Nufarm Australia Ltd v Dow AgroSciences LLC
[2014] APO 33
•27 May 2014
IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Nufarm Australia Ltd v Dow AgroSciences LLC [2014] APO 33
Patent Application: 2005240045
Title:Novel Herbicide Resistance Genes
Patent Applicant: Dow AgroSciences LLC
Opponent: Nufarm Australia Ltd
Delegate: Karen Ayers
Decision Date: 27 May 2014
Hearing Date: 5 March 2014
Catchwords: PATENTS - opposition to the grant of the patent under s 59 – opposed on the basis of novelty and inventive step – prior art did not disclose plant-optimised gene sequence for RdpA or a transgenic plant containing said sequence – obviousness argument is one of ‘obvious to try’ – with only a low level of activity against 2,4-D, there was no expectation that RdpA would confer multiple herbicide resistance in transgenic plant – specification is also directed to a manner of manufacture, is useful and meets section 40 requirements.
Representation: Patent applicant: Craig Smith of counsel instructed by Tom Gumley and Sarah Hennebry of Freehills (Melb)
Opponent: Helen Rofe and Clare Cunliffe instructed by Tony Davis and Amanda Stark of Griffith Hack (Melb). Matt Gallagher from the opponent company also attended.
IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Patent Application: 2005240045
Title:Novel Herbicide Resistance Genes
Patent Applicant: Dow AgroSciences LLC
Date of Decision: 27 May 2014
DECISION
The opposition is unsuccessful. Absent an appeal being filed within 60 days of this decision, I direct the application proceed to grant.
Costs awarded according to Schedule 8 against the opponent Nufarm Australia Ltd.
REASONS FOR DECISION
Background
Standard patent application 2005240045 was filed on 2 May 2005 under the provisions of the PCT in the name of Dow AgroSciences LLC. The application claimed priority from a US basic application (60/567,052) which had been filed on 30 April 2004. The Australian standard application was advertised accepted on 28 April 2011 and a notice of opposition was filed on 28 July 2011 by Nufarm Australia Ltd. Evidence was finalised on 20 March 2013 and the matter set for hearing in Melbourne on 5 March 2014.
Onus of Proof
The request for examination in relation to this patent application was filed on 27 March 2009. As a consequence, substantive amendments of the Patents Act brought about by the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 do not apply to the present application. This includes the amendment to subsection 60 (3A) that allows the Commissioner to refuse a patent application if satisfied on the balance of probabilities that a ground of opposition exists.
Consequently the former standard for opposition proceedings applies and the opponent must establish that it is clear or practically certain that the patent is invalid (F Hoffman La Roche AG v New England Biolabs Inc [2000] FCA 283 at [29], [67]; 50 IPR 305 at 311, 319; Commissioner of Patents v Sherman [2008] FCAFC 182 at [18], [22]; 79 IPR 426; Genetics Institute Inc v Kirin-Amgen Inc [1999] 92 FCR 106 at [17]).
Evidence
The following evidence was filed in this matter:
(a)Evidence in support
Statutory Declarations by:
·Amanda Stark dated 30 April 2012 together with Exhibits AJS-1 to AJS-26
·Dr Trevor Stevenson dated 30 April 2012 together with Exhibits TS-1 to TS-7 (Stevenson#1)
(b)Evidence in answer
Statutory Declaration by:
·Dr Gerhard Schenk dated 17 December 2012 together with Exhibits GS-1 to GS-2
(c)Evidence in reply
Statutory Declaration by:
·Dr Trevor Stevenson dated 19 June 2013 (Stevenson#2)
Specification
Herbicides (such as glyphosate) are a well-known means of controlling weeds in crops but have limited applicability if the crop plant is also sensitive to same herbicide. Glyphosate tolerant crops (GTC) were successfully developed in the mid-1990s by genetically modifying plants with glyphosate resistance genes[1]. Producers were quick to adopt these genetic engineered crops but this has led to an over-reliance on glyphosate (and GTCs) and the natural selection of glyphosate-tolerant weeds. As a consequence, there was a recognised need to develop other types of herbicide tolerant crops (HTCs).
[1] Opposed specification at [0004]
One of the most common herbicides for controlling broadleaf weeds is 2,4-D[2] which is part of a class of herbicides known as phenoxyauxins. This class of herbicides has a aryloxyalkanoate chemical sub-structure which it shares with a range of other classes of herbicides including the pyridyloxy auxins (such as fluroxypyr and triclopyr), aryloxyphenoxypropionates (AOPP), acetyl-coenzyme A carboxylase (ACCase) inhibitors (such as haloxyfop, quixalofop and diclofop) and 5-substituted phenoxyacetate protoporphyrinogen oxidase inhibitors (such as pyraflufan and flumiclorac)[3]. However, despite having a common sub-structure, these classes of herbicides are considered quite distinct and no evidence exists in the current literature for a common degradation pathway[4].
[2] 2,4-dichlorophenoxyacetic acid
[3] See appendix A for comparison of structures from a number of members of the phenoxy family of herbicides
[4] Opposed specification at [0015]
There has been extensive research into microorganisms which degrade 2,4-D and in particular the gram-negative bacterium Ralsotina eutropha. The gene that codes for the first enzymatic step in the mineralisation pathway in this microbe is known as TfdA[5]. TfdA catalyses the conversion of 2,4-D acid to dichlorophenol (DCP) via an α-ketoglutarate-dependent dioxygenase reaction. TfdA has also been used in transgenic plants to impart 2,4-D resistance in dicot plants (eg cotton and tobacco) normally sensitive to 2,4-D[6].
