Author’s copy
provided for non-commercial and educational use only
is s n 1 8 7 6 - 2 8 8 3
No material published in Beneficial Microbes may be reproduced without first obtaining written permission from the publisher. The author may send or transmit individual copies of this PDF of the article, to colleagues upon their specific request provided no fee is charged, and furtherprovided that there is no systematic distribution of the manuscript, e.g. posting on a listserve, website or automated delivery. However posting the article on a secure network, not accessible to the public, is permitted. For other purposes, e.g. publication on his/her own website, the author must use an author-created version of his/her article, provided acknowledgement is given to the original source of publication and a link is inserted to the published article on the Beneficial Microbes website (DOI at the Metapress website).
For additional information please visit
www.BeneficialMicrobes.org.
Editor-in-chief
Koen Venema, the Netherlands
Section editors • animal nutrition Isaac Cann, University of Illinois at Urbana-Champaign, USA • processing and application Knut Heller, Max-Rubner-Institute, Germany • medical and health applications Ger Rijkers, Roosevelt Academy, the Netherlands • regulatory and safety aspects Mary Ellen Sanders, Dairy and Food Culture Technologies, USA • food, nutrition and health Koen Venema, the Netherlands Editors
Alojz Bomba, Pavol Jozef Šafárik University, Slovakia; Robert-Jan Brummer, Örebro University, Sweden; Michael Chikindas, Rutgers University, USA; James Dekker, Fonterra Co-operative Group, New Zealand; Leon Dicks, University of Stellenbosch, South Africa; Ana Paula do Carmo, Universidade Federal de Viçosa, Brazil; Margareth Dohnalek, PepsiCo, USA; George C. Fahey, Jr., University of Illinois, USA; Benedicte Flambard, Chr. Hansen, Denmark; Melanie Gareau, University of Toronto, Canada; H. Rex Gaskins, University of Illinois at Urbana-Champaign, USA; Audrey Gueniche, L’Oreal, France; Dirk Haller, Technical University München, Germany; Arland Hotchkiss, USDAARS, ERRC, USA; Kikuji Itoh, The University of Tokyo, Japan; David Keller, Ganeden Biotech, USA; Dietrich Knorr, Technical University Berlin, Germany; Lee Yuan Kun, National University of Singapore, Singapore; Irene LenoirWijnkoop, Danone research, France; Takahiro Matsuki, Yakult Central Institute for Microbiological Research, Japan; Baltasar Mayo, CSIC, Spain; Eveliina Myllyluoma, Valio Ltd., Finland; Peter Olesen, ActiFoods ApS, Denmark; Maria Rescigno, European Institute of Oncology, Italy; David Topping, CSIRO Human Nutrition, Australia; Roel Vonk, University of Groningen, the Netherlands; Barbara Williams, University of Queensland, Australia; Zhongtang Yu, The Ohio State University, USA
Founding editors:
Daniel Barug, Ranks Meel, the Netherlands; Helena Bastiaanse, Bastiaanse Communication, the Netherlands
Publication information
Beneficial Microbes: ISSN 1876-2883 (paper edition); ISSN 1876-2891 (online edition) Subscription to ‘Beneficial Microbes’ (4 issues, calendar year) is either on an institutional (campus) basis or a personal basis. Subscriptions can be online only, printed copy, or both. Prices are available upon request from the Publisher or from the journal’s website (www.BeneficialMicrobes.org). Subscriptions are accepted on a prepaid basis only and are entered on a calendar year basis. Subscriptions will be renewed automatically unless a notification of cancelation has been received before the 1st of December. Issues are send by standard mail. Claims for missing issues should be made within six months of the date of dispatch. Further information about the journal is available through the website www.BeneficialMicrobes.org.
