question of whether the invention has been fully described has to be considered in the light of the complete specification as a whole and not part of it:
“it does not follow that the description is to be gleaned solely from one part (the body) and that it is forbidden to obtain any assistance by regard to the remainder (the claims) of the complete specification. Rather, the text indicates that the specification must be read as a whole and that reference to the claims may dispel ambiguity or uncertainty from the body of the specification concerning the description of the invention.”
In the current case, while there may be inconsistencies between certain parts of the specification, the nature of the invention (a Bacillus alpha-amylase mutant with altered activity and/or stability properties over the wild-type) is clear in light of the specification as a whole including the claims. Hence, I find that the opponent’s arguments have not been made out. The specification is fully described.
Novelty
I have already found that claim 18, and dependent claims 19-20 inadvertently define the known A4 form of the alpha-amylase rather than the mutant forms of the invention and therefore these claims fail to define the invention. However, in addition because of this problem, claims 18-20 also lack novelty in light of the known A4 form.
With regard to the remaining claims, the closest prior art raised by the opponent was patent application WO 91/16423 by Novo Nordisk (exhibit JFM-4) which was published on 31 October 1991. This citation broadly discloses a mechanism for improving the oxidative stability of detergent enzymes by mutating methionine residues in the enzyme into cysteine residues and then chemically modifying those cysteine residues so that they were less susceptible to chemical oxidation. The citation specifically exemplifies subtilisin but teaches that amylases, lipases, cellulases and proteases could also be protected by this means. According to the opponent, this is a clear recommendation to do something that falls within the scope of claims 1-4, 6, 11-23, 25 of the opposed application and therefore the citation deprives those claims of their novelty (as per Bristol-Myers Squibb Co v F H Faulding & Co (2000) 46 IPR 553).
However, in my view, while the prior art document clearly teaches how an methionine should be modified to improve stability, it does not provide a clear indication about which methionine residues should be targeted. The court in General Tire & Rubber Company v Firestone Tyre and Rubber Company Ltd, (1972) RPC 457 noted at pages 485, 486.
"To anticipate the patentees claim, the prior publication must contain clear and unmistakable directions to do what the patentee claims to have invented ... A signpost, however clear, upon the road to the patentee's invention will not suffice. The prior inventor must be clearly shown to have planted his flag at the precise destination before the patentee."
The list of possible target enzymes specified in the citation, including amylase, is only a guide to where a skilled worker might consider using the technique. Since all the enzymes listed have commercial applications in which oxidative stability is important, the list is not all that surprising. In fact, as acknowledged in another citation raised by the opponent [US 5,118,623 (JFM-1)- at column 2, lines 3-10], it would take an inordinate amount of experimentation to test every methionine site in every enzyme. Therefore, in absence of any specific indication of the importance of methionine residues in alpha-amylase (for example: from a 3-D structure of the enzyme, or by close homology to the exemplified subtilisin enzyme), the skilled worker would not necessarily target methionine in Bacillus alpha-amylase. Hence, the citation does not provide "clear and unmistakeable directions" to the claimed invention (as per General Tire & Rubber Co v Firestone [supra]).
As a result, all the claims (apart from claims 18-20) are novel in light of the citations provided by the opponent.
Obviousness
The opponent provided a number of citations in which methionine and/or tryptophan had been targeted by site-directed mutagensis to improve the oxidative stability of an enzyme.
The opponent contended (and the applicant conceded) that at the priority date that it was well known that:
Alpha-amylases were exposed to a wide range of conditions in commercial applications including conditions where oxidants such as bleach or peracids are present and that these enzymes were not always stable in such conditions; and
Certain amino acid residues (including methionine, tryptophan, cysteine, histidine, tyrosine and lysine) were known to be susceptible to oxidation and such residues had been targeted in other proteins in the technique of site-directed mutagenesis to attempt to improve oxidative stability of the protein;
The opponent argued that there was no inventive step based on this knowledge alone or in combination with any one of the following documents, all of which had been published prior to the earliest priority date and which the PSA would find, understand and consider relevant:
(a) US 5,118,623 (JFM-1) in relation to claims 1, 9, 10, 12-14, 16-18, 26, 28 and 29;
This citation disclosed an improved Bacillus licheniformis alkaline protease in which at least one tryptophan has been genetically substituted by tyrosine or phenylalanine to improve resistance to cleavage by hypochlorite.