[5] 2,4-dichlorophenoxyacetate/α-ketoglutarate-dioxygenase
[6] Opposed application at [0011]
The specification notes that a large number of TfdA-type genes have now been identified which have similar homologies (>85% amino acid identity) and hence similar enzymatic properties to TfdA. Such enzymes primarily degrade achiral phenoxyacetic acids (eg: 2,4-D)[7]. However there were also a number of enzymes containing the characteristic residues associated with α- ketoglutarate dioxygenase Fe2+ dioxygenases but with a significantly lower homology to TfdA (25-50%). It was not known what the substrate specificities of these divergent dioxygenases were[8].
[7] Opposed specification at [0013]
[8] Opposed specification at [0012]
One example of this is RdpA[9] from Sphingobium herbiciovorans which catalyses the first step in (R)-dichlorprop (and other (R)-phenoxypropionic acids) and 2,4-D (a phenoxyacetic acid) mineralisation (Westendorf et al, 2003)[10]. The RdpA sequence was identified from the NCBI data base (accession number AF516752) as having 28% homology with TfdA. Using primers developed from this sequence, the applicant isolated the RdpA gene from a known strain of Sphingobium herbiciovorans (ATCC #700291)[11]. The current invention uses this gene to transform plants making them resistant to both phenoxyherbicides (such as 2,4-D) and AOPP herbicides (such as dichloroprop, mecoprop, fluazifop, fenoxaprop, metamifop, cyhalofor and clodinofop[12]). The specification noted that prior to this invention there had been no expectation of success that a plant with resistance to both herbicides could be produced by the introduction of one gene.[13]
[9] (R)-phenoxypropionate/α-ketoglutarate-dioxygenase
[10] Opposed specification at [0013] and [0093]
[11] Opposed specification at [00172]-[00173]
[12] Opponent’s submissions at [45(l)-(m)] (page 10)
[13] Opposed specification at [0016]
10. The specification discloses the native amino acid (protein) sequence for RdpA which the specification now refers to as AAD-1[14] (SEQ ID NO:9). It also list three specific nucleotide sequences which encode this enzyme:
· SEQ ID NO:3 - the (native) nucleotide sequence [AAD-1 (v1)];
· SEQ ID NO:4 - the native AAD-1 gene with the internal NotI restriction site removed [AAD-1(v2)]; and
· SEQ ID NO:5 - the ‘plant-optimised’ DNA sequence for AAD-1 ([AAD-1(v3)] ].[15]
[14] Aryloxyalkanoate Dioxygenase
[15] Opposed specification at [0045] – [0050]
11. The specification ends with 68 claims only three of which (claims 1, 44 and 47) are independent. I have outlined claim 44 below in my discussion on fair basis. The other independent claims (claims 1 and 47) read as follows:
Claim 1: A transgenic plant cell comprising a polynucleotide that encodes an AAD-1 protein comprising the amino acid sequence of SEQ ID NO:9 or a variant of SEQ ID NO:9 wherein said variant has aryloxyalkanoate dioxygenase activity, at least one amino acid deletion of conserved substitution and at least 95% sequence identity with SEQ ID NO:9.
Claim 47: An isolated polynucleotide that encodes a protein that enzymatically degrades a herbicide selected from the group consisting of a phenoxy auxin and an aryloxyphenoxypropionate, wherein a nucleic acid molecule that encodes said protein hybridizes under stringent conditions with the full complement of a sequence selected from the group consisting of SEQ ID NO:3, SEQ ID NO:4 and SEQ ID NO:5, wherein the polynucleotide is operably linked to a promoter that is functional in a plant cell, wherein the stringent hybridization conditions comprise 1xSSPE and 42oC.
Novelty
Relevant Law
12. The question of novelty is determined under the provisions of subsection 7(1) of the Patents Act (Cth). The basic test for novelty is the "reverse infringement test" as set out in Meyers Taylor Pty Ltd v Vicarr Industries [1977] HCA 19; (1977) 137 CLR 228 at page 235 where Aickin J stated:
"The basic test for anticipation or want of novelty is the same as that for infringement and generally one can properly ask oneself whether the alleged invention would if the patent were valid, constitute an infringement."
13. Infringement is said to occur where "each and every one of the essential features of that claim have been taken" (Rodi and Wienenberger AG v. Henry Showell Ltd (1969) RPC 367).
Relevant Prior art
14. As both parties acknowledge, the closest prior art to the current claims is based on work by two research groups (the Kohler and the Babel groups). The Kohler work is summarised in the following citations:
D4: H. Kohler, “Sphingobium herbicidovorans MH: a versatile phenoxyalkanoic acid herbicide degrader” J. Industrial Microbiology & Biotechnology (1999) 23: 336-340 (exhibit AJS-4);
D16: K. Nickel, M.Suter and H. Kohler, “Involvement of two α-ketoglutarate-dependent dioxygenases in enantioselective Degradation of (R)- and (S)- Mecoprop by Sphingobium herbiciovorans MH J. Bacteriol. (1997) 179 (21): 6674-6679 (exhibit AJS-16);
D17: C. Zipper, M. Bunk, A. Zehnder and H. Kohler, “Enantioselective Uptake and Degradation of the chiral herbicide Dichloprop ((RS)-2-(2,4-Dichlorophenoxy)propanoic acid] by Sphingobium herbiciovorans MH J. Bacteriol. (1998) 180 (13): 3368-3374 (exhibit AJS-17).
