Metabolic dissimilarity determines the establishment of cross-feeding interactions in bacteria

Samir Giri, Leonardo Ona, Silvio Waschina, Shraddha Shitut , Ghada Yousif, Christoph Kaleta, Christian Kost

Preprint posted on October 10, 2020

Don’t say ‘no’ to strangers: Cross-feeding relationships between bacteria are driven by metabolic distance

Selected by Hiral Shah

Categories: ecology, microbiology


Microbial consortia are mixed microbial communities that may be constructed synthetically or derived from natural environments. Microbial consortia impact a variety of environmental, agricultural, and medical applications including bioremediation, polymer degradation, crop improvement, synthesis of nutrition supplements, and biofilm formation (1,2). However, the complex structure and dynamic nature of such communities makes designing and establishing potent microbial consortia far from trivial (3). A key step forward is to better understand the rules that determine the evolution of microbial interactions within these communities (4). A large proportion of microbial interactions rely on an exchange of essential metabolites between two or more bacterial partners. Many of those cross-feeding interactions are unidirectional interactions, meaning that one bacterium utilizes a metabolite that is provided by another one (5). Examples for this include  auxotrophic bacteria, which benefit by utilizing a metabolite (eg. an amino acid) that is provided by a prototropic neighbor. In such cases, the benefit is often asymmetrically distributed, with the prototroph bearing the cost of metabolite biosynthesis and release. While cross-feeding is ubiquitous, it remains unclear how these relationships are established.

The authors discuss two possible hypotheses for how these interactions may occur. First, the similarity hypothesis suggests that microbial cultures support the growth of close relatives diverting resources to kin. However, microbial communities are often heterogeneous involving very diverse bacterial groups. On the contrary, the dissimilarity hypothesis expects cross-feeding interactions should be more common amongst genetically more distantly-related organisms, thus favoring less similar donor-recipient pairs to establish cross-feeding interactions. To distinguish between these two possibilities, the authors study cross-feeding between phylogenetically diverse auxotroph-prototroph bacterial pairs.

Key Findings

What determines the establishment of these cross-feeding interactions? The authors address this question using a range of different bacterial species as potential amino acid donors to strains of  Escherichia coli and Acinetobacter baylyi that were auxotrophic for the two amino acids histidine and tryptophan. Prototrophic genotypes belonged to diverse phylogenetic groups closely and distantly related to the focal recipients. The selected donors lacked any previous coevolutionary history with the recipient, thus ruling out a pre-existing interaction other than the one being investigated. First, the growth of recipients in coculture with donors was compared to the growth they achieved in monoculture. 78% of instances showed a positive response to the presence of the donor (Fig. 1A), suggesting that cross-feeding relationships are easily established between donor and recipient.


Figure1 taken directly from Giri et al figure1 under CC-BY-NC-ND 4.0 international license.


In order to define metabolites supporting the growth benefit, donor culture supernatants were analyzed. Interestingly, the growth benefit was proportional to both the total amino acid concentration in addition to the concentration of the focal amino acid suggesting that the recipient may utilize other freely available amino acids apart from just the required one. Next the authors asked whether the difference in amino acid profiles could explain the differences in growth benefits provided by different donor groups. 

Could the different amino acid profiles of these diverse donors provide insight into which of the two hypotheses was at play? It turns out that the more different the amino acid profiles produced by donor genotypes were, the higher the growth benefit it provided to the recipient, thus supporting the dissimilarity hypothesis. Furthermore, the growth benefit of recipients also correlated positively with the phylogenetic distance to donors. With different amino acid profiles already favoring growth, was phylogenetic distance only a proxy for metabolic distance? Addressing this possibility, the authors next compared the metabolic networks between different donors. Once again growth benefit was directly proportional to metabolic distance of the donor-recipient pair. Thus, overall distances favored cross-feeding interactions.

Can these findings be extended to many more different microbial interactions or does it apply to just this subset analyzed? In silico simulation of over 330,000 pairwise interactions between 818 bacteria commonly found in the gut showed that the propensity of metabolic exchange follows a saturation curve in response to metabolic distance, once again supporting dissimilarity as a key determinant driving cross-feeding relationships.



  1. Zhang, S., Merino, N., Okamoto, A. & Gedalanga, P. Interkingdom microbial consortia mechanisms to guide biotechnological applications. Microb. Biotechnol. 11, 833–847 (2018).
  2. Qian, X. et al. Biotechnological potential and applications of microbial consortia. Biotechnol. Adv. 40, 107500 (2020).
  3. Kong, W., Meldgin, D. R., Collins, J. J. & Lu, T. Designing microbial consortia with defined social interactions. Nat. Chem. Biol. 14, 821–829 (2018).
  4. Giri, S., Shitut, S. & Kost, C. Harnessing ecological and evolutionary principles to guide the design of microbial production consortia. Curr. Opin. Biotechnol. 62, 228–238 (2020). 10.1016/j.copbio.2019.12.012 
  5. D’Souza, G. et al. Ecology and evolution of metabolic cross-feeding interactions in bacteria. Nat. Prod. Rep. 35, 455–488 (2018).

Tags: dissimilarity, interactions, metabolic exchange

Posted on: 23rd October 2020


Read preprint (No Ratings Yet)

Author's response

The author team shared


  1. How was tryptophan and histidine selected as the focal metabolite? Did the in silico analysis rely on exchange of many diverse metabolites? 

The idea here was to show that the response is not species or metabolite specific. Another reason was the amino acid biosynthesis cost is different, tryptophan is more costly than histidine. The in silico analysis relies on differences in the total metabolic fluxes between organisms.

2. Does cocultivation have any impact on donor growth? Do closely related donors experience any negative effects in cocultivation?

These are interesting questions and we did not discuss this issue in the current story. However, in most cases, they did not show cost. We are currently working on this.

1 comment

3 months

Christian Kost

Thank you very much Hiral for summarizing our work in this very nice piece!

Have your say

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here