Last summer gave a talk at a Gordon Conference about transmission and evolution of drug resistance in HIV and E. coli. When I was done with my talk, there was time for questions. Dmitri Petrov asked what I thought about why resistant and susceptible strains of bacteria co-exist. I had to admit that I hadn’t really thought about that.
This question of co-existence (why aren’t all bacteria resistant or all of them susceptible?) it not a new one. In fact, about a week after the Gordon Conference, I talked to Marc Lipsitch who has worked on this question for many years. It was just not on my radar. Until last summer.
The question of co-existence marinated in my head for over the summer. I read papers about it. Looked at the data I was analyzing. And then sometime in the fall it suddenly hit me! I saw a solution to the question that was real easy. This is what I think may be happening: Resistance evolves when a bacterial strain finds itself in a person who is treated with antibiotics. But because most of us aren’t on antibiotic treatment most of the time, these resistant strains tend to have lower R0 values than susceptible strains (that is, the resistant strains don’t spread as effectively). Therefore in the human population at large, existing resistant strains are losing against the susceptible strains.
I had been studying the many origins of resistance (resistance in E. coli and other bacteria evolves very often – lots of convergent evolution), and I had been studying the cost of resistance (I think most resistant E. coli strains tend to die out – thought this is not easy to prove). These two ingredients together can explain the co-existence of resistant and susceptible strains.
Early October I emailed Dmitri: “Dmitri, I think I have the answer to your question!”. Dmitri answered: “Exciting! But you forgot to attach the manuscript”. Me: “Oh, I didn’t write it up yet! It is just in my head.”
So I started writing because that’s how academic science works!
The manuscript now lives on Medrxiv. I have submitted it to Nature (desk-rejected) and to Science (reviewed and rejected). Traditionally, after a rejection from a high-profile journal, one would send a manuscript to another journal right away, but one of the reviewers from Science suggested using a particular Norwegian dataset, instead of the Enterobase data I had used for the manuscript. I really liked that idea (as well as some other ideas from the reviewers). So I decided to pause and do more analysis. Some of the new data made their way into my talk for the TAGC conference in Washington DC last week.
If you are curious to hear where I am at with this project, here is the video of my talk:
I am a big fan of Dr Paul Turner from Yale University. He is good at many, many things. He is a very successful scientist (member of the National Academy of Sciences), has held several different administrative positions, created a company and does important science outreach work, including lectures available on YouTube.
I first met Dr Turner at the GRC microbial population biology conference he chaired in 2013. This was my first GRC meeting and I was impressed and inspired. The science was super exciting and the meeting was more diverse than any meeting I had ever been too.
His work as an evolutionary biologist who studies microbes is just amazing. I first got to know him for his basic evolutionary work on adaptation in viruses. Later, I learned about his work using evolutionary principles to prepare phage cocktails to treat people who suffer from drug resistant bacterial infections. For me, that phage work is one of the coolest applications of evolutionary principles ever. Here is Dr Turner doing a TED talk about that work:
TED talk by Dr Paul Turner entitled “Phage therapy targeting antibiotic-resistant bacteria”.
The interview below is meant for a book chapter for a book about evolution for German high schools, but Dr Turner allowed me to print it here as well.
Pleuni: Can you tell me what your job is?
Dr Paul Turner: I am both a professional scientist as well as an educator, in my job as Professor of Ecology and Evolutionary Biology at Yale University in New Haven, Connecticut, USA. Most of my time is spent running a large and active research group, composed of postdoctoral, graduate student and undergraduate scientists doing projects on virus ecology, evolution and genetics. I educate them through mentoring and training, either working with them on projects one on one or in collaborative groups, often involving other scientists at Yale and elsewhere. My job as an educator also involves classroom instruction in undergraduate courses, and leading seminars taken by graduate students.
Pleuni: Did you always want to become an evolutionary biologist? If not, what made you become an evolutionary biologist?
