Microbial Biotechnology
John Wiley and Sons Inc.
image
Stochasticity in microbiology: managing unpredictability to reach the Sustainable Development Goals
Volume: 13 , Issue: 4
DOI 10.1111/1751-7915.13575
  • PDF   
  • XML   
  •       

Table of Contents

Highlights

Notes

Abstract

In this review paper, we highlight the impact of stochasticity in single culture and microbiome engineering. Stochasticity may give rise to some risks, such as the presence of pathogens in microbiomes. We argue here that by taking the necessary precautions and through clever monitoring and interpretation, these risks can be mitigated.

De Vrieze, De Mulder, Matassa, Zhou, Angenent, Boon, and Verstraete: Stochasticity in microbiology: managing unpredictability to reach the Sustainable Development Goals

Glossary

Axenic

A closed ecosystem maintaining a single monoclonal population

Deterministic

The absence of randomness in the development of the microbial community structure and composition, which is mainly influenced by manipulatable niche‐based factors

Drift

The random fluctuation in the number of gene variations within a population over a certain period of time

Dynamics

The change in microbial community composition in function of time

Immigration

The addition of a specific taxon to a resident community from the species pool of the metacommunity

Invasion

The establishment of an immigrant taxon within a certain niche or ecosystem

Speciation

Genetic divergence within clonal populations as a result of lateral gene transfer and incorporation of foreign DNA through recombination, mutational divergence and natural selection

Stochastic

Random processes that influence the development of the microbial community structure and composition, which cannot be steered or controlled

Introduction

For thousands of years, humans have been relying on microbial processes to produce a plethora of food products, such as bread, beer, vinegar, cheese and other fermented products. The industrialization of our society accelerated the development of novel technologies that directly exploit microorganisms and their unique abilities. For example, the treatment of wastewater by means of activated sludge and renewable energy recovery through anaerobic digestion is both established technologies, now in existence for over a century (McCarty, 1981; Sheik et al. , 2014). In our present society, drinking water and various foods contain microbial communities consisting of a variety of species that interact which each other. We define these cooperative assemblages of microorganisms as microbiomes (Burge, 1988). For future survival on this planet of our, at present 7.7 billion, but by 2050 predicted 9.7 billion co‐citizens (United Nations, 2019), we will depend on the production of these basic human goods through well‐designed and controllable systems (Kowalchuk et al. , 2008) in the framework of the Sustainable Development Goals. As long as we are unaware of the presence of a multitude of microorganisms in our surroundings and in what we eat and drink, we do not question the benefits of dealing with natural mixed and spontaneously evolving microbial systems or microbiomes.

Even though the discovery of microorganisms dates back to the 17th century with the design of a single‐lens microscope by Antonie van Leeuwenhoek (Lane, 2015), only in the last decades have we gained a view on the enormous microbial diversity in both natural and engineered ecosystems. The emergence of high‐throughput sequencing technologies has enabled us to characterize the microbial community composition, organization and dynamics, as reflected in the microbial resource management approach (Marzorati et al. , 2008), as well as identify potential key microorganisms in different systems. This has also allowed us to detect (potential) pathogenic species or genes in, for example, drinking water and microbial food products (Loman et al. , 2012; Marco et al. , 2017; Dowdell et al. , 2019). If we combine this knowledge with the fact that we are constantly in contact with microbiomes, for example through the consumption of fresh vegetables, and that such contact will increase due to the consumption of novel products of microbial fermentation, it becomes clear that societal and legislative concerns about the microbiome are imminent. This puts the use of microbiomes in engineered systems under pressure, despite their crucial role in our present society.

In the past decades, the fear for genetically modified organisms has had major consequences for the development of important applications of these powerful scientific tools. The issue of implementing microbiomes is entirely unrelated to the issue of GMOs, as no genes are modified, and all are 100% natural. Yet, one should be aware that the psychology of the consumer is sensitive to devious information. The purpose of this paper was to underscore the value of microbiomes that are not strictly defined nor strictly controllable in their evolvement. The analogy is that of a field of wheat: the farmer can never prevent that a poisonous plant species grows within his field. If regulators would impose such a request, normal agricultural practices would become impossible in terms of costs of preventing every single intruder and costs of controlling the absolute ‘purity’ of the crop (Chauhan et al. , 2017). The key feature is that one must deal with risks and this under conditions not entailing excessive costs.

Accurate control of microbial processes has been a central research topic for decades to improve process efficiency and/or accelerate production rates to maximize product quality and minimize operational costs. This ranges from basic grab‐sample pH measurements and subsequent base or acid addition to online sensorial analysis of multiple different parameters, for example in spontaneous fermented food products, to continuous steering of, for example, the oxygen supply in function of the incoming stream in activated sludge systems. However, there is a key gap between process stability, reflected in a constant output, and microbial community stability. A high degree of temporal variation has been observed in microbial community composition and organization, despite maintaining constant conditions, for example in anaerobic digestion (Fernandez et al. , 1999; De Vrieze et al. , 2013). A similar temporal variation was observed in full‐scale anaerobic digestion (De Vrieze et al. , 2016) and wastewater treatment systems (Meerburg et al. , 2016).

This temporal variation in microbial composition directly relates to the concept of deterministic vs. stochastic processes concerning microbial community organization. Deterministic or niche‐based factors strongly impact the way microbial communities are organized and functionally performing (Wang et al. , 2013; Zhou et al. , 2013b; Vanwonterghem et al. , 2014; Zhou et al. , 2014; Griffin and Wells, 2017; Zhou and Ning, 2017; Ning et al. , 2019), and these can be manipulated, to a major extent, by the process operators. In contrast, stochastic factors, such as genetic mutation, gene duplication, cell damage by radicals, die‐off, interspecies interactions, emigration, immigration and random drift also, have been shown to impact microbial community assembly and performance (Sloan et al. , 2006; Ofiteru et al. , 2010; Evans et al. , 2017) and are difficult to control. These stochastic factors are central to the neutral theory of biodiversity (Hubbell, 2001). At present, a consensus exists that both deterministic and stochastic factors shape microbial communities in natural and engineered systems (Van Der Gast et al. , 2008; Zhou et al. , 2013b; Xu et al. , 2019). Single (pure) or axenic cultures are also stochastic, since these may be subjected to various gene modifications, which cannot be predicted, as well as to various types of stress. Nonetheless, the level of unpredicted changes can be larger in systems open to the immigration of new species. Therefore, a key issue in engineered microbial processes, especially in food and drinking water niches, is monitoring and control of stochasticity to a high‐quality and safe end product.

The key objective of this study was to set a framework for the concept of stochasticity as an important process that shapes the microbiome in engineered systems. First, the concept of stochasticity in single (pure) cultures and microbiomes will be considered. Next, the mechanisms underlying stochasticity and the consequences (issues and opportunities) for engineered systems will be examined. Finally, the concept of stochasticity will be approached from a regulatory point of view for (in)direct human applications.

The concept and mechanisms of stochasticity in microbial ecology

Fuelled by rapid developments in (meta)genomics, transcriptomics, proteomics and metabolomics techniques, greater attention has been paid to study microbial communities in their entirety and to identify the mechanisms that shape and influence microbial ecosystems of both single cultures and microbiomes. Mainly in the last decade, scientists started to acknowledge the importance of ecological stochasticity in shaping the microbial community (Zhou and Ning, 2017). Stochasticity differentially impacts closed single cultures and open microbiomes, given the absence of immigration in closed single culture systems (Fig. 1).

Schematic overview of the differential impact of stochasticity with respect to closed single culture and open mixed culture (microbiome) systems. The key difference is the contribution of immigration to open systems, in contrast to closed systems.
Fig. 1
Schematic overview of the differential impact of stochasticity with respect to closed single culture and open mixed culture (microbiome) systems. The key difference is the contribution of immigration to open systems, in contrast to closed systems.

Stochasticity at the cellular level: drift in single (pure) cultures

Several industrial processes, e.g., to produce amino acids, organic acids, antibiotics and enzymes (Jiang et al. , 2017) and microbiology research studies are carried out using axenic cultures, cultivated in a closed and homogenous environment. Such a single‐strain culture is not challenged by interspecies interactions and immigration and could, therefore, be presumed more stable over time, provided that the efforts to maintain axenic conditions prove successful. However, this does not eliminate stochasticity, as multiple other stochastic processes can influence the microbial community.

Nicholson (2019) stepped away from a deterministic view of the cell as a neatly organized machine, but instead emphasized the dynamic nature of its constitution, the fluidity of its components and the non‐linear stochasticity of its underlying processes. Stochasticity, or noise, in gene expression is the driving force behind phenotypic variation within the same strain of species (Elowitz et al. , 2002; Raj and van Oudenaarden, 2008), which seems an inherent survival strategy offering resilience and flexibility to a clonal population. Kussell and Leibler (2005) proposed that clonal bacterial populations can use stochastic phenotype‐switching as a response to random and infrequent environmental changes. The randomization of the phenotype, i.e., phenotypic plasticity (Kussell and Leibler, 2005; Fusco and Minelli, 2010; Ackermann, 2015), as a survival strategy is favoured compared to responsive adaptation when the random switching rate mimics the statistics of environmental changes (Kussell and Leibler, 2005). Many bacterial species can navigate environmental change by utilizing their diverse metabolic capacities (Meyer et al. , 2004; Swingley et al. , 2007; Narancic et al. , 2012), which provides these populations with a distinctive advantage over non‐versatile populations. It allows them to (i) persist at fluctuating conditions and (ii) divide labour between individual cells to increase the overall population function (Ackermann, 2015). It is well established that environmental factors can promote metabolic diversification, but also cell‐inherent dynamics can cause metabolic heterogeneity within clonal microbial populations (Takhaveev and Heinemann, 2018).

