England and Wales High Court (Patents Court) Decisions

Novo Nordisk A/S v. DSM NV [2000] EWHC Patents 34 (21st December, 2000)

HC 1999 02226

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT

BEFORE: MR JUSTICE NEUBERGER

BETWEEN:

Mr. David Kitchin QC and Mr. Richard Meade (instructed by Messrs. Bird & Bird, London EC4) appeared on behalf of the claimant.
Mr. Simon Thorley QC and Dr. Justin Turner (instructed by Messrs. Lovells) appeared on behalf of the defendant.

This is an approved judgment of the Court and I direct that no further note or transcript be made.

The Hon Mr Justice Neuberger Dated: 21st December 2000

MR. JUSTICE NEUBERGER:

A. INTRODUCTION

1. The defendant, DSM NV ("DSM"), a Dutch company, is the registered proprietor of European Patent No. 0420358 ("the Patent"). In these proceedings, the claimant, Novo Nordisk A/S ("Novo"), a Danish company, seeks revocation of the Patent. In related proceedings, not before the Court, DSM is suing a company which imports Novo's products into this country, for infringement of the Patent.
2. The application for the Patent was filed on 27th September 1990, and it claims a priority date of 27th September 1989. The Patent purports to disclose for the first time the cloning, sequencing and recombinant expression of the gene for fungal phytase.
3. The grounds upon which revocation is sought are that the Patent is invalid on the grounds of:

(1) (a) Obviousness over common general knowledge;

(b) Obviousness over a specific disclosure in a paper published by Abul Ullah in 1988 ("Ullah");

(c) Obviousness over nature;

(2) Insufficiency, essentially arising from what Novo contends is "the extreme breadth of the claims of the Patent";

(3) Added matter, in that the disclosure of the Patent extends beyond that disclosed in the application for the Patent.

4. The scheme of this judgment is as follows. First, an explanation of what phytase is and does. Secondly, the general scientific and technical background. Thirdly, a summary of the relevant parts of the description and claims in the Patent. Fourthly, legal principles. Fifthly, a summary of the witnesses and factual evidence at trial. Sixthly, common general knowledge. Seventhly, obviousness over common general knowledge. Eighthly, obviousness over Ullah. Ninthly, a point on obviousness relating to the research which resulted in the Patent. Tenthly, obviousness over nature. Eleventhly, construction. Twelfthly, insufficiency. Thirteenthly, added matter. Lastly, conclusion.
5. No significant issues over the construction of the Patent arise in relation to the obviousness arguments. Accordingly, I deal with the construction of the Patent before sufficiency and added matter. It is unusual not to consider the construction of a patent in suit before proceeding with arguments of obviousness. However, as Mr Simon Thorley QC (who appears for DSM with Dr Justin Turner) says, it is convenient to deal with obviousness first. This is partly because it does not raise any significant issues on the construction of the Patent, but it is also because obviousness is the main and most fundamental basis for Novo's attack on the validity of the Patent.
   

B. PHYTASE

6. Phosphoros is a chemical element which, in normal conditions, is highly unstable, in that it ignites spontaneously. However, when found in chemical combination with other elements, it is an important mineral in the diet of farm animals. It is found in the soft tissues, membranes and body fluids of all animals. It plays an important role in the transfer of energy, the synthesis of amino acids (of which more below), and the formation of bones. A deficiency of phosphorus can lead to bone malformation, reduction in growth, lowered appetite, and reduced fertility.
7. In plants, phosphorus is stored in substances called phytates. However, to monogastric animals, such as poultry and pigs, phytate does not represent a nutritionally available form of phosphorus. That is because such animals lack the necessary enzyme (i.e., the necessary biological catalyst) which enables phosphorus to be released from phytate. That enzyme is phytase. Due to the inability of monogastric animals to make use of the phosphorus in phytase, it has been necessary to add an alternative source of phosphorus to their diets. This is not necessary in the case of ruminant animals (that is, those with more than one stomach, or with a stomach of more than one compartment) such as cows. They possess micro-organisms which produce phytase, and they are accordingly able to digest phytate.
8. If phytase is added to animal feed, it enables monogastric animals to digest and to metabolise phytate, thus avoiding the necessity of adding extra phosphorus to their feed. The main way in which phosphorus was given to animals until recently was through the addition of either inorganic phosphorus, (that is, inorganic compounds containing phosphorus) or animal by-products, such as bonemeal or bloodmeal, to their feed. However, by the mid-1980s, it was becoming increasingly clear that it was desirable to find an alternative source of phosphorus.
9. First, environmental problems were being caused by the use of phosphates in agriculture. Phosphorus used in this way leaches into rivers and lakes, causing a decrease in the amount of oxygen in the water (known as eutrophication) which leads to algal blooming and death of fish. Indeed, this leaching of phosphorus has caused irreversible pollution in the North Sea. The problem appeared so acute by 1987, that an Intergovernmental conference in London in that year agreed to achieve a reduction of the phosphorus content of the North Sea by approximately 50% in the decade to 1995. In the Netherlands, the government introduced legislation to reduce phosphorus output (and other countries have since followed its lead). The Dutch government also sponsored research into how the problem might be avoided, and the research included that conducted by Gist-Brocades, now a subsidiary of DSM.
10. Secondly, in the late 1980s, it seemed at least possible that the agent responsible for BSE could be spread through the carcasses of infected animals, and that one of the prime sources of this transmission might well be bonemeal. From 1988, the United Kingdom introduced a number of statutory restrictions on the use of such animal products in foodstuffs. This problem meant that, at least in this country, the cheapest sources of phosphorus, namely bloodmeal and bonemeal, were no longer available. The only source was the more expensive inorganic phosphorus, unless phytase was available in sufficiently large quantities and at a reasonable price.
11. Phytase is produced by a number of species of filamentous fungi. One of these is known as Aspergillus, which is one of the most common fungi in the environment. The ability of Aspergillus to produce organic acids was known at the beginning of the twentieth century, and this led to the discovery of phytase. There are more than 200 species in the Aspergillus genus. Each species is classified under a sub-genus, and a section. Thus, the Aspergillus genus has a sub-genus circumdati which itself is divided into sections, including negri, formerly known as Aspergillus niger.
    
    C. TECHNICAL BACKGROUND

Proteins and amino acids

12. Like most enzymes, phytase is a protein. Proteins are large molecules consisting of one or more polypeptides. A polypeptide is a chain of amino acids linked by peptide bonds. There are twenty different amino acids, although some are more common than others. Each amino acid has, at one end, a carboxyl group (—COOH) and, at the other end, an amino group (—NH₂). Each of the twenty
                                              H
                                              |
    amino acids can be written NH2—C—COOH, where R, in relation to that
                                              |
                                              R
    

particular amino acid, is its specific constituent, which varies from one amino acid to another. This specific constituent is known as the side chain of the amino acid, because, when the amino acid is in a polypeptide chain, this constituent is, in effect, a branch off the polypeptide chain. The side chains of amino acids are often divided into four categories, namely (i) acidic hydrophilic (water loving) side chains; (ii) basic hydrophilic side chains; (iii) neutral hydrophilic side chains; (iv) neutral hydrophobic (water hating) side chains.

13. In a polypeptide (just as in a protein), amino acids are linked by a peptide bond, which is formed with the loss of an -OH from the carboxyl group of one amino acid and an -H from the amino group of the other acid, i.e. with the effective loss of a water molecule. Thus, the smallest polypeptide possible (i.e. one with two amino acids) has one peptide bond and is represented thus:

            H  O  H   H
              |   ||  |   |
    NH2—C—C—N—C—COOH.
              |            |
              R₁            R₂

14. Many side chains, especially those which are hydrophilic, are subjected to modification during the process of their formation. An example of such modification is the addition of carbohydrates (or sugar residues), a process known as glycosylation.
15. While glycosylation of polypeptides does not occur in bacteria, it does occur in higher organisms including fungi. Glycosylation can involve addition of a single sugar residue, or a substantial number of sugar residues. Such residues can substantially vary in size, structure, and composition one from the other. There are two common types of glycosylation, the first known as N-glycosylation (normally encountered on side chains of the amino acid asparagine) and the second is called O-glycosylation (normally in side chains of the amino acids threonine or serine). The reference to N and O is a reference to the atom (nitrogen or oxygen respectively) to which the carbohydrate residue attaches. It is unnecessary to understand anything further about glycosylation for the purpose of this case, save that all or most of the sugars in an N-glycosylation can be removed by treatment with an enzyme known as Endo-H.
16. In a polypeptide, there are many (usually more than 100) amino acids joined together by peptide bonds. By convention, the chain is treated as starting at the amino end of the first amino acid, known as the N-terminus, and ending with a carboxyl group of the last amino acid, known as the C-terminus. As I have mentioned, a protein often has a single polypeptide chain, but it sometimes can have a number of polypeptide chains. In either case, the polypeptide only becomes a protein once it is folded into the correct shape. If it loses this shape, it ceases to be effective as a protein: it is denatured.
17. A given protein has a specific sequence of amino acids in its polypeptide chain, and if it consists of more than one chain, each chain has a specific sequence. However, I shall ignore proteins with more than one polypeptide chain, because phytase has only one chain.

The genetic code

18. In order to make a protein, the amino acids have to be assembled in the correct sequence. The sequence for a particular protein is specified by a gene which is present in a cell's genetic material. The expression "genetic code" is the term used to describe the relationship between a sequence of genetic material and the amino acid sequence of the corresponding protein.
19. The genetic material in a cell is made from deoxyribonucleic acid, known as DNA. DNA is made up of a strand or strands of nucleotides. Neither the nature nor the constituents of a nucleotide is germane for the purposes of this case, save that it should be appreciated that each nucleotide includes a nitrogenous base constituent and a sugar constituent. DNA contains only four different types of nucleotide, and each is identified by a different nitrogenous base, namely Adenine, Guanine, Cytosine and Thymine, abbreviated respectively to A, G, C and T. The nucleotide bases in DNA are linked by bonds between a certain carbon atom of a sugar of one nucleotide, known as a 3'-carbon, and another carbon atom, known as the 5'-carbon, of the neighbouring nucleotide. It is not necessary to understand what this means, but suffice to say that a strand of DNA will have one "5'end" and one "3'end".
20. In its normal state, DNA consists of two strands in the famous "double helix" configuration. Each of the two strands of DNA is linked to the other by the invariable rule that A always pairs only with T, and G always pairs only with C.
21. The order of the nitrogenous bases in a particular strand of DNA provides the blueprint for the production of a particular protein. In effect, the DNA blueprint is a succession of code words, each consisting of a set of three nucleotides, known as codons. Each codon (i.e. each set of three nucleotides) specifies a particular amino acid. In effect, a particular codon instructs the cellular machinery to add a particular amino acid to a growing polypeptide chain. However, there are three codons which instruct that machinery where to stop the chain.
22. As I have mentioned, there are twenty different amino acids, and there are four types of nucleotide or nitrogenous base (hereinafter "base") and a codon consists of three bases. Accordingly, there are 64 (or 61, if one ignores "stop" codons) possible permutations of bases to encode 20 amino acids. Two amino acids are encoded by only one group of three bases, i.e. one codon. Some amino acids are encoded by as many as six different groups of bases, or six codons. The majority of amino acids are encoded by between two and four different codons.
23. This feature of the genetic code, namely that (with two exceptions) each amino acid can be encoded by more than one codon, is known as degeneracy.
24. When a cell detects the need for a particular protein, the DNA that codes for that protein unravels from its double helical conformation, and exposes the relevant portion of the DNA. This causes the cell to produce a substance closely related to DNA, called ribonucleic acid, RNA. This is made from a slightly different nucleic acid from DNA, and consists of a single strand rather than a double strand. Further, in place of one of DNA's bases, Thymine, it uses a different base, Uracil: it contains U instead of T. The RNA manufactured in the cell is known as mRNA - i.e. Messenger RNA. mRNA is made by a process called transcription. Transcription involves an enzyme (called RNA polymerase) travelling along the relevant gene making the single stranded RNA with bases complementary to the bases on the DNA. Thus, A is reproduced as U (there being no T in RNA), G is reproduced as C, T is reproduced as A, and C is reproduced as G. At the start and end of each gene there is a promoter sequence and a terminator sequence respectively; the former is responsible for controlling the timing of the transcription and the amount of mRNA produced, and the latter instructs the RNA polymerase to stop copying the DNA.
25. Once the transcription of the mRNA is complete, its code must be translated to the encoded amino acid sequence. This takes place in a complex known as the ribosome, a microscopic structure in the cell. The ribosome effectively reads off the base sequence of the whole of the mRNA and translates it into polypeptide by use of another type of RNA called transfer RNA, tRNA. tRNA contains the appropriate codons for the amino acid residues of the particular protein, and can be joined by specific enzymes to the amino acid which corresponds to the particular codon. Those amino acids are then joined by peptide bonds, with the assistance of an enzyme called peptidyl transferase.
26. A complicating feature is that mRNA is normally shorter than the gene from which it is transcribed. This is because genes have non-coding sequences of bases, or "introns", between the portions which correspond to the portions of the mRNA which encode the protein. These introns are effectively edited out before translation of the mRNA, leaving only those regions of codons in the mRNA which are protein encoding. While not a feature of bacterial DNA, introns are virtually always present in higher organisms, including fungi.
    