Dr Paul Turner: My fascination with evolutionary biology occurred around the time that I attended middle school and high school, nearby Syracuse, NY. I’ve always been an avid reader of science fiction, but my go-to nonfiction books at that time usually concerned the evolution of biodiversity, especially collected essays by Stephen J. Gould, the deceased Harvard paleontologist who was a prolific essayist for Natural History magazine. I greatly enjoyed reading Gould’s essays on evolution, geared to a lay audience.
It was during graduate school that I became obsessed with microbial biodiversity, and the awesome power of microbes as models for rigorously studying evolution questions, and the need to understand the ecology and evolution of microbes, the most abundant and biodiverse inhabitants of Earth. Unlike most microbiologists, my lab conducts research on a broad diversity of microbes, including a wide variety of viruses and bacteria that do and do not cause diseases in humans. Essentially, my childhood fascination with biodiversity prompted me to establish an ever-expanding microbial zoo in my laboratory, harnessed for studying an equally diverse set of basic and applied questions relevant for microbial evolution.
Pleuni: Why are phages a good tool to treat antibiotic resistant infections?
Dr Paul Turner: Ever since phages were discovered around the early 1900’s, scientists started examining the successful abilities for phages to kill bacteria infections in animals and people. One important researcher in these early efforts to do phage therapy was the co-discoverer of phages, Felix d’Herelle. There are many benefits of this approach, especially the ability for lytic phages to kill specific bacterial cells while releasing hundreds of new virus particles that can repeat the process in other susceptible cells; this is a rare example of a self-amplifying drug. Also, phage therapy seems to be generally safe for humans and causes few if any side effects, as long as the phage preparation is carefully produced to remove any dangerous bits and pieces of the bacterial cells that were destroyed while preparing phage doses; otherwise, these cell remnants are endotoxins that can cause a bad physiological response in the patient. However, phage therapy became unpopular in Western countries only a few decades after d’Herelle’s work, as these scientists and physicians instead invested in the development antibiotics to treat bacteria diseases, following Alexander Fleming’s discovery of penicillin. Now that antibiotics are failing worldwide in their usefulness to treat antibiotic resistant bacteria, there is resurged interest in phage therapy as a possible alternative. Today we have much better microbiology tools and understanding of phage biology compared to the early 1900’s, when there was suspicion that the biological variability of phage doses was less trustworthy than purely chemical preparations of antibiotics. This gives us better confidence that these viruses can help treat infections, since antibiotics are waning in efficacy.
Dr Paul Turner working in the lab with a lab coat and gloves
Pleuni: What is the most fun part of your job?
Dr Paul Turner: I have always enjoyed the challenge of experimental design, which is a very fun and creative process. There are endless mysteries in the natural world, but it is not trivial to design a biological experiment properly to test a hypothesis. I greatly enjoy interacting with members of my lab group and with other scientists, as we grapple with different ways to test our ideas using data generated in the laboratory, and by analyzing biological samples from natural environments or the human body. The ease and efficiency of growing many microbes in this process allows us to generate results quickly, and to rapidly address different hypotheses in a relatively short amount of time.What part of your job do you find difficult?
Pleuni: What part of your job do you find difficult?
Dr Paul Turner: I greatly enjoy teaching in the classroom, but I am often amazed how much time and preparation is needed to create a well-organized and useful lecture. This is a difficult process, even though it is incredibly rewarding when I am successful in transferring knowledge to my students. It can be even more challenging to present complex scientific data and ideas to a lay audience of non-scientists, but I find this difficult task to be essential; I believe modern day scientists should embrace our role in trying to explain why science is important and how it impacts everyone’s daily lives.
Pleuni: You are from the United States, but worked in Spain too, right? Did you like living and working in a foreign country?
Dr Paul Turner: I am incredibly grateful that I had an opportunity to live and work outside of the USA during my professional training, as this gave me a glimpse into how scientists in other cultures may approach similar problems in different ways. Unraveling the complexities of the natural world benefits from applying different viewpoints and life experiences, which alone reminds us why it is important for science to embrace diversity of the individuals studying it.
Pleuni: Do you always do science or do you have other interests too?
