T 1587/17 (Production system/AMYRIS) 30-04-2021

Reasons for the Decision

Admissibility of the appeal

1. The time limit pursuant to Article 108, first sentence, EPC for filing an appeal against the interlocutory decision of the opposition division posted on 7 April 2017 concerning the maintenance of European patent No. 2 021 486 in amended form expired on 19 June 2017 (Rule 131(2), second sentence, and (4) EPC, Rule 134(1) EPC, and Rule 126(2) EPC). The only entry in the European patent register on 19 June 2017 for the opposed European patent No. 2 021 486 is a debit order for the appeal fee by the appellant-opponent's representative (EPO form 1010, "Payment of fees and expenses"). Payment of an appeal fee does not in itself constitute filing of an appeal (established case law following J 19/90 and T 371/92, OJ 1995, 324). Therefore, an appeal against the decision of the opposition division dated 7 April 2017 is deemed to have been filed only if a notice of appeal was received by the EPO on or before 19 June 2017.

2. The legal burden of proof is on the party who derives rights from an alleged positive fact. Accordingly, the party asserting a right must prove the facts giving rise to that right, while the burden of proving the facts nullifying a right lies with the party disputing its existence. Therefore, the legal burden of proof for the fact that a filing complying with a time-limit has been received by the EPO, rests on the sender. In the present case, the burden thus lies on the appellant to establish that their notice of appeal against the decision of the opposition division posted on 7 April 2017 concerning the maintenance of European patent No. 2 021 486 in amended form reached the EPO within the time limit pursuant to Article 108, first sentence, EPC (T 1200/01, Reasons, point 4; T 2454/11, Reasons, point 2.1).

3. With respect to the question of standard of proof, i.e. the degree of certainty that is required for the board to conclude that an alleged fact is made out, the respondent stressed that decision T 2454/11 establishes that the sender has to prove receipt of a document allegedly sent to the EPO "to the board's satisfaction" which implied a high standard. However, decision T 2454/11 does not explain what degree of certainty is required by the standard "satisfaction of the board" (see Reasons, points 2.5 and 2.6). It is therefore necessary to determine the required standard.

3.1 Decision T 2454/11 cannot be understood to require absolute certainty. Absence of slightest doubts would indeed be unreasonable since it would be impossible to meet such a stringent standard of proof. The board in case T 2454/11 nevertheless intended to distinguish the standard of proof applied by it from that applied in case T 1200/01. It appears that the board in case T 2454/11 took issue with a standard of "high probability" in the sense of a fact being "more likely than not". However, as becomes evident from the Reasons, point 4 of decision T 1200/01 (referring to "proof of higher probability") and the analysis of the assessment of the evidence by the board in that case (Reasons, points 6 to 8), the board in case T 1200/01 applied a more stringent standard of proof, namely that of "preponderant probability". Evaluating and weighing the conflicting evidence, the board assessed whether there remained reasonable doubts that the filing had occurred as alleged by the appellant.

3.2 The standard of "preponderant probability" requires a high degree of probability (see e.g. T 128/87, OJ EPO 1989, 406, point 6.1). The possibility that the alleged fact might not have materialised should only be remote or marginal. In other words, a fact is proven if the court no longer has serious doubts about the existence of the alleged fact, or if any remaining doubts appear to be slight. In the board's opinion, this equates to the standard requiring absence of "reasonable doubts" in which case the alleged fact is made out to the "satisfaction of the board" as in case T 2454/11.

4. In the present case, the appellant has proven "to the satisfaction of the board" that a notice of appeal regarding the decision of the opposition division posted on 7 April 2017 concerning the maintenance of European patent No. 2 021 486 in amended form reached the EPO on 19 June 2017.