    

Genetic engineering

27. The essence of modern DNA technology is the ability to isolate, copy, multiply, and, thereafter, to reproduce (or, to use the technical expression, to "express") individual DNA sequences of interest. A major aim of genetic engineering is to find a method of production of a given protein beyond that which can be found in nature.
28. Individual fragments of DNA can be copied, or, to use the technical, albeit well known, expression, cloned. Cloning technology involves putting together a new DNA sequence and placing it in a micro-organism, via a cloning vector, which is a substance which enables the DNA to be expressed by the host micro-organism (which is normally a bacterium). Cloning technology typically involves (i) the isolation of a DNA sequence which occurs naturally, (ii) combining it with another DNA sequence (known as a cloning vector), creating a so-called recombinant DNA sequence, and (iii) placing that combined sequence (which does not exist in nature) in a host micro-organism.
29. The placing of the DNA fragment in a micro-organism via a cloning vector was facilitated by the discovery of plasmids, which are the commonest type of cloning vector. They are small, discrete, circular portions of DNA, which have the ability of accepting portions of recombinant DNA, and also have the ability of being readily exchangeable between micro-organisms.
30. Owing to the possible existence of introns in a particular gene, direct copies of the preferred gene are not normally used for cloning, because the host micro-organism is usually a bacterium, and, as I have mentioned, they are not equipped to deal with introns. Accordingly, mRNA, transcribed from the target gene and containing no introns, is used. This mRNA is incubated with an enzyme known as reverse transcriptase which creates a reverse transcript of the mRNA into DNA. This type of DNA is known as cDNA, complementary DNA, which can be converted into double stranded DNA. It is those cDNA sequences which are cloned, inserted into hosts, and expressed to produce proteins.
31. Again, the technical details of this aspect do not matter for the purpose of this case. The essential point is that, once the specific gene for a particular protein has been found, the technology exists for inserting the gene into a micro-organism which will then produce the protein, using its normal system for gene expression.
32. In order to achieve this, of course, the particular gene has to be isolated. This can be done by using a DNA probe often known as an oligonucleotide probe. Such a probe is constructed by synthesising a comparatively short single strand of DNA with a base sequence complementary to one of the strands of the DNA sequence which one is trying to isolate. One normally attaches a "marker" (for instance a radioactive atom) in the probe. The process whereby one gets a DNA probe to attach to matching sequences within a comparatively large pool of different sequences is called hybridization.
33. In order to construct the probe, the normal course is to work backwards by obtaining a sample, ideally a pure sample, of the protein concerned and then obtaining at least part of its amino acid sequence. The technique of obtaining a pure sample is a central issue in this case, and I shall deal with it in connection with common general knowledge. Once one has all or some of the amino acid sequence of a protein, one takes part of that sequence (normally a minimum of six consecutive amino acids). From that, one can work out the possible mRNA sequences, and the possible DNA sequences, based on the codons for the amino acids in the sequence. It is then possible to build a short sequence, or a series of sequences, of nucleotides representing part of the DNA. This is a relatively easy task, because the construction of oligonucleotide sequences had been a routine task for some time by the claimed priority date. This short sequence (or series of short sequences) of DNA is the probe (or probes) which will then seek a complementary location on the DNA, which should, at least ideally, be the DNA of the gene which is being sought.
34. The target of the probe or probes will, of course, include the specific gene being sought or probed for. However, it will merely be one of the many (perhaps as many as 3,000) genes in the relevant micro organism. At the time of probing, one does not know which is the gene one is looking for: indeed, that is the whole point of probing. The probe or probes will attack, or "screen" what is known as the DNA library. This library effectively consists of all the DNA present in the micro organism; this DNA is cut, or cleaved, with restriction enzymes, so that it is in fragments. These are enzymes which will cleave only at specific base sequences in the DNA.
35. Probing is by no means always successful. One reason for this is that, as already explained, the DNA code is degenerate. Thus, if one is working from a known amino acid sequence to design a DNA probe, there can be a large number of different combinations of codons for the probe, only one of which may well be correct. For instance, take the case of a comparatively short polypeptide of six amino acids, three of which are represented by three possible codons and the other three of which are represented by four possible codons. In such a case, there would be as many as 1,728 different probes, only one of which may well work. Accordingly, the genetic engineer will seek to identify a sequence of amino acid sequences in the protein which contains a low degeneracy. In other words, he will be looking for a series of amino acids each of which is encoded by relatively few codons, ideally only one codon. The ideal amino acids for this purpose are methionine and tryptophan, each of which is represented by only one codon. Cysteine, tyrosine, asparagine and histidine are also attractive in this connection, because each of them is encoded by only two possible codons.
36. Apart from the problem of degeneracy, another potential problem arises from the conditions of the hybridization (which, it may be recalled, is the process of getting a DNA probe to attach to matching sequences within a large pool of different sequences of DNA fragments). In conditions of what are called low stringency (e.g. at low temperature or relatively high salt concentration) a probe may hybridise with DNA sequences which are not in fact closely matching, or, to use the technical expression, which have little homology. The more stringent the conditions in which hybridization takes place, the greater the required degree of homology between the probe and the target DNA before hybridization takes place.
37. Having, at least hopefully, isolated the gene one is searching for in this way, one then inserts it into a host micro-organism and grows it, with a view to seeing whether the desired protein is thereby produced. There are other ways of identifying that one has got the right gene, but they do not play any part in this case.
38. It is right to mention that there is an alternative approach to probing, and that is the use of antibodies. An antibody is a protein which recognises a foreign molecule, known as an antigen. When an antigen enters the system; the production of an antibody activates the immune system with a view to destroying the antigen. An antibody tends to be specific to a given antigen. Certain types of antibodies, known as monoclonal antibodies, are of particular use to molecular biologists, because they are much more specific than the more frequently encountered polyclonal antibodies.
39. It is common ground that, at any rate as at 1988 or 1989, a molecular biologist (with or without the assistance of a protein biochemist) would have used an oligonucleotide probe, and not the antibody screening technique, as the preferred technique for seeking the phytase gene, once phytase had been purified, and the whole or part of its amino acid sequence determined. However, the specificity of antibodies for antigens has a bearing on one of the issues in this case, and it is for that reason that I mention it.
40. Particularly from a commercial point of view, once the desired gene is identified, it is obviously important that there were certain further steps which had to be carried out. In summary, these steps were DNA sequencing of the gene, attempting to produce the protein with the gene, and using the gene to produce the protein on a commercially viable scale and basis. However, again, it does not appear to me that these aspects of the technology play any part in these proceedings.

D. THE PATENT

41. The description begins by stating that "The present invention relates to the microbial production of phytase" and goes on to set out the "Background of the Invention". This describes the effect and significance of phytase and includes reference to a number of previous published papers, including Ullah. It summarises what Ullah discloses and purports to identify a defect in his research of which Ullah was apparently unaware. In paragraph 19, there is this:

"The present invention addresses the need for an economical procedure for the production of phytase by providing the embodiments characterized in the claims. Thus, the present invention provides a purified and isolated DNA sequence coding for phytase. The isolation and cloning of this phytase encoding DNA sequence has been achieved via the use of specific oligonucleotide probes which were developed especially for the present invention. Preferred DNA sequences encoding phytases are obtainable from fungal sources, especially filamentous fungi of the genus Aspergillus."

42. The next three paragraphs set out two further "object[s] of the present invention", namely the provision of "a vector containing an expression construct which further contains at least one copy of at least one, preferably homologous DNA sequence encoding phytase" and the provision of a "transformant, preferably a microbial host which has been transformed by a vector", namely filamentous fungi of various genera, including Aspergillus. The description goes on to claim that the invention "provides nucleotide sequences encoding proteins exhibiting phytase activity, as well as amino acid sequences of these proteins".
43. The description goes on to explain the 19 different figures forming part of the Patent. Figure 1A purports to show the N-terminal amino acid sequences (that is, the specific amino acids, in the right order, which are found at the NH₂-end of phytase). Figure 1B shows some peptide amino acid sequences, which, again, are a series of amino acids, in the right order, found in polypeptides cleaved from phytase.
44. Figure 6 is described in the Patent as the "Compilation of nucleotide sequences of [certain identified] plasmids... encompassing the chromosomal phytase gene locus". It consists of a very long series of the letters A, C, T, and G, running to some 6,756 separate letters. The explanation states that "the phytase coding region is located from nucleotide position 210 to position 1713". In Figure 6, the introns are identified.
45. Figure 8 has two series of lines. The first is a much shorter version (running to some 1405 letters) of Figure 6. This again is, of course, a combination of letters A, C, G, and T. The second line contains a third of the number of letters which, by reference to a well established convention, represent different amino acids. The explanation of the Figure in the Patent is:

"Nucleotide sequence of the translated region of the phytase cDNA fragment and the derived amino acid sequence of the phytase protein..."

46. After explaining the Figures, the Patent goes on with a "Detailed Description of the Invention". There then follow thirteen Examples. These include, in Example 2, the "Purification and characterization of A. ficuum phytase", which begins with a description of the "phytase activity assay" and goes on to describe the "Purification of A. ficuum phytase". Example 3 is concerned with the "Determination of the amino acid sequence of phytase and design of oligonucleotide probes". Example 4 deals with "Hybridization of genomic blots and genomic libraries with a first set of oligonucleotide probes". Example 5 is "Isolation of the gene encoding phytase, using a second set of oligonucleotide probes". Example 13 is headed "Conservation of the phytase gene in different species". It includes a description of the hybridization exercise carried out in that Example as being:

"performed overnight at low stringency (6 x SSC; 50^oC) with a ³²P-labeled 5'-phytase cDNA fragment (described in Example 8). Blots were washed in 6 x SSC at room temperature and exposed to X-ray for 18 hours."

The results are then set out in Figures 19a and b. which show the consequent Southern blots.

47. Example 8 (referred to in Example 13) states that "four synthetic oligonucleotide primers were designed based on the genomic phytase sequence as presented in Figure 6". It then sets out those probes, each of which is a DNA strand varying in length between 7 and 12 codons (i.e. between 21 and 36 nucleotides), each of which begins with a 5'-end and ends with a 3'-end. Example 8 goes on to describe these four probes briefly.
48. There are seventeen claims, but of central relevance, for the purpose of these proceedings are Claims 1 and 2.
49. Claim 1 is in these terms:

"A DNA sequence encoding a fungal phytase which catalyses the liberation of at least one inorganic phosphate from a myoinositol phosphate, said DNA sequence being selected from the group consisting of

(a) DNA sequences comprising a nucleotide sequence encoding the amino acid sequence as depicted in Figure 8 from position -23 to 444, or from position +1 to 444;

(b) DNA sequences comprising the nucleotide sequence as depicted in Figure 6 or Figure 8; and

(c) DNA sequences hydridizing at low stringency conditions (6 x SSC; 50^o C; overnight; washing in 6 x SSC at room temperature) with a DNA fragment corresponding to a cDNA of the nucleotide sequence depicted in Figure 6 from position 210 to 1129."

50. Claim 2 is as follows:

"A DNA sequence which is related to the DNA sequence of claim 1 by the degeneration of the genetic code."

51. I should also refer to Claims 3 and 4, which are in these terms:

"3. A DNA sequence of claim 1 or 2, characterized in that it is derived from an Aspergillus source.

4. A DNA sequence of claim 3, characterized in that it is derived from an Aspergillus ficuum or an Aspergillus niger source."

E. THE LAW

52. Although there are a number of issues between the parties which involve law or the application of legal principles, there is no substantial dispute between them in relation to these aspects. However, although it is unnecessary to consider them in great detail, it appears to me right to set out the legal principles applicable in this case.

Construction

53. It is, of course, well established that the meaning and effect of a patent has to be construed through the eyes of a person skilled in the relevant art. In some cases, this is extended to a group of persons, each with somewhat different expertise, but skilled in their respective arts.
54. In the present case, it is common ground that the Patent is addressed to a team consisting of a biochemist, and in particular one skilled in protein purification, and a person skilled in the field of molecular biology, with some experience of cloning. There was a dispute as to the degree of skill and experience which this notional team would be expected to have, but if, as Novo contends, it would be at the level of experienced graduates or PhD's, then Novo accepts that they would have a senior and more experienced person who would either be a supervisor or who would, at least, be required to give approval before a particular set of experiments was undertaken.
55. Of course, the fact that a patent is to be construed through the eyes of a relevantly skilled person or persons does not alter the fact that, subject to one point, the construction of the specification is for the Court alone. However, expert evidence is admissible, indeed it is sometimes necessary, even on the issue of construction, for instance to explain technical terms: see per Lindley LJ in Brooks -v- Steele & Currie (1896) 13 RPC 46 at 73 and per Staughton LJ in Glaverbel SA -v- British Coal [1995] RPC 255 at 268.
56. Section 125 of the Patents Act 1977, which incorporates the protocol to Article 69 on the European Convention, gives authoritative guidance as to the more general approach to the construction of a claim in a European patent. Under the Protocol, the Court is enjoined to construe a European patent neither by confining itself to "the strict, literal meaning of the wording used in the claims" nor on the basis that "the claims serve only as a guideline". The interpretation to be arrived at is one that "defin[es] a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties."

Obviousness

57. I now turn to obviousness; a patent is invalid if it can be shown that, as at the claimed priority date, the claimed invention was obvious. In Windsurfing International Inc. -v- Tabur Marine (Great Britain) Ltd [1985] RPC 59 at 73-74, Oliver LJ laid down a structured, four stage, approach to be adopted in answering that question. Those steps are:

(1) Identifying the inventive concept embodied in the patent in suit;

(2) identifying the common general knowledge in the art in question so far as the normally skilled, albeit unimaginative, addressee in the art is concerned;

(3) identifying what, if any differences, exist between the matters cited as being "known or used" and the alleged invention;

(4) deciding whether, viewed without any knowledge of the alleged invention, those differences constitute steps which would have been obvious to the normally skilled, but unimaginative addressee, or whether they required any degree of invention.