In contrast to phenotypic heterogeneity, genetic drift can promote genome reduction in a population in low fluctuating environments. New mutations (with a higher tendency for deletions) are more likely to become fixed when a bacterial lineage has to adapt to a novel lifestyle that reduces its long‐term population size, such as obligate symbiosis or a limited habitat range (Kuo et al. , 2009). For example, under constant ambient conditions (e.g. controlled reactor systems), bacteria with less coping mechanisms may outcompete and replace those that need to invest energy maintaining mechanisms that they do not require to survive under those specific circumstances. Thus, genetic drift can be a survival strategy when a single culture is subjected to fluctuating conditions.

Stochasticity in natural mixed microbial communities or microbiomes

Nature is primarily stochastic. Flexibility is a key trait of microbial populations, consisting of a single species, as well as entire natural and engineered microbial ecosystems. In natural ecosystems, such as wetlands, soils, estuaries and marine environments, microorganisms occur in complex and multispecies microbiomes, which can vary greatly in composition, habitat, functionality and their relationships to the ambient environment. These communities can range from very few species to complex aggregates of trillions of cells and go through stages of community succession (Lyautey et al. , 2005; Datta et al. , 2016; Wright et al. , 2019). The complexity of community dynamics, even in low diverse communities, can be illustrated through the interactions of three common gut bacteria cultivated under well‐controlled in vitro conditions (D'Hoe et al. , 2018). The results showed the difficulty in predicting the winner without a predictive model, which should include both the internal metabolism of community members, as well as their response to interaction partners. Successes in modelling the behaviour of community dynamics (Song et al. , 2015; Song, 2018; Succurro and Ebenhöh, 2018) suggest that communities are shaped by deterministic and therefore predictive mechanisms. Such models (i) are limited by stringent biological assumptions (Succurro and Ebenhöh, 2018), (ii) group species in metabolic ‘guilds’ (Kettle et al. , 2015; Muñoz‐Tamayo et al. , 2016) or higher taxonomic levels (Cremer et al. , 2017) or (iii) only apply to select species within a composed (synthetic) community (D'Hoe et al. , 2018). This makes these models, despite their relevance to understand ecosystem functioning, often insufficient to accurately predict stochasticity in microbiomes.

The dynamics of any microbial community, reflected in successional shifts (both genetic and phenotypic), is powered by a broad array of deterministic factors, i.e., pairwise interactions that occur between species, such as (i) syntrophy (mutually dependent interaction) (Morris et al. , 2013; Schink and Stams, 2013), (ii) synergy (microbes supporting each other’s growth by creating favourable conditions) (Herschend et al. , 2018; Shaikh et al. , 2018), (iii) predation (Chen et al. , 2011; Fukami and Nakajima, 2011; Maslov and Sneppen, 2017) and (iv) competition for physical space and substrate (Hibbing et al. , 2010) and interactions that occur between microorganisms and their environment. Microorganisms can alter their environment through the production of various metabolites, the formation of flocs and biofilms, and precipitation or solubilization of reactive substances. Ambient conditions, such as pH, temperature and substrate availability, in turn, can influence the survival and growth rate of microorganisms. These multiple interactions and external influences, generally, do not destabilize microbial communities, because there is a constant pursuit for optimal functionality through a selection of species from a diverse pool of redundant microorganisms provided by the frequent influx of ‘immigrant species’ in an open system. This constant selection combined with a diverse microbial community allows that microbial assemblies, undergoing successions and evolving as complex communities, can tolerate a large level of stochasticity and still achieve constant high‐level performance. Several specific stochastic processes have been identified to govern microbial populations and induce structural fluctuations. Random immigration (Vuono et al. , 2016; Kirkegaard et al. , 2017; Mei and Liu, 2019) within open ecosystems is a key process that can alter community assembly through niche and fitness driven selection leading to invasion (Li et al. , 2019), combined with stochastic extinction (Kramer et al. , 2013), genetic drift and speciation (Hanage et al. , 2006). Limitation of dispersion can further increase stochasticity in community assembly and downplay the influence of environmental (deterministic) variables (Evans et al. , 2017).

It has been statistically concluded that stochastic processes shape the bacterial communities in the blood of pikas and their arthropod vectors (Li et al. , 2018), using the nearest taxon index (NTI) and mean nearest taxon distance (MNTD), as described by Stegen and colleagues (2013). The NTI values between −2 and 2 indicate stochastic community assembly, whereas NTI values less than −2 or higher than 2 indicate that deterministic processes play a more important role in structuring the community (Li et al. , 2018). On the other hand, Meyerhof and colleagues (2016) found that microbial communities in marine lakes were driven by the deterministic processes (all NTI > 4) of environmental selection, resulting from oxygen, salinity and pH gradients, which act as filters on microbial growth (Meyerhof et al. , 2016). Hence, the higher the harshness of environmental filters, the higher the dominance of deterministic processes (Chase, 2007; Meyerhof et al. , 2016). If a full physical barrier towards all other species is installed (i.e. in single culture systems), the level of determinism is at its maximum. However, even then, as indicated before, stochasticity cannot be fully excluded.

Microbiomes in industrial applications are governed by similar stochastic and deterministic mechanisms. Belgian Gueuze beers are traditionally produced by spontaneous fermentation, which is characterized by microbial succession over a 6‐month fermentation duration (Spitaels et al. , 2014). Many other food products, such as salami, sauerkraut, sourdough, surströmming and hákarl, wine, cheese, yoghurt and kefir, all rely on microbiomes for flavour, odour, texture and even shelf life. In vegetable fermentation facilities that produce spontaneously fermented sauerkraut, the raw vegetables and indoor environment and surfaces hosted distinct microbiomes, which were reflected in the final product (Einson et al. , 2018). In contrast, human contamination was found to have no effect on the final product. The general finding is that main bacterial species involved in such processes are re‐occurring, due to inoculum composition and fermentation conditions, resulting in highly similar community compositions within the same production process (Catzeddu et al. , 2006; Połka et al. , 2015; Shangpliang et al. , 2018). Nevertheless, diverse and dynamic communities are inherent to these fermentation processes and often contain a collection of low abundant (rare) and transient species (Połka et al. , 2015; Shangpliang et al. , 2018). Yet, the products itself are perceived as safe, and some even as health‐promoting (Marco et al. , 2017; Einson et al. , 2018), supported by a long history of trouble‐free consumption.

The implications of stochasticity: benefits versus challenges

Stochastic behaviour can be modelled both at the population and ecosystem level (Ning et al. , 2019), yet the direction in which populations and ecosystems shift cannot be predicted. The inherent uncontrollability and limited predictability of stochasticity are both a strength and a weakness of microbial communities used in engineered systems. Engineered systems conventionally and intrinsically provide predictability and stability, and uncontrollable variation is undesirable. Yet, the fact that a system can evolve in a multitude of ways (i.e. demonstrating flexible behaviour) can be a strong asset to assure optimal process performance under changing conditions.

The challenges of stochasticity

The limited degree of predictability of stochasticity can be considered at the genomic, phenotypic and ecosystems level and can lead to unforeseen deviations from the optimal process performance in an engineered ecosystem. This requires the need to take preventive actions to limit the effects of stochasticity and/or take necessary precautions to steer stochasticity towards a beneficial outcome.

Single cultures in the laboratory

At the genomic level, genetic drift as a consequence of multiple‐generation isolation leads to genetic impoverishment, which causes these so‐called ‘laboratory rats’ to lose at least part of their metabolic potential (Lenski and Travisano, 1994; Masel, 2011). This was demonstrated in a 10,000 generation experiment with 12 populations of Escherichia coli , in which a clear drift away from the ancestor was observed in the first generations (Lenski and Travisano, 1994). At the level of eukaryotes, 23 generations of Chironomus riparius resulted in a genetic impoverishment that could not be recovered by crossing different laboratory strains (Nowak et al. , 2007). Genetic drift in laboratory bacterial cultures could be prevented by (i) limiting the number of laboratory generations and/or (ii) keeping relatively large populations sizes (> 100) in the laboratory (Etzel and Legner, 1999).

Open systems are prone to stochasticity due to immigration and invasion

Open ecosystems that rely on microbiomes, such as activated sludge, anaerobic digestion, composting and the (human) digestive tract, imply that these are open to different stochastic processes. In that context, two features are of particular interest, which are (i) entry of outsiders in the microbial ecosystem and (ii) exit of insiders to the outside world. The potential downsides of such an open system are twofold.

