Application of the Scientific Method: Antibiotic Resistance

In this thread we will be dealing with the subject of how antibiotic resistance appears in bacteria in a specific experimental setup. What I would like to do is have the community come up with hypotheses at each step, and possible ways of testing those hypotheses with experiments. IOW, how can we apply the scientific method as it relates to the emergence of antibiotic resistance. Many of you are already familiar with this subject, so don't give away the punch line too early.We start with a simple observation in a laboratory setting. I have a standard strain of E. coli that I streak on a plate to get an isolated colony. This is a standard technique that spreads out bacteria so that a single bacterium gives rise to several hundred thousand bacteria. This allows us to start our experiment with a single, founding bacterium that is the single ancestor of the rest of the bacteria used in the experiment. This is an important part of the experiment, so don't forget it.We use this isolated colony to seed a liquid culture which produces trillions of bacteria from the hundreds of thousands found in the original colony that was itself founded by a single bacterium. We then spread some of that liquid culture onto two agar plates, one containing an antibiotic and the other one without antibiotic. We incubate it overnight and come back the next day to find a handful of colonies growing on the antibiotic plate, each of which was founded by a single bacterium. We compare this antibiotic plate to another that does not contain antibiotic and we find that the no antibiotic plate is overrun with bacteria without any isolated colonies. Obviously, only a handful of the billions of bacteria put on the antibiotic plate were antibiotic resistant.So how did this antibiotic resistance come about? What are your hypotheses?Suggested Forums: "Is It Science?" or "Biological Evolution"Edited by Taq, : No reason given.

Just as a clarification, I will be drawing from experimental results found in real peer reviewed scientific papers. However, I may not cite them in the early going because I don't want to give the punch line away. Rather, I would like this discussion to be about the process of figuring out how antibiotic resistance emerges in these experiments. So I am asking for a bit of faith on the part of the participants, but I promise to show you the papers once the discussion hits the proper stage. What I hope to do is allow people to understand the thought process and the historical path that science took where it concerns evolution and antibiotic resistance.

Bacteria have junk DNA as well. They use the same basic genetic and metabolic systems we do. I really don't see why this is a problem. That is certainly one hypothesis we can test. It is also one of the hypotheses that scientists seriously considered when faced with the observations found in the opening post.So let's develop that idea a little bit. To be "on cue" it would appear to me that the mutations should be a reaction to the presence of antibiotics. That is, we should be able to tie the introduction of antibiotics to the appearance of the mutation. In fact, we don't even have to say that it is a mutation. This could just be a hidden phenotype that is turned on by the presence of the antibiotic.Does this sound something like the hypothesis you would put forward?

This is all tangential to the real meat of the thread, so I will just comment briefly. I really don't see how having less junk DNA gives them more genetic possiblities. Their genomes are usually around 2 or 3 million bases compared to our 3 billion. Their gene count is also well below ours. Heredity is not alien. It is the same as ours, just with one copy of DNA instead of two. I really don't understand what your protestations are all about. The same way that it is hidden in animal species. Bacteria, just like animals, can change gene expression in response to a changing environment. They can alter their phenotype in response to physical cues just like animals do. So what we need to do is determine if the mutation conferring antibiotic resistance occurs before or after the bacteria are exposed to antibiotics, correct?

It is called phenotype plasticity, and it is a well understood phenomenon in biology. I'm not trying to sneak anything past you, I promise.Also, it is entirely possible that the presence of an antibiotic will turn on a resistance gene so that the resistance is not due to a mutation, wouldn't you agree? This isn't the case at all. Think of how you tan in the summer. This is a change in phenotype as a result of changing gene expression in response to environmental cues. The sun causes DNA damage in your skin cells. The cells respond by upregulating the production of melanin causing a change in your phenotype.Bacteria can also respond to environmental cues, even including the production of multicellular fruiting bodies in the case of the myxococcus bacterial species. The way in which humans communicate does often indicate a teleologic purpose when in fact there is none. Scientists are as guilty of this as anyone. The fact of the matter is that these are very mechanistic systems without any overarching intelligence to them. Also, each species has a set number of responses to a set number of stimuli, that is if we ignore the claim that mutations increase genetic diversity. In the view of biologists, that is the very definition of randomness. If a mutation occurs when there is no need for it then this is a random mutation with respect to fitness. This is contrasted with non-random mutations which are strongly guided by what the organism needs at that moment. So in our current system, a random mutation is one that occurs in the absence of antibiotics and a guided or non-random mutation is one that occurs only when the organism needs it. Does that sound good?

