Creating an Spontaneous Biological Computational Network from a Bacterial Biofilm: Part 1
It’s well known that when bacteria stick to a biological surface like surface of your teeth or to a non-biological surface, like rocks on a river bed, or medical equipment, they can form a complex layer of sticky sugar called a biofilm.
Did you also know that the bacteria within the biofilm also communicate with each other through chemical signals. And where there is communication between entities that can be controlled, there can be computation. And where there is computation, there can be artificial intelligence!
Why would you need to create an A.I. or do any computation from a colony of bacteria though? We have silicon computers to do that right?
Note: I’m thankful to live in the modern day and age when all this information is freely available and composable, thanks to the efforts of countless others who came before me. All relevant links to primary and secondary sources have been posted on the term, and for any that sources I have missed citing, I will work to update this to include those references.
Disclaimer: None of the literature contained in this article is medical advice or to be taken as medical advice. This article is for informational purposes and potentially subject to errors which I’m constantly working to correct. Consult a physician if you have a medical issue.
Well it’s already been shown that bacteria-logical computers can be used to solve so called NP-Hard problems in a much faster time than silicon computers can! This is due to their ability to replicate and spread partial solutions to a problem in parallel. This biological parallel processing becomes exponential and thus increase their ability to compute a problem much faster than a single silicon machine can.
Remember that bacteria are all over the planet as well. They are literally everywhere. If you rub your finger against the surface of a non-disinfected table, your finger will probably pick up millions of of bacteria. If you think of each bacteria as a computational node, you’ve just with a simple gesture run across a million nodes. Back of the envelope estimations demonstrate that there 5³⁰ bacteria on the planet at any given time. That’s a ludicrous number of computational nodes — one that we will never hope to match until the far future era where we can create ubiquitous molecular/atomic computation. Since there are 1.3 x 10⁵⁰ atoms on earth alone that figure amounts to an unfathomable number of computational nodes. Let’s not get ahead of ourselves!
Biofilms
Before we get into a hypothetical bacterial computation using biofilms, let’s dig into biofilms a bit more by talking about the its structure, and how bacteria work together to symbiotically survive in the confines of a biofilm.
A biofilm at it’s basic level is a large slimy mass of exopolysaccharide (sugar) that bacteria excrete in colonies. Each bacterium excretes a small amount of EPS as a capsule around itself which helps it adhere to surfaces.
Without the sticky capsule, a bacteria being so light, could easily be jostled off of its surfaces. With enough bacteria and enough slime, the biofilm congeals together creating a blob with a trillion or more microorganisms inside.
Biofilms form on virtually every non-sterile surface. They are known for forming on
- Lab equipment that hasn’t been cleaned properly
- Dental plaque that forms on teeth at night. Fun fact, the dental plaque biofilm can contain 700 different species and strains of bacteria. There is no requirement for a biofilm to be uni-species in nature
- At the interfaces of needles that have entered the skin through punctures and medical catheters that have been inserted into natural orifices for treatment of diseases.
- The scum that forms on the top of rocks that lie on the bottom of river beds
- Colonies of symbiotic bioluminescent bacteria that give nightlight properties to larger host organisms, like certain species of squid that intern kept the biofilm nourished
Many biofilms you see in nature, when small are harmless. Still other biofilms are virtual fortresses for bacteria that are causing potentially harm. These biofilms are so thick that antibiotics like penicillin or methicillin can’t get through them. This allows the bacteria to keep growing and propagating unchecked as long as a nutrient supply is is present. A well known dangerous biofilm protects the Staph Aureus bacteria when it becomes infectious and virulent causing many different conditions by releasing a multitude of exotoxins.
Bacterial Communication In Biofilms
Bacteria normally release various types of molecules when alone spontaneously. When in close proximity to other bacteria, for example in the context of a biofilm, these excreted hormone like molecules can have signaling properties.