Invasion of non‐desirable microorganisms could disturb process performance or compromise safety. Invasion can be defined as the establishment of an ‘immigrant’ species in a resident microbial community (Kinnunen et al. , 2016). This contrasts with immigration, which we define as the mere inflow, but not necessarily establishment, of species in a resident microbial community. Hence, invasion can be considered a consequence of immigration. Three key factors determine the potential degree of invasion. First, the presence or absence of environmental variables that induce stress in the microbial community strongly determines the susceptibility of the community to invasion. Under stress conditions, an invasive species could sustain process performance, while in the absence of such stress conditions, process performance can be impacted (De Roy et al. , 2013). Second, biodiversity in general (van Elsas et al. , 2012; Johnson et al. , 2013; Mallon et al. , 2015), and, more specifically, evenness (De Roy et al. , 2013), which also promotes process performance under stress (Wittebolle et al. , 2009), reduces the risk for and potential impact of invasion. Third, nutrient limitation also prevents the growth of potential invaders (Van Nevel et al. , 2013). Therefore, engineered ecosystems that are potentially susceptible to invasion‐induced process disturbance are those systems that have a low diversity with un‐occupied niches (i.e. tasks to be taken up) and a certain availability of nutrients. This implies that sterilized systems exposed to the open environment are extremely fragile, but also drinking water and composed culture‐based systems are prone to invasion and growth of outsiders (Fig. 2). An example is the natural birth of a ‘sterile’ human baby and the subsequent immediate colonization that follows until a mature human microbiome has developed (Koenig et al. , 2011). Another example is a well‐controlled system using composed cultures in the framework of the Micro‐Ecological Life Support System Alternative (MELiSSA) approach (Godia et al. , 2002). This approach (partially) relies on co‐cultures or single cultures in a bioregenerative life support system for complete recycling of gas, liquid and solid waste streams during space exploration (Hendrickx et al. , 2006). In such a system, heterotrophic strains need to be added to avoid/reduce invasion and subsequent process failure (Christiaens et al. , 2019). In contrast, even though immigration of species, which may lead to invasion, commonly takes place in open diverse systems, such as anaerobic digestion (Kirkegaard et al. , 2017) and activated sludge (Vuono et al. , 2016), mainly through the feedstock, this does not disturb and can even promote process performance.

Visualization of the potential impacts of immigration in open systems in relation to nutrient availability and indigenous microbial community diversity. A low nutrient availability prevents the growth of invasive species, while a high indigenous microbial diversity promotes homeostasis to eliminate potential invaders.
Fig. 2
Visualization of the potential impacts of immigration in open systems in relation to nutrient availability and indigenous microbial community diversity. A low nutrient availability prevents the growth of invasive species, while a high indigenous microbial diversity promotes homeostasis to eliminate potential invaders.

In the case invasion does not disturb process performance, the introduction of new microbial species could have a direct or indirect impact to the outside environment in general and to the human health in particular. Invasion in systems in which not all niches are filled is reflected by the growth of non‐tuberculous mycobacteria (Dowdell et al. , 2019) and Legionella sp. (Lu et al. , 2016) in drinking water systems. The global occurrence of the protozoan pathogens Cryptosporidium and Giardia species in Western drinking water systems in the 1990s (Lechevallier et al. , 1991), and still commonly present in drinking water supplies in developing countries (Omarova et al. , 2018), indicates that such vulnerability to invasion remains a pressing issue. Cryptosporidium and Giardia species are also commonly present in ‘rich’ and diverse systems, such as wastewater sludge (Ramo et al. , 2017) and animal manure (Vermeulen et al. , 2017), and their removal in anaerobic digestion remains poorly documented (Nag et al. , 2019). A similar observation can be made for other pathogens, such as Legionella sp. (Viau and Peccia, 2009) and Salmonella sp. (Kjerstadius et al. , 2013) in digestate. Once present and capable to proliferate, such putative pathogens may reach the levels and physiological conditions that could impose a risk to human health, particularly if spreading could occur by aerosols (Brooks et al. , 2012).

Overall, in open systems, and even in well‐controlled semi‐closed systems, stochasticity, especially through invasion, effectively poses important challenges, but numbers need to be taken into consideration. Process failure is rarely an issue, but the level of putatively pathogenic species developing inside the system can be of concern in case of high infectivity, i.e., the probability of infection from exposure to one cell or particle (Rose and Gerba, 1991). By operating under conditions with limiting energy and nutrient sources, and/or conditions that sustain a high indigenous microbial diversity, stochasticity is not conflicting with reliability and safety, particularly concerning species that require high numbers to be infective.

The benefits of stochasticity

Although stochasticity has downsides for single culture, composed culture and mixed culture (microbiome) engineered processes, the key advantages, especially towards microbiomes, are also twofold.

Homeostasis is the result of a dynamic balance of interactions between microorganisms in the community and with their environment (Li and Tian, 2016), which can be considered a key feature of invasion prevention. The constant threat of invasion keeps the system ‘alert’ towards hazardous invasions/disturbances that would impact process performance, due to the following two mechanisms. First, in a system that experiences a constant level of stochasticity, there is a higher probability that all niches are filled, thus, avoiding the intrusion of unwanted species (Ofiteru et al. , 2010). Second, stochasticity also maintains a certain level of process dynamics, even at highly stable and controlled operational conditions, thereby promoting the overall resistance of the process against disturbances (Briones and Raskin, 2003; De Vrieze et al. , 2013). Therefore, stochasticity can be considered a potential guardian of homeostasis, in addition to deterministic interactions between the microbiome and its environment.

Another key advantage of stochasticity resides in the fact that it sustains overall functional performance through the influx of novel species and/or creation of new niches, even though this strongly depends on the nature of the ecosystem (Marzorati et al. , 2008) and the spatial scale (Martiny et al. , 2011). It is generally assumed that a higher microbial diversity improves process stability and performance of engineered processes (Briones and Raskin, 2003; Beyter et al. , 2016) through a higher probability of positive species interactions, thus, positively affecting process performance (Cardinale et al. , 2002). Nonetheless, the benefit of diversity in microbial ecology is not a universal truth, as diversity assessment can be highly method‐dependent (Bent and Forney, 2008; De Vrieze et al. , 2018) and sampling‐related (Zhou et al. , 2011; Zhou et al. , 2013a). It only provides the onset of unravelling ecological mechanisms (Shade, 2017). A too high diversity could even increase antagonistic interactions (Becker et al. , 2012). The impact of stochasticity to functionality differs whether the microbial community is resistant, resilient or redundant in response to disturbances (Allison and Martiny, 2008). The key contribution of stochasticity to functionality resides in the increase of the redundancy potential of the microbial community. Community resistance (i.e. an unaltered microbial community in response to disturbances) appears to be independent of stochasticity. Community resilience (i.e. an altered microbial community that bounces back to its original composition) seems to be more depending on stochasticity though only if the temporal alteration sustains function. For community redundancy (i.e. a permanently altered community after a disturbance), stochasticity is key to create and sustain a plethora of novel community equilibrium compositions following a disturbance. Redundancy can be considered as key in microbiomes (Louca et al. , 2018). This is illustrated in anaerobic digestion (Langer et al. , 2015; De Vrieze et al. , 2017; Spirito et al. , 2018), where microbial communities continuously evolve as an inevitable consequence of stochasticity, but they nonetheless sustain functionality. Resilience and redundancy can both contribute to functionality within one microbiome. For example, syntrophic bacteria were found to be functional specialist and resilient, while higher level fermenters, such as Clostridia, were found to be functionally redundant within the microbiome of anaerobic granules for brewery wastewater treatment (Werner et al. , 2011).

Overall, stochasticity seems to be key to sustaining process performance by maintaining the microbial community ‘active’ (homeostasis) and ‘equipped to deal’ with to disturbances (redundancy). These examples show that stochasticity should be embraced for microbiome engineering as a natural mechanism of sustaining adequate process performance.

How to deal with stochasticity: the regulators’ challenge

Stochasticity is part of life when dealing with both single cultures and microbiomes in biotechnological applications and process engineering. Thus, it requires a suitable framework for the regulatory authority to deal with stochasticity in an economically feasible and safe way. This is a twofold issue, as it requires (i) accurate monitoring of stochasticity and (ii) setting of boundaries on stochasticity, regarding the system in which it is considered. A proposal to solve both issues is formulated here.

Monitoring stochasticity: a technological challenge

To monitor stochasticity, methods to detect changes in both single and mixed genetic and phenotypic traits are needed. Single culture phenotypic changes can be monitored through flow cytometry (Müller et al. , 2010) or matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry (MALDI‐TOF MS) (Nowakiewicz et al. , 2017; Zhao et al. , 2017) as high‐throughput alternatives of conventional plating. Concerning the genetic level and the expression thereof, knowledge of the complete genome and plasmid genetic code is essential to avoid the potential expression of genes related to, for example, the production of toxins or antibiotic resistance in advance.

In contrast to single cultures that enable accurate monitoring of stochasticity, amongst others due to the elimination of immigration, this is much more challenging in microbiomes in which invasion is a key stochastic process. In an ideal situation, pathogenic microorganisms with their related virulence traits or antibiotic resistance genes (ARGs) should be absent in microbiomes in biotechnological applications. However, given the open feature of these systems, the invasion of potentially pathogenic species cannot be avoided and requires accurate monitoring.

A first monitoring approach concerns the ‘fingerprinting’ approach in which the degree of microbial community dynamics, i.e., the change in microbial community composition in function of time can be evaluated. Such a fingerprint and its evolution in function of time can be obtained on the genetic level through 16S rRNA gene amplicon sequencing (Pilloni et al. , 2012) or even simpler/older techniques, such as terminal restriction fragment length polymorphism (T‐RFLP) (Pilloni et al. , 2012; Camarinha‐Silva et al. , 2014; De Vrieze et al. , 2018), automated ribosomal intergenic spacer analysis (ARISA) (Gobet et al. , 2014; van Dorst et al. , 2014) or denaturing gradient gel electrophoresis (DGGE) (Marzorati et al. , 2008). Fingerprinting can also be carried out at the phenotypic level through flow cytometry (De Roy et al. , 2012; Props et al. , 2018) and at the metabolic level through MALDI‐TOF MS (Sala‐Comorera et al. , 2016; Sandrin and Demirev, 2018). A key aspect in this approach concerns knowledge of the contribution of deterministic processes, which can be monitored and controlled, to the microbial community dynamics. Subsequently, the variance in community dynamics not explained by deterministic processes can be attributed to influence of stochastic processes.