58. In a sense, I suppose that it can be said that this four stage approach really involves ending up back where one started, namely with the original issue, embodied in the fourth question. However, I believe that it is appropriate to apply this four stage approach, not merely because it has been approved and applied in a number of previous cases, including in the Court of Appeal. It is also because it ensures that one does not go straight to the question of obviousness by reference to a general impression as to the evidence as a whole. By adopting the structured approach, one ensures that there is a measure of discipline, reasoning and method in one's approach. Indeed, it helps to ensure that there is consistency of approach in different cases involving the issue of obviousness.
59. When considering obviousness, it is technical obviousness, not commercial obviousness, that is relevant: see Hallen Company -v- Brabantia (UK) Ltd [1989] RPC 307 at 327 per Aldous J. In Brugger -v- Medic-Aid Ltd [1996] RPC 635 at 661, Laddie J said:

"First a route may still be an obvious one to try even if it is not possible to be sure that taking it will produce success, or sufficient success to make it commercially worthwhile. ...Secondly, if a particular route is an obvious one to take or try, it is not rendered any less obvious from a technical point of view merely because there are a number, and perhaps a large number, of other obvious routes as well. If a number of obvious routes exist it is more or less inevitable that a skilled worker will try some before others."

60. In Molnlycke AB -v- Procter & Gamble Ltd [1994] RPC 49, Sir Donald Nicholls V-C said at 113:

"In applying the statutory criterion and making these findings [sc. as to alleged obviousness] the court will almost invariably require the assistance of expert evidence. The primary evidence will be that of properly qualified expert witnesses who will say whether or not in their opinions the relevant step would have been obvious to a skilled man having regard to the state of the art. All other evidence is secondary to that primary evidence. ...In the nature of things, the expert witnesses and the court are considering the question of obviousness in the light of hindsight. What with hindsight, seems plain and obvious, often was not so seen at the time. It is for this reason that contemporary events can be of evidential assistance when testing the experts' primary evidence. For instance, many people may have been industriously searching for a solution to the problem for some years without hitting upon the allegedly obvious invention."

61. What constitutes the common general knowledge available or known to those skilled in the art, the test is as laid down in Beloit -v- Valmet (No. 2) [1997] RPC 489 at 494 by Aldous J:

"It has never been easy to differentiate between common general knowledge and that which is known by some. It has become particularly difficult with the modern ability to circulate and retrieve information. ...The notional skilled addressee is the ordinary man who may not have the advantages that some employees of large companies may have. The information in a patent specification is addressed to such a man and must contain sufficient details for him to understand and apply the invention. It will only lack an inventive step if it is obvious to such a man.

It follows that evidence that a fact is known or even well-known to a witness does not establish that that fact forms part of the common general knowledge. Neither does it follow that it will form part of the common general knowledge if it is recorded in a document."

Insufficiency

62. A patent is not valid if it is insufficient - see Section 72(1)(c) of the 1977 Act. In order to be sufficient, the specification of a patent must enable the invention to be performed across the full width of the claim, i.e. to the full extent of the monopoly claimed. It is not enough for a patent merely to disclose one way of performing the invention, while claiming many other ways as well, unless the disclosure within the four corners of the patent allows all those others to be performed. In this connection, see per Lord Hoffmann in Biogen Inc. -v- Medeva plc [1997] RPC 1 at 47-49 and 53.
63. Having observed that "the concept of an enabling disclosure is central to the law of patents", Lord Hoffmann said at 48:

"[T]he specification must enable the invention to be performed to the full extent of the monopoly claimed. If the invention discloses a principle capable of general application, the claims may be in correspondingly general terms. The patentee need not show that he has proved its application in every individual instance. On the other hand, if the claims include a number of discrete methods or products, the patentee must enable the invention to be performed in respect of each of them."

64. A little later (at 50)he said this:

"I think that in concentrating upon the question of whether Professor Murray's invention could, so to speak, deliver the goods across the full width of the patent or priority document, the courts and the E.P.O. allowed their attention to be diverted from what seems to me in this particular case the critical issue. It is not whether the claimed invention could deliver the goods, but whether the claims cover other ways in which they might be delivered: ways which owe nothing to the teaching of the patent or any principle which it disclosed."

65. In Mentor Corporation -v- Hollister Inc. [1993] RPC 7 at 10, Lloyd LJ said that the question as to whether or not the patent was sufficient:

"Involves a question of degree. Disclosure of an invention does not have to be complete in every detail, so that anyone, whether skilled or not, can perform it. Since the specification is addressed to the skilled man, it is sufficient if the addressee can understand the invention as described, and can then perform it. In performing the invention the skilled man does not have to be told what is self-evident, or what is part of common general knowledge, that is to say, what is known to persons versed in the art. But then comes the difficulty. How much else may the skilled man be expected to do for himself?"

66. The decision of the House of Lords in Biogen was recently applied by the Court of Appeal in American Home Products plc -v- Novartis Inc. (unreported, 27th July 2000). The patent in that case disclosed a new use for a drug called Rapamycin, but the patent also purported to cover the use of derivatives of Rapamycin, although it omitted to say how many there were, or even what they were. The skilled reader would have had to find out for himself by trial and error, and the patent was held to be invalid. Citing Mentor, the Court of Appeal held that the addressee of a patent cannot be called on to exercise invention in putting the patent into practice. Nor can he be required to carry out "any prolonged research, inquiry or experiment" or "to make a prolonged study of matters which present some initial difficulty". As was said in Mentor, the specification has to "lead the skilled person necessarily and directly towards success through the evaluation of initial failures".

Added matter

67. The relevant statutory provision is Section 72(1)(d) of the 1977 Act. If the matter disclosed in the specification of the patent extends beyond that disclosed in the applications of the patent as filed, then the patent is invalid.
68. In order to judge a claim of added matter, the Court must decide if, when viewed through the eyes of the relevantly skilled man, any subject matter relevant to the invention has been added (normally by way of addition, but also by way of deletion).
69. In Bonzel -v- Intervention Ltd (No. 3) [1991] RPC 553 at 574, Aldous J described the exercise required of the court in these terms:

"The task of the Court is threefold:

(1) To ascertain through the eyes of the skilled addressee what is disclosed, both explicitly and implicitly in the application.

(2) To do the same in respect of the patent as granted.

(3) To compare the two disclosures and decide whether any subject matter relevant to the invention has been added whether by deletion or addition. The comparison is strict in the sense that subject matter will be added unless such matter is clearly and unambiguously disclosed in the application either explicitly or implicitly."

70. The concept of implicit disclosure by an application, or indeed by a patent, which plays some part in the added matter argument in this case, has been recognised by the Court of Appeal - see AC Edwards Ltd -v- Acme Signs & Displays Ltd [1992] RPC 131.
    

F. THE EVIDENCE

The expert witnesses

71. In light of the fact that the parties have agreed that the Patent is addressed to a team consisting of a person skilled in protein purification and a person skilled in molecular biology and cloning, Novo and DSM each called two expert witnesses, one experienced in each field. There was, unsurprisingly, a degree of overlap in the areas on which each pair of experts gave evidence.
72. Novo's protein purification witness was Dr Robin Leatherbarrow, who has been at the Department of Chemistry, Imperial College since 1984, having been a lecturer in biological chemistry until 1999, when he became a Reader. He has been actively involved in the field of protein chemistry generally since 1980.
73. DSM's protein purification witness was Professor John Findlay of the School of Biochemistry and Molecular Biology at Leeds University. He has held that post since 1990, having been a Reader in protein chemistry for the previous two years. He has been involved in protein chemistry since 1968, but he has also been involved in DNA sequencing, and indeed cloning genes, since the mid-1980s.
74. The expert witness called by Novo in the field of molecular biology was Dr David James Ballance. He has been employed by Delta Biotechnology Limited of Nottingham since 1986, and was appointed its Director of Research and Development four years ago. He took a PhD from Bristol University in 1985 in the field of fungal molecular genetics. He cloned genes using oligonucleotide probes, including the Aspergillus genus, between 1984 and 1986.
75. DSM's molecular biology expert witness was Dr Iain Johnstone who has been working at Glasgow University since he graduated from there in 1982. He obtained his PhD in 1985, on the basis of a study of fungal molecular biology in a fungus of the Aspergillus genus. He worked in the University of California, and then went to work at the Wellcome Unit of Molecular Parasitology in 1987, and he remained working there. His research has relied predominantly on genetic and molecular biology techniques.
76. Although there was a divergence of views between these expert witnesses on the topics upon which they were called to give evidence, I have no doubt that each of them is not only a distinguished and experienced practitioner in his own field, but was giving the truth as he saw it.
    

The factual witnesses

77. Brief evidence was given by Dr Gregg Wallis who carried out hybridization experiments for Novo. There were statements from four other witnesses for Novo whose evidence was not challenged. They mostly related to an experiment carried out by Novo.
78. Experiments were in fact carried out not only by Novo, but also by DSM. Novo's experiments purported to follow the teaching of the Patent, and, in particular, screening with a probe in the stringency conditions for the hybridization and washing exercises as set out in Claim 1(c) on a genomic library prepared from phytal fungi other than Aspergillus ficuum. The experiment carried out by Novo was substantially along the same lines, save that the stringency of the washing conditions was significantly increased. I do not propose to go into the details of the experiments at this stage. I will discuss them when dealing with insufficiency (in Section M of this judgment).
79. I also heard evidence from Dr Peter van Paridon, currently the Business Manager in the Dairy Ingredients Unit of DSM Food Specialities, a Division of DSM. He has a degree in chemistry, and a PhD in biochemistry, from Utrecht University. After a year's post-doctoral research in the same university, he joined Gist Brocades, which is a company now in the DSM group. On joining Gist Brocades, Dr van Paridon immediately started supervising attempts to purify phytase, with a view to cloning the phytase gene. His evidence was very important because it is primarily on the basis of his contribution to the work at Gist Brocades that DSM is maintaining the Patent.
80. When Dr van Paridon joined the Gist Brocades team the project was already under way, but the difficulty which his colleagues appeared to be having was to obtain pure phytase. Various purification techniques had been attempted, including techniques for chromatofocusing and vertical IEF. However, Dr van Paridon had the idea of using a system known as Flat Bed Isoelectric Focusing ("Flat Bed IEF"). To cut a moderately long story short, he said that the use of this system, with the assistance of staining, succeeded in producing pure phytase. The staining he used was Coomassie Brilliant Blue total protein staining ("Coomassie blue staining") and acid phosphatase staining ("phosphatase staining"). Dr van Paridon explained how, having obtained a sample of pure phytase, he and his colleagues at Gist Brocades set about producing an appropriate probe and constructing an appropriate genomic library to be screened by the probe, with a view to identifying, and thereafter sequencing, the phytase gene. [There was also factual evidence from Dr. Beudeker].

G. COMMON GENERAL KNOWLEDGE

81. Before turning to the three specific issues of obviousness, I think it is convenient to explain, perhaps in slightly general terms, the state of common general knowledge in the relevant fields as at the priority date. Although, as I have mentioned, the expert witnesses differed on most of the fundamental issues between the parties, there was, perhaps not surprisingly, a considerable amount of common ground as to what was common general knowledge in this field as at September 1989.
    

Protein chemistry

82. So far as protein chemistry is concerned, the fact that proteins could be broken up, or cleaved, into polypeptides, by certain other chemicals was well known. In particular, it was common general knowledge that such cleaving could be effected by a chemical called cyanogen bromide, CNBr. The fact that CNBr would cleave a protein at the C-terminal of a methionine amino acid was something which could have been gathered from contemporary scientific literature.
83. Once one had a pure sample of a protein or a protein fragment (i.e. a polypeptide), the automatic or manual sequencing of the protein or polypeptide was also known to the relevantly skilled man. In other words, once one had a polypeptide or protein, working out the identity and order of the amino acids in the chain was a well known exercise.
84. There were also various techniques well known to protein biochemists for purifying a protein or a polypeptide, i.e. isolating the desired protein or polypeptide from a mixture of other, unwanted, proteins, polypeptides and/or other substances. These included ion-exchange chromatography (which extended to cationic exchange, anionic exchange, and a less traditional version known as HPLC), gel filtration or chromatography, cellulose phosphate chromatography, affinity chromatography, hydrophobic interaction chromatography, hydroxy apatite chromatography, chromatofocusing, and two types of IEF, namely vertical and column IEF, as well as the more traditional methods of precipitation.
85. Some of these techniques (e.g. gel filtration) work by separating different substances by reference to the molecular size of different substances. Other techniques (e.g. affinity chromatography) rely on the fact that some molecules are more attracted than others to certain substances (e.g. to lectin). Further techniques (e.g. chromatofocusing) separate substances by reference to the fact that molecules of different substances have different charges. In particular, IEF (like chromatofocusing) is a technique whereby molecules move to a location where the pH (i.e. the positively-charged hydrogen ion concentration) is such that the net electric charge of the molecule is neutral, known as the molecule's pI. As molecules of different substances tend to have more or less different electrical charges, this enables molecules of different substances to be separated from each other.
86. If one has a mixture of substances, from which one wishes to isolate a particular protein, then it was general common knowledge (and indeed common sense) that a combination of different purification techniques would normally be appropriate. It would be particularly worthwhile to use a combination of purification methods which rely on different characteristics - e.g. one method which relies on the fact that molecules of different substances tend to be different sizes, combined with another method which relies on the fact that such molecules are differently charged.
87. I have not so far mentioned in this context the third type of IEF, namely Flat Bed IEF. There was a dispute between the witnesses as to the extent to which, at least in 1989, protein biochemists would have regarded Flat Bed IEF, as opposed to the Vertical Slab IEF or Column IEF, as appropriate for the purpose of purifying proteins, as opposed to obtaining samples for analysis.
88. Once one believed that one had a pure, or acceptably pure, sample of a particular protein, it was generally known that one could (indeed should) test its purity by using various techniques. One such technique is know as SDS-PAGE. Its detailed technical characteristics do not matter. It involves a gel which is loaded with a sample of the (hopefully pure) protein. In principle, a pure protein can be expected to result in a single band on the gel. In practice such bands were often fuzzy rather than sharp.
89. Once one has separated out different proteins in this way, one can obtain a sample of the protein by cutting out that part of the gel in which it is isolated. One can then extract the protein from the gel and carry out work with it. Alternatively, in order to identify in grosser terms what is in a particular band, one can use staining reagents. Such a reagent is one which reacts with certain types of substance, but not with others, and ideally, when it so reacts, it produces a colour. Thus, some reagents (such as periodic acid-Schiff reagent, used by Ullah) stain on contact with glycosylated proteins. Accordingly, if one has separated proteins on an SDS-PAGE, treatment of the gel with such a reagent will enable one to discover which of the bands are glycoproteins and, depending on the density of the colouring, the degree to which the protein is glycosylated. If one has a reasonably pure sample, one would normally seek to identify the amino acid sequence at the N-terminal end of the protein. This can only be done once the protein was reasonably pure.
    

