Potential for enriching next-generation health-promoting gut bacteria through prebiotics and other dietary components

ABSTRACT

The human intestinal commensal microbiota and associated metabolic products have long been regarded as contributors to host health. As the identity and activities of the various members of this community have become clearer, newly identified health-associated bacteria, such as Faecalibacterium prausnitzii, Akkermansia muciniphila, Ruminococcus bromii and Roseburia species, have emerged. Notably, the abundance of many of these bacteria is inversely correlated to several disease states. While technological and regulatory hurdles may limit the use of strains from these taxa as probiotics, it should be possible to utilize prebiotics and other dietary components to selectively enhance their growth in situ. Dietary components of potential relevance include well-established prebiotics, such as galacto-oligosaccharides, fructo-oligosaccharides and inulin, while other putative prebiotics, such as other oligosaccharides, polyphenols, resistant starch, algae and seaweed as well as host gut metabolites such as lactate and acetate, may also be applied with the aim of selectively and/or differentially affecting the beneficial bacterial community within the gastrointestinal environment. The present review provides an overview of the dietary components that could be applied in this manner.

KEYWORDS:

• Prebiotics
• beneficial microbes
• health-promoting gut bacteria
• microbiota

Introduction

The human gut is estimated to contain approximately 10¹⁴ bacteria comprising more than 1000 species.^1,2 Intestinal commensal bacteria, as well as their metabolic products, have long been recognized as important contributors to host health, including nutrition/energy homeostasis, pathogen resistance, the regulation of intestinal epithelial proliferation and a variety of other factors.³ The corollary is that negative impacts on microbiota composition and functionality can contribute to illness.^4,5 Importantly, the gut microbiota can be modulated in a beneficial way, including through the selective manipulation of particular species of interest to maintain, restore or improve host health.

One approach to positively modulate the gut microbiota is through the administration of growth enhancing substrates that can be utilized selectively by health promoting bacteria, to encourage their growth and the production of associated desirable metabolites. The rationale of selectively enhancing beneficial microbes in the gut lead to the concept of prebiotics, initially coined in 1995 by Roberfroid and Gibson⁶ and described in greater detail below. Among the best characterized prebiotics are galacto-oligosaccharides (GOS) and inulin and its oligomer, fructo-oligosaccharides (FOS), as evidenced by numerous studies.^7–9 The vast majority of studies relating to these and other prebiotics have focused on elucidating their effects on Bifidobacterium and Lactobacillus as strains from these genera have been known for some time to confer a health benefit to the host.¹⁰ As our knowledge of the gastrointestinal microbiota expands, emphasis has now also been placed on the application of other bacteria present with possible health promoting effects and, in parallel, on the identification of prebiotics (targeted) and other dietary substrates (potentially less targeted) with the potential to stimulate the growth of these beneficial targets in the gut.

Newly identified health-associated microbes

Advances made in metagenomics and the application of emerging ‘multiomics’ technologies, in combination with cultivation techniques, have identified several putatively beneficial species of gut bacteria (Table 1). These microbes, including Faecalibacterium prausnitzii, Ruminococcus bromii and Akkermansia muciniphila, can be present in significant numbers in a healthy human gut. A single species such as F. prausnitzii may comprise more than 5% of the total intestinal community,³⁰ while there are also instances where up to 8% of the composition has been assigned to the phylum Verrucomicrobia,³¹ i.e., corresponding to A. muciniphila at the species level.

Table 1. Overview of some representative newly identified health-promoting bacteria and the associated benefits.

Family	Bacteria	Beneficial Impact	Reference(s)
Christensenellaceae	Christensenella minuta	SCFA producer and possible link with a lean phenotype.	J.K Goodrich et al., 2014^Citation11
Eubacteriaceae	Eubacterium eligens	SCFA producer and pectin utilizer.	M. Lopez-Siles et al., 2012^Citation12
Eubacteriaceae	Eubacterium hallii	Produces pseudovitamin B12 which can help ↑ SCFA production by surrounding bacteria e.g. A. muciniphila, lactate utilizer, butyrate producer	C. Belzer et al., 2017,^Citation13 S. Duncan et al., 2004^Citation14
Eubacteriaceae	Eubacterium rectale	Acetate consumer, butyrate producer. Can be involved in cross-feeding interactions with other beneficial bacteria e.g. Bifidobacterium longum.	A. Rivière et al., 2015,^Citation15 P. Louis et al., 2010,^Citation16 P. Louis & H.J. Flint 2009^Citation17
Lachnospiraceae	Anaerostipes caccae & Anaerostipes hadrus	Butyrate producers, lactate and acetate utilizers.	S. Duncan et al., 2004,^Citation14 E. Allen-Vercoe et al., 2012^Citation18
Lachnospiraceae	Coprococcus spp.	Butyrate and acetate producers closely related to Ruminococcus.	P. Louis & H.J. Flint, 2009^Citation17
Lachnospiraceae	Roseburia spp.	Butyrate and propionate producers. Decreased levels seen in those with ulcerative colitis.	K. Machiels et al., 2014,^Citation19 N. Reichardt et al., 2014,^Citation20 S. Duncan et al., 2002^Citation21
Ruminococcaceae	Faecalibacterium prausnitzii	Decreased levels observed in Crohn’s disease and minulcerative colitis indicating anti-inflammatory properties. One of the main butyrate producers within the gut.	M. Lopez-Siles et al., 2012,^Citation12 A. Heinken et al., 2014,^Citation22 S. Miquel et al., 2013,^Citation23 K. Machiels et al., 2014^Citation19
Ruminococcaceae	Oscillospira sp.	Enriched in those with a lean phenotype in comparison to obese subjects, decreased levels observed in those with inflammatory diseases.	T. Konikoff & U. Gophna, 2016,^Citation24 J.K Goodrich et al., 2014^Citation11
Ruminococcaceae	Ruminococcus bromii	Keystone species for degrading resistant starch enabling other bacteria to utilize the breakdown products.	X. Ze et al., 2012,^Citation25 A. Venkataraman et al., 2016^Citation26
Verrucomicrobiaceae	Akkermansia muciniphila	Mucin-degrading bacterium inversely associated with obesity and other metabolic diseases. Liberates oligosaccharides from mucin making them available to other bacteria. Produces acetate and propionate which some butyrate-producers can utilize.	PD. Cani & WM. de Vos, 2017,^Citation27 A. Everard et al., 2013,^Citation28 C. Belzer et al., 2017,^Citation13 M. Schneeberger et al., 2015^Citation29

