It looks like the next generation of GM wonder-plants is under development in Scotland.
About 5% of world crops, some $50 billion worth, are lost due to bacterial disease each year. One of the most common such infections is Pseudomonas syringae which attacks a wide variety of important crops, including tomato, kiwi, peppers, olive, soyabeans and fruit trees. Once the bacteria have gained a foothold in one part of a uniform commercial crop, they spread rapidly through the whole.
Breeding crops for resistance to bacterial disease has had only limited success. Chemical crop protectants are increasingly unpopular with consumers. Treating crops with conventional antibiotics is frowned upon as it fuels antibiotic resistance in human pathogens and compromises our ability to treat diseases.
Genetic engineers have hit on the idea of creating crops which generate 'bacteriocins'.
Bacteriocins are tiny antibiotic proteins produced by bacteria under environmental stress, such as low nutrient availability or over-crowding, to kill off the competition.
All major bacterial species produce bacteriocins: there's an enormous range for genetic engineers to draw from. These anti-microbial agents have a reputation for being natural, harmless, highly specific to the bug targeted, environmentally friendly, 'generally recognized as safe'*, and, since they're expressed in active form, simple and predictable.
*'Generally recognized as safe' is a get-out clause to avoid regulation. It has no scientific basis. It's acceptable for traditionally prepared and consumed foods but has no logical role in novel foods.
In a proof-of-concept study, Glasgow University scientists have shown that GM laboratory plant models expressing a bacteriocin against P. syringae, could provide effective resistance against a range of strains of the pathogen.
There's a long way to go before GM bacteriocin crops are commercialised, and the authors point to numerous knowledge gaps to be filled. For example, how bacteriocins actually kill isn't known, and the effects of bacteriocins on vital microbial ecosystems interlinked with the plant have still to be investigated. Also, there are paradoxical findings to be explained such as a lack of correlation between the number of pathogens infecting the GM plant and the severity of its disease symptoms.
Despite these gaps, the proof-of-concept has been enough to start the process of exploiting the technology: patent applications have been filed and prospective industrial partners have been identified.
It seems our agricultural future could include GM staple crops, such as potatoes and rice, in which a cocktail of bacteriocins is constantly being "expressed in active form at high levels".
GM 'Bt' insecticides were supposed to be highly specific too 
Natural bacteria in a natural, diversified environment, produce bacteriocins as and when required to keep competitors at bay. A whole monoculture constantly expressing these antibiotics, no matter how complex the cocktail and how high the expression, sounds suspiciously like the beginning of another very lucrative, patented GM pesticide treadmill.
Notably, when bacteria produce bacteriocins, they co-produce protective proteins to avoid being poisoned by their own toxin. This suggests they may possess a very good, in-built capacity to develop resistance in an environment of over-exposure to these antibiotics. In other words, a future spiral of increasing quantity and variety of GM bacteriocins is inevitable.
Since bacteriocins are also being developed to preserve food, as a replacement for conventional antibiotic treatment of disease, and for cancer therapy, cocktails of potentially related antibiotics in our food and environment could compromise our ability to treat disease in all sorts of ways.
Note that we're not even mentioning the potential health effects arising from off-target and epigenetic changes in the bacteriocin-laced GM food, nor the paradox of a GM 'cure' for a problem (lack of crop diversity) caused by humans in the first place.
Forewarned is forearmed.
 IS BT REALLY SPECIES-SPECIFIC? - February 2016
 BT TOXINS DON'T ACT IN A VACUUM - February 2016
- William M Rooney, et al., 2019, Engineering bacteriocin-mediated resistance against the plant pathogen Pseudomonas syringae, Plant Biotechnology Journal
- Jane Cassidy, Scots scientists develop antibiotics-producing crops, The National, 6.12.19
- Shih-Chun Yang, et al., May 2014, Antibacterial activities of bacteriocins: application in foods and pharmaceuticals, Frontiers in Microbiology 5