Genetic engineering

90. I have concentrated so far on common general knowledge in the field of protein purification. I now turn to common general knowledge in the field of molecular biology, and in particular in the field of genetic engineering. In this connection, the ordinary skilled molecular biologist would know that, in order to isolate a gene of interest, it would be necessary first to construct a DNA library, which would have to be screened with an appropriate oligonucleotide probe. Further, it was known, indeed it is self evident, that the success of such a screening programme would depend upon the quality of the available probes.
91. It was also generally known that the probes should be prepared based on polypeptide sequences which could have been obtained from the relevant protein through the cleaving and purification programmes to which I have referred. Ideally, one would be looking for a low degeneracy polypeptide sequence, which would reduce the number of possible DNA probes which would have to be constructed on the basis of the particular polypeptide sequence. If the N-terminal sequence was not degenerate, it would be a likely sequence on which to base a probe, unless one happened to find a particularly undegenerate internal sequence.
92. It was also known to the ordinary skilled molecular biologist, that, before screening with the probes, a so-called Southern blot experiment (referred to in the description in the Patent) might be carried out on the DNA library, with a view to assessing the likely success of the screening exercise, and with a view to assessing the likely appropriate stringency conditions for the hybridization exercise.
93. The Southern blot works in the following way. The DNA fragments in the library are separated, normally by a technique called electrophoresis, which separates molecules by relying on the principle that differentially charged or sized particles move differentially when subjected to an electric field. Because fragments of DNA tend to be similarly charged, the DNA will thereby be separated by size. The electrophoresis is normally carried out on agarose gel, a naturally occurring polysaccharide found in seaweed. The DNA fragments are then transferred from the gel onto a membrane. The membrane is then dipped into a solution containing the probes for the gene which is being sought. Ideally, at least one of the probes will then hybridise with those gene sequences which are similar enough to form a stable stretch of base-paired DNA. The probes will have a marker, as I have mentioned hence, it should be possible to detect whether any probe has hybridised with any of the DNA on the membrane. Further, in order to help identify what one has got on the membrane, staining techniques may be used.
94. Before attempting hybridization (namely getting a DNA probe to attach to the appropriate matching sequence of DNA in the large pool within the library), it was almost invariable to have a pre-hybridization step. That step included treating the membrane to be used in the hybridization so as to minimise the degree of direct binding of the probe to the membrane. Hybridization would then be carried out in a Southern blot procedure: this was common practice.
95. It was common general knowledge that the degree of hybridization would vary with the stringency conditions. Low stringency conditions would lead to a relatively high number of genes (over and above the gene one was looking for) hybridising with the probe (producing what are called "false positives"). On the other hand, if the stringency conditions were too high, one might lose any track of the gene being looked for. It was also well known that the higher the salt and the lower the temperature the lower the stringency conditions, and vice versa.
96. It was also known to the ordinary skilled man that non-specific binding (i.e. binding of the probe with genes one was not looking for, producing false positives) took place for two reasons. The first was that the probe would tend to stick to the membrane (and it was with a view to reducing this problem that the pre-hybridization step is taken). The second, and probably more important, reason was that the probe would be likely to stick to some extent to other DNA fragments which contained regions of homology to the probe (sometimes known as background binding).
97. After hybridization, particularly if carried out in low stringency conditions, it was common general knowledge that a wash, normally in rather higher stringency conditions than was used in the hybridization, would remove some (but probably not all) of the non-specifically bound DNA, that is, some of the false positives, and in particular those false positives with relatively little homology to the probe. The ordinary skilled man in this field would also know that there was, indeed that there would have to be, a degree of trial and error in varying the stringency conditions in which the hybridization and post-hybridization washing was carried out.
98. There were obvious complications and difficulties involved in the various steps from attempting to purify the protein to seeking to isolate the relevant gene fragment. Accordingly, there was a substantial risk that any particular attempt to identify the gene which encodes a particular protein may not succeed.



H. OBVIOUSNESS OVER COMMON GENERAL KNOWLEDGE

99. Applying the structured, four stage, approach identified in Windsurfing [1985] RPC 59 at 73-74, I turn first to consider the inventive concept which is claimed by DSM to be embodied in the Patent in Suit. As described by Mr Thorley in his closing speech, DSM's case is that the inventive step primarily arose from Dr van Paridon's idea, when all standard methods of purification had failed, of using Flat Bed IEF as a means of isolating pure phytase, together with Coomassie blue and phosphatase staining. In other words, the inventive step contended for by DSM in the present case is in relation to the purification of phytase.
100. I have considered in a little detail what constituted common general knowledge, indeed what was largely agreed in this case to constitute common general knowledge, so far as purification of proteins was concerned in 1989, in the first part of Section G of this judgment.
101. I turn then to what differences exist, or are at least said to exist by DSM, between the matters which were "known or used" to or by the relevantly unimaginative and skilled man in 1989 and the alleged invention. With all due respect to the evidence and arguments advanced on behalf of DSM, I find it hard to identify any significant addition to the common general knowledge and practice in protein purification in 1989 in Dr van Paridon's idea of using Flat Bed IEF together with Coomassie blue staining and phosphatase staining.
102. So far as the use of Flat Bed IEF is concerned, it is clear that, as a technique, Flat Bed IEF was well known in 1989. Indeed, commercially available equipment for Flat Bed IEF was available on the market in 1989, and Dr van Paridon made use of what may have been only one of severally commercially available systems, namely the LKB Multiphor. As I understand it, two points are made on behalf of DSM about the originality of the idea of using Flat Bed IEF for purifying and isolating phytase. First, use of Flat Bed IEF for the purpose of obtaining a purified sample of a protein was unusual, at least according to Professor Findlay. Secondly, the idea of using Flat Bed IEF when chromatofocusing and Vertical Column IEF, which worked substantially according to the same principles as Flat Bed IEF, had failed, was an original and clever notion.
103. Whether one takes those two factors together or separately, neither of them impresses me.
104. I accept that, to most (and possibly all) people skilled in the field of protein purification in 1989, Flat Bed IEF would not have been regarded primarily as a method for obtaining pure protein for further work. Chromatofocusing (which works on substantially the same principle, namely separating proteins, and indeed other substances, by reference to their different pI's) enables larger quantities of protein to be purified, and therefore obtained for further use. Indeed, Dr Ballance agreed that Flat Bed IEF was more frequently used for analytical purposes rather than preparative purposes.
105. Although Flat Bed IEF may not have been the first method of choice relying on differential pI's for separating a protein for further use, I consider that it would nonetheless have been an obvious technique to invoke for that purpose, if other methods, including methods relying on difference in pI's, had failed, for these reasons. First, it was by no means unknown as a method of separating proteins for preparative purposes. Secondly, it was a well known process, and, indeed, was referred to in text books dealing with protein purification and synthesis published in and before 1988 and 1989. Thirdly, it was neither particularly difficult nor particularly time consuming. It was a relatively simple technique which took no more than a day, or at the most two days, to undertake. Indeed, Dr van Paridon said that it "only took a couple of hours" when he carried it out. Fourthly, the necessary equipment, and in particular the LKB Multiphor, was readily available in the market. Fifthly, Dr Leatherbarrow said, and I accept, that it was reputed to be the purification technique which achieved the highest resolution, and in particular it was therefore a better method of purification than chromatofocusing. Indeed, Dr van Paridon described it as achieving "much better [resolution] than the vertical system [IEF]".
106. In his own report, Professor Findlay described IEF as "a very effective preparative method particularly [I interpose, to emphasise "particularly", not "exclusively"] if sequencing is the primary objective". While he also suggested that "some would say" that chromatofocusing produced "even better purity", I consider that the balance of the evidence before me clearly establishes that Flat Bed IEF, produced better purity than chromatofocusing, or, indeed, than Column or Vertical IEF, and that this would have been common general knowledge in 1989. That was the evidence of Dr Ballance, and, indeed, as I have said, it was the effect of the evidence of Dr van Paridon.
107. In these circumstances, it appears to me that it might well be natural to try chromatofocusing (because of its better yield) before considering Flat Bed IEF. However, I do not see why it was original to think of trying, and thereafter to try, Flat Bed IEF, a well known, reasonably speedy form of separating proteins, if chromatofocusing had not achieved the desired level of purification. This is particularly true, given that Flat Bed IEF was more effective at separation, albeit that it was less attractive because of its lower potential yield.
108. Some time at the hearing was given over in evidence as to the extent to which chromatofocusing had become the technique of choice for purifying proteins. I am not sure that that is a particularly helpful exercise in connection with the issue which has to be determined. There is no doubt that Flat Bed IEF was well known as a technique for separating proteins. I accept that it was thought of by many as being particularly appropriate if one was seeking a purified sample of a protein for the purpose of sequencing, rather than a larger quantity for further experimentation. However, as I have mentioned, Flat Bed IEF achieves, I think, a better level of purification than chromatofocusing. Further, as Professor Findlay accepted, once one had separated different proteins in the gel of the IEF system, one could cut out a band of gel containing the desired protein with ease.
109. I do not consider that, because chromatofocusing was the more normal method of purifying proteins (save if the sole purpose of obtaining a sample was for sequencing), to think of another method of purification (normally used for a slightly different purpose) represented an addition to common general knowledge, if that technique is already well known for that purpose. The fact that the Gist Brocades team had used chromatofocusing by the time Dr van Paridon had joined them may have made it a little less obvious to use the Flat Bed IEF technique, but that seems to me to involve starting at the wrong point for the same sort of reasons given by Laddie J in Brugger [1996] RPC 635 at 661. One does not ask whether an idea was original because the person who thought of it was hampered by research already carried out by members of the team which he had just joined. Even if that were a legitimate approach, I am firmly of the view that, although one understands why he feels pleased to have thought of it, Dr van Paridon cannot fairly be said to have added to common general knowledge when he suggested the use of, and thereafter used, Flat Bed IEF for the purpose of purifying phytase. Anyway, the fact that Flat Bed IEF may not have been (indeed, on the evidence, would not have been) in the first rank of purification methods which one might use to isolate phytase, does not render the idea of using it for that purpose inventive.
110. The same reasoning applies to the argument that Dr van Paridon was original when he conceived of Flat Bed IEF when the Gist Brocades team had already used Vertical IEF. Over and above that reasoning, there is evidence which suggests that, although relying on the same principles, Flat Bed IEF would have been known to be likely to achieve better results than Vertical IEF. My view is based partly on the evidence of Dr Ballance and Dr van Paridon. It is also based on material published by the manufacturers of the LKB Multiphor, who have a vested interest in praising Flat Bed IEF. However, there is no reason to think that the claims in that material are unjustified. When purifying proteins in gels, which is what is involved in any IEF system, the sharper the bands the better the purification, and the higher the voltage the more sharply focused the bands are likely to be. A higher voltage can be achieved in Flat Bed IEF than Vertical IEF. Furthermore, better application of samples can be achieved on a horizontal and open, Flat Bed system, than in a closed and vertical system, such as is involved in Vertical IEF. Additionally, efficient cooling is needed for the IEF system, and that is more easily achieved with the Flat Bed system than the Vertical system.
111. In my judgment, therefore, the main aspect of the claimed inventiveness involved in the Patent really fails when considering the third Windsurfer question. If that is wrong, it must a fortiori fail on the fourth question.
112. Whether considering the third or fourth Windsurfer questions, I must not overlook DSM's contention that the staining system invoked by Dr van Paridon was original, in the sense that it represented something over and above common general knowledge, either on its own or when linked to the use of Flat Bed IEF. I am not persuaded that there is anything in this point. Staining was a routine exercise in 1989. In his evidence, Dr van Paridon, when describing the work he carried out to purify phytase, mentioned staining in only a few words, and there is not the slightest suggestion that he regarded it either as critical to his success or as unusual. Indeed, it is not even clear from his evidence that anyone at Gist Brocades acted as a result of the staining. In his cross examination, Dr van Paridon agreed that the phosphatase staining (and the phytase assay which he then carried out) was "of a conventional form simply using phytate as a substrate to look for the liberation of phosphate". Dr van Paridon described it as "a well known assay". The Coomassie blue staining was carried out at an earlier stage, when purity was only around 45%, and it was not suggested that that type of staining was original or, indeed, vital.
113. As Mr David Kitchin QC (who appears for Novo with Mr Richard Meade) says, until a suggestion was made to that effect in Dr van Paridon's evidence, it was never a feature of DSM's case that the phosphatase staining or the Coomassie blue staining was inventive. It was never put to any witness called by Novo, and, other than coming from Dr van Paridon, there was no such suggestion from any witness called by DSM. As if this were not enough, staining is not even mentioned in the Patent.
114. In these circumstances, I have reached the conclusion that the claimed invention was in fact obvious over common general knowledge.
115. In support of its contention that the claimed invention was not obvious over common general knowledge, DSM relied on the perceived need, and consequent commercial desirability, of identifying and expressing the phytase gene for several years before the priority date. Although I see the force of that point, it does not cause me to doubt my conclusion.