15. The Kohler research found that the isolated S.herbicidovorans strain MH degrades phenoxylalkanoic acid herbicides such as mecoprop, dichlorprop, 2,4-D, MCPA and MCPB[16] and that the bacteria was capable of degrading both enantiomers of the phenoxyherbicide auxin mecoprop (R-mecoprop and S-mecoprop) and dichlorprop in an enantiomer specific manner[17].
[16] Opponent’s submissions at page 12 (dd)
[17] Opponent’s submissions at page 13 (b)
16. D4 was a short review which discussed the properties of the recently isolated microorganism Sphingobium herbiciovorans MH (strain MH) from a dichlorprop-degrading soil column. The bacterium was able to grow on multiple phenoxyalkanoic herbicides which the specification suggested made it a potential candidate for bioremediation and waste treatment[18]. When strain MH was offered racemic dichlorprop as a substrate, it degraded both enantiomers but with a preference for the S-enantiomer.
[18] Bioremediation is a specific waste treatment method which involves using organisms to remove or neutralize pollutants from contaminated sites (such as landfills or groundwater)
17. D16 found that cell extracts of Sphingobium herbiciovorans MH grown on (R)-mecoprop contained an enzyme activity that selectively converted (R)-mecoprop to 4-chloro-2-methylphenol whereas extracts of cells grown on (S)-mecoprop contained an enzyme activity selective for the S-enantiomer. Both reactions were dependent on α-ketoglutarate and ferrous ions. The results demonstrated that the conversion was initiated by two specific α-ketoglutarate-dependent dioxygenase activites (later known as RdpA and SdpA) which were distinct from TfdA.
18. D17 also investigated enzyme activities in crude cell extracts of Sphingobium herbiciovorans MH. The study found that the bacterium was able to completely degrade both enantiomers of the chiral herbicide dichlorprop. Their results also provide evidence that the first step in the degradation of dichlorprop, mecoprop and 2,4-D by Sphingobium herbiciovorans MH is active transport and that there were 3 separate active transport systems for the different herbicides (one for the (R)-enantiomer of dichlorprop and mecoprop, a second for the (S)-enantiomer of dichlorprop and mecoprop and a third for 2,4-D).
19. All of the Kohler work was performed on crude cell extracts. D4 did not identify or isolate any specific proteins or genes associated with the activity and while both D16 and D17 concluded that there were 2 separate enzyme activities associated with the enantioselective degradation of (RS) – mecoprop, they did not isolate or characterise the enzymes associated with that activity.
20. As a consequence, none of the Kohler citations disclose a DNA sequence encoding an RdpA (AAD-1) gene from Sphingobium herbiciovorans MH. There was also no suggestion to transform a plant containing this gene or to obtain an isolated AAD-1 sequence adapted for plant expression. Hence all of the claims are novel based on these three documents.
21. The Babel work is summarised in the following citations:
D6: A. Westendorf, D. Benndorf, R.H. Müller and W. Babel: (2002) “The two enantiospecific dichlorprop/α-ketoglutarate-dioxygenases from Delftia acidovorans MC1 – protein and sequence data of RdpA and SdpA” Microbial Res. 157: 317-322 (exhibit AJS-6);
D15: A. Westendorf, R.H. Müller and W. Babel: (2003) “Purification and characterisation of the Enantiospecific Dioxygenases from Delftia acidovorans MC1 Initiating the Degradation of Phenoxypropionate and Phenoxyacetate Herbicides” Acta Biotechnol. 23 (2003) 1, 3-17 (exhibit AJS-15);
D20: EMBL sequence listing #AF516752 containing the DNA and amino sequence for the Sphingobium herbiciovorans MH RdpA gene. The sequence was created on 6 August 2002 but replaced on 20 August 2003 and further updated on 15 April 2005 (this last date was post-priority date) (exhibit AJS-20)
22. D6 and D15 investigated two (separate) α-ketoglutarate dependent dioxygenases from Delftia acidovorans MC1 (RdpA and SdpA) carrying enantiospecific activity for the etherolytic cleavage of racemic phenoxypropionate herbicides (RS)-2-(2,4-dichlorophenoxy)propionate (dichlorprop) and (RS)-2-(4-chloro-2-methylphenoxy)propionate (mecoprop).
23. D6 resolved the two enzymes (RdpA and SdpA) on electrophoretic gels and compared their partial amino acid sequences with known sequences from commercial DNA databases. N-terminal sequence data from SdpA revealed its similarity with other 2,4-dichlorophenoxyacetate/α-ketoglutarate-dioxygenase. The N-terminal sequence data from RdpA showed no significant homology to any known protein. However internal sequence data showed homology to TfdA (32-37%) followed by TauD (21-30%).
24. D15 purified RdpA and SdpA from Delftia acidovorans MC1 and compared the activities of the purified enzymes against a range of substrates including the herbicides (R)-2,4-DP[19], (S)-2,4-DP, (R)-2,4-MCPP[20], (S)-2,4-MCPP and 2,4-D.
[19] (R)-4-(2,4-Dichlorophenoxy)propionate
[20] (R)-(2-(4-Chloro-2-methylphenoxy)propionate
25. Neither D6 nor D15 disclose the DNA sequence encoding an RdpA (AAD-1) gene from either the Delftia acidovorans MC1 or Sphingobium herbiciovorans MH strains nor can this be directly derived from the partial amino acid sequence data presented in D6. In addition, neither citation suggests that this gene should be adapted for the plant environment and used to transform a plant. As a result, all the claims are novel in light of D6 and D15.