4.1 On 19 June 2017, the opponent paid the appeal fee by way of debit order on EPO form 1010. This fee payment sheet refers to the opposed patent by both its application number (No. 07777278.8) and publication number (No. 2 021 486). The sheet is signed and dated by the appellant-opponent's representative and date stamped "19 June 2017" by the EPO. A scanned copy is part of the electronic file. The fee payment sheet was part of a grouped filing of the representative's office on 19 June 2017. The filing was accompanied by a list numbered 486/2017 specifying the documents enclosed in the grouped filing and the European patent applications and patents to which these documents relate. Item No. 17 in this list indicates the application number 07777278.8 of the opposed European patent. This entry specifies that the corresponding enclosure in the grouped filing is an "appeal" ("Beschwerde") and that payment is made by debit order ("ABB" which is the abbreviation for "Abbuchung"). The latter document undoubtedly reached the EPO while the document referred to as "appeal" is missing in the electronic file. However, by date stamping and returning the list of enclosures of the grouped filing No. 486/2017 the EPO acknowledged receipt of the complete enclosures. This holds even more true considering that it was explicitly requested on the list of enclosures to return the list immediately. Had an item been missing, the EPO should not have returned the stamped list without any proviso. As is evidenced by the date stamps on individual items of the grouped filing No. 486/2017, like the payment sheet for the appeal fee in the present case, the individual enclosures were sorted and processed by the EPO's mail room. Any missing item would have immediately become apparent. The list of the grouped filing bearing the date stamp of the EPO is confirmation and proof that the items on that list reached the EPO on 19 June 2017. Indeed, such a confirmation of receipt by the EPO was considered essential but missing in the case underlying decision T 2454/11 (see Reasons, point 2.3, first paragraph, and point 2.7, third paragraph: "Furthermore, neither the three documents nor D1 contain any indication to the mail room to acknowledge their receipt. Moreover, the appellant itself admits that document D1, which lists the documents contained in the UPS envelope, was neither stamped by the EPO nor returned to the office of its representative.", translation by the board).

4.2 The respondent argued that the reference to "Beschwerde" ("appeal") under item 17 of the list of enclosures was no proof that a notice of appeal had in fact been enclosed. However, the board has no serious doubts that the notice of appeal, a copy of which was filed on 14 August 2017, was part of the filing received by the EPO. The entry "Beschwerde" in the column "attached documents" refers to a document different from the fee payment sheet which is on file. When referring to a document, the term "Beschwerde" in German either means "Beschwerdeschrift" ("notice of appeal") alone or includes additionally the statement of grounds of appeal. It cannot be understood to refer to the payment sheet, let alone to any other document in this context. Therefore, there are no reasonable doubts that the payment sheet on file was sent together with a notice of appeal as part of item No. 17 of the grouped filing of 19 June 2017. The short signature of a secretary next to item No. 17 of said list shows that the respective enclosures had been checked at the representative's office. The possibility that the notice of appeal was not included in the grouped filing is therefore very remote. In any case, the question whether the envelope that left the representative's office contained all items listed in the list of enclosures of the grouped filing No. 486/2017 is not decisive. Since the EPO did not note on the returned list of enclosures that the document listed under item No. 17 as "Beschwerde" was missing, it must have reached the EPO. The board therefore disagrees with the respondent's argument that the evidence on file is merely circumstantial. As regards the further evidence in addition to the EPO's acknowledgement of receipt, the board agrees that it is circumstantial, but notes that it is not conflicting with receipt by the EPO of a notice of appeal but rather confirming this fact.

4.3 In sum, the board concludes that the appellant has discharged its burden of proof. The respondent has not succeeded in raising doubts that are serious enough to counter the evidence provided by the EPO's acknowledgement of receipt. The appeal has therefore to be considered as filed within the time-limit pursuant to Article 108, first sentence, EPC.

Admission into the proceedings of a new line of argument under added subject-matter

5. According to the established case law, the function of an appeal is to review in a judicial manner the decision under appeal taken by an examining or opposition division. The admission of inter alia new lines of argument into the appeal proceedings is at the board's discretion (see Articles 114(2) EPC and 25(2) RPBA 2020 in conjunction with Article 12(4) RPBA 2007).

6. It is uncontested that the appellant submitted in their statement of grounds of appeal for the first time in the proceedings a line of argument under added subject-matter against the feature "into at least one bio-organic compound having at least ten carbon atoms" in step c. of claim 1.

7. In essence the appellant submitted that this new line of argument was admissible because it focused on a specific aspect of a more generic objection under added subject-matter raised during the first instance proceedings.

8. The board does not agree. The contested feature in step c. of claim 1 set out above is already present in claim 1 as granted. Indeed present claim 1 differs from claim 1 as granted in the deletion of the range "40-99%" in step d. of claim 1 only, all other features, including the contested one, are identical in both claims.