I. OBVIOUSNESS OVER ULLAH

Introductory

116. Ullah was a paper published in a journal, which does not appear to have had a very wide circulation, called Preparative Biochemistry, in 1988, before the priority date, by Abul Ullah. He worked at a laboratory in New Orleans, Louisiana. The abstract of the paper begins with this sentence:

"Purified Aspergillus ficuum phytase's partial primary structure and amino acid and sugar composition were elucidated."

117. In the Introduction to Ullah, there are the following statements:

"The enzyme was recently purified to homogeneity and the physicochemical properties determined... The purification and general properties of a different fungal phytase produced by Aspergillus terreus have also been reported."

118. Ullah then went on to describe the "Materials and Methods". He explained how he thought that he obtained the purified phytase; his purification technique included chromatofocusing but not Flat Bed IEF. He also stated that he identified the N-terminal amino acid sequence (that is the identification of the amino acids, in the appropriate order, at the NH₂-end of the phytase protein chain). The paper explained how phytase was cleaved with CNBr, and also identified the results of an analysis aimed at showing the total number of each amino acid in phytase. Ullah then stated that four different peptides were identified. First, there was the N-terminal peptide to which I have referred. There were then three other internal peptides, which I will call UP2, UP3, and UP4 respectively ("UP" being short for "Ullah peptide").
119. Novo's case, supported by its expert witness, based on Ullah, is as follows. First, Ullah claimed that he had managed to obtain a pure sample of phytase. Secondly, there was good reason to accept his claim, partly because of the work he carried out, but also because he was able to sequence the amino acids at the N-terminus, which is not possible, or at least very difficult, unless one has a relatively pure sample of the relevant protein. Thirdly, one particular internal peptide which Ullah had identified and sequenced, namely UP2, was a very attractive peptide from the point of view of identifying the phytase gene, because it had relatively, indeed very, low degeneracy. Fourthly, and accordingly, the relevantly skilled person in the art, who was seeking to identify the phytase gene, and who read Ullah, would appreciate, without having to invoke any inventive step, that he should manufacture, or have manufactured, a selection of probes which contain the various degenerative codons representing the amino acid sequence of UP2, or part of UP2, and thereafter the exercise involved in seeking to identify and express the phytase gene was relatively routine.
120. If the first three propositions upon which this line of argument relies are correct, then I accept that the fourth proposition would follow, and that the Patent would fail on the grounds of obviousness over Ullah. Indeed, if the first three propositions are correct, I do not understand DSM really to challenge that conclusion. As my summary of the evidence of Dr van Paridon indicated, the most important contribution which he appeared to bring to the research being carried out by his colleagues was using Flat Bed IEF as the final, and crucial, purifying technique, whereby pure phytase was obtained. As I have also mentioned in the preceding section of this judgment, that is the most significant inventive step which DSM relies on as justifying this Patent over common general knowledge.
121. On behalf of DSM, two reasons are advanced as to why a skilled molecular biologist, seeking to identify and isolate the phytase gene, would not have set about constructing a probe based upon UP2. The first reason is that he would have had a strong suspicion that, despite the claims in his paper, Ullah had not in fact obtained a pure specimen of phytase, and that, accordingly, UP2, and indeed UP3 and/or UP4, may well be polypeptide sequences from some other protein which was mixed with the phytase, and from which phytase had not in fact been purified. The second reason for suspicion so far as UP2 was concerned would have been that it is a polypeptide sequence which begins with two methionines. DSM contends that this would have caused grave concern to anyone seeking to carry out work, the whole of whose validity rested on UP2. It is said that such concern would have arisen because this was a polypeptide supposedly formed following CNBr cleavage of a protein, and, as I have mentioned, CNBr cleaves proteins and polypeptides at the C-terminal of methionine, and it therefore says that it would appear quite inappropriate, indeed suspicious, that CNBr cleavage should have produced a polypeptide not merely with one, but with two, methionines at the N-terminus.
122. I start with the proposition that, to anyone wanting to isolate the phytase gene in 1988 or 1989, the contents of Ullah, at least if taken at face value, would have appeared very interesting indeed. Ullah claimed to have isolated a pure specimen of phytase, and to have obtained a polypeptide sequence, UP2, which was, owing to its low degeneracy, extremely attractive from the point of view of making probes. Indeed, Dr Ballance, not, I think, a man given to exaggeration, described UP2 as "a molecular biologist's dream".
123. Now that the Patent has revealed, in Figure 8, the amino acid sequence of phytase, it is clear that, whatever doubts the appropriately skilled and informed person reading Ullah may have had about its contents, UP2 is accurate in respect of each of its constituent nineteen consecutive amino acids (including the two initial methionines). That is emphatically not a factor which can be taken into account in favour of Novo's case of obviousness over Ullah. However, it is significant in that it establishes that the construction of a probe or probes based on the assumption that UP2 was an accurate representation of an amino acid sequence in phytase would have been sound. Indeed, one of the four helpful amino acid sequences identified by the Gist Brocades team, (including Dr van Paridon) and set out in Figures 1A and 1B of the Patent, is a polypeptide very similar to UP2, save that it omits the first two methionines of UP2, and contains one or two gaps (representing amino acids which had not been identified by the Gist Brocades team).
124. It is therefore necessary to consider carefully whether either, or a combination, of the two reasons advanced by DSM support its case that a molecularbiologist looking for the phytase gene who had read Ullah would not have carried out research based on Ullah, and in particular would not have constructed probes based upon UP2.

The purity problem

125. I turn first to the concern which DSM contends would arise in relation to the purity of the phytase obtained by Ullah. The points relied on by DSM, and which were regarded to be of particular significance by Professor Findlay, were these. First, Ullah reported that the allegedly pure phytase he obtained exhibited two bands on SDS-PAGE, which would have indicated to a molecular biologist (particularly if he sought the advice of a protein biochemist) that, far from having obtained a pure sample of phytase, there was at least one other substance present in significant proportions. I should mention that the existence of two bands is in fact clearer from a paper published by Ullah and his colleagues around a year earlier than Ullah, but nothing hangs on that). Professor Findlay also explained that the concerns of a molecular biologist would have been reinforced by the degree of purification claimed to have been obtained by Ullah (again in the earlier paper), which was apparently a low 22-fold. He said that there would be concern also about the fact that Ullah had apparently overlooked that one of the polypeptide sequences he identified, UP4, was in fact a polypeptide identified in a paper produced by Ullah and his colleagues in the same year, as being the N-terminal sequence of another protein called acid phosphatase.
126. Impressively though these points were articulated and adhered to, by Professor Findlay in his evidence, I am not persuaded by them. The point about the two bands produced on the SDS-PAGE is very considerably weakened, in my judgment, by the fact that Ullah revealed that treatment with Endo-H (which it will be remembered removes the N-glycosylated sugars) of the protein in the two bands resulted in subsequent SDS-PAGE treatment producing a single band with a lower molecular weight than the two bands. I am satisfied, in particular from the evidence of Dr Leatherbarrow, that a protein purification chemist would have concluded that, particularly in light of various other factors, Ullah had indeed probably purified phytase, but that there were two varieties of phytase, one of which had greater N-glycosylation than the other (hence the two SDS-PAGE variants before treatment with Endo-H) but these two versions of phytase were identical after deglycosylation treatment with Endo-H (hence the single lower molecular weight band on SDS-PAGEing after that treatment). This conclusion would have been consistent with the fact that the higher molecular weight (i.e., on this hypothesis, the more glycosylated) of the two bands before Endo-H treatment was the most darkly stained of the three bands, and the single band after Endo-H treatment was the most lightly stained.
127. There is obvious force in Dr Leatherbarrow's point that it seems inherently unlikely that there were two different proteins before Endo-H treatment, which, after such treatment, coincidentally were so similar in terms of molecular weight that they produced a single band. Dr Ballance and the Gist Brocades team had seen a case of a single protein in fungi producing two bands on SDS-PAGE treatment, probably due to differential glycosylation (the protein in question being glucoamylase). Dr Leatherbarrow had also seen cases of individual proteins which threw up more than one band on SDS-PAGE. These included ribonuclease and more than one instance of immunoglobulins which have more than one form due to differential glycosylation. Although he stuck to his view that the two bands on the SDS-PAGE, before Endo-H treatment, would raise a question in the mind of the person reading Ullah, Professor Findlay accepted that it could be due to two forms of phytase with different degrees of N-glycosylation.
128. Further, as Dr Leatherbarrow explained, there would have to be a double coincidence, if these two bands represented different proteins. On the basis of Ullah's results, the other protein (if there was another protein) appears to have reacted with the same monoclonal antibody. Thus, for there to be two proteins, there would have to be the double coincidence of their both having the same molecular weight after treatment with Endo-H, and both reacting with the same monoclonal antibody. I accept Dr Leatherbarrow's view that the relevantly skilled person reading Ullah in 1988 or 1989, and considering this aspect in depth, would have come to the conclusion that, although it was something to be borne in mind, the existence of two bands on the SDS-PAGE was not a significant factor standing in the way of proceeding to construct probes based on UP2.
129. I consider that this conclusion would have been reinforced by a number of factors. First, Ullah unreservedly reported that he had "purified [phytase] to homogeneity", and the reader would have presumed that this view was approved by the reviewers who approved Ullah for publication. Secondly, Dr van Paridon, in a signed declaration to the European Patent Office in December 1995, stated that he had initially accepted that phytase had been isolated in accordance with the description in Ullah, and in cross examination in this case he confirmed that that was his view on first reading and considering Ullah. Thirdly, it is self-evident from the detailed description in Ullah of the methods he used, that sensible steps were taken to check that purified phytase had been isolated. As I have mentioned, not merely did the two bands on SDS-PAGE collapse into a single band after Endo-H treatment, but the proteins represented by the two bands (like the single collapsed band after Endo-H treatment) were recognised by an appropriately produced monoclonal antibody. Fourthly, the N-terminal of phytase was sequenced, and the evidence before me indicates fairly clearly that it is hardly possible to analyse the N-terminal sequence of a protein unless it is at least 70% pure, and it is routine to require 90% purity for this purpose.
130. The fact that UP4 was a polypeptide already reported as having been obtained from acid phosphatase is not something which Professor Findlay claimed in his report would have been spotted by the appropriately skilled protein chemist reading Ullah in 1988 or 1989. DSM's other expert witness, Dr Johnstone, said it was "unclear whether or not an interested person reading these publications in 1989 would have noticed the similarity in these sequences", sc. the similarity between UP4 in Ullah and the amino acid sequence of the polypeptide in acid phosphatase revealed in the earlier paper. I do not consider this point would have been spotted by a person reading Ullah in 1988 or 1989 who was searching for the phytase gene. The point evidently was not spotted by Ullah, or, it would seem, by anyone working with him. Nor am I impressed with the point about the degree of purification claimed by Ullah. It depends on the degree of purity with which he started.



The two methionines problem

131. I turn then to the second factor which, according to DSM, would have been of grave concern to a person contemplating constructing a probe based on UP2, namely that it is a polypeptide beginning with two methionines, which ought to have been cleaved off by the CNBr treatment. I accept that one would not expect a polypeptide fragment obtained from CNBr cleavage of a protein to commence, at the N-terminal, with a methionine. That is clear, once the point is made and one examines the literature as at 1989 dealing with CNBr cleavage. However, I am not satisfied that the point would in fact have struck a reasonably skilled molecular biologist contemplating using UP2 for the construction of a probe in 1989, even if (which I think would be correct) one assumes that he would have consulted a protein chemist, and would have required approval from a relatively senior molecular biologist (or indeed protein chemist) before proceeding with his work.
132. First, there is no suggestion anywhere in Ullah itself that the author considered the presence of a methionine, let alone two methionines, at the N-terminal of UP2 as worthy of note. Secondly, Professor Findlay did not comment on this aspect in his detailed report prepared for the purpose of these proceedings. Thirdly, although the Gist Brocades team considered Ullah, there is nothing in the working papers they have produced to suggest that they thought the presence of the methionines at the N-terminal of UP2 worthy of doubt or worry. Fourthly, Dr Leatherbarrow said that, when he first read Ullah, the presence of methionine at the N-terminal of UP2 did not strike him as odd. Fifthly, in a paper published by Professor Findlay (in connection with a wholly different protein) and in Gist Brocades' own workings, a polypeptide sequence, containing a methionine residue, was reported without comment as having been produced by CNBr-cleavage of a protein. It is fair to say that Professor Findlay explained, in relation to his paper, that the conditions in which the cleavage was carried out were unusual, and that Gist Brocades' later work showed that the reported methionine was in fact a mistake.
133. Nonetheless, taking into account these various factors, I consider that it is more likely than not that it would not have occurred to a molecular biologist or even a protein chemist, reading Ullah in 1989 and considering whether UP2 would be a sensible basis for constructing a probe, to have doubts about UP2 because it began with two methionines.
134. However, matters do not stop there. Assume that I am wrong and the skilled person reading Ullah did wonder why UP2 begins with two methionines if it was achieved through CNBr cleavage. On the basis of the evidence in this case, there could have been three explanations. The first was that the CNBr cleavage did not go to completion; whether it does so depends, among other things, upon the conformation of the protein (thus, if the protein is folded in such a way as to render a methionine particularly inaccessible, the CNBr may not effect its cleavage) and the level of pH. If that were the explanation, then it would in no way call into question the accuracy of UP2, and that would be the end of any worry so far as someone contemplating constructing a probe based on UP2 was concerned.
135. The second possible explanation was that, as sometimes happens, the two methionines were "miscalls", i.e. the wrong amino acid was identified. It is clear from the expert evidence, and indeed from the work of Gist Brocades, not merely that this can occur in the context of polypeptide purification, but that it did occur in some cases in connection with phytase purification. However, if this was the explanation, and there was no other reason to doubt UP2, one would simply construct a probe based on its amino acid sequences ignoring the first two methionines. Although methionine is a particularly attractive amino acid, because it has no degeneracy, UP2 would still be an attractive basis for a probe on the grounds of comparatively limited degeneracy, even if one ignored the first two methionines.
136. The third possible explanation for the two methionines was the presence of an impurity. In this connection, the presence of the two methionines in no way calls into question the purity of the phytase obtained by Ullah. If there was an impurity present after the final stage of purification, then one would expect the CNBr to have cleaved the impurity in the same way as it cleaves any other protein, namely at the C-terminal of methionine. Accordingly, this explanation would have to be based on an impurity which was permitted or overlooked by Ullah after CNBr cleavage had been completed. It seems to me that this is inherently implausible, and I note that it was not suggested in any of the reports prepared by the four experts for the purpose of this case. Given that there are other explanations which do not involve criticism of the way in which the work described in Ullah was carried out, and given that there is no basis other than the existence of the two methionines in UP2, which even suggest any impurities, I do not consider that this explanation would be one which would strike a reasonably skilled reader as at all likely.

