Points to bear in mind
- That the biological coupling of an energetically favourable and unfavourable reaction (I would avoid using the term spontaneous†) is done through a composite reaction involving all the components of the two separate reactions. The reactants and products are the same as in the sum of the separate reactions — so it is valid to calculate the overall free energy change from the separate reactions — but we are talking about a different reaction or serious of reactions that occur on the enzyme. For example in the glutamate synthetase reaction that you mention there are the two reactions shown below. Gamma-glutamyl phosphate can be considered as an activated intermediate to which the free energy of hydrolysis of ATP has been transferred, rather than dissipated in heat, as would occur in the isolated reaction with inorganic phosphate as the product.

- Such coupled reactions are no different from non-coupleld thermodynamically favourable reactions in a cell in that they require an enzyme to catalyse them. They do not proceed spontaneously† because there is an activation energy barrier. If you regard this as “forcing” a reaction, then the answer to the question you pose in your comment to @VonBeche is ‘yes’. However I would describe it as an enzyme having evolved to ‘permit’ or ‘facilitate’ a reaction. It does this by providing binding sites for the substrates and (often) a different reaction pathway that involves a lower activation energy.
So why no reaction Gln + ATP ➞ Glu + NH3 + ADP + Pi ?
This can be regarded in two ways.
(1) Why hasn’t an enzyme evolved to do this sort of thing?
Either there is no selective pressure on an organism for an unneccesary enzyme to evolve, or (in this case) if the organism needs an enzyme to deaminate glutamine one evolves, but this enzyme (glutaminase) doesn’t need to involve ATP as the reaction is already thermodynamically favourable.
(2) Why doesn’t the glutamate synthetase enzyme catalyse this reaction
From your comment, this seems to be your concern. There are three binding sites on the enzyme, one for glutamate, one for ATP and one for ammonia; and I presume you are thinking why can’t glutamine bind to the glutamate site (after all it leaves from there) and react with ATP to form γ-glutamyl phosphate. The answer to this is the chemistry of the reactions shown above. The mechanism of the glutamate synthetase reaction is complex, but at the simplest level the first stage is the nucleophilic attack of the negative acid oxygen (yellow) of glutamate on the γ-phosphate of ATP:
In glutamine there is an amino group instead of the negative oxygen, so the reaction will not occur.
Furthermore, you should be aware that the enzyme is an active participant in the reaction, and the various substrates and products are oriented specifically in relation to residues at the active site of the enzyme to allow the reaction to proceed. To give a flavour of this I reproduce a portion of the summary of a paper by Liaw and Eisenberg on the mechanism of action of glutamine synthetase. I don’t expect you (or anyone else) to digest this — it merely serves to illustrate that enzymes are highly sophisticated machines in which many components participate in producing a specific product.