8.1 In the board's opinion, focusing on a specific aspect of a more generic objection previously raised requires at least that the contested features are linked, for example, structurally or functionally. In the first instance proceedings the appellant raised three objections of added subject-matter against the combination of three specific features in claim 1 (see point VII, pages 7 to 9 of the notice of opposition; page 1, last paragraph of the minutes). These three lines of argument have been maintained by the appellant in the present appeal proceedings (see below).

8.2 The new objection of added subject-matter concerns a feature in step c. of claim 1 which relates to a functional property of genetically modified host cells, namely their ability to convert a specific carbon source into a bio-organic compound (i.e. an isoprenoid) of a defined minimal length ("at least ten carbon atoms"). This feature is structurally and functionally independent of the three other features in claim 1 against which the previous objections have been raised, i.e. the specific combination of genetically modified host cells, a spontaneously formed organic phase, which moreover, comprises an amount of isoprenoids of at least 90%.

9. Since the line of argument raised against the contested feature in step c. of claim 1 identified above is new, although the feature is present in claim 1 from the onset of the opposition proceedings, the appellant should have submitted an objection against it already in the first instance proceedings. Reasons for submitting it with the statement of grounds of appeal only have not been brought forward by the appellant and are not apparent to the board.

10. Accordingly, exercising its discretion under Article 12(4) RPBA 2007, the board decided not to admit this new line of argument under added subject-matter into the proceedings.

Main request

Claim interpretation - claim 1

11. Claim 1 relates to a product which is directed to a production system for bio-organic isoprenoids having at least 10 carbon atoms in length. The system is characterised by the following structural and functional features:

(i) a vessel of a defined minimum size ("at least 100 liters") (hereinafter the "fermenter size" feature);

(ii) a first phase of an aqueous medium in the vessel (hereinafter the "aqueous phase" feature) that contains:

(a) genetically modified bacteria or (b) fungi as host cells,

(c) which are further functionally defined as being "capable of converting a carbohydrate carbon source present in the aqueous medium into at least one bio-organic compound having at least ten carbon atoms";

(iii) a second phase of a liquid organic (hereinafter the "organic phase" feature) characterised by:

(a) its contact with the first phase (i.e. the "aqueous phase"),

(b) its spontaneous formation, and

(c) its isoprenoid concentration "in an amount of at least 90%".

12. Claim 1 encompasses as a product any production system for bio-organic isoprenoids having at least 10 carbon atoms in length with the properties defined in steps a. to d., irrespective of how the production system is provided. In other words, although claim 1 comprises a process feature for generating the organic phase ("spontaneously formed"), the claimed production system is limited by that process only in so far as it results in properties of the system that can be distinguished from other systems, irrespective of the way in which they have been provided.

13. Contrary to the appellant's view, the fermenter size feature in claim 1 is not an arbitrary design option, but a structural feature that limits the claimed system to those with a minimum vessel size of at least 100 l.

14. The aqueous phase feature in claim 1 is specified as an aqueous medium at a particular location (within the vessel) that contains a plurality (i.e. at least two of a single species, or a mixture of at least two different species) of "genetically modified" bacteria and fungi. Accordingly, natural isolates or wild-type host cells that are not genetically modified do not fall within the scope of claim 1. These cells are further functionally limited in that they have to be suitable for a particular purpose, i.e. the conversion of a "carbohydrate carbon source" into an isoprenoid compound of a certain minimum length ("at least 10 carbon atoms"). Since the term "carbohydrate" is normally construed by the skilled person to relate to a saccharide or sugar, bacteria or fungi that convert carbon sources other than carbohydrates into isoprenoids do not fall within claim 1.

15. The organic phase feature in claim 1 is defined by various structural and functional features.

15.1 Firstly, the phase is liquid, i.e. a volatile phase is not comprised by the claim.

15.2 Secondly, the organic phase in claim 1 is characterised in that it is "in contact" with the aqueous phase. This requires that both phases are in direct physical contact with each other. Other restrictions are not imposed by that term, for example, a particular order, position, or form of the two phases.

15.3 Thirdly, the organic phase is "spontaneously formed". In the board's view, the skilled person construes a spontaneous formation to take place without active external intervention, for example, by technical means (e.g. centrifugation), or chemical means (e.g. the addition of organic solvents, de-emulsifiers or nucleating agents, see paragraph [0058] of the patent application). In other words, this functional feature characterises an intrinsic property of isoprenoids of at least 10 carbon atoms in length at a concentration of at least 90% in an aqueous medium. The appellant submitted that 24 hours for forming an organic layer by gravity settlement as disclosed in documents D9 to D11 was too long to qualify as spontaneous. The board does not agree, since "spontaneously" does not equate to instantly and, hence, imposes no time restriction. Accordingly, organic phases according to claim 1 form immediately, or over some time.