Discussion

137. In my judgment, therefore, particularly bearing in mind that UP2, given its low degeneracy, was a very attractive polypeptide so far as constructing a probe was concerned, I consider that the notional team of a reasonably skilled protein biochemist and a reasonably skilled molecular biologist, setting out in 1988 or 1989 to identify and sequence the phytase gene, would have accepted Ullah's conclusions, and would have used UP2 for that purpose.
138. Dr Johnstone was very dubious about using UP2 as a basis for forming probes. However, his counterpart, the expert molecularbiologist called by Novo, Dr Ballance, was strongly of the view that, in 1988 or 1989, a molecularbiologist wishing to clone the phytase gene would have been prepared, indeed keen, to press ahead with probes based on UP2, not least because of its unusually low degeneracy. He said this:

"I believe it would have been obvious and routine to synthesise probes based on [UP2] and also to take the teachings of Ullah to purify phytase in order to generate additional sequence data, ... by analysing further CNBr-generated peptides."

139. As is probably clear from what has already been said, I prefer the evidence of Dr Ballance. First, it seems to me that Dr Johnstone was over impressed by the concerns which were expressed in relation to Ullah based on the level of purity and the two methionines at the beginning of UP2. Secondly, unlike Dr Ballance, Dr Johnstone appears to have had little, if any, experience in the field of screening genomic libraries with DNA probes based on polypeptides, although it is fair to mention that he has used other hybridization techniques with DNA fragments.
140. That is not quite the end of the matter, however. There are two supporting reasons for reaching this conclusion. First, in research to which I have not so far referred, (namely that of Piddington and others in Gene 133 (1993) pp55-62, carried out between 1990 and 1992 according to the evidence of John Rambosek) two groups of researchers in the United States and Finland carried out work to clone the phytase gene, and did so on the basis of the reliability of Ullah, and, in particular, of UP2. The fact that they were successful is scarcely surprising, because, as I have said, UP2, including the first two methionines, is indeed an accurate sequence of part of the phytase protein. However, as I have also mentioned, that is not the point for the purpose of considering Novo's reliance on Ullah, and in particular on UP2. The point is in 1990 that two groups of scientists proceeded on the basis that UP2 was a reliable peptide sequence in phytase, which obviously provides some support for Novo's case that that is the view which would have been taken in 1988 or 1989.
141. The second point is that, if the reader of Ullah had the sort of doubts which DSM suggests, then it seems to me that it is unlikely that he would simply have put Ullah on one side. Given that Ullah's claims do not appear, to put it at its lowest, unreasonable, and given the very attractive nature of UP2, I consider that any person who wished to construct probes based on UP2, but had doubts about Ullah's results, would have repeated Ullah's work. On the basis of the evidence I have, it seems to me very likely that that work would have resulted in a polypeptide which was either identical to UP2 or (as Gist Brocades appears to have found after carrying out CNBr treatment to phytase) a polypeptide with a sequence very similar to that of UP2, but without the first two methionines. As I have said, in connection with the possibility of the two methionines being a miscall, that would still have produced a polypeptide which was valuable for the purpose of constructing probes, because of its low degeneracy.
142. Mr Thorley, on behalf of DSM, emphasised that a researcher in this field would not be a gambler. To my mind, that point raises as much of a question as it seeks to answer: what does one mean by gambler? In a sense, any research is a gamble, because it involves embarking on a voyage which either is one of uncertainty, in the sense that the researcher has no notion of what he will find, or is one which has a certain destination, but the researcher is nonetheless seeking an end which, almost by definition, no one has so far achieved. In either case, there must be a degree of uncertainty as to whether there will be a result which can be characterised as successful. To that extent, any research is a gamble. Having said that, I accept that any researcher, and anyone funding his research, will want to keep the prospects of failure and the risk of wasting money to a minimum. Equally, as Mr Kitchin says, the degree of acceptable uncertainty or risk of failure will depend upon the attractiveness (including financial attractiveness) of the ultimate goal.
143. In the present case, if a researcher was considering whether to rely on Ullah for the purpose of setting out to identify the phytase gene, he would be aware that a significant part of his work would not be wasted even if Ullah turned out to be unreliable. Thus, he would have to build a DNA library anyway (as Professor Findlay accepted) and that work would have to be done, even if probes based on any of the peptides in Ullah failed. The disadvantage of unsuccessfully using probes based on any of Ullah's peptides would be the time (and consequent cost) involved in that exercise, and, on the evidence I have heard, that would not be particularly extensive, a matter of a few months. The unchallenged evidence called by Novo (which I have summarised in Section B of this judgment) would have indicated that the commercial requirement for phytase by 1988 or 1989 was significant, and that the potential reward of success could be quite substantial.
144. DSM's last stand on Ullah was based on the unchallenged proposition that a person seeking to clone the phytase gene would wish to use probes based on more than one polypeptide. In this case, it is common ground that, if the probes were based on Ullah, the peptide which would be used in addition to UP2 would have been UP3. Unlike UP2, it turns out that UP3 is not, in fact, a peptide sequence found in phytase. Of course, that is not something which would be known to the molecular biologist and/or protein chemist carrying out work based on Ullah in 1989. However, the important point is that the fact that probes based on UP2 would have succeeded, while probes based on UP3 would have failed, would not, in my judgment, have altered the fact that the phytase gene would have been cloned if one had used probes based upon UP2. Dr Johnstone accepted that the gene "could probably [be] clone[d] easily if one screened the relevant [DNA] library with probes based upon UP2". I also accept Dr Ballance's evidence that "if you are using two probes in which you had the utmost confidence, there is still a possibility that you will have clones that do not hybridise to both. It is entirely within the normal experience".
145. Accordingly, it would not be unusual or discouraging if it transpired that one or more probes based on UP2 hybridised with a gene fragment, which was part of the phytase gene, whereas one or more probes based on UP3 either did not hybridise with such a fragment or hybridised with fragments from different genes. If the former happened, one would take the genes which hybridised with the UP2 probes, and that would, on the evidence, fairly easily lead one to the phytase gene. Alternatively, if there were other genes which hybridised with the UP3 probes, while one might initially think that they were part of the fragments of the phytase gene, subsequent analysis would lead one to discard them. However, that would not impinge upon the fact that, with use of the UP2 probes, one would identify the phytase gene. That, again, was consistent also with the evidence of Dr Johnstone.
146. In these circumstances, even if, contrary to my view, the Patent was not obvious over common general knowledge, I consider that it was obvious over Ullah. Given that Ullah was published shortly before the priority date, the contention that there was a demand for the phytase gene sequence is of no relevance.



J. A FURTHER POINT ON OBVIOUSNESS

147. I have so far been proceeding on the assumption that Dr van Paridon's evidence, to the effect that his idea of using Flat Bed IEF represented a breakthrough so far as the Gist Brocades team was concerned, is correct. I am, however, by no means satisfied that it is, in fact, right. Dr Leatherbarrow's evidence, which seemed to me to be convincing, was that, prior to Dr van Paridon joining them, the Gist Brocades team had substantially followed the protocol suggested in a paper presented by Abul Ullah and various people working with him, in Japan in 1987, which involved using chromatofocusing, and which produced two types of phytase. There was a sharp difference of opinion between Dr Leatherbarrow and Professor Findlay as to whether pure phytase was in fact thereby produced. The issue, which I have not found so far necessary to determine, hangs on whether the two bands on SDS-PAGE, which had been reported in Ullah were, as Ullah seems to have thought, and as Dr Leatherbarrow believes, both bands of phytase, albeit differentially glycosylated, or whether they are different proteins, as Professor Findlay thinks.
148. I do not believe that I have to decide this point, having come to the conclusion that the Patent is obvious over common general knowledge and, separately, over Ullah. However, in case it is necessary to form a view on this topic, which is difficult, I have come to the conclusion that Dr Leatherbarrow's view is to be preferred. The two proteins had the same molecular weight after Endo-H treatment. They also had the same pI, in circumstances where, according to Dr Leatherbarrow's unchallenged evidence, a difference of four amino acids on a polypeptide with 444 amino acids is sufficient to produce separate bands with IEF. Further, they each produced a single band on Gist Brocades' experiments, at two different positions (at pH 5.2 and pH 5.4). Further, like Ullah, the Gist Brocades team, before Dr van Paridon joined them, was able to obtain the N-terminal sequence, which requires reasonably good purification, at least 70%, as I have already mentioned.
149. I also note that, according to Dr van Paridon, Gist Brocades had obtained what one now knows were authentic amino acid sequences for phytase from material which had been purified other than through Flat Bed IEF. For a substantial amount of time, indeed until about the time that Dr van Paridon joined the Gist Brocades team, they had not considered using CNBr to cleave the protein they had obtained, with a view to obtaining polypeptide fragments. That course was proposed in June 1988, not by Dr van Paridon, but by Gist Brocades' advisers or co-researchers, known as TNO.
150. Accordingly, albeit very much on the balance of probabilities, I am not in fact satisfied that Dr van Paridon's idea of using Flat Bed IEF was even an essential step in the work carried out by Gist Brocades to purify phytase.
151. At first sight, this conclusion might merely appear to be effectively the same reasoning which led me to the conclusion that the Patent is obviousness over Ullah, but I think it differs from that in two respects. First, in considering obviousness over Ullah, one is directing ones mind to the work Ullah reported. On this point, however, I am considering the work carried out by Gist Brocades, albeit that it appears to have been pretty similar, and indeed had similar results (although their equivalent of UP2 did not begin with the double methionine sequence, and Gist Brocades' work went well beyond purifying phytase and identifying some of its characteristics). Secondly, when considering obviousness over Ullah, one is assessing the effect Ullah would have had on the appropriately skilled person or team as at 1989, (and in particular what view would have been taken about the presence of the two bands). However, I am here considering the correct inference from the existence of the two bands in the Gist Brocades research.



K. OBVIOUSNESS OVER NATURE

152. Obviousness over nature is not part of Novo's pleaded case, but I need not dwell on that. Novo basically contends that the phytase activity of Aspergillus ficuum was known before the priority date, and therefore that it was clear that a DNA sequence which produced phytase was present in its cells.
153. At first sight, this argument raises a nice point, in that it could be said to be raising the contention that what the Patent claims is not so much an invention as a discovery. It could also be said to raise the issue which has been of some public debate, to the effect that it is immoral that "the secret of life" can be patented.
154. However, whatever may be the arguments in terms of morality, benefit to mankind and the public interest, it appears to me that the nature of the claim in the present case is an invention, in the sense of an inventive step, without which it had not been previously possible to get from the phytase activity of Aspergillus ficuum, upon which Novo relies, to coding and making the phytase gene.
155. Subject to the other issues of obviousness which I have considered, and subject to any question of insufficiency and added matter, I consider that that type of claim in the Patent is something which can properly be claimed in a patent, particularly in light of the reasoning of the Court of Appeal in Genentech Inc's Patent [1989] RPC 147 at 203. To hold otherwise would also seem to be inconsistent with the underlying assumption and reasoning of the House of Lords in Biogen [1997] RPC 1.
156. The same point appears to me to dispose of Novo's further argument (in so far as it is still maintained) that the claims of the Patent are no more than mere obvious desiderata.
     