26. The last of the Babel citations (D20) contains the DNA sequence listing for the native RdpA from Sphingobium herbiciovorans MH. I note that this was the same sequence listing (AF516752) used in the opposed specification to generate primers for the AAD-1 gene. The citation contains no suggestion that this should be adapted or used in the plant environment to confer herbicide resistance. Hence D20 can be distinguished from current claims 44-49 (which are all define RdpA sequences which have been modified for expression in plants) and claims 1- 43, 50-68 (which define (or use) genetically engineered plants containing the RdpA gene). As a consequence, D20 does not deprive any of the claims of their novelty.
Inventive Step
Relevant Law
27. The question of inventive step is determined under the provisions of subsections 7(2) and 7(3) of the Patents Act (Cth). For standard applications with an examination request filed before 15 April 2013, a claimed invention will lack an inventive step if it is obvious in the light of:
(a)common general knowledge; or
(b)common general knowledge considered together with information in a single document or through doing a single act, provided that the document or act could reasonably be expected to have been ascertained, understood and regarded as relevant to work in the relevant art in the patent area by the person skilled in the art.
28. The general principles regarding inventive step were recently discussed in the Ranbaxy Laboratories Ltd v AstraZeneca AB [2013] FCA 368 (23 April 2013) ([203]-[218]). In that decision, Justice Middleton cited Jagot J’s comments in Apotex Pty Ltd v AstraZeneca AB (No 4) [2013] FCA 162 with approval:
“in Danisco at [326] one principle which Bennett J identified as orthodox having regard to the reasons “as enunciated by the High Court in Aktiebolaget, Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) [2007] HCA 21; (2007) 235 CLR 173 [(Lockwood v Doric (No 2)] and Wellcome and by the Full Court in Lundbeck [H Lundbeck A/S v Alphapharm Pty Ltd (2009) 177 FCR 151 ; [2009] FCAFC 70] and Apotex”, and which was not disturbed in Novozymes A/S v Danisco A/S [2013] FCAFC 6, is that:
In assessing obviousness, it is necessary first to determine the nature of the claimed invention and the inventive step described in the Patent. This may involve ascertaining the “starting point” of the inventive step, sometimes described in terms of an existing problem for which the inventor found a solution. The obviousness of the invention as claimed is then assessed by reference to common general knowledge in Australia at the priority date.
29.Justice Middleton then concluded that a key element of the test for inventive step is that the required comparison is between the relevant prior art base and the “invention, so far as claimed in any claim” (rather than each of the individual integers thereof). This was emphasised in Aktiebolaget Hässle v Alphapharm Pty Limited [2002] HCA 59; (2002) 212 CLR 411 at 441 [72] in the context of discussing the relevant section of the Patents Act 1952 (Cth) (this passage was cited in Eli Lilly [2013] FCA 214 at [460]):
[The relevant provision does not] direct an inquiry respecting each integer of the claimed combination. The paragraph asks whether “the invention ... as claimed”, here the combination, was obvious, not each of its integers.
30.As noted by the High Court in Aktiebolaget Hässle [2002] HCA 59; (2002) 212 CLR 411, referring to Diplock LJ’s comments in Technograph Printed Circuits Ltd v Mills & Rockley (Electronics) Ltd [1972] RPC 346 at 362 (at 423 to 424 [21]):
Once an invention has been made it is generally possible to postulate a combination of steps by which the inventor might have arrived at the invention that he claims in his specification if he started from something that was already known. But it is only because the invention has been made and has proved successful that it is possible to postulate from what starting point and by what particular combination of steps the inventor could have arrived at his invention. It may be that taken in isolation none of the steps which it is now possible to postulate, if taken in isolation, appears to call for any inventive ingenuity. It is improbable that this reconstruction a posteriori represents the mental process by which the inventor in fact arrived at his invention, but, even if it were, inventive ingenuity lay in perceiving that the final result which it was the object of the inventor to achieve was attainable from the particular starting point and in his selection of the particular combination of steps which would lead to that result.
Problem to be Solved
31.The problem in the art to be solved is the improvement of herbicide resistance in crop plants. The skilled worker is a plant biotechnologist.
Relevant Prior Art
32.For inventive step, the opponent relied on the Kohler and Babel citations discussed above under novelty. These citations relate to bacterial enzymes capable of degrading a major class of herbicide. The skilled person in trying to develop herbicide tolerant crops HTC would be very interested in bacterial genes capable of degrading herbicides. The citations are in well-respected, peer reviewed journals and the opposed specification acknowledged the work as admitted prior art[21]. Dr Stevensen specifically noted in his evidence that he kept up to date with scientific papers and conferences and that he was aware of the work of the Kohler group (in particular) at the priority date[22]. I therefore accept that the skilled worker would have found and regarded these documents relevant in light of the stated problem and that these documents are therefore all part of the prior art data base for inventive step under section 7(3).
33.The opponent also argued that the documents were also part of the CGK[23] and therefore discussed the citations as a combined disclosure. While I am not convinced this is necessarily the case, the citations are similar and both parties argued their respective cases based on the combined disclosure. The outcome would also be the same regardless of whether (or not) the citations were part of the CGK. Therefore, for convenience, I will discuss the prior art as a combination.