That is certainly a viable hypothesis worth testing. We would seem to have three possible hypothese to test at this point.1. Random mutations where the antibiotic resistance is due to a mutation that appears in the absence of antibiotics.2. Non-random mutations where the antibiotic resistance is due to a mutation that appears in response to the presence of antibiotics and is guided by the specific needs of the organism.3. The induction of a antibiotic resistance gene that is able to save the bacteria in rare cases. The resistance is innate, but the chances that it works are low.Before we move on, we should probably give other people a chance to add to these hypotheses or modify them.

The short answer is that E. coli don't have legs. They stay right where they are born (or close to it). After some time they use up all of the food in that immediate area causing replication to slow down. This limits the overall size of the colony. Even on plates without antibiotics you will still get small round colonies that look like this: As an aside, there are bacteria that use "slime" or flagella to spread over a plate, but E. coli lack this ability so they make nice separate colonies if the progenitors are far enough apart.

Moving on from our hypotheses, we need to get a better feel for the tools that we can use in our experiments. The advantages of using bacteria like E. coli is that they are clonal organisms, they have a very short generation time, and they are easily cultivated. So let's put these characteristics to work.First, we need to go back the opening post. As I stressed, the entire experiment started with a single bacterium that formed a single colony. What we do next is transfer this single colony to some liquid culture and let them multiply overnight. This will give us a ton of bacteria to work with, and they will still all share a common ancestor. However, these bacteria have been accumulating mutations as they replicate so we will have a genetically diverse population over these many generations that have occurred since we founded that original colony with a single bacterium.As noted in the post above, E. coli don't have legs. When a founding bacterium is placed on a plate the descendants stay right next to the founder. Think of it like the children of farmers using the land adjacent to their parent's farm. As long as you keep the founders far enough apart you will get nice separate colonies like those in the picture above. However, if you put the founders close together you will get a "lawn" of bacteria where the borders of one colony butt up against the colony next to it. What you get is a continuous layer of bacteria much like your lawn of grass at home, and this plate of bacteria will now be referred to as the master plate. The master plate will contain colonies founded by genetically diverse descendants of a single bacterium.Now we will introduce replica plating. This process can be described as bacteria stamping where our bacterial lawn serves as our ink pad. What you do is take your round pad that is just a little smaller than the diameter of the plate and press it down on the bacterial lawn. Once you have "copied" the master plate you stamp it onto the replica plates making sure to record the orientation of the stamp so you can keep track of where the bacteria came from on the original master plate.For this experiment, we create a master plate on media without antibiotic. We then replica plate onto media containing antibiotics.What predictions do each of these hypotheses make?1. Random mutations: If resistance is due to mutations that occur prior to antibiotic exposure then they will already exist on the master plate, and they will be clonal. Therefore, we should find resistant colonies on the replica plates at the same position on each replica plate since they came from the same position on the master plate.2 and 3. Non-random mutations and/or resistance induction: Since each bacteria has the same chance or same mechanism of producing resistance we should NOT see colonies appear at the same position on each of the replica plates.What are the results? We see colonies at the same position on each of the replica plates. This is consistent with random mutations, and inconsistent with non-random mutations or induced resistance.Any questions?

I did fail to spell it out in a concise manner. The steps are there in the posts, but they are spread out and not put together in an obvious way.Observation: Bacteria become resistant to antibioticsHypothesis: Random mutations produce a new phenotype that is resistant to antibiotics.Null hypothesis: The new phenotype is not due to random mutations.Experiment: Produce antibiotic replica plates from drug free master plate.Predictions: The hypothesis predicts that resistance will be clonal and the mutation will arise prior to exposure to antibiotics. If the colonies on the replica plates come from the same position on the master plate then the hypothesis is supported. If the resistant colonies do not come from the same position on the master plate then the null hypothesis is supported.Experimental results: Resistant colonies on the antibiotic plates come from the same area of the master plate.Conclusion: The results of the experiment are consistent with random mutations producing a resistant phenotype in the absence of antibiotics.Further experiments: . . . coming soon

It is not reasonable without further experimentation. If, however, disparate phenotypes arise in a similar manner as seen with antibiotic resistance then you can start to hypothesize that there is a basic mechanism that ties all of the together. For example, what if bacteriophage resistance can also be shown to come about in a similar fashion, or mutations that revert a negative mutation in a lactase gene? Even more, what if it is shown that deleterious mutations arise in a similar manner, such as those that knockout an important amino acid synthesis pathway?I think it is also important to point out that the simple experiment discussed in the previous posts was first done before DNA had even been discovered. Since then we have a much better understanding of the molecular basis of genetics. However, I think it is important to start with these foundational experiments to show how we got to the theory we have now. I think it is also important to show that we do not just assume that mutations are random. We have reasons why we conclude that they are random.