Some of these newly identified health-promoting bacteria are associated with various benefits to their respective hosts. One example of such an apparent relationship is the observation that the abundance of F. prausnitzii is inversely correlated with incidence of inflammatory bowel disease (IBD), in particular Crohn’s disease.^19,23,32 As a result, F. prausnitzii has been investigated with respect to its potential to alleviate inflammation.³² Another intriguing host-microbe interaction relates to that between A. muciniphila and obesity. Accumulating evidence has shown that the abundance of A. muciniphila is inversely proportional to body weight and type-1 diabetes.^27,33 A. muciniphila is specialized as it colonizes the mucus lining and can use mucin as a sole carbon and nitrogen source, releasing growth substrates for other beneficial bacteria.^34,35 Interestingly, in a recent study, Roseburia hominis and Roseburia intestinalis have also been found to utilize mucin as energy source indicating that mucin degradation is more widespread than was previously appreciated.³⁶ Mucin is a major component of intestinal mucus layer. The ability to utilize mucin offers an ecological benefit to these bacteria over those that are dependent on dietary nutrients, providing a source of host-derived nutrients while the geographical location provides a means of interacting with the immune system.³⁷ Notably, Christensenella minuta and Oscillospira spp. have also been linked with a lean phenotype and may be useful targets when combating obesity.^11,24,28 The establishment of potential relationships such as these support the continued efforts in identifying approaches to encourage the growth of target microbes in the gut.

The breakdown and metabolism of complex carbohydrates, particularly non-digestive carbohydrates, by gut bacteria, hugely benefits primary degraders as well as the community which lack the ability to use the parent components. R. bromii has been suggested to be a key species with respect to the breakdown of resistant starch, with the resultant products being utilized as substrates for syntrophic growth by other beneficial microbes in the gut.²⁵

While in some instances the benefits attributed to specific gut microbes are thought to be direct, in other cases they can be indirect, such as via their metabolites. This can include the ability to utilize fibers and complex carbohydrates to produce short chain fatty acids (SCFAs),^23,25,27 the most abundant of which are acetate, propionate and butyrate. SCFAs have numerous benefits including; lowering the luminal pH, contributing to increased calcium and magnesium absorption, helping to reduce potential pathogenic bacteria, serving as an energy source for epithelial cells and also involvement in immune modulation. Butyrate, as mentioned previously, is produced mainly by components of the major phylum, Firmicutes, such as Clostridium clusters XIVa and IV,³⁸ and is particularly important as it is utilized as an energy source for colonocytes and has a proposed role in protection against diseases such as colon cancer and colitis.^39–42 It is therefore important that these health-promoting bacteria are present in sufficient numbers and are maintained within the gut.

Other species that are abundant in a healthy gut, including Lachnospiraceae such as Anaerostipes caccae, Anaerostipes hadrus, Roseburia spp., Coprococcus catus, and Eubacteriaceae such as Eubacterium rectale and Eubacterium hallii, are acetate consumers and produce butyrate via butyryl-CoA:acetate CoA-transferase,⁴³ which will be discussed in more detail later. There are a number of other examples of substrate cross-feeding that can have a substantial impact on the survival of certain bacteria and production of useful metabolites in the gut.⁴⁴ For instance, lactate is readily available in the gut as a result of the activity of lactic acid bacteria and can be utilized by Anaerostipes sp. and Eubacterium sp. to produce SCFAs.⁴⁵ Similarly, C. catus and the ruminal bacterium Megasphaera elsdenii can convert lactate to propionate via the acrylate pathway.²⁰ The metabolic diversity and dependency of these health associated microbes is complex and, while our understanding of the pathways involved in carbohydrate fermentation and cross-feeding are improving (Figure 1), there is likely to be still much to learn.