Dr Paul Turner: I love science, but I am passionate about many other things as well. I have a spouse and two daughters, and we enjoy getting outside for hikes and other recreational activities in natural lands, such as forests and the beach. Also, I have always enjoyed listening to music, and have a large collection of vinyl records that has grown ever since my childhood. I also like to read books for pleasure (not just for work), and especially enjoy science fiction novels and look forward to seeing movies made from them.
Pleuni: You have published many research articles and given many lectures. Do you like writing articles and giving lectures?
Dr Paul Turner: Yes, I find writing to be very enjoyable and perhaps the only downside is that scientific writing can be rather formulaic. Therefore, I also try to write review and opinion articles, where I can express my creativity a bit more. My goal is to write a science fiction novel some day!
Pleuni: Do you think drug resistance will ever be solved?
Dr Paul Turner: I believe that drug resistance will always prove challenging for humans, as we grapple with emerging pathogens that threaten the health of humans, as well as that of domesticated plants and animals, and species that we are trying to preserve through conservation biology efforts. Thus, I admit that I am pessimistic that we will ever truly conquer the problem of drug resistance. However, I am increasingly optimistic that better science and engineering tools and expanded perspectives on the natural world will allow us to use creative approaches to solve difficult societal problems, including the challenges of drug resistant pathogens.
I love making plans and I love the beginning of the new year and the new semester. I actually think that the yearly rhythm of semesters and breaks is a huge benefit of being in academia.
Today I spent some time thinking about the writing I plan to do in the coming semester and the talks I will be giving in the near future. The first talk that’s coming up is an invited talk at Stanford. I am very honored to be an invited speaker at the CEHG symposium alongside Anne Stone (ASU) and Graham Coop (UC Davis). I want to try and use the opportunity of the talk to think about what stories I want to tell and I plan to write the story up for publication in addition to talking about it at Stanford.
So what are the stories I want to tell? There are many! But here are some thoughts:
“The evolution of HIV evolution.”
Recently I’ve given seminars in which the overarching storyline was “The evolution of HIV evolution.” I focused on the evolution of drug resistance within patients and explained how drug resistance evolution used to be very fast, but became slower over time. When people were treated with a single drug (AZT) in the late 80s and early 90s, the virus would evolve drug resistance very quickly and the treatment would quickly become useless. Freddy Mercury probably died because of a very fast evolving virus. Over time, treatments improved (using combinations of drugs and using better drugs) such that it became harder for the virus to evolve and nowadays, drug resistance evolution is so slow in patients on treatment, that it is no longer a big worry, and people can stay on the same drugs for many years.
A slide from a talk I gave at SMBE and ESEB in the summer of 2017.
“HIV does it all”
Here is another storyline. Any field in science needs some systems that are looked at in detail. In evolutionary genetics, these systems are fruitflies, yeast, humans, mice etc. HIV is a great system as well in part because we know so much about it and data is abundant. One of the things we have learned in recent years, thanks to work by people like Richard Neher and Kathryna Lythgoe, is that HIV evolution can surprise us again and again. For example, HIV evolution, even in absence of drugs, can be fast within patients and slow at the epidemic level. It can be happening with a lot of recombination, or showing clonal interference (unpublished, Kadie-Ann Williams and PSP), and sweeps can be hard or soft. Within host populations can be panmictic or structured. So if everything can happen, how can we make sense of this all?
“Drugs to prevent HIV”
I like the story of how drugs were and are used to prevent HIV infection. In the 90s and well into the 2000s, drugs were used to prevent mother-to-child-transmission of the virus during child birth. In fact, this was one of the big successes in the world of HIV before treatment was really working to keep infected people alive. Nowadays, drugs are available for HIV-negative people who are at increased risk of HIV infection. Pre-exposure-prophylaxis (Prep, marketed as Truvada) is probably contributing to the shrinking of the HIV epidemic in San Francisco as many of the HIV-negative gay men in the city are taking Prep. When drugs were used to save babies, they were uncontroversial, but when they are used to save gay men, they continue to be controversial and there are many places where Prep is not available (for example, in my home country, The Netherlands).
How is this story linked to evolutionary genetics? When someone is taking drugs to prevent HIV, but they end up getting infected anyways, there may be a high risk of drug resistance evolution (this happened in the babies, in their already infected mothers, and it is happening occasionally in Prep users). Also, at an epidemic level, if a large part of the population is on Prep, this may lead to sub-epidemics of drug-resistant viral strains. There is some interesting modeling work by Sally Blower on these questions.
———————————————-
OK, that’s enough brainstorming for today. I’ll develop one of these stories into a presentation for CEHG and for an article to be published somewhere. If you have any questions or suggestions, let me know!
Last year I read some really cool books that are somehow related to my work. I also read books that were so annoying, I didn’t even finish them. I wanted to share some of my thoughts here.
Jim Ottaviani, Maris Wicks: Primates
Lovely comic book about three women researchers who study primates (Jane Goodall, Dian Fossey, and Biruté Galdikas). Great gift idea! Link to book
Steven Strogatz, The joy of X.
Highly recommended! Great book with essays about fun math. Made me want to learn more. Link to book.
Jennine Capó Crucet: Make Your Home Among Strangers
I very much enjoyed this novel about a young cuban woman who is the first of her family to go to college. It’s an easy read, but it has some insights that may be useful for those of us who teach. Link to book.
Vanessa Woods: Bonobo handshake
Well written memoire by traveler, writer and bonobo researcher Woods, with a lot of background on Congo and neighboring countries. The descriptions of awful violence during the wars in Congo may be upsetting to some. Link to book.
Bill Nye: Undeniable
The topic of this book, evolution, is dear to my heart, but I didn’t manage to finish it. It is simply not well written / edited. Link to book
Frank Ryan: Virolution
This book was definitely worse than Bill Nye’s book! It is not well written and it is full of nonsense about evolution. Disappointing, because it would have been nice to have a good popular book on viruses and evolution. Here Carl Zimmer explains why the book is not recommended: link to book review.
In the fall semester of 2014 I taught a reading seminar for master students at SF State on contemporary evolution in human viruses. This blog post contains a list of the papers we read in the seminar.
I posted about this seminar previously here (about the seminar format) and here (no powerpoint allowed), and here (about being nervous for a talk).
The students’ work can be read and seen here (about H1N5), here (polio outbreak), here (Dengue), here (Ebola), here (HIV in court), here (doing my own homework), here (the origin of HIV), here (on bad small things) and here (Hep B).
These are the papers we read:
1. Fast evolution of drug resistance in HIV patient the 1980s
I recently reread this paper by my colleagues Philipp Messer (used to be my office mate at Stanford) and Richard Neher (who works on the population genetics of HIV, just like I do). I thought it’d be worth writing a short blog post about this paper because it has some really nice ideas but it is quite technical and you may not have read it.
Selective sweeps in HIV
Selective sweeps happen in HIV when the virus fixes immune escape mutations or drug resistance mutations. Often, we don’t have good enough time series data to determine the frequency path of the beneficial mutation (i.e., how fast does the beneficial mutation increase in frequency in the viral population). Without frequency path it is hard to quantify the selection coefficient of the beneficial mutation; how much fitter are they than the virus they replace?
The authors of the paper present a new method to estimate the selection coefficient of a beneficial mutation. The method requires deep sequencing data from a population in which a beneficial mutation has recently gone to fixation. The method is applied to HIV sequences from patients in which a drug resistance mutation or an immune escape mutation has just gone to fixation. It seems to me that the method may be especially useful for drug resistance mutations because they may go to fixation rapidly and at unpredictable times, so that it is hard to follow their frequency path. The proposed method just requires a sample after fixation has happened.
The idea
The method is based on the following idea: If the selection coefficient of a beneficial mutation is very high, then the selected allele will quickly reach a high frequency without accumulating many new mutations. But if the selection coefficient is not so high, then it will take more time for the selected allele to reach a high frequency, during this time it will accumulate new mutations.
New, neutral, mutations that occur on the background of the beneficial mutation, will be dragged to a higher frequency by the beneficial mutation. If a new mutation occurs on the background of the beneficial mutation very early when there is only one copy of the beneficial mutation, then the frequency of the new mutation will always be the same as the frequency of the beneficial mutation. They likely fix in the population together. If, however, the new mutation occurs when there are already 8 copies of the beneficial mutation, then the new mutation will likely reach approximately 12% frequency (like the red fraction of the population in the figure).
This figure shows how earlier mutations on the background of the beneficial mutation reach higher frequencies. (Fig 1 A in the paper)
In a fast sweep, the “5 copy moment” goes by quickly
For a new, neutral, mutation on the background of the beneficial mutation to ultimately reach frequency 20% in the population, it needs to occur when the beneficial mutation is present at approximately 5 copies. The new mutation then occurs on one of the 5 copies, and is thus present on 20% of the viruses with the beneficial mutation. If the beneficial mutation fixes, the new mutation will have a population frequency of around 20%. In a slow sweep, the beneficial mutation may spend several generations at around 5 copies, whereas in a fast sweep, the “5 copy moment” goes by quickly. A mutation that happens when there are 10 copies may reach 10% freq, at 100 copies 1%. If we have many sequences from the population (say, 1000), we can look at all the new mutations and their frequencies and determine how fast the sweep went, or what the frequency path of the beneficial mutation was. If we know the frequency path, we can estimate the selection coefficient of the beneficial mutation.
Richard and Philipp used their method on HIV data because these data are deep enough to do this.
This is a sweep of a drug resistance mutation. The inset shows the genetic distances between the most common haplotypes in the dataset. All haplotypes have just one new mutation, except haplotype 13 which has 2. The main figure shows the ranks of the haplotypes on the x-axis vs their abundance (relative to the haplotype that had no new mutations) on the y-axis. Haplotype 1 (with 1 new mutation) has approximately frequency 0.05, so it must have occurred when there were around 20 copies of the beneficial mutation. The estimated selection coefficient is 0.07. This is figure 6 A in the paper.
Use the method to study new infections?
I wonder whether this method can be used to see how quickly a new HIV infection is growing in a person if we’d have deep sequence data from a newly infected person.
Almost three years ago, in early 2012, I attended a talk by Martin Nowak. He talked about cancer and one of the things he said was that treatment with multiple drugs at the same time is a good idea because it helps prevent the evolution of drug resistance. Specifically, he explained, when treatment is with multiple drugs, the pathogen (tumor cells in the case of cancer) needs to acquire multiple resistance mutations at the same time in order to escape drug pressure.
As I listened to Martin Nowak’s talk, I was thinking of HIV, not cancer. At that time, I had already spent about two years working on drug resistance in HIV. Treatment of HIV is always with multiple drugs, for the same reason that Martin Nowak highlighted in his talk: it helps prevent the evolution of drug resistance.
However, as I read the HIV drug resistance literature and analyzed sequence data from HIV patients, I found evidence that drug resistance mutations in HIV tend to accumulate one at a time. This is contrary to the generally accepted idea that the pathogen must acquire resistance mutations simultaneously.
There seemed to be a clear mismatch between data and theory. Data show mutations are acquired one at a time, and theory says mutations must be acquired simultaneously. One of the two must be wrong, and it can’t be the data![1]
Interesting!
After Martin Nowak’s talk, I went up to him and told him how I thought data didn’t fit the theory. Martin’s response: “Oh, that is interesting!” (Imagine this being said with an Austrian accent). “Let’s meet and talk about it.”
So, we met. Logically, Alison Hill and Daniel Rosenbloom, then grad students in Martin’s group, were there too. I had already met with Alison and Daniel many times, since they were also working on drug resistance in HIV. John Wakeley (my advisor at Harvard) came to the meeting too.
Between the five of us, we brainstormed and fairly quickly realized that one solution to the conundrum was to assume that a body’s patient consisted of different compartments and that each drug may not penetrate into each compartment. Maybe we found this solution quickly because Alison and Daniel had already been thinking of the issue of drug penetration in the context of another project. A body compartment that has only one drug instead of two or three would allow a pathogen that has acquired one drug resistance mutation to replicate. If a pathogen with just one mutation has a place to replicate, this makes it possible for the pathogen to acquire resistance mutations one at a time.