Benchmarking of microbial community dynamics is essential to distinguish between a system with acceptable background dynamics, as observed in anaerobic digestion (Klang et al. , 2015) and activated sludge (Mielczarek et al. , 2012; Mielczarek et al. , 2013; Ju and Zhang, 2015) systems, and uncontrollable domination of stochasticity over deterministic effects. Conventional microbial community dynamics at steady state ranges between 20 and 30% change in microbial community composition in a 14‐day window in anaerobic digestion (Pycke et al. , 2011; De Vrieze et al. , 2016) and activated sludge (Wang et al. , 2010) systems. However, the background degree of change depends on multiple factors and should be evaluated for each case specifically. For example, the dynamics of the ammonia oxidizing bacterial community was twice as high in a membrane bioreactor, compared to a sequential batch reactor, both showing functional stability and operating under similar conditions (Wittebolle et al. , 2008).

Drinking water supply systems, as an example of a process related to direct human usage, require a more direct monitoring of stochasticity beyond the DNA level to phenotypic and/or activity changes for which flow cytometry (De Roy et al. , 2012) and MALDI‐TOF MS (Sala‐Comorera et al. , 2016) could be suitable methods. Such flow cytometry methods can even be extended to more complex systems, such as anaerobic digestion (Dhoble et al. , 2016) and activated sludge (Günther et al. , 2012), indicating the progress in accurate fingerprinting techniques. Overall, the key aspect of these fingerprinting techniques is their moderate level of cost and rapid nature to gain a real‐time view on the status of the stochasticity impact.

Key to the validity of the fingerprinting approach is knowledge on (i) the initial status of the microbial community, (ii) the deterministic processes (potentially) influencing the microbial community and (iii) the (change in) input streams, e.g., the absence of pathogens and ARGs, for which the ‘omics’ or targeted methods are needed, as discussed above. Taking these three conditions into consideration, if fingerprinting methods do not indicate a deviation from the background variation, no additional analyses nor direct actions are needed. Deviations from the background dynamics profile warrant further investigation to pinpoint potential stochastic hazards and take necessary actions.

Even though these fingerprinting techniques provide an important overall view on the microbial community dynamics and distinguish background dynamics from strong deviations or sudden changes, they fail to detect the gradual potential influx of pathogens. Hence, identification at the lowest most precisely possible taxonomic level is mandatory to identify potential pathogens. For this, 16S rRNA gene amplicon sequencing falls short on two levels, i.e., (i) the limited depth of phylogenetic identification (Poretsky et al. , 2014) and (ii) the inability to accurately identify rare (pathogenic) species (Huse et al. , 2010; Kunin et al. , 2010), in contrast to metagenome analyses. Metagenome analysis has the potential to identify genes that are possibly related to antibiotic resistance or to the production of toxins, e.g., in reclaimed water distribution systems (Garner et al. , 2018), with co‐occurrences across ecosystems (Li et al. , 2015). However, their presence in the metagenome does not necessarily imply their actual expression.

Monitoring of expression or translation of undesirable genes, e.g., related to toxicity, can take place at the RNA, protein and general metabolic level, either focussed on the general microbial community level or to specific targets. To gain an overall view on the microbial community, ‘omics’ approaches are essential to map the actual ‘performance’ of the microbial community to stochasticity. A metatranscriptomic approach has been shown useful to identify the transcription of ARGs in activated sludge systems (Liu et al. , 2019) and to characterize host–microbe pathogenic interactions (Hampton‐Marcell et al. , 2013). Yet, this still presents multiple methodological challenges (Asante and Osei Sekyere, 2019). Similarly, metaproteomic and metabolomic analyses could assist in the characterization of stochastic hurdles in microbiomes. Based on this information, specific targets can be identified for which suitable methods can be optimized, for example using real‐time PCR methods, for detecting ARGs and/or pathogens in wastewater treatment systems (Volkmann et al. , 2004; Tao et al. , 2014), manure treatment systems (Yu et al. , 2005) and artificial groundwater recharge systems (Böckelmann et al. , 2009), as well as their expression and/or activity.

Overall, monitoring of stochasticity, especially related to the key impact of invasion, can be considered challenging in microbiomes, in contrast to single culture‐based systems. A combination of technologies is essential to monitor stochasticity in microbiomes. Fingerprinting at different levels as such can be considered insufficient, yet can serve as the first stochastic ‘smoking gun’ to push for further elucidation, for which the ‘omics’ can provide subsequent suitable information. This information then needs to be translated into accurate methods that can pinpoint stochasticity‐related microbial flaws in the system. An integrated case‐by‐case approach should be considered, as discussed in Setting the boundaries on stochasticity: tolerance versus assurance, to monitor stochasticity for which the application range needs to be investigated.

Setting the boundaries on stochasticity: tolerance versus assurance

Technological developments in the field of high‐throughput sequencing in the last decades have provided us with the tools for detailed monitoring of stochasticity in a microbiome, as discussed above. However, the time‐consuming and complex nature, as well as associated costs of high‐throughput techniques, are bottlenecks for full‐scale application and warrant consideration for the frequency and depth of application. Nonetheless, high‐throughput sequencing methods are becoming cheaper, faster and more directly accessible, e.g., with the development of the MinION™ device (Quick et al. , 2014) and gradually find their way into different fields, such as clinical microbiology (Deurenberg et al. , 2017). Here, we propose an integrated approach (Fig. 3) to monitor stochasticity to keep the potential risks towards human health and process disturbance as low as reasonably achievable (ALARA).

Schematic overview of the different steps for stochasticity monitoring and control in microbiomes. Full lines represent actual decisions and actions, while dashed lines represent flows of information.
Fig. 3
Schematic overview of the different steps for stochasticity monitoring and control in microbiomes. Full lines represent actual decisions and actions, while dashed lines represent flows of information.

Initial screening of the degree of stochasticity in function of time (e.g. dynamics) can take place through basic fingerprinting methods, as mentioned earlier. The selection of the appropriate fingerprinting method depends on the system, as amplicon sequencing at the DNA level can be considered suitable for processes in which the outcome does not directly link to human consumption, such as anaerobic digestion and activated sludge systems.

At present, such an approach that relies on microbial fingerprinting represents a balance between the need for costly and time‐consuming analyses and the reliability of timely screening, even though this does not provide 100% assurance of process/product safety. The industrial practice of using microbiomes generally allows that the ‘contaminants’ can be present at a factor 10 000 lower than the functional species of interest. A clear example of how the complete absence of unwanted species is an unfeasible imposition for full‐scale industrial processes using microbiomes is provided by the production of microbial protein from natural gas (Hamer, 2010). In this case, the final product (BioProtein®) obtained by growing a single culture of Methylococcus capsulatus under semi‐sterile conditions was eventually characterized by the presence of heterotrophic bacterial strains, which fill a key functional niche necessary to maintain acceptable process performances, i.e., the degradation of the organic carbon produced by Methylococcus capsulatus (Bothe et al. , 2002). Such cooperation and the possibility that newly appearing species might change over time did not hinder the regulatory approval of BioProtein® as a feed additive. The same product is now being developed for human food purposes (Strong et al. , 2015). Other less obvious examples are reflected in Belgian Gueuze beers (Spitaels et al. , 2014), traditional Icelandic fermented fish, i.e., hákarl (Osimani et al. , 2019), traditional cheeses (Montel et al. , 2014) and salami (Połka et al. , 2015), none of which contain a defined microbiome, but also contain few or no pathogenic interference in the processing stage or final product.

Overall, engineered processes involving microbiomes are essential for waste(water) treatment, energy recovery, drinking water supply and food processing, and, therefore, cannot be eliminated from our present society. Complete elimination of all risks in microbiomes is impossible (tolerance), due to the stochastic nature of the microbiome. However, if the necessary precautions and above‐mentioned level of monitoring, e.g., on site via the MinION™ device (Maestri et al. , 2019), are considered (assurance), microbial processes will continue to play a crucial role in our efforts to pursue the Sustainable Development Goals for a more sustainable future.

Conflict of interest

None declared.

Acknowledgements

Jo De Vrieze is supported as postdoctoral fellow by the Research Foundation Flanders (FWO‐Vlaanderen). Silvio Matassa is supported as postdoctoral fellow by the Biofeedstock project, funded under the National Operational Programme (PON‐Italy) 2014/2020. Jizhong Zhou is supported by ENIGMA—Ecosystems and Networks Integrated with Genes and Molecular Assemblies (http://enigma.lbl.gov), a Scientific Focus Area Program at Lawrence Berkeley National Laboratory is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Biological & Environmental Research under contract number DE‐AC02‐05CH11231. Largus Angenent acknowledges support from the Alexander von Humboldt Foundation in the framework of the Alexander von Humboldt Professorship endowed by the Federal Ministry of Education and Research in Germany. This work was also supported through the Geconcerteerde Onderzoeksactie (GOA) from Ghent University (BOF15/GOA/006).

References

1 

    Ackermann, M. (2015) . A functional perspective on phenotypic heterogeneity in microorganisms. Nat Rev Microbiol13: , pp.497–508.