(b) EP 0130756 (JFM-2) in relation to claims 1, 2-8, 9-14, 16-29;
This citation inter alia disclosed a point mutation in Bacillus subtilisin at Met-222 which resulted in improved oxidative stability (see example 17).
(c) EP 0410498 (JFM-3) in relation to claims 1, 12-14, 30-33;
This citation disclosed mutant Bacillus alpha-amylases with substitutions at Ala-111, His-133 and Thr-149 which were thermostable and acid stable.
(d) WO 91/16423 (JFM-4) in relation to claims 1, 2-4, 6, 16-18, 23, 24 and 26-29;
This citation (also cited under novelty) disclosed a general method to improve oxidative stability in detergent enzymes in which one or more methionines were mutated into cysteines which were then subsequently chemically modified. Subtilisin was exemplified but the method also suggested other enzymes could be used, including alpha-amylases.
(e) US 4,760,025 (JFM-8) in relation to claims 1, 11-13, 16-24, 26-29;
This citation disclosed Bacillus subtilisin was modified at specific sites (+32, +155, +104, +222, +166, +64, +33, +169, +189, +217, or +157) to alter stability properties.
(f) Estell et al, J. Biol. Chem. 260 (11) pp 6518-6521 (JFM-9) in relation to claims 1, 4, 7, 12, 13, 16-18, 23 and 24;
This citation used looked at Bacillus site-directed mutagenesis in subtilisin at position Met-222 to investigate effects of mutations. It reported the feasiblity of improving oxidative stability in proteins by site-directed mutagenesis.
(g) EP 0285123 (JFM-5) in relation to claims 1-4, 6, 7, 9-17, 22, 23, 25-30;
This citation disclosed a general method of introducing base substitutions randomly to any nucleic acid. It exemplified B. stearothermophilis alpha-amylase showing alterations at a number of sites but not including any methionine or tryptophan sites.
(h) Gray et al, J. Bacteriol. 166(2) pp 635-643 (JFM-58) in relation to claims 1-4, 6, 7, 9-17, 22, 23, 25-30;
This citation disclosed the DNA sequence and complete amino acid sequence of B. stearothermophilis and Bacillus licheniformis alpha-amylases.
Joyet et al, Biotechnology 10 pp 1579-1583 (JFM-7) in relation to claims 1, 9, 12-17, 25, 27-29, 31 and 32.
This citation disclosed thermostable variants of a Bacillus licheniformis alpha-amylase with substitutions at the 209 (Alanine) and 133 (Histidine) positions.
According to the opponent, faced with the problem of improving oxidative stability in an enzyme, the skilled worker would selectively target amino acid residues which were known to be susceptible to oxidation and either delete these or substitute them for non-oxidative residues in the technique of site-directed mutagenesis. The opponent argued that similar approaches had been used successfully in the subtilisin enzyme and would motivate the skilled worker to try the same approach in the alpha-amylase enzyme. Further, the primary structure (ie: the amino acid sequence) of alpha-amylase was known at the priority date of the current specification and therefore there would have been no practical difficulties in applying the technique of site-directed mutagenesis to alpha-amylase.
I note that it appears to have been known since as early as 1969 that methionine was susceptible to oxidation (see JFM-1, column 4 lines 10-15). Despite this, and although there were a number of citations raised where methionine and/or tryptophan were targeted in an enzyme (see JFM-1, JFM-2, JFM-4, JFM-8, and JFM-9), there appear to be very few examples of types of enzymes being modified at those residues. The applicant even argued that there was really only one example amongst the citations since JFM-2, JFM-4, JFM-8 and JFM-9 all relate to subtilisin and the alkaline proteases described in JFM-1 are serine proteases like the subtilisins.