We decided to start a collaboration to analyze a formal model to see whether our intuition was correct. Over the following three years, there were some personnel changes and several moves, graduations and new jobs. Stefany Moreno joined the project as a student from the European MEME Master’s program when she spent a semester in Martin’s group. When I moved to Stanford, Dmitri Petrov became involved in the project. Next, Alison and Daniel each got their PhD and started postdocs (Alison at Harvard, Daniel at Columbia), Stefany got her MSc and started a PhD in Groningen, I had a baby and became an assistant professor at SFSU. No one would have been surprised if the project would never have been finished! But we stuck with it and after many hours of work, especially by the first authors Alison and Stefany, and uncountable Google Hangout meetings, we can now confidently say that our initial intuition from that meeting in 2012 was correct. Compartments with imperfect drug penetration indeed allow pathogens to acquire drug resistance one mutation at a time. And, importantly, the evolution of multi-drug resistance can happen fast if mutations can be acquired one at a time, much faster than when simultaneous mutations are needed.
Our manuscript can be found on the BioRxiv (link). It is entitled “Imperfect drug penetration leads to spatial monotherapy and rapid evolution of multi-drug resistance.” We hope you find it useful!
[1]Of course, it could be my interpretation of the data!
Stefany Moreno (in large window), Alison Hill, Daniel Rosenbloom and myself in one of the many Google Hangout meetings we had.
One of the most fun things about teaching a grad seminar last semester was reading the homework assignments. Seriously!
Before I move on to the next semester (teaching genetics for undergrads), I wanted to share one more homework assignment. This one by Emily Chang, a graduate student in Scott Roy’s lab. The paper about viral quasispecies in Hep B was one of the harder ones for the students, but this graphical abstract very neatly sums up the main results. I also love that Emily used old fashioned paper and pens to make the abstract, knowing that using fancy drawing software isn’t needed to communicate science.
Here is some of the homework from the students in my class.
Make a graphical abstract of the paper
Cameron Soulette
What kind of data are used in the paper?
Most of the data used in this paper were viral isolates obtained from the stool samples of two patients in the Dominican Republic and Haiti. These were collected (presumably) by the authors because the patients were exhibiting signs of Acute Flaccid Paralysis (AFP). Individuals can have nonpolio AFP, but these two patients were exhibiting characteristics that led the authors to believe their AFP was caused by wild poliovirus, of which very few infections had been observed since the 1980’s.
Nucleic acid probe hybridization identified Vaccine-Derived Polio Virus (VDPV) in these samples. They then performed a sequence characterization of the major capsid surface protein VP1 and compared the isolates from samples to wild-type. The authors looked for more polio cases in the area, and obtained 31 more samples from which they isolated VDPV. They used bioinformatic approaches to analyze their data, including maximum-likelihood and neighbor-joining trees. Using these methods they were able to figure out the timeline for this outbreak of the virus.
Jennifer Gilbert
How much impact did this paper have?
According to Google scholar, this paper has been cited 441 times with the most recent being this year. I found two articles about the paper: one from the Telegraph and the other was a story from Reuters Health (I could not find the original article, but I found it on two separate forums). I think this paper is very influential, but it also has the potential to be used in ways that the authors did not intend. The paper emphasizes the need for continued vaccination and increased surveillance to further the effort of poliomyelitis eradication. However, it seems that this paper has also been picked up as fodder for those in the anti-vaccination movement (one of the forums was hosted by a group called The American Iatrogenic Association – a group focused on raising awareness for illness/ injury caused by physicians).
This week I decided to do some of my own homework. Just for fun.
It’s a graphical abstract of a classic paper we read in class.
Turns out, making a graphical abstract is no easy task! Next week, there’ll be students’ work here again.
What I found most surprising about this paper is that they had to sequence the chimps’ MtDNA to find out what subspecies they were. I would have expected that experts could simply look at a chimp and know what subspecies it is.
Being A Better Scientist 