These class of molecules are called auto-inducers because they induce a trigger a cascade of chemical events in the same organism that created them. In sufficient concentration, which occur when the bacteria excrete enough of these molecules, the bacteria are able to express certain genes which activate particular traits when these molecules bind to certain receptors on the exterior wall of the bacteria.
In other words, the bacteria under go a process called “quorum sensing”. The bacteria can also be deactivated as well when the concentration of the bacteria reduces, causing the trait to be turned off because the auto-inducer concentration has also reduced. This is where the idea of “quorum” comes into play because a quorum of bacteria are needed to activate the trait.
Both gram negative and gram positive bacteria use quorum sensing to activate a community trait, but they use very different mechanisms to do this.
Gram Negative Communication
The main family of auto-inducer for gram negative bacteria is called an N-Acyl-Homoserine Lactone or AHL.
These molecules are produced in the bacteria by an enzyme called AHL Synthase that exists in the cytoplasm of the bacteria. When they get excreted out of the bacterial cell, they free float in the environment.
When enough of them are present, due to a large population of the producer bacteria, then the AHL molecules bind to bacterial outer-membrane receptors. If enough receptors are bound, the bacteria will bring them into the cell and have them shuttled to the bacterial chromosome where they are bound to the promotor regions of DNA for the transcription of specific proteins, that lead to a bacterial ‘community function’
There are many kinds of auto-induction processes in gram-negative bacteria, all driven by different kinds of AHL molecules. The HL group is bonded to an R’ group, which has a different chemical structure based on the bacteria that uses the specific HL. The wide variety of molecular attachments means that AHL can be bacterial species specific: if a bacterial receptor makes contact with the wrong type of AHL on a different species of bacteria, it wont be activated, which prevents the auto induction cascade.
Gram Positive Communication
The mechanism of quorum sensing in utilizes peptides in gram positive bacteria rather than Homoserine-Lactone as auto-inducers, partially due to the structure of the cell wall, which if you recall from my previous articles, or your science knowledge in general, consists of no outer membrane, but rather a super thick peptidoglycan layer.
First The peptides are encoded in genes and made as precursor proteins. Next a processing system cleaves portions of the proteins away to make them structurally appropriate to be secreted out a the gram positive cell, and past the cell wall. The peptide secretion machinery is very complex compared to the method of secretion of Acyl-Homoserine Lactones from Gram Negative bacteria, and part of the reason for this is that AHL molecules have a much easier time diffusing in and out of the gram negative cell’s inner membrane, outer membrane, and much thinner peptidoglycan layers.
Finally the peptides that are free floating in the environment encounter sensor transmembrane enzymes embedded in the inner phospholipid layer of the bacteria called sensor kinases. When enough of the peptides have interacted with enough of the sensor proteins, then a transcription factor enzyme pulls the peptide down into the cell and places it on a promoter region of the DNA which allows the bacteria to express the ‘community trait’. The molecules and enzymes involved are different but the end results are the same as in the Gram Negative case.
Example of AutoInduction
A neat example of quorum sensing via auto-inducer activity is as follows: glowing bacteria can express their bioluminescence gene when there are enough bacteria present in the biofilm that they reside in. This has specific uses for some types of nocturnal squid that increase their bacterial concentrations at night when they come out to hunt for prey. By consuming more nutrients, and supplying the nutrients to the bacteria to cause them to grow in an internal biofilm, the bacteria produce enough auto-inducers to grow, causing the squid itself to glow at night and use that light to hide its own shadow that’s produced by star and moonlight.
The phenomenon of quorum sense was first discovered in 1970 by Woody Hastings and his team who were studying sea-born photo-bacteria that have an enzyme called Luciferase which activate their ability to glow. Literally called Luciferase! Sounds sinister, but it’s just nature!
Typical uses for auto inducers by bacteria are:
- Virulence factor production - when enough bacteria are present, they will collaborate to form stronger exotoxins. This is actually an insidious strategy to sneak up on the immune system which would easily target and destroy individual bacteria that were detected to be harmful, but be overwhelmed by large amounts of bacteria. So the bacteria active their attack when they are legion
- Biofilm expansion - Sometimes the bacterial colonies know they are under threat due to an external factor, so they will form even more biofilm when enough of them are present so they can ward off the threat more effectively.