2 

    Allison, S.D., and Martiny, J.B.H. (2008) . Resistance, resilience, and redundancy in microbial communities. Proc Natl Acad Sci USA105: , pp.11512–11519.

3 

    Asante, J., and Osei Sekyere, J. (2019) . Understanding antimicrobial discovery and resistance from a metagenomic and metatranscriptomic perspective: advances and applications. Environ Microbiol Rep11: , pp.62–86.

4 

    Becker, J., Eisenhauer, N., Scheu, S., and Jousset, A. (2012) . Increasing antagonistic interactions cause bacterial communities to collapse at high diversity. Ecol Lett15: , pp.468–474.

5 

    Bent, S.J., and Forney, L.J. (2008) . The tragedy of the uncommon: understanding limitations in the analysis of microbial diversity. ISME J2: , pp.689–695.

6 

    Beyter, D., Tang, P.‐Z., Becker, S., Hoang, T., Bilgin, D., Lim, Y.W., et al (2016) . Diversity, productivity, and stability of an industrial microbial ecosystem. Appl Environ Microbiol82: , pp.2494.

7 

    Böckelmann, U., Dörries, H.‐H., Ayuso‐Gabella, M.N., Salgot de Marçay, M., Tandoi, V., Levantesi, C., et al (2009) . Quantitative PCR monitoring of antibiotic resistance genes and bacterial pathogens in three European artificial groundwater recharge systems. Appl Environ Microbiol75: , pp.154–163.

8 

    Bothe, H., Møller Jensen, K., Mergel, A., Larsen, J., Jørgensen, C., Bothe, H., and Jørgensen, L. (2002) . Heterotrophic bacteria growing in association with Methylococcus capsulatus (Bath) in a single cell protein production process. Appl Microbiol Biotechnol59: , pp.33–39.

9 

    Briones, A., and Raskin, L. (2003) . Diversity and dynamics of microbial communities in engineered environments and their implications for process stability. Curr Opin Biotechnol14: , pp.270–276.

10 

    Brooks, J.P., McLaughlin, M.R., Gerba, C.P., and Pepper, I.L. (2012) . Land application of manure and class B biosolids: an occupational and public quantitative microbial risk assessment. J Environ Qual41: , pp.2009–2023.

11 

    Burge, M.N. (1988) Fungi in Biological Control Systems. Manchester, UK: Manchester University Press.

12 

    Camarinha‐Silva, A., Jauregui, R., Chaves‐Moreno, D., Oxley, A.P.A., Schaumburg, F., Becker, K., et al (2014) . Comparing the anterior nare bacterial community of two discrete human populations using Illumina amplicon sequencing. Environ Microbiol16: , pp.2939–2952.

13 

    Cardinale, B.J., Palmer, M.A., and Collins, S.L. (2002) . Species diversity enhances ecosystem functioning through interspecific facilitation. Nature415: , pp.426–429.

14 

    Catzeddu, P., Mura, E., Parente, E., Sanna, M., and Farris, G.A. (2006) . Molecular characterization of lactic acid bacteria from sourdough breads produced in Sardinia (Italy) and multivariate statistical analyses of results. Syst Appl Microbiol29: , pp.138–144.

15 

    Chase, J.M. (2007) . Drought mediates the importance of stochastic community assembly. Proc Natl Acad Sci USA104: , pp.17430.

16 

    Chauhan, B.S., Matloob, A., Mahajan, G., Aslam, F., Florentine, S.K., and Jha, P. (2017) . Emerging challenges and opportunities for education and research in weed science. Front Plant Sci8: , pp.1–13.

17 

    Chen, H., Athar, R., Zheng, G., and Williams, H.N. (2011) . Prey bacteria shape the community structure of their predators. ISME J5: , pp.1314–1322.

18 

    Christiaens, M.E.R., De Paepe, J., Ilgrande, C., De Vrieze, J., Barys, J., Teirlinck, P., et al (2019) . Urine nitrification with a synthetic microbial community. Syst Appl Microbiol42: 126021.

19 

    Cremer, J., Arnoldini, M., and Hwa, T. (2017) . Effect of water flow and chemical environment on microbiota growth and composition in the human colon. Proc Natl Acad Sci USA114: , pp.6438.

20 

    Datta, M.S., Sliwerska, E., Gore, J., Polz, M.F., and Cordero, O.X. (2016) . Microbial interactions lead to rapid micro‐scale successions on model marine particles. Nat Commun7: , pp.11965.

21 

    De Roy, K., Clement, L., Thas, O., Wang, Y., and Boon, N. (2012) . Flow cytometry for fast microbial community fingerprinting. Water Res46: , pp.907–919.

22 

    De Roy, K., Marzorati, M., Negroni, A., Thas, O., Balloi, A., Fava, F., et al (2013) . Environmental conditions and community evenness determine the outcome of biological invasion. Nat Commun4: , pp.5.

23 

    De Vrieze, J., Verstraete, W., and Boon, N. (2013) . Repeated pulse feeding induces functional stability in anaerobic digestion. Microb Biotechnol6: , pp.414–424.

24 

    De Vrieze, J., Raport, L., Roume, H., Vilchez‐Vargas, R., Jáuregui, R., Pieper, D.H., and Boon, N. (2016) . The full‐scale anaerobic digestion microbiome is represented by specific marker populations. Water Res104: , pp.101–110.

25 

    De Vrieze, J., Christiaens, M.E.R., Walraedt, D., Devooght, A., Ijaz, U.Z., and Boon, N. (2017) . Microbial community redundancy in anaerobic digestion drives process recovery after salinity exposure. Water Res111: , pp.109–117.

26 

    De Vrieze, J., Ijaz, U.Z., Saunders, A.M., and Theuerl, S. (2018) . Terminal restriction fragment length polymorphism is an “old school” reliable technique for swift microbial community screening in anaerobic digestion. Sci Rep8: 16818.

27 

    Deurenberg, R.H., Bathoorn, E., Chlebowicz, M.A., Couto, N., Ferdous, M., García‐Cobos, S., et al (2017) . Application of next generation sequencing in clinical microbiology and infection prevention. J Biotechnol243: , pp.16–24.

28 

    Dhoble, A.S., Bekal, S., Dolatowski, W., Yanz, C., Lambert, K.N., and Bhalerao, K.D. (2016) . A novel high‐throughput multi‐parameter flow cytometry based method for monitoring and rapid characterization of microbiome dynamics in anaerobic systems. Bioresour Technol220: , pp.566–571.

29 

    D'Hoe, K., Vet, S., Faust, K., Moens, F., Falony, G., Gonze, D., et al (2018) . Integrated culturing, modeling and transcriptomics uncovers complex interactions and emergent behavior in a three‐species synthetic gut community. eLife7: e37090.

30 

    van Dorst, J., Bissett, A., Palmer, A.S., Brown, M., Snape, I., Stark, J.S., et al (2014) . Community fingerprinting in a sequencing world. FEMS Microbiol Ecol89: , pp.316–330.

31 

    van Elsas, J.D., Chiurazzi, M., Mallon, C.A., Elhottovā, D., Krištůfek, V., and Salles, J.F. (2012) . Microbial diversity determines the invasion of soil by a bacterial pathogen. Proc Natl Acad Sci USA109: , pp.1159–1164.

32 

    Dowdell, K., Haig, S.‐J., Caverly, L.J., Shen, Y., LiPuma, J.J., and Raskin, L. (2019) . Nontuberculous mycobacteria in drinking water systems – the challenges of characterization and risk mitigation. Curr Opin Biotechnol57: , pp.127–136.

33 

    Einson, J.E., Rani, A., You, X., Rodriguez, A.A., Randell, C.L., Barnaba, T., et al (2018) . A vegetable fermentation facility hosts distinct microbiomes reflecting the production environment. Appl Environ Microbiol84: e01680‐01618.

34 

    Elowitz, M.B., Levine, A.J., Siggia, E.D., and Swain, P.S. (2002) . Stochastic gene expression in a single cell. Science297: , pp.1183.

35 

    Etzel, L.K., and Legner, E.F. (1999) CHAPTER 7 ‐ Culture and colonization In Handbook of Biological Control. Bellows T.S., and Fisher T.W. (eds). San Diego, CA, USA: Academic Press, pp. , pp.125–197.

36 

    Evans, S., Martiny, J.B. H., and Allison, S.D. (2017) . Effects of dispersal and selection on stochastic assembly in microbial communities. ISME J11: , pp.176–185.

37 

    Fernandez, A., Huang, S.Y., Seston, S., Xing, J., Hickey, R., Criddle, C., and Tiedje, J. (1999) . How stable is stable? Function versus community composition. Appl Environ Microbiol65: , pp.3697–3704.

38 

    Fukami, T., and Nakajima, M. (2011) . Community assembly: alternative stable states or alternative transient states?Ecol Lett14: , pp.973–984.

39 

    Fusco, G., and Minelli, A. (2010) . Phenotypic plasticity in development and evolution: facts and concepts. Philos Trans R Soc B Biol Sci365: , pp.547–556.

40 

    Garner, E., Chen, C.Q., Xia, K., Bowers, J., Engelthaler, D.M., McLain, J., et al (2018) . Metagenomic characterization of antibiotic resistance genes in full‐scale reclaimed water distribution systems and corresponding potable systems. Environ Sci Technol52: , pp.6113–6125.