15.4 It was common ground between the parties that "spontaneously formed" in claim 1 implied a separation of the organic phase from the aqueous phase. This separation relates to any form, including layers (see documents D9 and D10). Since a phase separation is implied in this feature, the board does not agree with the respondent that an emulsion of an aqueous and an organic phase is encompassed by claim 1. Although in an emulsion both phases are mixed, i.e. in physical contact, they are not separated from each other.

15.5 The organic phase of step d. in claim 1 is further characterised in that it comprises "the at least one bio-organic compound in an amount of at least 90%". The appellant submitted that it is not clear to what the "at least 90%" refers, i.e. the compound or the organic phase. Therefore claim 1 encompassed an organic phase that contained (i) at least 90% of the total amount of at least one isoprenoid, leaving 10% or less in the aqueous phase, and (ii) an organic phase of at least 90% of this compound, i.e. a phase formed of at least 90% of at least one isoprenoid, and 10% or less of other organic materials only. In the board's view, the skilled person would normally construe the "at least 90%" in this feature to refer to the organic phase. This construction is in line with the description (see paragraph [0045] of the patent), and makes technically sense, since isoprenoids with a length of at least 10 carbon atoms are normally hydrophobe, and hence, form "spontaneously" an organic phase in an aqueous medium (see above).

16. In a nutshell, claim 1 is directed to a system for the production of isoprenoids of a minimum length of 10 carbon atoms by biotechnological means, in particular by using genetically modified microorganisms in an aqueous culture, which at an undefined point in time during the fermentation of a carbohydrate as the carbon source spontaneously form an organic phase of at least 90% isoprenoids. In other words, claim 1 provides a snapshot of the production system in operation since it is defined by ingredients that are not all present in the vessel at the beginning of the fermentation but only at a later stage.

Added subject-matter

17. It is established case law that any amendment, for example, in a claim can only be made within the limits of what a skilled person would derive directly and unambiguously, using common general knowledge, and seen objectively and relative to the date of filing, from the whole of the patent application as filed.

18. The appellant submitted three objections under added subject-matter against various features in claim 1.

19. Firstly, the appellant argued that the combination of "a liquid organic second phase, comprising the at least one bio-organic compound in an amount of at least 90%", wherein the compound "is an isoprenoid" in step d. of claim 1 added subject-matter. While claims 7 and 15 as filed disclosed these features in individualised form, both claims referred back to claim 1 as filed only. Thus, claims 7 and 15 as filed provided no basis for the combined use of these features in present claim 1.

19.1 The board does not agree. The patent application discloses as the sole exemplified production system for bio-organic compounds that for isoprenoids (see e.g paragraphs [0022], [0051], [0054], [0092], claims 1, 7, 16, 21, 34 and 35). This, in the board's opinion, points to production systems for isoprenoids as preferred.

19.2 A concentration of "at least 90% bio-organic compound" in the organic second phase of the system is disclosed in claim 15 as filed, which likewise indicates that this feature is preferred. Since the production of isoprenoids as bio-organic compounds is likewise preferred, the patent application provides a direct and unambiguous basis for the combination of these two features in step d. of claim 1.

20. Secondly, the appellant submitted that the combination of "genetically modified" and "bacterial or fungal host cells" in step c. of claim 1 was obtained after performing "several selection steps".

20.1 The board again does not agree. Paragraph [0021] of the patent application mentions that the terms "Host cell" and "microorganism" are interchangeably used which refer "to any archae, bacterial, or eukaryotic living cell into which a heterologous nucleic acid can be or has been inserted" (emphasis added). In this context, paragraph [0083] of the patent application further discloses that any "suitable host cell can be used in the practice of the present invention. In some embodiments, the host cell is a genetically modified host microorganism" (emphasis added). In other words, both paragraphs disclose that genetically modified archae, bacterial or eukaryotic cells are an embodiment of the invention.

20.2 As regards eukaryotic host cells, the patent application mentions fungal cells only (see paragraphs [0088] to [0090]), which implies their preferred use as host cells in the patent application.