- General Cell to Cell communication within a species: and community activation of “socially coordinated” traits in bacteria e.g. glowing when in a colony and not alone
- General Cell to Cell communication across species: Bacteria typically have auto-inducers to communicate with members of the same species, and the chemical structure of the auto-inducer (AHL) as we discussed, is usually specific to the particular species of bacteria. But many bacteria have a secondary quorum sensing system that allow for interspecies communication. Bacteria likely use this system to detect the presence of competitor strains, which allow them to trigger warnings, defenses, or even symbiotic signal factors which drive interspecies collaboration
Cellular and Bacterial Computation
At its basically level, a standard silicon based computer works by flipping bits on and off via simple electronic gateways called transistors. This should probably start to ring some bells when we look at the quorum sensing mechanisms described above.
Now before we talk further about bacterial computation, I wanted to make a quick reference to eukaryotic computation. Eukaryotic organisms like single celled amoebas and complex multicellular organisms are far more advanced than bacteria, evolutionarily speaking. Thus, it stands to reason they have, and are apart of much more complex signaling machinery and systems than colonies of bacteria are.
It’s even speculated that bacterial biofilms are the prehistoric precursors to multicellular tissues. They behave similarly in terms of encapsulating multiple cellular units (the bacteria) and allowing for the creation of nutritional channels to supply water and essenetial nutrients directly to the bacteria, much like Eukaryotic tissue cells induce angiogenesis to route blood vessels to them in multicellular bodies.
Take for instance one of the most famous systems is the human immune system, whose immune cells communicate almost exclusively through the use of cytokines and hormones. These signaling molecules control the immune system’s response in coordination with the nervous system in a hyper complex biological and chemical computational network.
While the chemical computational potential for multicellular organisms is widely studied and speculated on, bacterial computation interests me because bacteria are prokaryotic organisms that are ubiquitous in nature and found almost everywhere, including extreme environment. Bacteria also transfer capability, like antibiotic resistance, to each other very easily through conjugation and plasmid transduction.
This ubiquitous distribution gives rise the potential to create adhoc networks virtually anywhere on earth. Just pass the correct plasmid “programs” to bacteria and let them do all the work! Their quorum sensing capabilities allow them to have a switching capacity much like a transistor, and because different species of bacteria can activate each other through cross-induction quorum sensing, the ability to create heterogenous “bacterial circuitry” also exists.
The economics of this spontaneous dynamic, genetically engineered, multi-spanning bacterial computer are also attractive in that bacteria replicate very easily, and at a very low cost of a few environmental peptides and amino acids which are presumably available enough for them to grow anyway. The biofilms they form give them natural containers and help them maintain the chemical and population environments to help sustain and grow the “computer”.
Of course it’s far from that simple, but the principle of ubiquity and exponential growth of computational power are limitless.
Imaginative Uses for Spontaneous Biofilm Computers
- Bacterial Internet!
- Bacterial Spy Networks!
- Bacterial Sensing Networks for Environmental Changes!
- Bacterial AI/Machine Learning Networks!
- Bacterial space exploration networks — yes because extremophile bacteria that can live unhindered in a vaccuum present an unprecedented opportunity to study the extraterrestrerium at a far lower cost than mechanical Rovers and Explorers
- Bacterial Tiktok! Just Kidding…or am I?
Conclusions and Next Steps
This was a high level introduction to how quorum sensing works in bacteria and touches on some of the concepts needed to take advantage of quorum sensing through genetic engineering to create human utilizable biological computational machinery. This is one of the holy grails of synthetic biology.
Next time I’ll delve deeper into some actual chemical and biological mechanism that can be used in such computers, and some examples of the research being done by universities and labs to help usher forth this new and radical age of biological computation.