41 

    Gobet, A., Boetius, A., and Ramette, A. (2014) . Ecological coherence of diversity patterns derived from classical fingerprinting and Next Generation Sequencing techniques. Environ Microbiol16: , pp.2672–2681.

42 

    Godia, F., Albiol, J., Montesinos, J.L., Perez, J., Creus, N., Cabello, F., et al (2002) . MELISSA: a loop of interconnected bioreactors to develop life support in space. J Biotechnol99: , pp.319–330.

43 

    Griffin, J.S., and Wells, G.F. (2017) . Regional synchrony in full‐scale activated sludge bioreactors due to deterministic microbial community assembly. ISME J11: , pp.500–511.

44 

    Günther, S., Koch, C., Hübschmann, T., Röske, I., Müller, R.A., Bley, T., et al (2012) . Correlation of community dynamics and process parameters as a tool for the prediction of the stability of wastewater treatment. Environ Sci Technol46: , pp.84–92.

45 

    Hamer, G. (2010) . Methanotrophy: from the environment to industry and back. Chem Eng J160: , pp.391–397.

46 

    Hampton‐Marcell, J.T., Moormann, S.M., Owens, S.M., and Gilbert, J.A. (2013) Chapter nine – Preparation and metatranscriptomic analyses of host‐microbe systems In Methods in Enzymology. DeLong E.F. (ed). Academic Press, pp. , pp.169–185.

47 

    Hanage, W.P., Spratt, B.G., Turner, K.M. E., and Fraser, C. (2006) . Modelling bacterial speciation. Philos Trans R Soc B Biol Sci361: , pp.2039–2044.

48 

    Hendrickx, L., De Wever, H., Hermans, V., Mastroleo, F., Morin, N., Wilmotte, A., et al (2006) . Microbial ecology of the closed artificial ecosystem MELiSSA (Micro‐Ecological Life Support System Alternative): reinventing and compartmentalizing the Earth's food and oxygen regeneration system for long‐haul space exploration missions. Res Microbiol157: , pp.77–86.

49 

    Herschend, J., Koren, K., Røder, H.L., Brejnrod, A., Kühl, M., and Burmølle, M. (2018) . In vitro community synergy between bacterial soil isolates can be facilitated by pH stabilization of the environment. Appl Environ Microbiol84: e01450‐01418.

50 

    Hibbing, M.E., Fuqua, C., Parsek, M.R., and Peterson, S.B. (2010) . Bacterial competition: surviving and thriving in the microbial jungle. Nat Rev Microbiol8: , pp.15–25.

51 

    Hubbell, S.P. (2001) The Unified Neutral Theory of Biodiversity and Biogeography (MPB‐32). Princeton, NJ, USA: Princeton University Press.

52 

    Huse, S.M., Welch, D.M., Morrison, H.G., and Sogin, M.L. (2010) . Ironing out the wrinkles in the rare biosphere through improved OTU clustering. Environ Microbiol12: , pp.1889–1898.

53 

    Jiang, L.‐L., Zhou, J.‐J., Quan, C.‐S., and Xiu, Z.‐L. (2017) . Advances in industrial microbiome based on microbial consortium for biorefinery. Bioresour Bioprocess4: , pp.11–11.

54 

    Johnson, P.T. J., Preston, D.L., Hoverman, J.T., and Richgels, K.L. D. (2013) . Biodiversity decreases disease through predictable changes in host community competence. Nature494: , pp.230–233.

55 

    Ju, F., and Zhang, T. (2015) . Bacterial assembly and temporal dynamics in activated sludge of a full‐scale municipal wastewater treatment plant. ISME J9: , pp.683–695.

56 

    Kettle, H., Louis, P., Holtrop, G., Duncan, S.H., and Flint, H.J. (2015) . Modelling the emergent dynamics and major metabolites of the human colonic microbiota. Environ Microbiol17: , pp.1615–1630.

57 

    Kinnunen, M., Dechesne, A., Proctor, C., Hammes, F., Johnson, D., Quintela‐Baluja, M., et al (2016) . A conceptual framework for invasion in microbial communities. ISME J10: , pp.2773–2779.

58 

    Kirkegaard, R.H., McIlroy, S.J., Kristensen, J.M., Nierychlo, M., Karst, S.M., Dueholm, M.S., et al (2017) . The impact of immigration on microbial community composition in full‐scale anaerobic digesters. Sci Rep7: , pp.11.

59 

    Kjerstadius, H., Jansen, J.L., De Vrieze, J., Haghighatafshar, S., and Davidsson, A. (2013) . Hygienization of sludge through anaerobic digestion at 35, 55 and 60 degrees C. Water Sci Technol68: , pp.2234–2239.

60 

    Klang, J., Theuerl, S., Szewzyk, U., Huth, M., Tolle, R., and Klocke, M. (2015) . Dynamic variation of the microbial community structure during the long‐time mono‐fermentation of maize and sugar beet silage. Microb Biotechnol8: , pp.764–775.

61 

    Koenig, J.E., Spor, A., Scalfone, N., Fricker, A.D., Stombaugh, J., Knight, R., et al (2011) . Succession of microbial consortia in the developing infant gut microbiome. Proc Natl Acad Sci USA108: , pp.4578–4585.

62 

    Kowalchuk, G.A., Jones, S.E., and Blackall, L.L. (2008) . Microbes orchestrate life on Earth. ISME J2: , pp.795–796.

63 

    Kramer, A.M., Lyons, M.M., Dobbs, F.C., and Drake, J.M. (2013) . Bacterial colonization and extinction on marine aggregates: stochastic model of species presence and abundance. Ecol Evol3: , pp.4300–4309.

64 

    Kunin, V., Engelbrektson, A., Ochman, H., and Hugenholtz, P. (2010) . Wrinkles in the rare biosphere: pyrosequencing errors can lead to artificial inflation of diversity estimates. Environ Microbiol12: , pp.118–123.

65 

    Kuo, C.H., Moran, N.A., and Ochman, H. (2009) . The consequences of genetic drift for bacterial genome complexity. Genome Res19: , pp.1450–1454.

66 

    Kussell, E., and Leibler, S. (2005) . Phenotypic diversity, population growth, and information in fluctuating environments. Science309: , pp.2075–2078.

67 

    Lane, N. (2015) . The unseen world: reflections on Leeuwenhoek (1677) 'Concerning little animals'. Philos Trans R Soc B Biol Sci370: , pp.10.

68 

    Langer, S.G., Ahmed, S., Einfalt, D., Bengelsdorf, F.R., and Kazda, M. (2015) . Functionally redundant but dissimilar microbial communities within biogas reactors treating maize silage in co‐fermentation with sugar beet silage. Microb Biotechnol8: , pp.828–836.

69 

    Lechevallier, M.W., Norton, W.D., and Lee, R.G. (1991) . Giardia and Cryptosporidium spp. in filtered drinking water supplies. Appl Environ Microbiol57: , pp.2617–2621.

70 

    Lenski, R.E., and Travisano, M. (1994) . Dynamics of adaptation and diversification: a 10,000‐generation experiment with bacterial populations. Proc Natl Acad Sci USA91: , pp.6808.

71 

    Li, Y.‐H., and Tian, X.‐L. (2016) Microbial interactions in biofilms: impacts on homeostasis and pathogenesis In Microbial Biofilms Importance and Applications. Dhanasekaran D., and Thajuddin N. (eds). IntechOpen.

72 

    Li, B., Yang, Y., Ma, L.P., Ju, F., Guo, F., Tiedje, J.M., and Zhang, T. (2015) . Metagenomic and network analysis reveal wide distribution and co‐occurrence of environmental antibiotic resistance genes. ISME J9: , pp.2490–2502.

73 

    Li, H., Li, T., and Qu, J. (2018) . Stochastic processes govern bacterial communities from the blood of pikas and from their arthropod vectors. FEMS Microbiol Ecol94: , pp.1–14.

74 

    Li, S.‐P., Tan, J., Yang, X., Ma, C., and Jiang, L. (2019) . Niche and fitness differences determine invasion success and impact in laboratory bacterial communities. ISME J13: , pp.402–412.

75 

    Liu, Z., Klümper, U., Liu, Y., Yang, Y., Wei, Q., Lin, J.‐G., et al (2019) . Metagenomic and metatranscriptomic analyses reveal activity and hosts of antibiotic resistance genes in activated sludge. Environ Int129: , pp.208–220.

76 

    Loman, N.J., Misra, R.V., Dallman, T.J., Constantinidou, C., Gharbia, S.E., Wain, J., and Pallen, M.J. (2012) . Performance comparison of benchtop high‐throughput sequencing platforms. Nat Biotechnol30: , pp.434–439.

77 

    Louca, S., Polz, M.F., Mazel, F., Albright, M.B. N., Huber, J.A., O’Connor, M.I., et al (2018) . Function and functional redundancy in microbial systems. Nat Ecol Evol2: , pp.936–943.

78 

    Lu, J., Struewing, I., Vereen, E., Kirby, A.E., Levy, K., Moe, C., and Ashbolt, N. (2016) . Molecular detection of Legionella spp. and their associations with Mycobacterium spp., Pseudomonas aeruginosa and amoeba hosts in a drinking water distribution system. J Appl Microbiol120: , pp.509–521.

79 

    Lyautey, E., Jackson, C.R., Cayrou, J., Rols, J.‐L., and Garabétian, F. (2005) . Bacterial community succession in natural river biofilm assemblages. Microb Ecol50: , pp.589–601.