In addition, in the citations where subtilisin was modified (ie JFM-2, JFM-4, JFM-8 and JFM-9), only one methionine residue was targeted (M222) rather than all methionine residues. M222 had already been well-established as being responsible for the lability of subtilisin towards oxidative agents (see JFM-4, page 2, lines 12-15). This suggests that there is no general principle to randomly target any methionine and/or tryptophan residues to improve oxidative stability in an enzyme, rather there has to be some motivation to target a particular residue in a particular enzyme. In other words, an amino acid position would need to be pre-identified as being important for stability in a given enzyme before being targeted in site-directed mutagenesis.
This is consistent with the applicant's view that without knowing the tertiary structure of an enzyme, it would not be possible to know which or whether any of the Met or Trp residues were close to the active site and choose these as the first targets for mutagenesis [Scopes, in evidence in answer (paragraph 20.13)]. This view is supported by a number of statements in the citations provided by the opponent. JFM-5 at page 2, lines 32-41, for example, indicates that unless a skilled worker has extensive knowledge of an enzyme including three dimensional structure, they would not be motivated to try site directed mutagenesis:
"A single replacement of only one functionally critical amino acid by site directed mutagenesis …..has been demonstrated to be sufficient to bring about the desired modification of protein properties in, for example, the following cases: an altered specificity of trypsin, an altered pH dependence or enhanced resistance to chemical oxidation of subtilisin and an increased thermostability of lysozyme or dihydrofolate reductase …However the above enzymes all belong to the exceptionally rare examples where extensive information is available concerning enzymology and three dimensional structure" (my emphasis)
Similarly, paragraph 1 of JFM-9 states:
"One of the primary sources of protein stability is their susceptibility to oxidation and subsequent inactivation or denaturation …..This is especially true for proteins containing methionine, cysteine or tryptophan residues in or around the active site …..It would therefore be useful to investigate the functional consequences of replacing activity critical residues which are sensitive to oxidation" (my emphasis)
And paragraph 1, column 2 of page 1579 in JFM-7 states:
"A rational strategy for increasing the stability of a specific protein by site-directed mutagenesis requires a thorough knowledge of its three-dimensional structure and of the molecular mechanisms responsible for its heat inactivation. There are still a limited number of enzymes that fulfil these requirements and examples of successful design of stabilised proteins are rare". (my emphasis)
Given this evidence, I accept the applicant's view that a skilled worker would require a tertiary structure, or some other motivation, before targeting any methionine or tryptophan residues in a given enzyme to try and improve oxidative stability.
In the current case, there was no three-dimensional structure available for alpha-amylase at the priority date. There is also no homology between subtilisin and the alpha-amylase of the current specification, and therefore it is not possible to directly extrapolate from the subtilisin results to alpha-amylase. Further, none of the citations raised by the opponent which specifically attempted to improve the properties of Bacillus alpha-amylases by mutagenisis, actually targeted methionine or tryptophan residues - JFM-5 used random mutagenesis to investigate mutations at a number of positions but none of these were methionine or tryptophan; JFM-3 targeted Ala-111, His-133 and Thr-149; and JFM-7 targeted Ala-209 and His-133. As a result, there would be no motivation therefore to target methionines or tryptophans in alpha-amylase. Although JFM-4 did list alpha-amylase as one of a number of enzymes where the skilled worker might consider using protecting methionine residues, this is no more than a guide of where the technique might be useful provided a key methionine residue existed near the enzyme's active site.
Given the lack of knowledge about the location and importance of the methionine and tryptophan residues or any other reason to target methionine and/or tryptophan in the Bacillus alpha-amylase, there was no reasonable likelihood of success to motivate a skilled worker to try to alter any of the methionine or tryptophan residues in alpha-amylase. In light of this, the opponent's argument that it was obvious to alter any and all methionine and tryptophan residues in any and all enzymes, including alpha-amylase, based on a particular methionine residue in a particular enzyme, appears to be ex post facto analysis. The opponent has therefore not established that any of the claims lack an inventive step based on the common general knowledge alone, or in combination with any one of the above documents.
