A Letter to the Blog by Julian Davies

More numerous than stars in the firmament....
(Van Gogh's Starry Night)

A number of recent analyses have led to estimates of the number of prokaryotes in the Biosphere. This is currently around 10³¹, but many believe that this is an underestimate. In any event, it is a huge number that makes the US national debt seem miniscule and far surpasses the number of stars and heavenly bodies in outer space. I'd like to point out that since every prokaryote is likely to make at least one bioactive small molecule (for signaling, ion complexing, and other purposes such as antibiotics) we can assume that a significant number of different small molecules (<3000 daltons) are represented.

Antibiotics are the best known class of microbial molecules because we have spent the most time looking for such activities (although quorum sensing autoinducers are sexier and are on a roll at the moment!). The world pharmaceutical industry was essentially founded on the production of antibiotics, but now eschews this important source of therapeutic agents because new compounds are difficult to find and increasingly expensive to develop and bring to market. It is widely considered that all the "low-hanging fruit" have been harvested.

I would like to argue that the industry is "vertically-challenged". The compounds are out there, and with the right approaches can be found and used to treat human, animal and plant diseases. My good friend, Dick Baltz, who is one of the most creative and perceptive scientists in the field, calculated that the low rate of antibiotic discovery from nature is to be expected and that new approaches, such as the metagenomic cloning of biosynthetic pathways from the environment, followed by expression in surrogate hosts, is not going to work (1,2). I am much more optimistic about the future of small molecule research and respectfully disagree with Dick's assessment for the following reasons

1. Most of the estimates that have been made (they don't deserve to be called calculations) are Streptomycete- or Actinomycete- centric. We cannot even guess at the number of Actinomycete genera or strains, but, on average, each one of these beautiful bacteria must have the genetic capacity to make at least 10 distinct bioactive small molecules. This is based on a small number of genomic studies of streptomycetes that show as many as 25 pathways per strain in some cases.
2. The Actinomycete class is but one component of the Actinobacteria family, one of the largest-known phyla in the kingdom Bacteria. The Actinobacteria are very diverse and universally distributed. Of those studied, many have the capacity to produce bioactive small molecules.


3. All estimates of small molecule diversity completely ignore the Gram-negative bacteria (such as the Myxococci and Pseudomonads) and seem to forget that the first small molecules with antibacterial activity to be detected were probably from Gram-positive Bacilli, in the late 19th century. The use of gramicidin and bacitracin pre-dated the use of streptomycin, and are still widely used.
4. Fungi are prolific in bioactive small molecule production; plants, animals and even humans make biologically-active small molecules ( think cationic peptides!).

Streptomycete Moonscape (Davies lab)

My point is that bioactive small molecules are critical to life. It has been suggested that they should be ranked equal in importance with the dogmatic three: DNA, RNA and protein. There must be an enormous number of different small molecules in nature, waiting to developed and used to our advantage. We have to do a better job of finding these intriguing molecules, determining their structures, and developing ways of identifying their beneficial functions. This is an enormous task and, obviously, it cannot be left to the pharmaceutical industry alone. However the industry and government could combine to fund the innovative academic research that is required to make sure that the compounds and their uses are found.

References

1. R. Baltz, Microbe, 3, 1 (2007)
2. R. Baltz, SIM News, 55, 186 (2005)

Julian Davies is Professor Emeritus at the University of British Columbia, Fellow of the Royal Society, and a former President of the American Society for Microbiology.

Comments

[Cesar Sanchez]

I completely agree with Julian Davies' opinions. On the other hand, I understand Dick Baltz's point of view. Perhaps it is just a normal misunderstanding between academia and industry.

We all probably agree that the microbial world (and the biosphere in general) still keeps a huge under-explored chemical space. On the other hand, the new approaches (metagenomics, etc.) are not going to work as long as they remain low-throughput processes. I'm optimistic that technical advances will soon convert these processes into high-throughput techniques.

But, still, I think the central problem is the one commented at the end of Julian's letter: "This is an enormous task and, obviously, it cannot be left to the pharmaceutical industry alone. However the industry and government could combine to fund the innovative academic research that is required to make sure that the compounds and their uses are found". Are our societies (political systems, etc) ready to do this in a near future? Here I'm pessimistic.
*********************************************
Elio's response: Thanks for your comment. I agree with it, generally speaking. I have a feeling that funding decisions for the quest of antimicrobial drugs will be guided in part by future developments in infectious diseases. This is a movable feast, which makes it hard to predict what the need for novel antimicrobials may be in the near future.

[Robert Murray]

Julian Davies is probably correct in assuming that there are a lot of small molecules of antibiotic interest out there. It has been argued that nearly all bacteria live in communities and that it is a rough tough world to them and they have to produce, by accident or make-up or lucky mutations, protective molecules for their own survival or modify their surfaces as armour against aggression. So keep on looking. The real trouble is that it is expensive and rewards come slowly to researcher and industrialist.