International Human Microbiome Congress 2011 – Vancouver

This week I attended the IHMC. It was my first scientific conference and imposter syndrome levels reached an all time high. I’d like to thank the organizers at the Foundation for the National Institutes of Health who kindly provided me with a free pass in exchange for some promotion on social media. The conference was amazing! I learned a lot.

.

Microbiomics is still a very young and still maturing field. While people have speculated for decades that the microbial flora was important only recently with the availability of low cost high throughput sequencing has research become possible. Back in the day the tools simple didn’t exist to examine microbial community. Culturing has never been viable as we didn’t and to a large extent still don’t understand how to create the conditions to make bacteria grow. You could isolate DNA and sort by size and to detect change but it wouldn’t be able to interpret anything, and even if you did go to the trouble of getting sequence data back then you wouldn’t have anything to compare to know what you were looking at. Now the field is growing fast. In 2006 the first study of the human microbiome made the cover of science, a study of the microbial composition of only three individuals(grads students). Today we have access to the reference sequences of hundreds of people. Sequencing is just getting much cheaper and much deeper, and will make research possible that would have been absolutely ridiculous  if not literally impossible even a few years ago. One of my favorite people I got to meet at the conference, a scientist from Guelph, has been discovering and culturing hundreds of new novel strains, and doesn’t know what to do with them or who to give them to. The problem that’s already arising is keeping up with the analysis of all the new data that is and will be becoming available

It’s a golden age of research. It’s an amazing time to be a grad student, maybe one of the best since a long while. This generation will go down as heralding an extremely productive era of scientific discovery where many fundamental truths were found having radical positive impacts on humanity. We are now at an intersection in time where our tools are allowing us to look at things we’ve never before been able to look at, and the amount of discoveries waiting to be made are staggering. Just counting the things we know that we don’t know, the things that will for sure be found out one way or anouther, will provide us with immense knowledge. It’s just different than it used to be. Science for better or worse ain’t what it used to be. Not that I’d know.

There are basic methods almost everyone uses to look at the microbiome, the most common being sequencing the 16S ribosomal RNA gene, a very conserved gene in bacteria, and comparing that information to a refernce genome. There are still a lot of bacteria found in many studies that do not match anything in the reference. No one does anything else because 1) we assume that everything has 16S so we shouldn’t be missing anything and a different approach wouldn’t provide new information and 2) There’s a massive reference database that exists for 16S. Once you have all these sequences comes the difficult bioinformatic challenge of knowing how to arrange this data to get the most accurate picture of species composition. Many studies combine this approach with something like culture or gnotobiotic animal models.

So lets break down some of the discoveries presented at the conference. For anyone whose interested in specifics I’d recommend looking over the IHMC11 program guide, which provides a brief abstract of all the talks and poster presentations. It’s good and it’s thick. While the findings are significant there’s a common caveat to keep in mind throughout almost all the studies: We can not yet determine causative agents from correlations. For example, if a study found that in people with a particular illness had less commensal bacteria and more of other types of bacteria what can we say? Are the bacteria found only in the ill or found in greater numbers are pathogenic and causing the illness? The lack of normal bacteria is the problem? The lack of normal bacteria allows bad bacteria to come in? Some complicated situation of greater diversity/heterogenoity or lower total population causing problems? Is the microbial community just responding to something odd about the host? Are the new bacteria doing the same thing as the old bacteria? While these questions are answerable at this time it’s very hard to dissect the information apart, especially since we don’t know the primary function of many of these bacteria and even if we had a clue it’s often  the case that a species could be doing two completely different things – A weakness of looking only at DNA is that we wouldn’t be able to detect this. It’s often a very complicated story, even still there’s a lot we do now know.

A healthy micriobiome composition is essential to health, and is affected by our lifestyle and diet. The conference was right next to a macdonals and I’m sure everyone at the end was thinking “hmm that would probably be bad for my microbiome”. We have 10 times more bacteria in us than human cells, and maybe 100 times more bacterial genes than human genes. We contribute 23,000 genes, they contribute over a million. We can not live without them. The microbiome is our second genome, and this genome is highly unique to us and temporal. All the studies found that interpersonal variation was massive, we are much more different from eachother in our microbiome than our genome, it’s a highl personal thing and raises questions of “what makes of us?””Are these bacteria part of us?””What makes an individual?”.  Philiphosical discussions ran deep.

Cystic Fibrosis: It’s a disease marked by excaberations and periods of relative stability, the usually bacteria blamed; Pseudomonas aeruginosa, Staphylococcus aureus, Burkholderia cepacia did not correspond with exaberations. Other currently unknown pathogens may be responsible. Evidence for the interactions of multiple species being responsible. The streptococcus milleri group was found in half of all exaberations and hospitalization of CF adults, a group not detected by standard clinical tests. While aviral themselves they make. P. aeruginosa pathogenic. Treatments agaisnt this group has been developed and are effective. Due to the synergestic effect personalised thearpy based on patients individual microbiomes is needed. See Michael Surette.