[21] Opposed specification at [0093], [0094] and [0013] for example
[22] Stevensen#1 at [2.14]
[23] Opponent’s submissions at [43] et seq
Inventive step based on the Prior Art
34. As the specification admits, TfdA (an α-ketoglutarate dependent dioxygenase) was known to degrade 2,4-D and had previously been used to impart 2,4-D resistance in transgenic dicot plants[24]. TfdA was part of a ‘superfamily’ of related enzymes which is responsible for a wide variety of reactions (eg: protein side-chain modifications, biosynthesis of antibiotics and plant products and metabolism of lipids)[25]. Some of these enzymes showed significant homology to TfdA (>85%) and had similar enzymatic properties. However there were also less related enzymes in the same family (25-50% homology) whose function and substrate specificities could not be predicted from TfdA[26].
[24] Opposed specification at [0011]
[25] Applicant’s submissions at [51]
[26] Opposed specification at [0012]
35. One of these enzymes is RdpA (AAD-1) which has 28% homology with TfdA[27]. It was known that RdpA was a α-ketoglutarate dependent dioxygenase which targeted (R)-enatiomers of phenoxypropionate herbicides. However it had not been recognised that this enzyme could also degrade (achiral) 2,4-D. The specification noted that although aryloxyalkanoate side chains are a common entity of many commercialised herbicides (including 2,4-D and AOPP), the classes of herbicides are quite distinct and there was no evidence for a common degradation pathway[28]. According to the applicant, a single enzyme imparting multiple herbicide resistance was against the expectation of the art and conferred a surprising (inventive) advantage over other possible options.
[27] Opposed specification at [0013]
[28] Opposed specification at [0015]
36. The opponent’s expert (Dr Stevenson) suggested that RdpA and SdpA were the only α-ketoglutarate dependent dioxygenases (apart from TfdA) known to degrade 2,4-D and related herbicides and which also possessed the “well-defined sequence motifs present in TfdA or other α-ketoglutarate dependent dioxygenases” [29]. His view was that the next “obvious and straight forward step” for a skilled worker would have been to transfer and express each of these genes into a plant to determine what level of herbicide resistance could be imparted. He then suggested that because RdpA was the only gene sequence available from a commercialised database, this would have been the obvious (indeed the only) choice[30].
[29] Stevenson#2 at [3.3]
[30] Stevenson#2 at [4.2]
37. Dr Stevenson’s argument presupposes that the 2,4-D degrading properties of RdpA were known at the priority date. However the key citations (D15 and D16) show RdpA to have either little or no activity against 2,4-D[31]. Such low activities seem unlikely to have motivated the skilled worker to use the gene in transgenic plants to confer resistance to 2,4-D. Given this and the lack of overall similarity with TfdA (28%), I am not convinced that a skilled worker at the priority date would have recognised a ‘well-defined sequence motif’ as indicating functional similarity between RdpA and TfdA. I therefore agree with the applicant (and the specification)[32] that without clear guidance elsewhere, the skilled worker would more likely focus their attention on genes with more homology to TfdA.
[31] D15, page 14, last line to page 15, first line and D16, page 6677 (table 2) and last paragraph on page 6678, column 1
[32] Opposed sspecification at [0012]-[0013], applicant’s submissions at [74]-[79]
38. In my view, the opponent has provided Dr Stevenson with a small selection of documents which discuss RdpA as an option has directed their expert’s attention towards a different (and inventive) solution in hindsight. Therefore while I accept that the skilled worker could have followed that research path suggested by Dr Stevenson ‘to see what phenotype might result’[33], I am not convinced that they would have directly been led there without the benefit of hindsight.
[33] Stevensone#2 at [3.5]
39. None of the citations progress the opponent’s argument any further. Most of the research about RdpA’s enzymatic activity was performed by Kohler using crude cellular extracts (D4, D16 and D17). While the skilled worker would generally understand from these documents that the microbe is able to degrade multiple herbicides, they would not know whether this activity was associated with one or more enzymes. As the applicant noted, with seven different herbicides being tested, there could have been seven different enzymes involved[34].
[34] Applicant’s submissions at [83] and [90]
40. I accept that D16 implies that the multiple herbicide degrading activities are associated with two specific enantiospecific α-ketoglutarate dependent dioxygenases (now known as RdpA and SdpA)[35]. However the citation shows RdpA to have a low activity against 2,4-D (11-15%), MCPA[36] (10-39%) and MCPB[37] (4-5%)[38]. It seems unlikely that the skilled worker would have considered this activity sufficient to confer resistance in a transgenic plant particularly given that TfdA (which does not confer multiple resistance) appears to have a higher activity against some of these substrates[39].
[35] See the sentence on D16, page 6677, column 2, paragraph 1 (for example) which discusses ‘substrate specificities’ for different enzymes including TfdA
[36] 4-Chloro-2-methylphenoxy)acetic acid
[37] 4-(4-Chloro-2-methylphenoxy)butanoic acid
[38] D16, page 6677 (table 2) and last paragraph in discussion on page 6678, column 1
[39] D16, page 6677 (table 2)-TfdA had 73% activity with MCPA (73%) and 23% with (S)-dichlorprop
41. The Babel work (D6, D15 and D20) had recognised that certain strains of bacteria (Sphingobium herbiciovorans MH, Delftia acidovorans MC1, and Rhodoferax) were able to degrade racemic phenoxypropionates as well as utilise a broader spectrum of herbicides[40]. The aim of their research was to elucidate the enzymatic basis of the broad herbicide consumption profile. The most relevant citation is D15 which reported properties of the purified RdpA enzyme. However, while it noted that RdpA had the capacity to cleave the ether bond of a number of substrates, it clearly suggests that RdpA does not cleave 2,4-D[41]. Based on this teaching, the skilled worker would be led away from using RdpA to impart multiple herbicide resistance in plants.