Paper submission
http://mc.manuscriptcentral.com/bm
Editorial office
Orders, claims and back volumes Wageningen Academic P u b l i s h e r s
P.O. Box 179 3720 AD Bilthoven The Netherlands
[email protected] Tel: +31 30 2294247 Fax: +31 30 2252910
P.O. Box 220 6700 AE Wageningen The Netherlands
[email protected] Tel: +31 317 476516 Fax: +31 317 453417
Wageningen Academic P u b l i s h e r s
Beneficial Microbes, September 2013; 4(3): 219-222
Is it who you are or what you do that is important in the human gut? E. Avershina and K. Rudi Norwegian University of Life Sciences, Department of Chemistry, Biotechnology and Food Science, Universitetstunet 3, 1432 Ås, Norway;
[email protected] Received: 17 February 2013/ Accepted: 22 April 2013 © 2013 Wageningen Academic Publishers
OPINION PAPER Abstract The current leading view is that functionality and not phylotype is the most important determinant for the services provided by the gut microbiota. Here we present an alternative opinion, advocating the importance of phylotype in addition to function. We believe the literature is misled by technical artifacts in defining operational taxonomic units (OTUs), which are binned groups of bacteria based on sequence homology. Furthermore, the current metagenomic approaches where the total DNA in a sample is mixed prior to sequencing and subsequently resolved by a bioinformatics approach, are highly error prone with respect to both functional and phylotype assignments. We argue that the directions of the OTU and metagenome errors are such that stable phylotypes are overlooked, while functional stability is overestimated. Taking these errors into account, we propose that phylotype represents an interface for functionality, and is for this reason an important determinant for the services provided by the gut microbiota to the host. Keywords: ecology, evolution, phylotype, gut microbiota, symbiosis,
1. Introduction The gut microbiota serves three main basic functions to the host: (1) extracting energy from the food; (2) production of vitamins and other (essential) biomolecules; and (3) protection towards pathogens (Tremaroli and Backhed, 2012). The underlying host-bacterial interactions in obtaining these services, however, are not yet well described. A major question that has not been settled is if it is what you do (functioning) or who you are (the phylotype) that is most important for the mutualistic relation between gut bacteria and the host (Lozupone et al., 2012). Potentially, due to methodological constraints and artifacts the literature is now tipping towards the view that only the function, and not phylotype, is crucially important for the host-bacterial relations in the gut (Lozupone et al., 2012; Turnbaugh et al., 2009; Yatsunenko et al., 2012). However, we theorise that it is the interplay between function and phylotype. We believe that this one-side view stems partly from the artifacts in binning bacterial phylotypes into socalled operational taxonomic units (OTUs), which are groups of bacteria with shared identity over a certain
threshold. When reanalysing the lead article (Turnbaugh et al., 2009) advocating the view of functional stability and phylotype instability using an OTU independent approach, we did indeed identify stable phylotypes (Figure 1) (Sekelja et al., 2011). Furthermore, it has recently been realised that the metagenomic data should be interpreted with extreme care both due to artifacts (Gomez-Alvarez et al., 2009) and low degree of functional annotations (Prakash and Taylor, 2012).
2. Challenges with OTU binning The OTU binning is in itself an artificial process, since bacteria are asexual organisms with no defined taxonomic barriers (Hamady and Knight, 2009). The main challenges with OTU binning relate to the fact that bacterial diversity is more or less continuous and not strictly categorical. It is well known that the evolutionary rates among microorganisms can differ dramatically, even among closely related taxa (Holder and Lewis, 2003). There are also major challenges with correct distance estimates for OTU binning (Avershina et al., 2013). Therefore, the assignments of sequences to OTUs are not crisp, that is, there will be several alternatives
ISSN 1876-2833 print, ISSN 1876-2891 online, DOI 10.3920/BM2013.0016219
E. Avershina and K. Rudi
V2 region of 16S rRNA 154 individuals (Turnbaugh, 2009)
Identify core 1 and core 2 phylotypes
Define phylotypes by binning sequences with more than 97% similarity
Two core phylotypes shared among all individuals over time defined (cores 1 and 2 within Lachnospiraceae)
Cores 1 and 2 present in all individuals
Identify phylotypes shared among all individuals
Define phylotypes based on nucleotide pentamersignatures*
No shared phylotypes detected
Full-length 16S rRNA 12 individuals, 4 time points (Ley et al., 2006)
Sekelja et al., 2011
Turnbaugh et al., 2009
Figure 1. OTU-independent reanalysis of the human gut microbiota sequencing data from 154 individuals (Turnbaugh et al., 2009). At first, full-length 16S rRNA genes data from stool sampled from 12 individuals over time (Ley et al., 2006) were used to identify most stable phylotypes. Two groups belonging to Lachnospiraceae family (cores 1 and 2) satisfied the criteria set. Then, sequence data from the study of 154 individuals (Turnbaugh et al., 2009), where the absence of stable operational taxonomic units was claimed, were screened for the identified cores 1 and 2. Sequences belonging to both cores were detected in every individual. A schematic explanation of the phylotype identification method is given in Figure 2.
