dissenting.
*316NEWMAN, Circuit Judge.
I respectfully dissent. There is no suggestion or motivation in the prior art to make the claimed combination, and no suggestion that, if made, the combination would be successful to achieve the highly useful result that was achieved, for the first time, by these inventors.
The invention is directed to the discovery of chemical/ biological compounds that modulate the activity of a G protein-coupled receptor (“GPCR”) on the surface of a cell. The specification teaches that a compound binding to a cell surface GPCR induces a complex cascade of events, including the production of cyclic adenosine monophosphate (“cAMP”) and ultimately produces a change in reporter gene expression. The complexity of this mechanism was oversimplified by the Board, leading the Board to combine, in perfect hindsight, a reference showing that GPCR can cause changes in cAMP, and a reference showing that cAMP can cause changes in reporter gene levels. From these references, the Board adduced the claimed invention. The Board found no evidence of a motivation to combine these references, or any suggestion of an expectation of success in achieving a reliable screening method for compounds that affect GPCR activity.
No such motivation or suggestion can be found in the cited Bonner reference. Bonner describes a particular GPCR receptor that causes changes in cAMP levels, and speculates that such changes “could form the basis of a functional assay” for cAMP, by “standard procedure.” The panel majority overstates that Bonner taught any procedure or “measured cAMP levels,” for Bonner did neither. Sibia accurately points to the cAMP measuring procedures then known, as “cumbersome,” and the statements that such procedures were expected to be replaced by methods that measured bioluminescence or fluorescence indicate that Bonner thus teaches away, not toward, the reporter gene approach of the Sibia invention.
The Sassone-Corsi reference, relied on by the Board as teaching that cAMP can affect reporter genes, contains no suggestion that GPCR receptors can be affected and activity screened with reference to receptor genes. Sassone-Corsi merely explores the relationship between cAMP and reporter gene expression, without any suggestion to apply this teaching to receptor signaling, or that such application would be successful. The majority reasons that because Sassone-Corsi “covers the steps” between Bonner and the Sibia invention, the fact that Sassone-Corsi does not discuss or mention GPCR is “irrelevant.” I cannot agree. It is highly relevant that Sassone-Corsi makes no mention of GPCR activity as related to reporter genes, and it is highly relevant that Bonner makes no mention of possible use of reporter genes to indicate GPCR activity. No reference contains any motivation or suggestion to combine these teachings; that comes only from PTO hindsight using the Sibia disclosure as the template.
Sibia points out that Sassone is a mechanistic study in which cAMP is shown to be responsive to extracellularly applied forskolin, and does not contain any suggestion that receptor-mediated cAMP operating as a secondary messenger could be produced at a sufficient level or be correctly localized so as to function as a signal and thereby be used in a screening method for a large number of candidate molecules. That discovery came from Sibia.
Finally, the Evans reference is clearly inapposite, for it relates to a completely different receptor technology, that of an intracellular receptor. The Board found that this difference is not “significant”; however, this conclusion is without sup*317port, as against Sibia’s overwhelming showing. The structural and functional differences between intracellular and GPCR receptors are striking. The intracellular steroid receptors in Evans have a DNA binding domain but no trans-membrane domain, whereas GPCRs have a trans-membrane domain but no DNA binding domain. The intracellular steroid receptors localize and form a complex in the nucleus or cytoplasm of the cell, whereas the GPCRs localize and form a complex at the cell surface. The steroid receptors bind directly to the promoter of the reporter gene being monitored, whereas the GPCRs induce a complex sequence of events that ultimately results in a change in reporter gene expression — any missing link in this sequence may result in nonresponsiveness of the reporter gene. Sibia well demonstrated that persons experienced in this field would recognize the large differences between the receptor studies in Evans and the cell surface receptor mechanisms of the claimed invention, and the absence of predictable interchangeability of methodologies.
In this complex case of biological advance, the Board’s decision is not supported by substantial evidence, and should be reversed. From the panel majority’s contrary holding, I must respectfully dissent.