20.3 Examples of bacterial host cells are disclosed in paragraphs [0085] to [0087] of the patent application.

20.4 Furthermore, the working examples of the patent application disclose solely genetically modified bacterial and fungal host cells (see page 44, paragraph [0181] to page 80, first paragraph). This is a further indication that in the production system of claim 1 the use of genetically modified bacterial and fungal cells are preferred.

20.5 In view of these considerations, the board concludes that the host cells cited in step c. of claim 1 are indicated as preferred in the patent application, and hence, directly and unambiguously derivable therefrom.

21. Thirdly, the appellant submitted that the feature "the organic second phase is spontaneously formed" in step d. of claim 1 lacked a proper basis in the patent application.

21.1 Paragraph [0058] of the patent application states: "In some embodiments, the organic second phase occurs spontaneously as a result of chemical and molecular interactions such as differences in solubility, or hydrophobicity, density, concentration or any other spontaneous phase separation mechanism. In other embodiments, separation of the first and second phases is induced in a separation vessel or vessels or system that may be the same or a different vessel or vessels or processing system as the fermentation vessel or vessels" (emphasis added).

21.2 In other words, this paragraph teaches in general that the organic second phase forms either spontaneously or is actively separated from the aqueous first phase. Thus, this paragraph discloses two generic principles or alternatives to achieve a phase separation. A selection of one generic alternative out of two does not constitute a selection from a list.

21.3 The appellant argued that the patent application disclosed various alternative mechanisms for separating organic and aqueous phases. However, all of the examples referred to by the appellant concern actively induced phase separation mechanisms (see page 12, last line to page 13, first paragraph of the patent application), i.e. these examples are embodiments of the second phase separation principle set out above, which is not cited in claim 1. In these circumstances, the appellant's argument does not convince the board.

22. Accordingly, the board concludes that claim 1, and hence, the main request complies with the requirements of Article 123(2) EPC.

Sufficiency of disclosure

23. The appellant submitted that the patent application lacked any information for the skilled person how in an industrial-scale production system a spontaneously formed organic isoprenoid phase could be separated from an aqueous phase. Example 23 of the patent application related to a hypothetical working example only. Although it mentioned settlement and decantation for separating an isoprenoid organic phase from the aqueous phase in a small-scale cultivation vessel ("250 mL"), it was silent on how decantation could be performed in an at least 100 l fermenter as cited in claim 1. Furthermore, Example 23 provided no information about the minimal isoprenoid titer to form an organic phase. Since the patent provided no information how to perform the claimed invention, the burden of proof to substantiate insufficiency by verifiable facts was not on the appellant, but on the respondent. The respondent had not discharged their burden of proof by submitting the supplementary experimental data (see document D9 and D10), since these documents provided no evidence that the claimed invention could be carried out. The vessel sizes used therein were smaller than the size referred to in claim 1. Moreover, the data was obtained from a single experimental set-up, and hence, not sufficient to support the whole breadth of the claim. The required 24 hours for obtaining an organic layer was also too long for a spontaneous phase separation.

24. According to the established case law the provision of experimental evidence in the patent application for a claimed effect is not a prerequisite for patentability, since it suffices that based on the data in the patent application, or from common general knowledge it is plausible that a product (here an isoprenoid of a certain minimal length and concentration) is suitable for the claimed technical effect (here the spontaneous formation of an organic phase) (see Case Law of the Boards of Appeal of the EPO, 9th edition 2019, (hereinafter "Case Law"), II.C.7.2.).

25. The claimed invention, as set out above, is directed to a system for the production of isoprenoids of a minimum length by using genetically modified microorganisms in an aqueous culture utilising a carbohydrate as carbon source, which at an undefined point in time during the fermentation spontaneously form an organic phase of at least 90% isoprenoids.

26. The patent application discloses in paragraph [0058] that the spontaneous formation of an organic phase in an aqueous phase is the "result of chemical and molecular interactions such as differences in solubility, or hydrophobicity, density, concentration or any other spontaneous phase separation mechanism".

27. Due to their chemical nature, isoprenoids are hydrophobic, in other words, immiscible with water. In the board's opinion, the skilled person taking common general knowledge into account would therefore reasonably assume that isoprenoids due to their chemical properties form at a certain concentration inherently, and hence "spontaneously", an organic phase in an aqueous medium.