for OTU assignment of a single sequence. In its simplest form, this can be exemplified with three sequences, were sequence 1 and 2, and 2 and 3 differ by one nucleotide, respectively, while 1 and 3 differ by two nucleotides. Given an OTU depth of one nucleotide, then sequence 2 can equally be OTU assigned with sequence 1 and 3, while sequence 1 and 3 cannot be within the same OTU. A further challenge relates to the fact that the sequence level for binning is not connected to biology. Thus, the binning in itself may be at a phylogenetically non-relevant level. For instance, in host-bacterial relations relevant levels for phylotype depths would be that for co-evolutionary events. For insects such events are well characterised (Moran, 2006). It has clearly been shown that symbiotic bacteria providing essential functions coevolve with the host (Dale and Moran, 2006). For humans, however, very few studies have actually addressed the issue of coevolution of mutualistic bacteria. As a proof of concept, we therefore did a recent reinvestigation of the literature with the aim of 220
identifying stable phylotypes using an OTU-independent searching approach (Figure 2). This lead to the discovery of two ancient coevolution events (Sekelja et al., 2011).
3. Phylogeny vs. functionality Although not yet completely described, there is a set of conserved functions that are needed for an organism to survive (Qin et al., 2012). Therefore, it is inherently difficult to differentiate between intrinsically (basic needs for cell survival) or environmentally selected (give competitive advantage in a particular environment) functionality. As an analogous system, one could imagine the metagenome investigation of large animal diversity at different locations. Such a study would probably reveal a high degree of functional stability among the locations, since most animals need the same functions to survive. This would, however, certainly not mean that the different animal species do not have different ecological functions. We Beneficial Microbes 4(3)
Gut microbe service providers
Seq 2 ACTGAT
1.
Seq 3 GAGCTG
CTGAA
CTGAT GAGCT AGCTG
1
1
0
0
0
1
0
1
0
0
0
0
0
1
1
2.
PC2
Seq 1 ACTGAA
ACTGA
PC1 3. Phylotype 1 PC2
Steps description: 1. Create nucleotide profiles based on pentamer frequency counts. 2. Perform Principal Component Analysis (PCA) on pentamer frequency table. 3. Define phylogroups based on the PCA score plot. Close placement on the plot (even scores of two main principal components) indicates many shared nucleotide pentamer signatures, i.e. phylotypes.
Phylotype 2 PC1
Figure 2. Schematic representation of the phylogeny-independent search for stable phylotypes. The approach is based on a principal component analysis of the nucleotide pentamer frequency counts with a subsequent search for groups of bacteria that are given similar scores for two main principal components.
therefore believe that, despite functional stability among the bacteria in the human gut (Turnbaugh et al., 2009), this does not necessarily mean that the stable genes represent the ecologically relevant ones. Unfortunately, most of the current metagenomic approaches actually entangle the functional and phylogenetic identity – making it very difficult to identify which phylotypes are associated with which function. Using a bioinformatics approach, however, we recently found that basic functionality, such as phosphorylation or methylation, showed a systematic distribution difference among the bacterial phylotypes investigated (Figure 3) (Rudi and Sekelja, 2013). This suggests that regulatory networks rather than the actual presence of functional genes are the main determinants for phylotypes and potentially ecotypes, as is the case for larger animals.
327 genomes belonging to 20 phyla
Identify gene clusters that occur only in subsets of genomes
4. Phylotypes represent an interface for functionality Given systematic differences in regulatory networks among phylotypes, one could argue for a model where the phylotype represents an interface for functionality. Although a given function such as antibiotic resistance can be specifically defined through a set of genes, it would need a range of other genes for proper phenotypic expression in a given environment. The concept of interface has recently been illustrated by the demonstration that an Archaean transformed the lifestyle to an oxygen rich environment through the acquirement of more than 1000 genes from the bacterial domain (Nelson-Sathi et al., 2012), indicating that complex environmental adaptation requires multiple gene interactions.
Correlate discriminative clusters to full-length 16S rRNA gene data
Strong correlations of Gammaproteobacteria to methylases and Actinobacteria to kinases
Figure 3. Schematic representation of the study providing a link between phylogeny and function (Rudi and Sekelja, 2012). Discriminative co-occurring gene clusters in different bacterial groups were identified and then correlated to their phylogeny based on 16S rRNA gene data. Strong correlations between Actinobacteria and kinases as well as between Gammaproteobacteria and methylases were detected.
Beneficial Microbes 4(3)
221
E. Avershina and K. Rudi
Although functions can be widely transmitted among phylotypes (Smillie et al., 2011), we believe the phylotypes allowed in a given environment, such as in the human gut, represent an important determinant for the ecological functioning in that environment, and not only the genes present. For example, imagine there are two hypothetical bacteria A and B that have very similar functional genes sets providing the same services. However, bacterium A belongs to a microbial phylotype that has an opportunistic interaction with the host, whereas B is a representative of a rather ‘friendly’ phylotype that is host selected. Given that both of them acquire a pathogen island, bacterium A would have a much higher chance of developing into a pathogen than B. B can simply be eradicated by the host through shutting down the positive host selection, while A has already passed most of the host’s defences through its opportunistic life style. And therefore here, phylotype provides a ‘green card’ for the microbe in its relation with the host, whereas its functional genes provide the actual services. This view is also in accordance with recent gut microbiota transplantation experiments, suggesting that a host specific microbiota is needed for proper immune development, despite the similar functionality of the microbiota used for the transplantations (Chung et al., 2012).
5. Conclusions Functions are more or less universally distributed in all environments, while the phylotype provides the environmentally selected interface for the functions. We argue that it is the phylotype that will allow providing certain services to the host. Therefore, phylotype in itself will be an important determinant for the service providers in the gut. In our opinion the key question for the future is to link functionality to phylotype. This would require further development in the direction of understanding the properties important for defining phylotype, rather than just trying to understanding functionality independent of the phylotype interface.
References Avershina, E., Frisli, T. and Rudi, K., 2013. De novo semi-alignment of 16S rRNA gene sequences for deep phylogenetic characterization of next generation sequencing data. Microbes and Environments, in press. DOI: http://dx.doi.org/10.1264/jsme2.ME12157. Chung, H., Pamp, S.J., Hill, J.A., Surana, N.K., Edelman, S.M., Troy, E.B., Reading, N.C., Villablanca, E.J., Wang, S., Mora, J.R., Umesaki, Y., Mathis, D., Benoist, C., Relman, D.A. and Kasper, D.L., 2012. Gut immune maturation depends on colonization with a host-specific microbiota. Cell 149: 1578-1593. Dale, C. and Moran, N.A., 2006. Molecular interactions between bacterial symbionts and their hosts. Cell 126: 453-465.
222
Gomez-Alvarez, V., Teal, T.K. and Schmidt, T.M., 2009. Systematic artifacts in metagenomes from complex microbial communities. ISME Journal 3: 1314-1317. Hamady, M. and Knight, R., 2009. Microbial community profiling for human microbiome projects: Tools, techniques, and challenges. Genome Research 19: 1141-1152. Holder, M. and Lewis, P.O., 2003. Phylogeny estimation: traditional and Bayesian approaches. Nature Reviews Genetics 4: 275-284. Ley, R.E., Turnbaugh, P.J., Klein, S. and Gordon, J.I., 2006. Microbial ecology: human gut microbes associated with obesity. Nature 444: 1022-1023. Lozupone, C.A., Stombaugh, J.I., Gordon, J.I., Jansson, J.K. and Knight, R., 2012. Diversity, stability and resilience of the human gut microbiota. Nature 489: 220-230. Moran, N.A., 2006. Symbiosis. Current Biology 16: R866-R871. Nelson-Sathi, S., Dagan, T., Landan, G., Janssen, A., Steel, M., McInerney, J.O., Deppenmeier, U. and Martin, W.F., 2012. Acquisition of 1,000 eubacterial genes physiologically transformed a methanogen at the origin of Haloarchaea. Proceedings of the National Academy of Sciences of the USA 109: 20537-20542. Prakash, T. and Taylor, T.D., 2012. Functional assignment of metagenomic data: challenges and applications. Briefings in Bioinformatics 13: 711-727. Qin, J., Li, Y., Cai, Z., Li, S., Zhu, J., Zhang, F., Liang, S., Zhang, W., Guan, Y., Shen, D., Peng, Y., Zhang, D., Jie, Z., Wu, W., Qin, Y., Xue, W., Li, J., Han, L., Lu, D., Wu, P., Dai, Y., Sun, X., Li, Z., Tang, A., Zhong, S., Li, X., Chen, W., Xu, R., Wang, M., Feng, Q., Gong, M., Yu, J., Zhang, Y., Zhang, M., Hansen, T., Sanchez, G., Raes, J., Falony, G., Okuda, S., Almeida, M., LeChatelier, E., Renault, P., Pons, N., Batto, J.M., Zhang, Z., Chen, H., Yang, R., Zheng, W., Yang, H., Wang, J., Ehrlich, S.D., Nielsen, R., Pedersen, O. and Kristiansen, K., 2012. A metagenome-wide association study of gut microbiota in type 2 diabetes. Nature 490: 55-60. Rudi, K. and Sekelja, M., 2013. High or low correlation between co-occuring gene clusters and 16S rRNA phylogeny. FEMS Microbiology Letters 339: 23-29. Sekelja, M., Berget, I., Naes, T. and Rudi, K., 2011. Unveiling an abundant core microbiota in the human adult colon by a phylogroupindependent searching approach. ISME Journal 5: 519-531. Smillie, C.S., Smith, M.B., Friedman, J., Cordero, O.X., David, L.A. and Alm, E.J., 2011. Ecology drives a global network of gene exchange connecting the human microbiome. Nature 480: 241-244. Tremaroli, V. and Backhed, F., 2012. Functional interactions between the gut microbiota and host metabolism. Nature 489: 242-249. Turnbaugh, P.J., Hamady, M., Yatsunenko, T., Cantarel, B.L., Duncan, A., Ley, R.E., Sogin, M.L., Jones, W.J., Roe, B.A., Affourtit, J.P., Egholm, M., Henrissat, B., Heath, A.C., Knight, R. and Gordon, J.I., 2009. A core gut microbiome in obese and lean twins. Nature 457: 480-484. Yatsunenko, T., Rey, F.E., Manary, M.J., Trehan, I., Dominguez-Bello, M.G., Contreras, M., Magris, M., Hidalgo, G., Baldassano, R.N., Anokhin, A.P., Heath, A.C., Warner, B., Reeder, J., Kuczynski, J., Caporaso, J.G., Lozupone, C.A., Lauber, C., Clemente, J.C., Knights, D., Knight, R. and Gordon, J.I., 2012. Human gut microbiome viewed across age and geography. Nature 486: 222-227.
Beneficial Microbes 4(3)