Co-cultures: If you throw random bacteria species in pure culture  maybe 1% of them will grow. If you throw them in combination you might get 3%.  Symbiotic relationships are very important to bacteria, many rely not only on their host but also their neighbors. Complicated co-occurance and co-exclusion relations are being worked out. This is a function of their nature. They are small and specialized and reproduce very quickly so evolution works very quickly. The microbiome is the greatest site of evolutionary ecology in the world. Several groups have developed microfluid coculture techniques that I want to play with. Also everything is cultured with poo, so if you want to work in this field the chances are high you’ll have to work with poo(but you’ll call it feces). See Xiaoxia Nina Lin

Bacterial Vaginosis: 8 genera are strongly linked including Gardnerella, Atopobium, Megaspahaera, Eggerthella, Aerococcus, Leptotrichia/Sneathia, Prevotella and Papillibacter. The study was done only in Chinese women, and since we are still figuring out how populations differ and why it may not just be these 8 or these same bacteria worldwide. See Charlie Xiang

Resistant starch: Eat more resistant starch(bran) and you’ll probably change your microbiome to have more clostridial cluster IV ruminococci. Their specialists at attaching to the fibres and probably extract energy from it. Huge interpersonal variation. Not everyone will respond the same way to a change in diet. See Alan Walker

Vaccines: When there are a lot of pathogenic strains and a vaccine only confers immunity to some of them, does widespread use of the vaccine alter the microbiome composition of these strains in vacicnate people? In non vaccinated people? And if these strains have differing relationships with other bacteria, does altering their composition alter the microbiome composition. Theres a study underway on the temporal effects of the 7-valent pneumococcal conjugate vaccine on nasopharyngeal microbiome among Gambian infants to answer these questions. See Martin Antonio, who couldn’t be at the conference due to visa issues. Vaccines may have unintended consequences on a population level microbiome ecology. Which brings me to one of the coolest things presented.

Antibiotics: They don’t do what we thought they did. This was the most suprising and shocking discovery to me at the entire conference. We’ve used them for decades and assumed they only had short term effects, and it’s pretty clear they have been messing with our micriobiome balance. In particular there’s this thing called STAT, sub therapeutic antibiotic treatment, farmers have known for decades that if you give livestock little doses of antibiotics they grow bigger quicker. Until recently no one knew why, the mechanism was completely unknown but hey we’re capitalists and STAT has a great profit incentive

STAT increased adiposity(fatness), early bone growth and hepatic infiltration by fat, significantly altered microbial composition – functionally altered it because the new composition wasn’t doing the same things as the old composition. Early life use of antibiotics is important, if theres one practical tip I got out of is don’t let just your doctor give infants antibiotics without considering the consequences. The antibiotics in your meat may be making you fatter than you should be, the organic meat or veggie burgers may be worth it. I thought the hormones would be the problem. Antibiotics affect gut flora which affect you. I came out of the conference resolved to be more hippy-dippy, I’ve got a mechanism and datasets for that holistic mind-body balance stuff now. See Martin Blaser

Irritable Bowel Disorder(I’m going to clump all the talk about IBD, Ulcerative colitis and crohns here since they all seem to work the same way. There were many talks and posters about these disease, if you are interested in the specific results on the program guide will be a good place to start):

Significant differences in microbiota between affected and healthy controls, but the causation problem is big here. Complicating things is that one of the symptoms, diarrhea would be affecting composition. People with IBD have significantly reduced levels of microbial genes in remission than healthy first degree relatives, and reduced diversity and temporal instability with active disease(although this could be due to the diarhea) See Francisco Guarner for UC, Mark Morrison for clinical managment of IBD, Brandi Cantarel for Crohns proteonomics

Necrotizing Enterocolitis: A novel bacteria most closely related to y-proteobacteria but not matching any known sequence more than 97% should up more frequently in NEC. See Volker Mai.

Microbiota-targeted preventative Healthcare: One of the best talks by one of the most charismatic presenters I’ve ever seen. Variation in microbiota and metabolites can be used as biomarkers. variation can come from the host, a viral infection or carcinogen affecting microbiota or from diet disrupting microbes that increase antigen load and cause chronic low level inflammation.  Health can be predicted from interactions between the human genome and the microbial genome. Toilets of the future may analyze your #1s and 2s and recommend specifically designed diets, or drugs for diseases that you don’t even have yet. Startrek didn’t even have that. See Liping Zhao

Bacteroides fragilis: B. fragilis, a common commensal, induces the devolpment of Foxp3+ regulatory t cells, reducing inflamation, by the immunomodulatory molecule Polysacharride A(PSA). PSA not only prevent but cured colitis in mono culture germ free animal experiments

Probiotics: People seemed reluctanct about this word and I never really understood why. Bottom line: If you take a probiotic within a few hours your guts will be taken over by lactobacilus or bifidobacterium, however within around 24 hours the composition will rapidly return to it’s previous state. Probably.