28. Example 23 of the patent application "describes the production of linalool, limonene, beta-pinene, beta-phellandrene, carene, or sabinine in Escherichia coli host strains", i.e. of various isoprenoids with at least 10 carbon atoms, in a bacterial host (see paragraph [0319]). A genetic modification of the E. coli cells used in Example 23 is implied by the addition of "IPTG" (a known inducer of heterologous gene expression) into "250 mL" culture vessels to start the production of isoprenoids. The isoprenoids are separated from the culture medium inter alia "by settling and decantation if the titer of the compound of interest is large enough to saturate the media and to form a second phase" (see paragraph [0321]).

29. A simple decantation of the organic phase by using, for example, taps in the vessel is disclosed in paragraph [0059] of the patent application.

30. In the board's opinion, based on the information in the patent application and the considerations indicated above, the skilled person would consider it at least plausible that isoprenoids produced by microbes and secreted into the aqueous medium spontaneously form an organic phase, depending on their concentration. Moreover, since this is an inherent property of isoprenoids in water, it occurs independently of the vessel or sample size used.

31. It is further established case law that an effect which is at least rendered plausible by the teaching of the patent application can be backed-up by supplementary experimental evidence (see Case Law, II.C.6.8.). Documents D9 and D10 disclose two experiments using a genetically modified yeast cell for the production of farnesene, i.e. an C15-isoprenoid (see "Experimental Summary"). After a settling time of 24 hours, the culture samples form an organic layer of farnesene that overlays the culture medium (see documents D9 and D10, Figures, and document D11, point 4). Further, the respondent submitted a report of an integrated biorefinery pilot project that started on 28 December 2009 (see document D8, front page). Document D8 discloses a pilot plant that houses "two 300L fermentors" (see page 8, first paragraph, below the heading "Task A: Pilot plant upgrades and operations"), and mentions that the purity of crude farnesene using various sugar syrups as carbon source for fermentation is consistently above 90% (see page 38, Table A.7). The appellant submitted that the crude farnesene did not spontaneously form but had been separated from the medium by centrifugation (see page 37). The board does not agree, since the centrifugation step described on page 37, last paragraph to the first paragraph on page 38 of document D8 concerns the clarification of the sugar syrup used as carbon source prior to sterilization and fermentation.

32. The appellant further submitted that the supplementary data did not demonstrate that the claimed production system could be put into practise. The board does not agree. Although documents D9 and D10 use cultivation vessels that are smaller than the minimum size of 100 l cited in claim 1, and use samples of the fermenter medium instead of the whole medium to demonstrate the spontaneous formation of an isoprenoid organic phase by gravity settlement, the board has no doubts that these experiments are representative for the claimed production system, since as set out above, isoprenoids at a certain concentration spontaneously form an organic phase as an inherent property in an aqueous medium, irrespective of the vessel size or sample size used. Moreover, the spontaneous formation does not impose any time restrictions.

33. The appellant has not submitted any evidence of non-working embodiments falling within the scope of claim 1, or other verifiable facts in support of doubts that the claimed system could be carried out across the whole breadth of claim 1. In inter partes proceedings, the burden of proof initially rests upon the opponent to establish, based on the balance of probabilities, that a skilled person reading the patent application/patent using his/her common general knowledge would be unable to carry out the invention. The burden of proof can be reversed, however, when the patent application/patent does not contain detailed information of how to put the invention into practice, and hence, a weak presumption exists that the invention is sufficiently disclosed. It is then up to the patent proprietor to prove the contrary (see Case Law, II.C.9.).

34. As set out above, the board is convinced that based on the information provided in the patent application, it is at least plausible that the microbial production of isoprenoids with the properties as defined in claim 1 results in the spontaneous formation of an organic phase in an aqueous medium. In the absence of any evidence to the contrary, the board concludes that the main request complies with Article 83 EPC.

Novelty

35. The appellant submitted that the subject-matter of claim 1 lacked novelty over documents D1 to D3.

36. As set out above, the board does not agree with the appellant that the "fermenter size" feature of claim 1 relates to an arbitrary design option, but is a structural feature of the claim. Since, the fermenters/cultivation vessels in documents D1 to D3 have sizes of "10-L", "1.5 l", or of "40 ml"/"125 ml", respectively (see page 685, column 2, second paragraph of document D1; page 720, column 2, third paragraph of document D2; column 3, lines 24 and 25, and column 4, lines 5 to 7 of document D3), which are all smaller than 100 l, the board concludes that the subject-matter of claim 1 is novel over the cited documents, at least because of this feature.