80 

    Maestri, S., Cosentino, E., Paterno, M., Freitag, H., Garces, J.M., Marcolungo, L., et al (2019) . A rapid and accurate MinION‐based workflow for tracking species biodiversity in the field. Genes10: , pp.11.

81 

    Mallon, C.A., van Elsas, J.D., and Salles, J.F. (2015) . Microbial invasions: the process, patterns, and mechanisms. Trends Microbiol.23: , pp.719–729.

82 

    Marco, M.L., Heeney, D., Binda, S., Cifelli, C.J., Cotter, P.D., Foligné, B., et al (2017) . Health benefits of fermented foods: microbiota and beyond. Curr Opin Biotechnol44: , pp.94–102.

83 

    Martiny, J.B. H., Eisen, J.A., Penn, K., Allison, S.D., and Horner‐Devine, M.C. (2011) . Drivers of bacterial beta‐diversity depend on spatial scale. Proc Natl Acad Sci USA108: , pp.7850–7854.

84 

    Marzorati, M., Wittebolle, L., Boon, N., Daffonchio, D., and Verstraete, W. (2008) . How to get more out of molecular fingerprints: practical tools for microbial ecology. Environ Microbiol10: , pp.1571–1581.

85 

    Masel, J. (2011) . Genetic drift. Curr Biol21: , pp.R837–R838.

86 

    Maslov, S., and Sneppen, K. (2017) . Population cycles and species diversity in dynamic Kill‐the‐Winner model of microbial ecosystems. Sci Rep7: 39642.

87 

    McCarty, P.L. (1981) . One hundred years of anaerobic treatment. Anaerobic Digest, pp.3–22.

88 

    Meerburg, F.A., Vlaeminck, S.E., Roume, H., Seuntjens, D., Pieper, D.H., Jauregui, R., et al (2016) . High‐rate activated sludge communities have a distinctly different structure compared to low‐rate sludge communities, and are less sensitive towards environmental and operational variables. Water Res100: , pp.137–145.

89 

    Mei, R., and Liu, W.‐T. (2019) . Quantifying the contribution of microbial immigration in engineered water systems. Microbiome7: , pp.144.

90 

    Meyer, A.F., Lipson, D.A., Martin, A.P., Schadt, C.W., and Schmidt, S.K. (2004) . Molecular and metabolic characterization of cold‐tolerant alpine soil Pseudomonas sensu stricto. Appl Environ Microbiol70: , pp.483–489.

91 

    Meyerhof, M.S., Wilson, J.M., Dawson, M.N., and Michael Beman, J. (2016) . Microbial community diversity, structure and assembly across oxygen gradients in meromictic marine lakes, Palau. Environ Microbiol18: , pp.4907–4919.

92 

    Mielczarek, A.T., Kragelund, C., Eriksen, P.S., and Nielsen, P.H. (2012) . Population dynamics of filamentous bacteria in Danish wastewater treatment plants with nutrient removal. Water Res46: , pp.3781–3795.

93 

    Mielczarek, A.T., Nguyen, H.T. T., Nielsen, J.L., and Nielsen, P.H. (2013) . Population dynamics of bacteria involved in enhanced biological phosphorus removal in Danish wastewater treatment plants. Water Res47: , pp.1529–1544.

94 

    Montel, M.‐C., Buchin, S., Mallet, A., Delbes‐Paus, C., Vuitton, D.A., Desmasures, N., and Berthier, F. (2014) . Traditional cheeses: Rich and diverse microbiota with associated benefits. Int J Food Microbiol177: , pp.136–154.

95 

    Morris, B.E. L., Henneberger, R., Huber, H., and Moissl‐Eichinger, C. (2013) . Microbial syntrophy: interaction for the common good. Fems Microbiol Rev37: , pp.384–406.

96 

    Müller, S., Harms, H., and Bley, T. (2010) . Origin and analysis of microbial population heterogeneity in bioprocesses. Curr Opin Biotechnol21: , pp.100–113.

97 

    Muñoz‐Tamayo, R., Giger‐Reverdin, S., and Sauvant, D. (2016) . Mechanistic modelling of in vitro fermentation and methane production by rumen microbiota. Anim Feed Sci Technol220: , pp.1–21.

98 

    Nag, R., Auer, A., Markey, B.K., Whyte, P., Nolan, S., O'Flaherty, V., et al (2019) . Anaerobic digestion of agricultural manure and biomass – critical indicators of risk and knowledge gaps. Sci Total Environ690: , pp.460–479.

99 

    Narancic, T., Djokic, L., Kenny, S.T., O’Connor, K.E., Radulovic, V., Nikodinovic‐Runic, J., and Vasiljevic, B. (2012) . Metabolic versatility of Gram‐positive microbial isolates from contaminated river sediments. J Hazard Mater215–216: , pp.243–251.

100 

    Nicholson, D.J. (2019) . Is the cell really a machine?J Theor Biol477: , pp.108–126.

101 

    Ning, D.L., Deng, Y., Tiedje, J.M., and Zhou, J.Z. (2019) . A general framework for quantitatively assessing ecological stochasticity. Proc Natl Acad Sci USA116: , pp.16892–16898.

102 

    Nowak, C., Vogt, C., Diogo, J.B., and Schwenk, K. (2007) . Genetic impoverishment in laboratory cultures of the test organism Chironomus riparius. Environ Toxicol Chem26: , pp.1018–1022.

103 

    Nowakiewicz, A., Ziółkowska, G., Zięba, P., Gnat, S., Trościańczyk, A., and Adaszek, Ł. (2017) . Characterization of multidrug resistant E. faecalis strains from pigs of local origin by ADSRRS‐fingerprinting and MALDI ‐TOF MS; Evaluation of the compatibility of methods employed for multidrug resistance analysis. PLoS ONE12: e0171160.

104 

    Ofiteru, I.D., Lunn, M., Curtis, T.P., Wells, G.F., Criddle, C.S., Francis, C.A., and Sloan, W.T. (2010) . Combined niche and neutral effects in a microbial wastewater treatment community. Proc Natl Acad Sci USA107: , pp.15345–15350.

105 

    Omarova, A., Tussupova, K., Berndtsson, R., Kalishev, M., and Sharapatova, K. (2018) . Protozoan parasites in drinking water: a system approach for improved water, sanitation and hygiene in developing countries. Int J Environ Res Public Health15: , pp.18.

106 

    Osimani, A., Ferrocino, I., Agnolucci, M., Cocolin, L., Giovannetti, M., Cristani, C., et al (2019) . Unveiling hákarl: a study of the microbiota of the traditional Icelandic fermented fish. Food Microbiol82: , pp.560–572.

107 

    Pilloni, G., Granitsiotis, M.S., Engel, M., and Lueders, T. (2012) . Testing the limits of 454 pyrotag sequencing: reproducibility, quantitative assessment and comparison to T‐RFLP fingerprinting of aquifer microbes. PLoS ONE7: e40467.

108 

    Połka, J., Rebecchi, A., Pisacane, V., Morelli, L., and Puglisi, E. (2015) . Bacterial diversity in typical Italian salami at different ripening stages as revealed by high‐throughput sequencing of 16S rRNA amplicons. Food Microbiol46: , pp.342–356.

109 

    Poretsky, R., Rodriguez‐R, L.M., Luo, C.W., Tsementzi, D., and Konstantinidis, K.T. (2014) . Strengths and limitations of 16S rRNA gene amplicon sequencing in revealing temporal microbial community dynamics. PLoS ONE9: e93827.

110 

    Props, R., Schmidt, M.L., Heyse, J., Vanderploeg, H.A., Boon, N., and Denef, V.J. (2018) . Flow cytometric monitoring of bacterioplankton phenotypic diversity predicts high population‐specific feeding rates by invasive dreissenid mussels. Environ Microbiol20: , pp.521–534.

111 

    Pycke, B.F. G., Etchebehere, C., Van de Caveye, P., Negroni, A., Verstraete, W., and Boon, N. (2011) . A time‐course analysis of four full‐scale anaerobic digesters in relation to the dynamics of change of their microbial communities. Water Sci Technol63: , pp.769–775.

112 

    Quick, J., Quinlan, A.R., and Loman, N.J. (2014) . A reference bacterial genome dataset generated on the MinION (TM) portable single‐molecule nanopore sequencer. Gigascience3: , pp.6.

113 

    Raj, A., and van Oudenaarden, A. (2008) . Nature, nurture, or chance: stochastic gene expression and its consequences. Cell135: , pp.216–226.

114 

    Ramo, A., Del Cacho, E., Sanchez‐Acedo, C., and Quilez, J. (2017) . Occurrence and genetic diversity of Cryptosporidium and Giardia in urban wastewater treatment plants in north‐eastern Spain. Sci Total Environ598: , pp.628–638.

115 

    Rose, J.B., and Gerba, C.P. (1991) . Use of risk assessment for development of microbial standards. Water Sci Technol24: , pp.29–34.

116 

    Sala‐Comorera, L., Vilaro, C., Galofre, B., Blanch, A.R., and Garcia‐Aljaro, C. (2016) . Use of matrix‐assisted laser desorption/ionization‐time of flight (MALDI‐TOF) mass spectrometry for bacterial monitoring in routine analysis at a drinking water treatment plant. Int J Hyg Environ Health219: , pp.577–584.

117 

    Sandrin, T.R., and Demirev, P.A. (2018) . Characterization of microbial mixtures by mass spectrometry. Mass Spectrom Rev37: , pp.321–349.