[40] D5, page 4, paragraph 1
[41] D15, page 10 line 15, page 15, line 1
42. Given the low levels of activity against certain herbicides and the general expectation in the art at that time that most enzymes do not degrade multiple herbicides[42], my view is that it had not been recognised at the priority date that RdpA could confer multiple herbicide resistance in plants. As a consequence, there was no clear motivation to transform a plant using the RdpA gene from either the CGK or any of the citations raised by the opponent. I therefore find that all the claims are inventive in light of the citations provided by the opponent.
[42] Opponent’s submissions at 45(n) (page 10)
Manner of Manufacture
43. The opponent argued that the specification was not a manner of manufacture because there was no invention on the face of the specification. According to the opponent, the specification had conceded that the enzyme RdpA was known to degrade 2,4-D and dichlorprop[43]. They considered that the use of this enzyme to impart this resistance in a transgenic plant was nothing more than the use of a known material for a purpose for which its properties made it suitable[44].
[43] Opponent’s submissions at [54]
[44] Commissioner of Patents v Microcell Ltd (1959) 102 CLR 232
44. The opponent relied on 2 specific passages in the specification (paragraphs [0013] and [0094]) to support their argument. These paragraphs both referred some of the citations used in the opponent’s inventive step argument (D4, D6 and D15). The manner of manufacture argument is therefore effectively a restatement of the opponent’s inventive step case. While the admitted prior art may have speculated that RdpA had some 2,4-D degrading activity, the low levels of activity seem unlikely to have motivated the skilled worker to use the gene in transgenic plants to confer resistance to 2,4-D.
45. This is consistent with the teaching of the specification (as a whole) which highlights the unexpected nature of a single enzyme conferring multiple herbicide resistance. For example, the specification notes at [0015] that no evidence exists in the current literature for common degradation pathways of the different classes of herbicides. At [0016], it states that prior to the current invention there was no expectation or suggestion that resistance to 2,4-D and AOPP herbicides could be conferred by the one gene and a plant the specification. Finally, at [0018] it concludes that ‘no α-ketoglutarate-dependent dioxygenase enzyme has previously been reported to have the ability to degrade herbicides of different chemical classes and modes of action’.
46. While the low levels of 2,4-D activity observed in the prior art might explain the multiple resistance activity (in hindsight), this had not been recognised before the priority date and cannot be seen as a concession in the specification that the RdpA was known to degrade multiple herbicides. As a consequence, I find that the claims are not directed to the use of a known material for its known purpose and hence the claims define a manner of manufacture.
47. The opponent also argued that the method claims related to a method of weed control (claims 13-39, 40-43, 60 and 64-68) are to a mere collocation of integers and not to a combination. All these claims are dependent claims which use a transgenic product of an earlier claim. An example is claim 13 which reads as follows:
13. A method of controlling at least one weed in a field, wherein said field contains at least one plant of claim 5, wherein said method comprises applying to at least a portion of a field a first herbicide selected from the group consisting of a phenoxy auxin herbicide and an aryloxyphenoxypropionate herbicide.
48. According to the opponent, the transgenic plant in this claim does not interact with the other integers of the relevant claim to produce a new method of weed control. In their view, the plant or seed may be herbicide resistant but it does not have any role in controlling weeds (which is a function of the herbicide itself). They therefore suggested that the presence of the plant is therefore an inessential integer and without it the claims are directed to a method of weed control which has been practised for decades.[45]
[45] Opponent’s submissions at [55] and [56]
49. I note that as a general principle, if an independent claim defines a product which is a manner of manufacture, then this would flow through to a dependent claim which uses that product. Otherwise the definition of ‘exploit’ in schedule 1[46] would make no sense. In any event, the opponent’s ‘collocation’ argument depends on the only result of the method being a ‘dead weed’. However the presence of the transgenic (herbicide resistant) plant which survives the method of treatment means that the claimed method also produces a ‘weed free’ field of the targeted plants. This is a ‘new result’ which depends on the (resistant) transgenic plant being present (whether one or multiple plants). The claims are therefore not simply a ‘collocation’ of known integers but they define a manner of manufacture.
[46] The definition provides a patentee with protection from infringement which covers an inventive product and any use thereof.
Sections 40 and 18(1)(c)
50. The opponent raised a number of particulars under sections 40 (full description, clarity and fair basis) and section 18(1)(c) (utility). With a significant overlap in the particulars, some arguments were repeated under multiple grounds. I have dealt with these in the context and the order they were presented by the opponent. In cases where an argument was repeated, I have only discussed under the ground where it first appeared.
Utility [section 18(1)(c)].
Relevant law
51. As noted in Ranbaxy Australia Pty Ltd v Warner-Lambert Company LLC (No 2) [2006] FCA 1787, the question of lack of utility under s18(1)(c) is whether the invention enables the addressee to attain the result promised by the patentee in the patent specification. In Austal Ships Pty Ltd v Stena Rederi Aktiebolag (2005) 66 IPR 420, Bennett J generally accepted the proposition outlined in Norton and Gregory Ld v Jacobs (1937) 54 RPC 271 (the Norton principle) at 275-6 that a claim will fail for inutility if within its scope there is subject matter claimed which will not achieve the desired or promised result.
52. However, she did not believe this extended to alternatives within the claim that any sensible person would appreciate would lead to unworkability. In her view:
“The claims are not directed to readers in a vacuum, they are directed to and are to be understood by the skilled workers in the field. That is the person who construes them, in a commonsense way (Populin at 476-477). It would be artificial to assess utility in a way that ignores the fact that a design that is theoretically or mathematically within the parameters of the claims would never be contemplated for use by the skilled naval architect wishing to design a hull for a multi hulled vessel capable of speeds greater than 30 knots. A design that no naval architect would adopt would not be the appropriate test on the question of utility” [at 240]
Application of facts
53. The opponent argued that the claim 1 merely defined a transformed plant cell with a particular nucleotide sequence without any requirement that it contains a particular level of aryloxyalkanoate dioxygenase activity to render a plant herbicide resistant. As a consequence, the opponent argued that the transgenic plant defined in claim 1 is not useful.
54. The opponent also argued that claims 1-4 lacked utility because there was no requirement that the plant as a whole was resistant (just one cell of a plant is resistant). Similar problems exist for claims 57-60 which define seeds, plants grown from seeds and parts of the plant without necessarily being resistant to a herbicide.
55. The applicant argued that all these arguments involved a de minimus claim construction where the opponent deliberately construed the claims to their outer extremes (a ‘worst case scenario’) to argue that the invention would not work. I agree. As Bennet J noted above in Austal Ships Pty Ltd v Stena Rederi Aktiebolag [supra], the claims have to be construed in a common sense way. The skilled worker would clearly understand that the purpose of the transgene was to impart herbicide-resistance whether (or not) this was explicitly defined in the claim. They would therefore not consider making a plant which did not have (some) herbicide-degrading activity due to the presence of the exogenous gene. As this would achieve the promise of the invention, the claims as drafted are useful under section 18(1)(c). The same argument applies to a single cell within a plant. The skilled worker would clearly understand that a herbicide resistant cell should impart some level of resistance to the plant.
Section 40
Sufficiency [section 40(2)]
56. Under section 40(2) of the Patents Act 1990 (Cth), a complete application must describe the invention fully, including the best method known to the applicant of performing the invention. This requires a specification to provide clear and intelligible directions to the skilled addressee of how to perform the invention so that the addressee can produce something within each claim without new inventions or additions or prolonged study[47].
[47] As per Kimberly-Clark Australia Pty Ltd v Arico Trading International Pty Ltd (2001) 207 CLR 1
57. The opponent’s first insufficiency argument relates to claim 5 which reads as follows:
“A transgenic plant comprising a plurality of cells according to claim 1, wherein the expression of said polynucleotide renders said cells tolerant to an aryloxyalkanoate herbicide”
58. The opponent argued that claim 5 was not fully described because it is apparent from the specification [00264] that transformed plants had no resistance to fluroxypyr (a type of aryloxyalkanoate herbicide) and therefore the specification does not provide:
“sufficient disclosure to enable the skilled worker to identify whether the claimed sequences will be effective in rendering plants tolerant or resistant to other aryloxyalkanoate herbicides, except for those herbicides which are the subject of the examples”[48]
[48] Opponent’s submissions at [85]
59. Related to this argument was the opponent’s suggestion that the claimed invention was clearly not effective in rendering plant cells resistant to all aryloxyalkanoate herbicides and it would take a prolonged study for the skilled addressee to determine which herbicides could be used.
60. I note that there is no direct evidence supporting the opponent’s arguments and I am not convinced that testing herbicide resistance in a specific plant using a limited number of known herbicides would require anything more than routine experimentation. In any case, the invention as described in the specification is the transformation of a plant cell with a specific polynucleotide sequence (SEQ ID NO:9) to render the cell resistant to a herbicide. There is no suggestion that the cell would be resistant to any and all aryloxyalkanoate herbicides. This is reflected in the claim itself which merely requires a cell to be resistant to one (ie: ‘an’) aryloxyalkanoate herbicide. Claim 5 would therefore be fully described, fairly based and useful irrespective of whether the cell was (or wasn’t) resistant to a particular aryloxyalkanoate herbicide (fluroxypyr).
61. Similarly, working the invention is not dependent on knowing which aryloxyalkanoate herbicide could be used in the field as long as the plant is resistant to one aryloxyalkanoate herbicide. The specification identifies a specific gene useful in herbicide resistance. There are a limited range of herbicides which are likely to be targeted by this gene and it would be a matter of routine steps to determine which ones. It is not essential that the skilled person understands the full extent of the herbicide resistance being imparted to work the invention. I therefore disagree with the opponent that the claims should specify particular aryloxyalkanoate herbicides.
62. The opponent second insufficiency argument relates to the word ‘tolerant’ as being a relative concept with the specification referring to plants being ‘more tolerant’ in places (see [0005] and [0006]) and ‘highly tolerant’ in other places (see [0074]). According to the opponent, this means that the meaning of the term is ‘insufficient’[49]. However as noted by Gummow J in Rehm Pty Ltd v. Websters Security Systems (International) Pty Ltd (1988) 11 IPR 289, a specification will be fully described despite inexact expressions as long as it:
“… makes the nature of the invention plain to persons having reasonably competent knowledge of the subject and also makes it plain, to persons having reasonable skill, how to perform the invention.”
[49] Opponent’s submissions at [87(c)]
63. In the current case, there is no evidence that difficulties in construing the word ‘tolerant’ would prevent the skilled worker from understanding or performing the invention. In my view, this term is clear and the specification is sufficient under section 40(2).