     L. CONSTRUCTION
157. The issues of construction of the Patent that have been raised relate to Claim 1 (c) and Claim 2.
     
     Claim 1(c)
158. So far as the opening three lines, and paragraphs (a) and (b) of Claim 1 are concerned, there is little room for dispute. The reference to "a fungal phytase" is specific in so far as it ties the claim to fungi, but, as Dr Ballance explained, that taxonomic kingdom extends to a large number of organisms, including filamentous fungi (such as the various aspergilli) but also yeast-like fungi, and even slime moulds. Paragraphs (a) and (b) are precise, in that they limit the claim to DNA comprising a sequence of nucleotides as shown in Figures 6 and/or 8 of the Patent.
159. The effect of paragraph (c) is, on any view, to widen the scope of Claim 1 considerably. Relying on the evidence of Dr Ballance, Novo contends (to quote from Mr Kitchin's opening argument) that "the skilled person would think that this [aspect of the] claim is defining a set of further DNA sequences which are sufficiently similar (homologous) to hybridise under the specified conditions; that is to say that those closely related to the DNA sequence will specifically hybridise when other DNA sequences will not".
160. Subject to one point, this construction is not disputed on behalf of DSM. However, DSM emphasises the importance of the word "specifically" in Novo's interpretation (and, indeed, it appears that Novo considers that it is of some significance, because the word is italicised in Mr Kitchin's skeleton argument). The point is that, although Claim 1(c) refers simply to "hybridising", the relevantly skilled addressee would appreciate that this means a binding of the probe to target DNA which is stronger than background binding to non-homologous DNA. This accords with the evidence of both Dr Johnstone and Dr Ballance.
161. Where the parties part company is essentially in relation to the low stringency conditions described in Claim 1(c). Novo's contention is simple: any DNA sequence which hybridises under the conditions there described is within the scope of Claim 1(c), and that is the end of it.
162. Accepting that the stringency conditions described in Claim 1(c) are, as is recorded in the claim, low, DSM contends that the skilled man would read those conditions in conjunction with Example 13 (which has the resultant bands on the Southern blot). Consequently, argues DSM, the skilled person would see that the purpose of paragraph (c) of Claim 1 was to differentiate homologous DNA sequences of fungal phytases from unrelated fungal phytases, and that the purpose of low stringency hybridization conditions was to maximise the sensitivity of the assay, so as to embrace related phytases but to exclude unrelated DNA.
163. Subject to one possible point, I consider that the construction advanced by Novo is to be preferred. The one point is that it appears to me that, in order to fall within Claim 1(c), the DNA must encode for fungal phytase, in light of the fact that the paragraph is governed by the opening three lines of Claim 1. As indicated, I do not think that this limitation was in fact disputed by Novo, but, for the avoidance of doubt, it is right to spell it out.
164. The concept of a "related" phytase, for which DSM contends, is not expressed in Clause 1(c). Accordingly, it would have to be implied. Classically, a term will only be implied into a contract or other document, if certain conditions are satisfied, including a requirement that the term is clear. Having heard the evidence and the argument in this case, I must admit to real uncertainty as to what is meant by a "related" phytase, even accepting the proposition that it would be related to DNA falling within paragraphs (a) and/or (b) of Claim 1. Further, as Mr Kitchin points out, there is no disclosure in the Patent as to what a "related" phytase might be. Quite apart from this, it is right to add that, it is a very long way from obvious that the draftsman, when drafting Claim 1(c) of the Patent, had in mind only "related" DNA (whatever that may mean).
165. It is also argued on behalf of DSM that the skilled person reading the Patent would, if wishing to apply its teaching, increase the stringency of the hybridization conditions. I did not understand that to be a point relating to the construction of the Patent: it was more directed to the issue of sufficiency, and I will deal with it in the next section of this judgment. As Mr Thorley submits, one should not allow the assessment of construction and infringement to become blurred with the assessment of sufficiency.
166. Apart from that, in light of the specific reference to the hybridization, and indeed the washing, conditions in paragraph (c) of Claim 1, I can see no reason to imply any further qualification. In particular, as already mentioned, I can see no justification for reading paragraph (c) of Claim 1 as having the intention of excluding "unrelated DNA", if, by that expression, DSM is intending to suggest that there is some sort of implied qualification over and above those expressed in the first three lines of Claim 1, and the stringency and washing conditions identified in Claim 1(c). If (as I shall have to consider) this means that Claim 1 fails for insufficiency, I do not consider that that would call this construction into question. While I believe that, as with any other legal document, the Court should lean in favour of a construction which validates a patent rather than invalidates it, that rule of construction is only normally of use where there is real doubt as to which of two alternative constructions is correct. So far as Claim 1(c) of the Patent in the present case is concerned, I see no room for implying into Claim 1(c) a term that it is only intended to refer to "related" phytases.
     
     Claim 2
167. I turn then to Claim 2. It is common ground that this claim is concerned with degeneracy, that is, the fact that, with two exceptions (methionine and tryptophan) each amino acid can be coded for by between two and six different codons. DSM contends that, to the relevantly skilled addressee, it would be clear that Claim 2 was intended to be limited to cover all degenerate versions of fungal phytase genes falling within Claim 1. Novo contends that there are no words in Claim 2 which are limited to fungal phytase genes, as opposed to other genes.
168. On this issue, I prefer DSM's case. I accept that, read literally, there is a strong argument for saying that, in order to fall within Claim 2, a gene need only be "related" to a gene falling within Claim 1 in terms of degeneracy; in other words, that the only requisite connection between a Claim 1 gene and a Claim 2 gene is that the DNA sequence in the latter gene can vary in its codons from the former, only to the extent that the variations encode for the same amino acid. In other words, as an example, if the codon at positions 100-102 (because, it may be remembered, a codon consists of three nucleotides) of a Claim 1 gene encodes for the amino acid asparagine, then a gene which is otherwise identical but which has different nucleotides at positions 100-102 would only be within Claim 2 if that codon encodes for asparagine. That is common ground, but there is obviously a powerful argument for saying that there is nothing in Claim 2 which expressly requires a gene otherwise falling within its ambit to be a fungal phytase gene.
169. Nonetheless, in my judgment, it can fairly be said that the whole thrust of the Patent is directed towards fungal phytases, and that the classic requirement of an implied term is satisfied in relation to DSM's construction of Claim 2. In other words, I consider that the appropriately skilled reader of the Patent would regard it as so obvious as to go without saying that Claim 2 is only intended to extend to fungal phytases. In this connection, I consider that, unlike in Claim 1, it is legitimate to take into account the rule of construction that, so far as reasonably possible, one interprets a document in a way which validates, rather than in a way which invalidates. It is self-evident that the description in the Patent is directed towards fungal phytases, and Claim 2 applies potentially to any gene, whether a fungal phytase or not, then there is simply no relevant teaching or assistance in the Patent in relation to a substantial element of Claim 2.
     
     M. SUFFICIENCY
170. Novo's attack on the Patent on the grounds of insufficiency is targeted to Claim 1(c) and Claim 2. I shall deal with the arguments in turn.
     
     Claim 1(c): Introductory
171. So far as Claim 1(c) is concerned, Novo starts with the proposition that its effect is to claim all fungal phytases, provided that they hybridise in the conditions therein stipulated. For the reasons I have just given, I agree with that contention. On the face of it, that is a very wide claim, bearing in mind that there are apparently 100,000 species of fungi now known, and, according to Dr Johnstone, DSM's witness, many of them will produce phytase. The breadth of the claim is scarcely cut down by the hybridization and washing conditions specified in Claim 1(c). The hybridization conditions are both by their nature and by their description in the claim "low", and, unusually, the wash conditions appear to be less, rather than more, stringent.
172. In these circumstances, as Dr Johnstone accepted, "the great majority" of phytase DNA would bind with a probe on hybridization.
173. Support for the view that the hybridization and wash conditions of Claim 1(c) will result in a very substantial number of gene fragments hybridising is to be found in an experiment carried out by Novo. This established that use of a probe prepared in accordance with the teaching of the Patent in the hybridization and subsequent wash conditions identified in Claim 1(c) resulted in the probe binding to a large portion of the DNA in a library prepared from each of a number of various organisms including Aspergillus aculeatus, Aspergillus terreus, Paxillus involutus, and Thermomyces lanuginosus. Southern blots coupled with staining showed a promiscuous degree of binding, at least in the hybridization and washing conditions described in Claim 1(c).
174. It is right to say that Dr Johnstone pointed out a number of criticisms, at least some of which appeared to me to be probably cogent, in relation to this experiment. However, it is unnecessary to evaluate their effect, because the degree of binding revealed by the experiment was said by Dr Johnstone to be what he would have expected, and that this was because of the low stringency conditions. He explained that "the overwhelming majority" of the DNA which appeared to bind in these conditions was not phytase. I should add that Dr Johnstone did not suggest that his criticisms had no effect on the results of the experiments, but he did say that they had "a very marginal effect on the results presented".
175. Dr Johnstone also accepted that the results obtained by the experiment, and the results he would have expected to have been obtained if hybridization with a probe in accordance with the teaching of the Patent was carried out in the conditions described in Claim 1(c), were "not sufficiently significant to take you to the next stage of cloning the gene". On the face of it, this presents DSM with a severe problem so far as the sufficiency of Claim 1(c) is concerned. DSM's answer to the problem, as advanced by Mr Thorley, is that, to put it in summary form, having obtained a potentially phytase-containing fungus, the skilled man would be perfectly competent to select appropriate probes based on the amino acid sequence, optimise the hybridization and wash conditions by a familiar trial and error exercise, and then screen the DNA library with the probes. In particular, argues Mr Thorley, he would not regard himself as bound by the specific disclosures in relation either to the probes or the stringency conditions. He criticises the experiment carried out by Novo on the not unfamiliar ground that, far from it being carried out by a person with a will to make it work, it was being carried out by a person who wished to show that the Patent in Suit was insufficient.
176. Attractively though that argument was presented, I cannot accept it. So far as Novo's experiment is concerned, I accept that the Court should always approach any experiment designed to show that a patent is insufficient (or, indeed, one designed to show that it is sufficient) with care. It is almost inevitable that, however much expert witnesses are enjoined to adopt a neutral stance, the experiments they design (and, even more, the experiments which the parties themselves design or commission) will have a degree of "spin". However, it is not enough for the other party to invite the Court to disregard an experiment on this ground. Furthermore, and crucially in this case, it seems to me that the validity of the experiment was effectively ratified by DSM's expert witness, Dr Johnstone.
177. DSM's more specific point is that the appropriately skilled person (a) will design his probe appropriately for the particular fungal phytase, and (b) will experiment, essentially on a "trial and error" basis, in relation to the hybridization and wash conditions. I do not think that will do.

Claim 1(c): the probe

178. So far as the teaching of the Patent as to the probe is concerned, it is limited to Claim 1(c) and Example 13. Dr Ballance took the view that, in light of the terms of the Patent and common general knowledge, the skilled person would use such a probe with a view to identifying other fungal phytase genes. He explained that the problem with increasing the length of the probe is that it may be no more homologous to the target, and could simply result in a greater number of false positives.
179. Dr Johnstone thought that the appropriately skilled person might use the whole of the disclosed gene as a probe, which would increase his chances of finding a phytase gene, but he accepted that there were advantages and disadvantages in using such a long probe. The longer the probe, the greater the risk of some part of it hybridising with wholly unrelated DNA, but the longer the probe the more likely one might be to isolate the gene one is seeking in high stringency conditions. He described this as "very much... a swings and roundabout technical situation".
180. Dr Johnstone's view was that a skilled worker seeking to implement the teaching of the Patent might well start with a probe in accordance with Claim 1(c), but he would expect such a person to experiment with other probes if such a probe does not work. He said this:

"You may not get there. Until you actually do the experiment, you do not know how long it would take you to get there. I am confident, looking at the level of homology, that the gene of interest would be among one of the more highly hybridising sequences from that genome. There will be others, and you may have to screen through a large number of false positives to get there. I cannot tell how many without doing the experiment."

181. In light of the guidance given in the authorities to which I have referred as to what constitutes sufficiency, and bearing in mind that sufficiency must be satisfied over the whole breadth of the claim, I consider that the evidence of Dr Ballance and Dr Johnstone is fatal to the sufficiency of Claim 1(c). It is clear from this evidence, and from other evidence, that the exercise involved would require the skilled worker to depart from the express teaching of the Patent, to experiment over what may be a long passage of time (which could be well over three months), and, even after that, he may not achieve the desired result.
     
     Claim 1(c): hybridization
182. So far as the hybridization conditions are concerned, the experts agreed that the skilled worker would modify the stringency conditions from those described in the Patent, on the sort of "trial and error" basis I have mentioned. However, the Patent gives no guidance as to how this is to be done: the skilled person has to work it out for himself.
183. In the hybridization conditions described in Claim 1(c), it is clear that the probe will hybridise with many phytase sequences, and a large number of bands are likely to appear. The skilled man has no guidance from the Patent, or indeed from anything else, as to which, if any, of those bands represents phytase.
184. The skilled man presumably is expected to increase the stringency with a view to reducing the number of DNA fragments which hybridise with the probe. However, this could easily have the result of the probe no longer hybridising with the phytase gene. Against that, if he does not increase the hybridization, he will continue to have a large number of false positives. Even on Dr Johnstone's evidence, it seems to me that this aspect of Claim 1(c) presents another insurmountable difficulty for DSM so far as insufficiency is concerned. Dr Johnstone accepted that it was a matter of judgment as to whether, and if so to what extent, to increase stringency conditions. He accepted that the phytase band may not be the strongest band, and that he thought it could take up to six months to obtain the sequence of one phytase gene, and that that sort of exercise would have to be carried out in relation to each species.
185. DSM has itself organised an experiment for the purpose of effectively illustrating how a skilled person, with a will to make the teaching of the Patent work, would set about following its teaching. However, to my mind they did not really answer any of the points which cause me to conclude that Claim 1(c) is insufficient. I accept Dr Ballance's evidence, after considering DSM's experiment, that he would have been very reluctant to construct a DNA library, and to screen it, in light of the results produced by the experiment, and that he would not have given such work to a PhD student. Consistently with Dr Johnstone, as I see it, he thought that the whole exercise could take a substantial amount of time.
186. Quite apart from this, as Mr Kitchin put it in his skeleton argument, "as the stringency of the conditions... is increased so the scope of the claim is narrowed. And the extent to which it is narrowed will depend upon the conditions of stringency adopted". In other words, once one departs from the specific conditions described in Claim 1(c) the claim has variable scope, depending upon the conditions adopted by the person following, or, perhaps more accurately, adapting, its teaching.
187. In these circumstances, in agreement with Mr Kitchin, it appears to me that DSM finds itself in a "squeeze" from which it cannot escape, so far as the sufficiency of Claim 1(c) is concerned. If, as I believe to be correct, Claim 1(c) is to be taken literally, it is extraordinarily broad in its effect and, as Dr Johnstone accepts, unworkable. On the other hand, if one can treat it as being subject to increased stringency conditions to a degree appropriate for a particular fungal phytase probe, then not only does it give no teaching as to what stringency conditions are appropriate, but its interpretation depends upon the stringency conditions selected by the person carrying out the work.
     