118 

    Schink, B., and Stams, A.M. (2013) Syntrophism among prokaryotes In The Prokaryotes. Rosenberg E., DeLong E., Lory S., Stackebrandt E., and Thompson F. (eds). Berlin Heidelberg, Germany: Springer, pp. , pp.471–493.

119 

    Shade, A. (2017) . Diversity is the question, not the answer. ISME J11: , pp.1–6.

120 

    Shaikh, H., Patil, S., Pangam, T., and Rathod, K. (2018) . Polymicrobial synergy and dysbiosis: an overview. J Ind Soc Periodontol22: , pp.101–106.

121 

    Shangpliang, H.N. J., Rai, R., Keisam, S., Jeyaram, K., and Tamang, J.P. (2018) . Bacterial community in naturally fermented milk products of Arunachal Pradesh and Sikkim of India analysed by high‐throughput amplicon sequencing. Sci Rep8: , pp.1532.

122 

    Sheik, A.R., Muller, E.E. L., and Wilmes, P. (2014) . A hundred years of activated sludge: time for a rethink. Front Microbiol5: , pp.1–7.

123 

    Sloan, W.T., Lunn, M., Woodcock, S., Head, I.M., Nee, S., and Curtis, T.P. (2006) . Quantifying the roles of immigration and chance in shaping prokaryote community structure. Environ Microbiol8: , pp.732–740.

124 

    Song, H.S. (2018) . Special Issue: Microbial community modeling: prediction of microbial interactions and community dynamics. Processes6: , pp.3.

125 

    Song, H.S., Cannon, W.R., Beliaev, A.S., and Konopka, A. (2015) . Mathematical modeling of microbial community dynamics: a methodological review (vol 2, pg 711, 2014). Processes3: , pp.699–700.

126 

    Spirito, C.M., Daly, S.E., Werner, J.J., and Angenent, L.T. (2018) . Redundancy in anaerobic digestion microbiomes during disturbances by the antibiotic monensin. Appl Environ Microbiol84: e02692‐02617.

127 

    Spitaels, F., Wieme, A.D., Janssens, M., Aerts, M., Daniel, H.M., Van Landschoot, A., et al (2014) . The microbial diversity of traditional spontaneously fermented Lambic beer. PLoS ONE9: , pp.13.

128 

    Stegen, J.C., Lin, X., Fredrickson, J.K., Chen, X., Kennedy, D.W., Murray, C.J., et al (2013) . Quantifying community assembly processes and identifying features that impose them. ISME J7: , pp.2069–2079.

129 

    Strong, P.J., Xie, S., and Clarke, W.P. (2015) . Methane as a resource: can the methanotrophs add value?Environ Sci Technol49: , pp.4001–4018.

130 

    Succurro, A., and Ebenhöh, O. (2018) . Review and perspective on mathematical modeling of microbial ecosystems. Biochem Soc Trans46: , pp.403–412.

131 

    Swingley, W.D., Sadekar, S., Mastrian, S.D., Matthies, H.J., Hao, J., Ramos, H., et al (2007) . The complete genome sequence of Roseobacter denitrificans reveals a mixotrophic rather than photosynthetic metabolism. J Bacteriol189: , pp.683–690.

132 

    Takhaveev, V., and Heinemann, M. (2018) . Metabolic heterogeneity in clonal microbial populations. Curr Opin Microbiol45: , pp.30–38.

133 

    Tao, C.‐W., Hsu, B.‐M., Ji, W.‐T., Hsu, T.‐K., Kao, P.‐M., Hsu, C.‐P., et al (2014) . Evaluation of five antibiotic resistance genes in wastewater treatment systems of swine farms by real‐time PCR. Sci Total Environ496: , pp.116–121.

134 

    United Nations (2019) . Department of Economic and Social Affairs, Population Division. World Population Prospects: Highlights (ST/ESA/SER.A/423).

135 

    Van Der Gast, C.J., Ager, D., and Lilley, A.K. (2008) . Temporal scaling of bacterial taxa is influenced by both stochastic and deterministic ecological factors. Environ Microbiol10: , pp.1411–1418.

136 

    Van Nevel, S., De Roy, K., and Boon, N. (2013) . Bacterial invasion potential in water is determined by nutrient availability and the indigenous community. FEMS Microbiol Ecol85: , pp.593–603.

137 

    Vanwonterghem, I., Jensen, P.D., Dennis, P.G., Hugenholtz, P., Rabaey, K., and Tyson, G.W. (2014) . Deterministic processes guide long‐term synchronised population dynamics in replicate anaerobic digesters. ISME J8: , pp.2015–2028.

138 

    Vermeulen, L.C., Benders, J., Medema, G., and Hofstra, N. (2017) . Global Cryptosporidium loads from livestock manure. Environ Sci Technol51: , pp.8663–8671.

139 

    Viau, E., and Peccia, J. (2009) . Survey of wastewater indicators and human pathogen genomes in biosolids produced by class A and class B stabilization treatments. Appl Environ Microbiol75: , pp.164–174.

140 

    Volkmann, H., Schwartz, T., Bischoff, P., Kirchen, S., and Obst, U. (2004) . Detection of clinically relevant antibiotic‐resistance genes in municipal wastewater using real‐time PCR (TaqMan). J Microbiol Methods56: , pp.277–286.

141 

    Vuono, D.C., Munakata‐Marr, J., Spear, J.R., and Drewes, J.E. (2016) . Disturbance opens recruitment sites for bacterial colonization in activated sludge. Environ Microbiol18: , pp.87–99.

142 

    Wang, X., Wen, X., Criddle, C., Yan, H., Zhang, Y., and Ding, K. (2010) . Bacterial community dynamics in two full‐scale wastewater treatment systems with functional stability. J Appl Microbiol109: , pp.1218–1226.

143 

    Wang, J.J., Shen, J., Wu, Y.C., Tu, C., Soininen, J., Stegen, J.C., et al (2013) . Phylogenetic beta diversity in bacterial assemblages across ecosystems: deterministic versus stochastic processes. ISME J7: , pp.1310–1321.

144 

    Werner, J.J., Knights, D., Garcia, M.L., Scalfone, N.B., Smith, S., Yarasheski, K., et al (2011) . Bacterial community structures are unique and resilient in full‐scale bioenergy systems. Proc Natl Acad Sci USA108: , pp.4158–4163.

145 

    Wittebolle, L., Vervaeren, H., Verstraete, W., and Boon, N. (2008) . Quantifying community dynamics of nitrifiers in functionally stable reactors. Appl Environ Microbiol74: , pp.286–293.

146 

    Wittebolle, L., Marzorati, M., Clement, L., Balloi, A., Daffonchio, D., Heylen, K., et al (2009) . Initial community evenness favours functionality under selective stress. Nature458: , pp.623–626.

147 

    Wright, R.J., Gibson, M.I., and Christie‐Oleza, J.A. (2019) . Understanding microbial community dynamics to improve optimal microbiome selection. Microbiome7: , pp.85.

148 

    Xu, R.H., Yu, Z., Zhang, S.Q., and Meng, F.G. (2019) . Bacterial assembly in the bio‐cake of membrane bioreactors: stochastic vs. deterministic processes. Water Res157: , pp.535–545.

149 

    Yu, Z., Michel, F.C., Hansen, G., Wittum, T., and Morrison, M. (2005) . Development and application of real‐time PCR assays for quantification of genes encoding tetracycline resistance. Appl Environ Microbiol71: , pp.6926–6933.

150 

    Zhao, H., Jia, J.T., Chen, Y., Tang, J., Jiang, Y.H., Wang, J., et al (2017) . Strain‐level characterization of cold‐stressed Listeria monocytogenes using maldi‐tof mass fingerprinting. J Food Saf37: , pp.7.

151 

    Zhou, J., and Ning, D. (2017) . Stochastic community assembly: does it matter in microbial ecology?Microbiol Mol Biol Rev81: e00002‐00017.

152 

    Zhou, J., Wu, L., Deng, Y., Zhi, X., Jiang, Y.‐H., Tu, Q., et al (2011) . Reproducibility and quantitation of amplicon sequencing‐based detection. ISME J5: , pp.1303–1313.

153 

    Zhou, J., Jiang, Y.‐H., Deng, Y., Shi, Z., Zhou, B.Y., Xue, K., et al (2013a) . Random sampling process leads to overestimation of β‐diversity of microbial communities. mBio4: e00324‐00313.

154 

    Zhou, J.Z., Liu, W.Z., Deng, Y., Jiang, Y.H., Xue, K., He, Z.L., et al (2013b) . Stochastic assembly leads to alternative communities with distinct functions in a bioreactor microbial community. mBio4: , pp.8.

155 

    Zhou, J., Deng, Y., Zhang, P., Xue, K., Liang, Y., Van Nostrand, J.D., et al (2014) . Stochasticity, succession, and environmental perturbations in a fluidic ecosystem. Proc Natl Acad Sci USA111: , pp.E836–E845.
https://test.researchpad.co/tools/openurl?pubtype=article&doi=10.1111/1751-7915.13575&title=Stochasticity in microbiology: managing unpredictability to reach the Sustainable Development Goals&author=Jo De Vrieze,Thijs De Mulder,Silvio Matassa,Jizhong Zhou,Largus T. Angenent,Nico Boon,Willy Verstraete,&keyword=&subject=Editorial: Microbial Biotechnology for Sustainable Development Goals,Editorial: Microbial Biotechnology for Sustainable Development Goals,