Manner of manufacture
The law on manner of manufacture has been examined in a number of recent decisions by Australian courts: the High Court in NV PhilipsGloelampenfabriken v Mirabella International Pty Ltd (1995) 183 CLR 655 and Advanced Building Systems Pty Ltd v Ramset Fasteners (Aust) Pty Ltd (1998) 194 CLR 171, the Federal Court in Bristol-Myers Squibb Co v FH Faulding & Co Ltd (2000) 46 IPR 553.
In the Bristol-Myers case, the majority summarised the effect of the Philips case as:
"Philips stands for the proposition (as a matter of construction of the 1990 Act) that if, on the basis of what was known, as revealed on the face of the specification, the invention claimed was obvious or did not involve an inventive step - that is, would be obvious to the hypothetical non-inventive and unimaginative skilled worker in the field (Minnesota at CLR 260 per Barwick CJ) - then the threshold requirement of inventiveness is not met."
[page 564]
However, as noted in the office decision of Novo Nordisk v Eli Lilly and Company [supra], what is known is not everything that is stated in the specification:
"Some elaboration, however, is required in relation to what the specification reveals as 'known'. If a patent application, lodged in Australia, refers to information derived from a number of prior publications referred to in the specification or, generally, to matters which are known, in our view the court - or the Commissioner - would ordinarily proceed upon the basis that the knowledge thus described is, in the language of s 7(2) of the 1990 Act, part of 'the common general knowledge as it existed in the patent area'."
[Bristol-Myers at page 564]
The assessment is based on the common general knowledge as it is presented on the face of the specification.
In this case, the common general knowledge is the same as that alleged by the opponent under inventive step and is not contested. The admitted prior art of the opposed specification includes the citations raised by the opponent under inventive step. The arguments raised by the opponent under the ground of manner of manufacture are therefore the same in substance as those already considered under inventive step. I have already found that the claims have an inventive step based on the prior art cited by the opponent, and for the same reasons, there is an invention on the face of the specification. As a result, all the claims define a manner of manufacture.
CONCLUSION
Claim 18, and dependent claims 19-20 do not define the invention and are not novel because they define the known A4 form of the alpha-amylase rather than the mutated form of the invention.
Apart from claims 18-20, none of the other claims have been found to lack novelty or an inventive step. However, there is a serious fair basis problem in the specification.The invention does not lie in the choice of any methionine or tryptophan residues as the applicant has argued but in the identification of the particular methionine or tryptophan residues which the applicant has shown to affect stability (ie: M15 and M197 and W138).Claims 1, 2, 9, 12-21, 25, 27- 29, 31, 32 which encompass sequences containing substitutions other than these are not fairly based.
Further, the evidence shows that the B subtilis alpha-amylase is unrelated to the B licheniformis enzyme. Without homology between the enzymes, and given the inherent uncertainty in the field, there is no basis to extrapolate from B licheniformis to the B subtilis alpha-amylase. The same problem will occur if there are other unrelated alpha-amylases from Bacillus. As a consequence, claims which encompass the B subtilis alpha-amylase (or other unrelated Bacillus alpha-amylases) go beyond the scope of the invention and are not fairly based (ie: claims 1-12, 15-17, 22-32).
I believe that there is patentable subject matter in the specification and that the deficiencies noted above can be resolved by amendment. I therefore allow the patent applicant 60 days from the date of this decision in which to file appropriate amendments.
COSTS
Costs normally follow the event. In this case, I have found that the opponent has been partly successful in respect of a major problem with fair basis of the claims. I therefore award costs against the applicant.
Karen Ayers
Delegate of the Commissioner of Patents
Patent attorneys for the applicant : Phillips Ormonde & Fitzpatrick, Melbourne
Patent attorneys for the opponent : Spruson & Ferguson, Sydney
0
7
0