37. Accordingly, the main request complies with the requirements of Article 54 EPC.

Inventive step

Closest prior art and technical problem

38. The appellant selected either documents D1, D2 or D3 as closest prior art in combination with document D5, or alternatively document D6 as closest prior art combined with documents D1, D2 or D3. The respondent selected document D1.

39. According to established case law, a document aiming at the same purpose or effect and having most of the relevant technical features in common with the claimed invention normally represents the closest prior art (see Case Law, I.D.3.1).

40. Document D1 discloses a method for the production of amorpha-4,11-diene (a C15 isoprenoid) by an engineered E. coli strain using a two-phase partitioning bioreactor (TPPB) (see abstract). The cells are grown on glycerol as carbon source (see page 685, column 2, first paragraph). Although glycerol is a polyol, it is commonly used in foods as sweetener, and hence, in the present case can be regarded as a carbohydrate. The bioreactor in document D1 has a volume of "10-L" (see page 685, column 2, second paragraph). Although the first sentence in the abstract of document D1 mentions "large-scale fermentations", this statement does not relate to the production process or the system actually disclosed in document D1, but rather indicates a general technical context. Document D1 further mentions that an overlay of the organic solvent dodecane is added to the fermenter vessel for traping volatile amorpha-4,11-diene. The dodecane phase contains more than 97% of the generated amorpha-4,11-diene (see page 686, column 1, first paragraph, page 687, column 1, second and third paragraphs, page 690, column 1, third paragraph, column 2, second paragraph). Alternatively, volatile amorpha-4,11-diene is isolated by condensation in a cold trap cooled with dry ice and ethanol (see page 686, column 1, second paragraph). The appellant submitted that a spontaneous formation of an organic phase related to any phase separation mechanism, including the use of a cold trap. Therefore no difference existed between the organic phase obtained by the cold trap in document D1 and claim 1. The board does not agree. While an "organic second phase" that "is spontaneously formed" as cited in claim 1 without further specifications indeed relates to organic phases that form without an external activity (see above) in any physical state, claim 1 further defines that the organic phase spontaneously formed is "liquid" and "in contact with the first [aqueous] phase". However, the amorpha-4,11-diene in document D1 is volatile, and moreover, no longer in contact with the aqueous phase.

41. In summary, the subject-matter of claim 1 differs from document D1 in the "fermenter size" feature (100 l vs 10 l), and in the "organic phase" feature (no disclosure that the organic phase consists of at least 90% isoprenoid, and is in contact with the aqueous phase).

42. Document D2 discloses an aqueous-organic two-phase system for the bacterial conversion of carveol (an isoprene) into the C10 isoprenoid carvone by using Rhodococcus erythropolis (see title and abstract; also document D3, Figure 1). Document D2 is silent on whether this Rhodococcus strain is genetically modified, or non-modified (see page 720, column 1, fourth and fifth paragraph, column 2, second paragraph). The fermenter in document D2 has a "working volume of 1.5 l" (see page 720, column 2, third paragraph). Different carbon sources are used in the growth medium for optimising the conversion reaction, including an C10 isoprenoid ("limonene"), alcohols such as "ethanol", or "cyclohexanol", an alcoholic derivative of cyclohexane (i.e. a cycloalkane) (see page 720, column 2, third paragraph). None of them is a carbohydrate. The bacterial conversion of carveol into carvone is either performed in a single-aqueous phase system (see page 720, column 2, last paragraph), or in a two-phase system by adding an overlay of "iso-octane" or "n-dedecane", i.e. two organic solvents (see page 721, column 1, second paragraph). The one-phase system adds "ethylacetate" at a later stage to extract the carvone from the aqueous phase (see page 720, column 2, last paragraph, and page 721, column 1, fourth paragraph).

43. Accordingly, the subject-matter of claim 1 differs from document D2 in the "fermenter size" feature (100 l vs 1.5 l), the "aqueous phase" feature (no genetically modified bacteria, no carbohydrate as carbon source), and the "organic phase" feature (no disclosure that the organic phase comprises at least 90% isoprenoid).