     Claim 1(c): conclusions
188. While, of course, each case of alleged insufficiency in connection with a patent has to be judged by reference to the terms of the patent in suit, and the factual and expert evidence, it appears to me that the conclusion I have reached is consistent not only with the general principles enunciated by Lord Hoffmann in Biogen [1997] RPC 1, but also with the more particular reasoning in that case. The Biogen patent related to a DNA sequence for encoding the hepatitis B virus; although it provided a route to the gene, and thence to the virus, it did not disclose the sequence of the virus or of the gene. Although the House of Lords (in agreement with Aldous J) accepted that the inventor, Professor Murray, was indeed inventive, the Biogen patent was nonetheless held to be invalid on grounds of insufficiency. At 51-52, Lord Hoffmann said that:

"[The] excessive breadth [of the Biogen patent] is due, not to the inability of the teaching to produce all the promised results, but to the fact that the same results could be produced by different means. Professor Murray had won a brilliant Napoleonic victory in cutting through the uncertainties which existed in his day to achieve the desired result. But his success did not in my view establish any new principle which his successors had to follow if they were to achieve the same results."

189. Later on the same page he said:

"The metaphor used by one of the witnesses was that before the genome had been sequenced everyone was working in the dark. Professor Murray invented a way of working with the genome in the dark. But he did not switch on the light and once the light was on his method was no longer needed."

As Mr Thorley puts it in argument here, Professor Murray "had enabled getting the material in your hand, but he had not told you what the material was. The only way you could get more was to follow his route again."

190. It seems to me that, at least viewed in the round, the only difference between the Patent here and the Biogen patent, at least in very broad terms, is that the patentee in the present case has "gone the extra mile" in sequencing the protein and sequencing the gene, in respect of which he claims a monopoly (indeed, his claim to monopoly extends to many other phytases and genes, as I have mentioned). I find it hard to accept that what is, and was at the priority date, the comparatively routine exercise of sequencing the amino acids in a protein and the nucleotides in a gene, should make the difference between invalidity, as in Biogen, and validity, as is claimed by DSM here. Of course, as in any other field of law, the fact that there has to be a boundary between success and failure, with no intervening grey area, means, as a matter of logic, that there will be cases close to, and on either side of, the borderline, which must mean that in some cases there is a comparatively small difference between success and failure. Nonetheless, for the reasons I have given, including this analysis of Biogen, it is my view that Claim 1(c) is insufficient.
191. In reaching this conclusion, I have not overlooked the point that it would be unrealistic to confine DSM to the specific amino acid and nucleotide sequences disclosed in Figures 6 and 8 of the Patent. I quite accept that, if the Patent was not invalid because of obviousness, it would (at the very least arguably) not be right to reject it for insufficiency in such a way as it could only be valid if it were so restricted. It would give the Patent no practical or commercial effect. Proteins of a similar nature, such as phytases, can vary subtly from each other - e.g. by a single amino acid or a few amino acids in a long polypeptide sequence. If the Patent in this case had to be limited to the single source - e.g. as in Claims 1(a), 1(b) and 4, it would be of no sensible practical effect. The relevantly skilled person could follow the teaching of the Patent and, with a minor modification (such as changing one amino acid in the polypeptide revealed on the second line of Figure 8) achieve effectively everything the Patent claims with the assistance of its teaching, while not being an infringer.
192. The draftsman of a patent such as the Patent in Suit therefore faces this difficulty: he must not formulate his claim as broadly as Claim 1(c), as he will thereby fall foul of insufficiency; on the other hand, it is pointless to formulate the claims as narrowly as Claims 1(a), 1(b) and 4. There should be a way round this problem, and there may be more than one way. Indeed, by raising his argument based on "related" fungal phytases, Mr Thorley has implicitly suggested one possible route. The draftsman would be faced, I think, with a difficult problem of identifying the "relation" so as to be clear and sufficient, on the one hand, while, on the other hand, putting forward a formula which has commercial efficacy.
193. The problem I have been discussing is one which no doubt will have to be considered in the future in relation to other claims of this sort. No doubt, the law will have to develop on a case by case basis. I suspect that, to some extent at least, each case inevitably will depend on its own facts.
     
     Claim 2
194. So far as Claim 2 is concerned, it must follow that it is insufficient. If any aspect of Claim 1 is insufficient, then it must follow that Claim 2, which is wider than Claim 1, must itself be insufficient. If Claims 1 and 2 failed on this ground alone, it may well be that Claim 4 and even Claim 3, which may well not run into insufficiency problems, could be saved, were it not for my conclusion on obviousness.
     
     N. ADDED MATTER
195. As with the insufficiency arguments, Novo's case concentrates on Claim 1(c) and Claim 2. I shall deal with the argument on Claim 2 first.
     
     Claim 2
196. Claim 2 was added to the Patent after the Application. Novo contends that it constitutes added matter on the basis that, as I have mentioned, it covers DNA sequences which do not hybridise to a probe falling within Claim 1 but which are related to Claim 1 DNA sequences by virtue of degeneracy. This class of DNA sequences was not disclosed in the Application as filed; indeed the concept of degeneracy is not disclosed, or even referred to, in the application.
197. There is no doubt that the appropriately skilled addressee of the Patent would have known about degeneracy. However, Mr Kitchin contends that that is irrelevant, on the basis that obviousness is not the test as to whether or not something is added matter: the test is whether there is anything in the Application as filed to indicate that DSM was intending to extend its claim to a class of degenerate DNA genes as part of its claimed invention. On the other hand, the mere fact that a claim is broadened after the filing of the application does not mean that there is added matter.
198. Not without hesitation, I have come to the conclusion that Claim 2 does not represent added matter. In my judgment, it is an oversimplification to say that, because a claim in a patent is added or amended after the application is filed, there is added matter merely because the addition or amendment relies on technical matter which is not claimed or otherwise referred to in the patent. In general, it may well be that such a proposition is a good general basis from which to start, but I do not consider that it is always the right conclusion. Each case must be looked at on its own facts.
199. Take a rather childish example. The description in a patent may refer to carrying out experiments or other work at a certain acidity, and a claim may be added or amended to refer to the pH conditions in which the claimed invention is to be carried out. I do not think it could seriously be suggested that the addition or amendment in such circumstances amounted to added matter because there was no teaching or reference anywhere in the patent in question to the fact that pH is a measure of acidity. To appreciate that point could be said to involve technical knowledge which is not revealed in the hypothetical patent, but I do not consider that it would therefore amount to added matter.
200. Although degeneracy in the context of the DNA code may be something known to far fewer members of the public than the meaning of pH, it seems to me that in the context of considering those who practice in the fields of protein purification and genetic engineering, degeneracy of the genetic code is something which needs to be explained as little as the meaning of pH. It is taken for granted, and that was as true in 1989 as it is now.
201. Mr Thorley suggests that Claim 2 was in any event unnecessary, because, if the Patent had otherwise been valid, a gene which fell within the ambit of Claim 2 would infringe Claim 1. To put the point another way, he says that the reasonably skilled man would appreciate that, even without Claim 2, the Patent extends to degenerate versions of DNA falling within Claim 1. Little argument was developed as to whether that was correct, and, at least as at present advised, I am inclined to think that it is not. However, the fact that the point could have been seriously advanced, and, indeed, that fact that the point as it appears to me to have real force, tends to support the conclusion, Claim 2 does not represent added matter.
202. Mr Kitchin says that the view that Claim 2 does not represent added matter is difficult to reconcile with two decisions of the EPO Technical Board of Appeal, namely Leland/Lightsource T187/91 [1995] EPOR 199 at 202 and PPG Coating T823/96 [1999] EPOR 417. However, those were cases decided on their own particular facts, and do not cause me to change my view.
     
     Claim 1(c)
203. Finally, I turn to the added matter argument on Claim 1(c). Novo points out, correctly, that there was no explicit claim for a probe in the Application as filed, whereas in Claim 1(c) of the Patent, the probe is identified by the closing part (namely a reference to positions 210 to 1129). In answer to that, DSM rely on Example 13 which, it will be recalled, refers to the "5'-phytase cDNA fragment (described in Example 8)", which takes one to Example 8, which refers to four cDNA fragments. Examples 8 and 13 were included in the Application as filed.
204. Novo attacks DSM's reliance on Examples 8 and 13 in the Application as filed as supporting the contention that the closing part of Claim 1(c) is not added matter, on two grounds.
205. The first ground is that it is not in fact possible to determine from Example 13 which of the two of the four cDNA fragments referred to in Example 8 was intended to be identified. In his cross examination, Dr Johnstone said that it was not clear to him which of the two was intended and that "it could be either". However, almost immediately after that he changed his mind. He came to the conclusion that he could identify which of the two fragments in Example 8 was being referred to, because one was a 5'-phytase fragment and the other was a 3'-phytase fragment, and Example 13 plainly directed one to the 5'-phytase fragment.
206. Although, as Mr Kitchin says, it took Dr Johnstone a little time to get there, and although his final view was never put to Dr Ballance, I have come to the conclusion that it does fairly dispose of Novo's first argument as to why Claim 1(c) is added matter. Dr Johnstone was, as I have mentioned, an honest witness, and in any event, once he had reached his final conclusion, there was no reason for doubting its correctness: his explanation for it seemed convincing, and he was not seriously challenged on it. The fact that there was some delay before he got there does not alter the fact that, having got there, it was clear to him (and indeed was clear to me, and would have been clear to the appropriately skilled person in 1989) that Example 13 in the Application as filed did specifically identify one fragment in Example 8. I am unimpressed by the argument that Dr Ballance did not have an opportunity to deal with the point: if Mr Kitchin had been concerned about that, he could have recalled him.
207. I turn to Novo's second line of attack on the argument that the closing part of Claim 1(c) represents added matter. Accepting that the Application as filed identified a probe through Example 13 and Example 8, Novo says that that probe is not a cDNA fragment "depicted in Figure 6 from position 210 to 1129" (which is what is, of course, specified in Claim 1(c) of the Patent). When Mr Kitchin put to him that the Example 8 fragment identified in Example 13 "will certainly not be anything like that described within Claim 1(c)", Dr Johnstone accepted that it would have approximately 27 bases (or 9 codons) more than the Claim 1(c) probe. Novo also relies on the fact that he also accepted that a probe with the extra bases might hybridise differently from a probe without those extra bases. However, he said that this would only occur "in extreme cases" and then only to a "marginal" extent. Indeed, he described the possibility as "theoretical".
208. DSM argues that, since Figure 6 of the Patent (which was, of course, included in the Application as filed) discloses the whole phytase gene, the skilled man would know that he could use any part of it to construct a probe. Accordingly, DSM argues that the closing part of paragraph (c) of Claim 1 is not so much adding matter, as defining the scope of the claim.
209. I have reached the conclusion that DSM is correct on this aspect of the added matter argument also. I accept that it would have been obvious to the addressee of the Patent as filed that a probe may be constructed from any part of the gene which was disclosed in full in Figure 6 (which was in the Application). Although Dr Johnstone accepted that different sequences may produce different results during hybridization, and that this might lead to a requirement for different conditions of stringency, it appears to me that that goes to insufficiency, not to added matter. Although Dr Johnstone accepted that there could be a difference between the hybridising of the probe identified by Example 13 and a probe as identified in Claim 1(c), it seems to me that the likelihood of that occurring is very slight, on the evidence.
210. In my judgment, the essential point is that, if one reads the description in the Patent, the way in which one constructs a probe (effectively based on the DNA sequence of the gene one is searching for) is made clear. Therefore, it is inherent in the fact that the nucleotide sequence for the phytase gene is set out in Figure 6, that an appropriate probe can be constructed by taking an appropriate sequence of nucleotides within the gene. To my mind, Mr Thorley is correct in his contention that the reference to a specific probe in Claim 1(c) does not add technical information to the Patent: it serves to define the extent of the monopoly claimed. In so far as Dr Ballance's evidence conflicts with that of Dr Johnstone on this issue, I prefer the evidence of Dr Johnstone, but it is fair to say that, as it struck me, there was no significant inconsistency between the two experts' respective views on this topic.
211. Standing back and considering the arguments on Added Matter more broadly, I am fortified in my conclusion that the various points raised by Novo fail. The purpose of the principle that adding matter between the filing of the application and the grant of a patent invalidates the patent is not so much to prevent the broadening or altering of the scope of the claim or claims. That is, in principle, permissible. It is to prevent the patentee adding new teaching to the application. Looking at each of the additions relied on by Novo, it seems to me that, if the Patent were otherwise valid, they would not represent additional teaching to the mind of the appropriately skilled addressee.
     
     O. CONCLUSION
212. In these circumstances, although the attack on the Patent fails on the grounds of obviousness over nature and Claim 1(c) or Claim 2 constituting added matter, the attack succeeds on the grounds of obviousness over common general knowledge, obviousness over Ullah, and insufficiency in relation to Claim 1(c) and Claim 2. Accordingly, I propose to revoke the Patent.