64. The opponent also suggested that the level of aryloxyalkanoate dioxygenase activity required to achieve herbicide resistance has not been described and hence the specification is not fully described. However, the skilled worker would clearly understand the relationship between enzymatic activity and the level of herbicide resistance and therefore would recognise the need to adjust the former to ensure the levels of resistance required. This would require routine experimentation well within the capabilities of the skilled worker. In my view, the absence of specific levels of AAD activity being set out in the specification does not lead to a problem with sufficiency.
Fair basis [section 40(3)]
65. Section 40(3) of the Patents Act 1990 (Cth) requires that the claim or claims must fairly based on the matter described in the specification. The opponent noted that the test for fair basis was whether there was a ‘real or reasonably clear disclosure’ of the invention as claimed. They argued however that many of claims travelled beyond the matter disclosed in the specification and in particular that:[50]
(a)The invention disclosed in the specification is a method of producing herbicide resistant plants using specified gene sequences but some of the claims merely relate to weed control (claim 13 for example); and
(b)Some of the claims (claim 47 and dependent claims) extend to an undefined range of polynucleotides.
[50] Opponent’s submissions at [90]
66. The opponent explained their first argument with reference to claim 13[51] which they noted was a method of weed control by applying a herbicide to a field containing a plant made in accordance with the invention. They argued that the principle of weed control by means of the application of herbicide was well known long before the priority date. In their view, the invention does not make any contribution to weed control: it merely assists ensuring that the plant is resistant or tolerant to the herbicide used. A similar argument was raised with respect to regard to claims 40-43 and 64-68.
[51] Opponent’s submissions at [91]
67. The former argument is an extension of the opponent’s manner of manufacture case and assumes that the presence of the transgenic plant is not critical to the scope of the claim. However the invention described by the specification is a transgenic plant containing a specific bacterial gene (AAD-1) which is optimised for the plant environment. This feature is explicitly defined in each of the claims including the ones (such as claim 13) which refer to methods of treatment. The claims are therefore all consistent with what the specification as a whole describes as the invention (as per Lockwood v Doric [2004] HCA 58) and in my view are fairly based.
68. The opponent explained their second argument by reference to claims 44-49[52] which they argued defined an undefined range of polynucleotides with the only requirement being that the nucleic acid molecule encoding the aryloxyalkanoate dioxygenase activity hybridises under strict conditions with the specified sequences. In their view, the claims were broad enough to capture pre-existing polynucleotides that were not created by reference to the invention.
[52] Opponent’s submissions at [91]
69. The argument is best understood by reference to a particular claim, an example of which is claim 44:
Claim 44: A polynucleotide optimised for expression in a plant wherein said polynucleotide encodes a protein having aryloxyalkanoate dioxygenase activity wherein a nucleic acid molecule that encodes said protein hybridizes under stringent conditions with the full complement of a sequence selected from the group consisting of SEQ ID NO:3, SEQ ID NO:4 and SEQ ID NO:5, wherein the polynucleotide is operably linked to a promoter that is functional in a plant cell, wherein the stringent hybridization conditions comprise 1xSSPE and 42oC.
70. I note that these claims do not define any polynucleotide which hybridises to specific sequences but only those which:
(a) Are optimised for expression in a plant (eg: operably linked to a plant promoter); and
(b) Encode a protein having aryloxyalkanoate dioxygenase activity.
71. The skilled person knows that changes can be made to a gene sequence to create identical or homologous proteins and/or to adapt the gene to its expression host. These are routine adjustments which are well within the capabilities of a skilled worker and would be understood by them as being a non-inventive variation from the original gene. As a consequence, the claims are consistent with what the specification as a whole describes as the invention. I therefore find claims 44-49 fairly based.
Conclusion
72. None of the citations deprive any of the claims of their novelty or inventive step. The prior art had shown the RdpA containing extracts were not only highly active against (R)-enantiomers of chiral phenoxyalkanoate herbicides but also had a low activity against 2,4-D. Given this low level of 2,4-D activity, the skilled worker would not expect RdpA to confer multiple herbicide resistance in a plant as this was against the general expectation of the art. They would therefore not be directly led to use the gene for this purpose. As a consequence, each of the claims is novel and inventive based on the prior art and the claims define a manner of manufacture.
73. None of the remaining issues raised by the opponent result in sections 40 or 18 problems with the specification. The opposition is therefore unsuccessful and the patent application is in order for grant.
Costs
74. As the opposition has been unsuccessful on all grounds, I award costs in accordance with Schedule 8 of the Patents Regulations 1991 against the opponent, Nufarm Australia Ltd.
Karen Ayers
Delegate of the Commissioner of Patents
Annex A
Six members of the phenoxy family of herbicides showing the common aryloxyalkanoate chemical sub-structure
From Bruce T Lyon “Engineering Microbial Herbicide Detoxification Genes” (chapter 5 of “Plant Gene Research: Molecular Approaches to Corp Improvement” Ed E.S. Dennis and D.J. Llewellyn (1991) (Published by Springer-Verlag) (Exhibit TS-1)]
Annex B
Initial step in the degradation of chiral phenoxypropanoic acids in S. herbicidovorans MH catalysed by the α-ketoglutarate dependent (R)–dichlorprop dioxygenase (RdpA) and the α-ketoglutarate dependent (S)–dichlorprop dioxygenase (SdpA)
From: Müller et al “Purification and Characterisation of two enantioselective α-ketoglutarate dependent dioxygenases, RdpA and SdpA from Sphingomonas herbicidovorans MH pplied Environ Microbiol July 2006: 72(7):4853-4861
0
2
0