44. Document D3 discloses a genetically modified E. coli that is used in a two-phase aqueous-organic system for producing various isoprenoid derivatives (alpha-terpineol, carveol, carvone, and perillyl-alcohols) of the C10 isoprenoid limonene (see column 1, lines 10 to 20, 26, 42, column 2, lines 51 and 52, and Figure 1). The isoprenoid derivatives are extracted from the aqueous medium as soon as they are formed by adding an overlay of an organic solvent, preferably limonene (see column 1, lines 10 to 17, column 2, lines 47 to 50, column 5, lines 5 to 14, and Figure 3). Aside its function for extracting the derivatives, limonene is also used as carbon source for the E. coli cells to produce the derivatives (see column 2, lines 47 to 51). Document D3 uses "40 ml serum bottles" or "125 mL screw cap bottles" as cultivation vessels (see column 3, lines 24 and 25, and column 4, lines 5 to 7, respectively).

45. Thus, the subject-matter of claim 1 differs from document D3 in the "fermenter size" feature (100 l vs 40/125 ml), and the "aqueous phase" feature (no carbohydrate as carbon source). Furthermore, since in document D3 an isoprenoid serves as extraction medium and carbon source, the two-phase system is based on a concept different from the claimed production system.

46. Document D6 discloses a further two-phase aqueous-organic system for the production of beta-carotene, a C40 isoprenoid, by the microalgae Dunaliella salina. Since beta-carotene is intracellularly produced in chloroplasts and not secreted into the medium, alkene-based organic solvents, including decane and dodecane, are added to the medium for its extraction from the cells (see abstract, and Figure 3). The cells are cultivated in 50 ml growth medium (see page 178, column 2, fourth paragraph), which implies the use of a fermenter that is significantly smaller than 100 l. Document D6 is further silent on the use of genetically modified organisms, let alone a bacterium or a fungus, including the use of carbohydrates as carbon source. Furthermore, since beta-carotene is extracted from the cells by adding an organic solvent, document D6 does not unambiguously disclose that the organic phase comprises at least 90% beta-carotene and 10% or less of the alkene solvent.

47. In summary, document D6 differs from the claimed subject-matter in the "fermenter size" feature, the "aqueous phase" feature (no genetically modified bacteria or fungi, no carbohydrate as carbon source), and the "organic phase" feature (no disclosure of an organic phase formed of at least 90% isoprenoid).

48. Consequently, documents D1 to D3 and D6 are all directed to the same purpose underlying the claimed subject-matter, namely the provision of a biotechnology-based isoprenoid production system. The system disclosed in document D1 shares most of the relevant technical features with the claimed subject-matter (since it differs in two features only), while documents D2 and D6 differ in three features, and document D3 relates to a different concept (use of an isoprenoid as extraction medium and carbon source).

49. In line with the case law (see above) the board considers that document D1 represents the closest prior art. The larger vessel size of the claimed system enables the microbial production of isoprenoids in higher amounts. The direct use of a spontaneously formed organic phase for separating isoprenoids from the aqueous phase instead of adding an organic solvent or using a cold trap results in the provision of a simplified production system, since less reagents/technical means and process steps are required.

50. The technical problem to be solved is thus defined as the provision of a simplified production system for isoprenoids.

51. The board is convinced that this problem is solved by the subject-matter of claim 1 in light of Example 23 and paragraph [0048] of the patent.

Obviousness

52. It remains to be assessed whether or not the skilled person starting from document D1 and faced with the technical problem identified above, would have arrived at the claimed production system in an obvious manner.

53. Document D1 explicitly suggests the use of large-scale fermentations (see abstract), which obviously implies the use of larger fermenter vessels, including for example, a volume of 100 l. Accordingly, the distinguishing "fermenter size" feature alone does not render the claimed production system inventive over document D1. Document D5 only discloses that large-scale aerobic fermentation reactors are known since the 1930s (see page 167, column 1, second paragraph), and that such fermenters are applicable for fermentation volumes from 1 L to several 1000 L (see page 168, column 2, last paragraph, and Tables 7.2 and 7.3 on pages 169 and 170). Therefore, none of documents D1 to D3, D5, or D6 suggest or hint at the spontaneous formation of an organic phase of microbially-produced isoprenoids during fermentation. Nor that this property of the isoprenoids could be exploited for directly separating an isoprenoid phase from an aqueous cultivation phase, and hence be useful in the provision of a simplified production system.

54. Accordingly, the subject-matter of claim 1 is not obvious in the light of the teaching of document D1 alone, or in combination with document D5. The same applies for the combination of any of the other documents cited by the appellant.

55. Thus, the main request complies with Article 56 EPC.