Viruses, the ‘virome’, have the greatest genetic diversity on earth and are also the thing we know the least about. Both human and bacterialphages. Every gram of healthy human poop has 10^10 viral sequences in it, most of which we can’t classify. Is the virome the next big thing, the microbiome of the microbiome? Will we find that just like the microbiome, a balanced viral composition is needed for human health and homeostsis? Does diet affect the virome? This is a field in it’s infancy. See Curtis Suttle and Frederic Bushman.

This list is not nearly comprehensive, merely the topics and talks that stood out the most to me. I’m sorry to all those with incredible research I neglected to mention.

Along with learning lots of raw dirty hardcore science, I got to learn how science is orchestrated. I was extremely impressed by how well the Human Microbiome Project is organized., I have a lot of respect for the talented and dedicated individuals at the top of the field who act as show runners and general managers. It’s a massive endeavor. The organization and business of science is hard to pin down, it’s a loose hierarchy not really comparable to any other large enterprise,  governed by many factors I’m still oblivious to, but I greatly enjoyed observing the pulse of a modern scientific community discussing ideas and posturing itself. There are massive efforts to coordinate and collaborate all this data and research, people dedicated to building and providing the tools and analysis for other scientists to advance the field with. Organizations like HMP, the DAAC, MetaHIT are essential to the field, individualistic efforts to solve such a large bioinformatics problem would be foolhardy and redundant. Additionaly much of the talks were reflections on the field, critical analysis and discussion on things like how confident we can trust the methods and analysis, what bias exist, where theres a need be viligant for possiblly misleading data, and projections on how new tools should be integrated and will contribute to the field. Researchers are starting to use diverse techniques, like proteonomics and transcriptonomics, to look at the microbiome. This is important because the early results are showing that we can’t assume that gene presence correlates with protein presence, even if the exact same species are found they could be doing two radically different things. There are specialists in everything: ethics, technology development, bioinfromatics, statistics, ecological theory, etc. Coming into the conference I’ve never been immersed in science in such a way, I only had my naive optimistic expections of how the big picture of science works and moves. I was very impressed by the entire thing, it’s an open and coherent cooperative effort/atmosphere that appears to work pretty efficiently

image

All the talks except for the concurrent sessions were held in this large conference room, usually with with 20 minute talks followed by a round of questions. Pictured: Rosie posing a question to Karen Nelson

imageRosie blogging away

image

The food was excellent. I averaged around five cups of coffee per day. Frederick Bushman presented an unexpected result in one of his studies that all his subjects unexpectedly showed a convergence of microbiome composition from several bacterial species during the study, likely due to all the treatments staying in the same hospital for a few days. Perhaps all the attendants not only grew more closer as a community, but also grew more alike in their second genomes throughout the duration of the meeting.

image

Many posters were presented, and there were lots of opportunities to talk to the authors. Listening in on these discussions was fascinating but overwhelming as they tended to go into much greater detail than the seminar talks. There were a lot of similarities in research, as almost everyone used the same approaches and tools to sampling and analyzing the microbiome – usually 454 sequencing on 16S followed by heat mapping and some type of principle component analysis. By the end of the conference I felt a lot more comfortable with the research and was able to understand the data quicker and more fully.

image

A dry wit accompanied some of the studies. It was a little relieving.  The vast majority of the talks were on the large intestine, the vaginal microflora representing a distant second, and nasal/lungs, oral, small intestine, viruses, and skin filled out the rest of the research presented, along with many posters and presentations on the technical tools of the field. The high ratio of large to small intestie studies reflected the relative ease of collecting colon samples and the difficulty in testing small intestine microflora(you start at the mouth). If anything I though studies on vaginal and especially breast micrbiomes were under represented considering how important they are in the vertical transmission of flora.

image

No comment.

image

Friendships were made! I got to meet a lot of really smart and knowledgable scientists all of who were very friendly but it was a little intimidating, I definitely felt out of place a lot. In retrospect in was good that I came in as green as I did, I never knew who the bigwigs were until someone pointed it out to me later.

image

Josh posing with his and Rosies posters

image

Drinks were had afterwards. Jan was nice enough to cover the bill. Everyone was very scientific.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: