OOPS...
Well this is awkward. I promised a research report in one week’s time and now it’s been three...All I can really say is that my work on this blog does not directly impact my GPA, while well, my classes do. So while I have somewhat less enthusiasm for commenting on the artistic intricacies of Langston Hugh’s poetry (I apologize, Professor Fritsch) than doing research and writing for this blog, “D’s,” in fact, do not earn degrees, much to the chagrin and dismay of some of my former classmates.
Let’s get back to actual research
You did not misread the title. Crazy right? The implications of the title, I mean, not your literacy. You’ve probably (accurately) been told that antibiotics are only effective in treating bacterial infections (here’s a lazy, not entirely necessary Wikipedia citation to back this up). While that is true, recent (well, if you count a paper 2 years old “recent”) literature suggests that some antibiotics, especially in high concentrations, negatively affects the replicative capacity of bacteriophages.
Let me qualify my remarks a bit. Some, but not all antibiotics appear to affect the replication of bacteriophages. Also, keep in mind that the evidence presented by the paper whose data I am going to walk through in a moment does not appear to suggest that antibiotics are directly affecting bacteriophages (such as “killing them”). Again, the data only appears to suggest that the replicative ability of bacteriophages is impacted by some antibiotics.
Skeptical? I don’t really blame you.
So it’s time for some experimental data!
The Paper: Synergy and Order Effects of Antibiotics and Phages in Killing Pseudomonas aeruginosa Biofilms [Chaudry et al. 2017]
In brief: this paper examines the effectiveness of combined phage and antibiotic treatments compared to only-phage or only-antibiotic treatments, in addition to changes in effectiveness when “staggering” the introduction of phage and antibiotic treatments (in other words, not simultaneously).
The experiments (What was done?): In the set of experiments described by this paper, the researchers wanted to determine how treatment effectiveness differs when antibiotics are used alongside phages. So they took two different types of bacteriophages (one is called “NP1,” the other “NP3”) that can kill the type of bacteria mentioned in the title, Pseudomonas aeruginosa, and used them to kill those bacteria in combination with a bunch of different antibiotics.
The results of this experiment were interesting...to say the least. “How so?” you ask. It’s better if you see the data yourself, so let’s take a look at it.
Experiment 1 data (combined, simultaneous treatment with phage and antibiotic)
Breaking this data dooown
As I said in the section header, this is the data for an experiment in which the researchers treated bacterial colonies with the two phages, NP1 and NP3, alongside various antibiotics. You could squint to read the figure caption (sorry again), but I’ll do you the favor of just saying what the antibiotics were. That’d be: Ceftazidime, ciprofloxacin, colistin, gentamicin, and tobramycin.
In the experiment, the researchers grew the bacteria for 48 hours, applied the respective treatments, then recorded bacterial density, which is the data that you see on the bar graph above. Y-axis is for (bacterial) cell density, that’s the easy part. But then you’ve got that confusing jumble of abbreviations and numbers. What does it mean? It’s pretty simple. The bars with x-axis label like “Cip1,” “Tob1,” “Tob8” represent the cell densities of cultures treated with ciprofloxacin 1x concentration, tobramycin 1x, tobramycin 8x, respectively. The first bit is the antibiotic used, the number is the concentration used of that particular antibiotic. Simple right?
Then what about the ones that say things like “Con,” “NP1,” NP3,” “N,” “N-Cef1,” and “N-Cip8”? Con, as you might have guessed, stands for control, the corresponding bar represents the concentration of bacterial cells that have not been exposed to any treatment at all. Of course, this bar is a fair bit higher than the other ones, which is a good sign of course, as it means the treatment is actually making a difference. As I said earlier, NP1 and NP3 are the names of the phages used, so the bars labeled as such represent concentrations of bacterial cells treated only with the respective phages. N means that both phages have been used as treatment. “N-Cef1” and “N-Cip8,” logically, just mean that the treatment was both phages, alongside 1x concentration ceftazidime and 8x ciprofloxacin, respectively. If this is too confusing to figure out just reading, here’s an annotated version of the figure with the axis label abbreviations alongside what they mean:
So now that you know what the x-axis labels mean, let’s see if we can make sense of what the actual values are telling us. A few things to note:
Point of clarification: Even though there are single bars for each label, the values shown by the bars represent the mean/average of many identically treated cultures.
(1.) Pretty much all of the bars are lower than the control (con) bars, which is good, as again, it signifies that all of the treatments resulted in lower concentrations than the untreated control, suggesting that they all did something (although the NP1 and Col1 culture concentration are very close to the control culture concentration)
(2.) When comparing pairs of combined treatment using the same antibiotic but different concentrations, such as N-Cip1 to N-Cip8, one finds that most times, the higher antibiotic concentration treatment is the most effective, resulting in a “lower” bar, or lower recorded bacterial cell concentration.
However, if you look back at the first chart, you’ll see that this isn’t always the case. While Tobramycin 8x kills more bacteria than Tobramycin 1x, with the combined treatments, it appears that N-Tob1 works better than N-Tob8. In other words, the combined treatment using phages and a higher concentration of antibiotic actually performed worse than the lower concentration one. How does this work?
How does this work?
It seems complicated and confusing, but watch, this can be reasoned out fairly easy.
Compare the value of Tob1 and N-Tob1. There’s a big difference, right? Because the only difference is of course that the phages were added to the treatment, the boost in bacteria killing can be attributed to the phages with confidence. The phages add to the effectiveness of the tobramycin treatment at 1x concentration, no big surprise there. But again, looking at N-Tob8, the treatment works worse than N-Tob1. The thing that ties it all together though, is the fact that N-Tob8 approximates the value for plain Tob8, without the phage. The reason I say approximate, is because the value for Tob8 falls within the error bar of N-Tob1. I’m not incredible at statistics stuff like that yet, so I’ll leave it at that and not make a fool of myself. Anyhow, the point is that for N-Tob8, it’s almost as if the phage isn’t active or has been interfered with somehow by the tobramycin. We know that phage can contribute to the treatment at lower concentrations, as in N-Tob1, but the logical conclusion here is that higher concentrations of tobramycin mess with the phage, negating its bacteria-killing abilities.
Based on the results of this experiment, the authors of this paper made pretty much the same inference. A second experiment and set of data support this conclusion. Take a look at it here.
Experiment 2 data (combined, staggered treatment with phage and antibiotic)
...Aaaand the caption...I annotated these myself, by the way, since figure captions can be confusing at times
This experiment is just a little more complicated than the last. Same antibiotics (only used at 8x concentration though), same phages, same bacteria. The twist though is that for each antibiotic-phage treatment combo, such as tobramycin (8x) and NP1 and NP3 phages, there were 4 variants in terms of the delay with which antibiotics were applied. That is, after the 48 hours allowed for the bacterial cultures to grow, the phages were applied to the cultures, but the amount time that the researchers waited before applying antibiotics was varied. If you look at the legend for the chart, you can see that the delays for antibiotic application were 0, (simultaneous) 4, and 24 hours (You may notice that there is an antibiotic group only, but that isn’t really relevant to the results of the experiment).
Keep in mind that even though the bars are only labeled according to the antibiotic used, both phages are still being used (except for the antibiotic only groups of course)
The second row of bars provides a new type of information: concentrations of phages measured at the end of treatment, that is, 48 hours after the phages were first introduced, and 96 hours since the bacterial cultures were started. Note that the bars of the second row represent phage density values corresponding to the cultures represented by the bars of the same color and antibiotic in the first row. For example, the red bar in the second row represents the values of the “Antibiotic + Phage Simultaneously” (no delay for antibiotic) treatment group just as it does in the first row. Of course, the y-axis in the second row represents phage density instead of bacterial cell density, but the point that I wanted to make was that the same colored bar represent data values from the same set of cultures. This is important.
Remember that in the first experiment, the cultures treated by 8x concentration of tobramycin and phage were the ones that yielded weird results. So we’re going to pay special attention to the “Tob” groups as we look at the data from this experiment. Before we look at the Tob groups, let’s examine the “normally-behaving” antibiotics.
The data from the ceftazidime set of treatments definitely seems to suggest that ceftazidime is one of these “normally-behaving” antibiotics. What do I mean by a “normal” antibiotic exactly again? Well, recall that I found the interference of high concentration antibiotic with phages to be pretty strange, so I’m calling any antibiotic that doesn’t do this “normal”. Since these antibiotics don’t interfere with phages, the longer the normal antibiotic application is delayed from phage application, the less effective the treatment should be.
Why would that be the case?
Assuming a certain amount of phage and antibiotic can only kill a set amount of bacteria, and knowing that bacteria reproduce exponentially, if one wants to have as low a concentration of bacteria at the end as possible, it is advantageous to kill as many bacteria as early as you can. Take a look at the diagram I drew below. For the sake of example, we have a species of bacteria whose doubling time is 1 hour (they divide once per hour), a set of phages that kill 2 bacterial cells, and an antibiotic that kills 1. As for this example, a “normal” antibiotic is used, the function of the phages is not affected by the antibiotic. When we use both phages and antibiotic immediately, as in Treatment 1, we get a much lower final bacteria count than in Treatment 2, when antibiotic application is delayed. This is what will happen when you have two effective treatments that do not interfere with each other, and stagger or don’t stagger the application of both.
So one reason why I think the data from the ceftazidime treatments suggests it to be a normally behaving antibiotic is because when the delay between phage and ceftazidime application is increased, observed bacteria concentration increases as well, fitting the profile of a normal antibiotic as we saw above.
A second, more easily explained reason would be the values for final observed phage density for the ceftazidime treatment group. Take a look at them for a moment. They’re pretty close to each other, despite the fact that the delay for antibiotic application changes betwen the subgroups, which suggests the fact that the antibiotic does not interfere with the phage. Were that the case, we would see a lower concentration of phage as the delay for antibiotic decreased (as the phage would have less time unintefered to replicate and stuff).
Now let’s look at the data from the tobramycin group. You can see that the concentration of bacteria sharply drops off (I’m not particularly certain why the decrease is so sudden though) as the delay for tobramycin application increases. However, I think that it is the phage density data that provides the most incontrovertible evidence of the antibiotic’s interference with the phages. See that solid black line in the second row of the chart? It marks the initial concentration of phages added. The red bar representing the final phage concentration (48 hours after initial phage application, mind you) for the subgroup with no delay for tobramycin application matches extremely closely with the solid line. In other words, it appears that the increase in phage concentration was extremely tiny, even with 48 hours elapsing.
This strongly suggest that tobramycin somehow interferes with phage replication.
Well that’s great, but how exactly does it do this?
Chaudry et al. seem to think that this could happen because antibiotics reduce the amount of bacteria, reducing the speed at which phages can replicate (as less bacteria = less potential hosts = less phages made), because antibiotics mess with the cellular machinery that allows the bacterial hosts of phages to produce them, or maybe both.
So there you have it. Thanks for sticking through this text wall, if you did. I didn’t expect it would take this many words to explain how the data leads to the conclusions made by the paper, but it did. Huh. I’ll try to slim down future reports if I can. Thanks for reading.
Best,
BJW
Commentaires