Figure 1. Schematic representation of selected gut bacteria involved in carbohydrate fermentation and cross-feeding interactions resulting in the production of major microbial metabolites.

Pathways leading to the production of the three main SCFAs, acetate, butyrate and propionate, are depicted here. Acetate can be produced from acetyl-CoA or by acetogens using H₂ and CO₂ or formate. It can also be used for the formation of butyrate. Butyrate can be formed in two ways; either through the formation of butyryl-P or more commonly through the use of butyryl-CoA:acetate CoA-transferase which many Firmicutes possess. The main route by which propionate is generated is via the succinate pathway. However, two other pathways have also been found; i.e., the acrylate pathway which involves lactate and the propanediol pathway which utilizes deoxyhexose sugars. DHAP, dihydroxyacetonephospate; PEP, phosphoenolpyruvate. Adapted from Louis et al., 2014.⁴³

Prebiotic criteria and considerations

A large number of studies have been conducted to investigate a range of prebiotics and the effects that they have on the gut microbiota. With growing interest and knowledge, the definition of prebiotics has evolved and in December 2016, the International Scientific Association for Probiotics and Prebiotics (ISAPP) proposed that the definition be revised to ‘a substrate that is selectively utilized by host microorganisms conferring a health benefit’.⁴⁶ This definition expands the concept to include non-carbohydrate substances. It also was suggested that the revised definition does not limit the selective stimulation of beneficial microorganisms to the gut if in the future prebiotic targets are beyond this environment. Currently, in order to be classified as a dietary prebiotic, three broad criteria need to be fulfilled. These include (i) resistance to gastric acid and hydrolysis by mammalian enzymes and gastrointestinal absorption; (ii) ability to be utilized by the intestinal microbiota; and (iii) selectively stimulate the growth and/or the activity of intestinal bacteria associated with health-promoting effects.⁴⁷

A major prerequisite to the success of a particular prebiotic is that the target bacteria must be present or present at a specific threshold. If levels are below a certain threshold, due to factors such as ageing or antibiotic usage, then the prebiotic may not be effective in increasing the numbers of these desirable bacteria to the necessary level in order to confer the anticipated health benefits.²⁶ Thus, having an understanding of such factors in target populations can enhance the likelihood of beneficial prebiotic-related outcomes. However, there are a number of other factors to consider. While prebiotics, and other beneficial food bioactives, can enable the growth of targeted health-promoting bacteria, more detailed microbiome analysis has highlighted that these substrates are not always as specific as once thought.^46–48 It is also necessary to consider that the benefits of prebiotics can sometimes be secondary, e.g., production by health promoting bacteria of metabolic products that can inhibit the growth of enteric pathogens and/or attenuate their virulence.⁴⁹ One can determine prebiotic impacts/specificity through the use of ex vivo fermentation studies, with fecal samples or mock gut microbial communities, or through in vivo studies. However, even then, care should be taken when interpreting outcomes as, due to inter-individual differences, the same effect may not be observed in all cases. It is also important to consider that some populations, when present at sufficiently high levels, may produce gases from prebiotic substrates that can result in bloating and abdominal discomfort.^50,51 Additionally, some prebiotics such as inulin-derived oligosaccharides can have mildly laxative effects.⁵² All of these considerations highlight the importance of understanding the interactions between prebiotics (as well as other dietary components), target microbes and the metabolites generated in order to determine ultimate resultant impacts on the host.

Targeting newly identified health-associated microbes through well-established prebiotics

There are a few extensively studied prebiotics, including FOS, GOS and inulin, which have been examined in detail using in vitro assays and in vivo animal models and human studies. These investigations have in particular highlighted the growth-promoting impact of prebiotics on members of the genera Bifidobacterium and Lactobacillus,^6,9,46,53 as a consequence of the health-associated status of many strains from these taxa. However, there are many ways in which prebiotics can positively impact the host (Figure 2). A direct benefit of prebiotics is enhancing the growth of target micro-organisms which in turn compete with harmful species for energy sources and exclude them proving protection or through facilitating the production of beneficial fermentation products, such as SCFAs.

Figure 2. An overview of some beneficial impacts of prebiotic supplementation on the gut microbiota.

Clinical trials and human studies are pivotal when assessing the benefits of newly identified prebiotics. Of the many studied potential prebiotics, only a few substrates, including inulin, FOS and GOS, have been validated through several human studies and only those that looked at a broader gut community, and thus are of greater relevance to this review, are presented in Table 2. These studies have highlighted that the extensively examined prebiotics, for example inulin, can also have a positive effect on the level of F. prausnitzii and Anaerostipes sp. within the gut, which may explain some of the butyrogenic effects that ensue when inulin is consumed.^53,56,57 Likewise, FOS and GOS have also been demonstrated to enhance F. prausnitzii levels.^53,55,56 Of the few human studies assessing the impact of prebiotics on microbial diversity in the colon, contradictory observations with regards to SCFA levels have been reported. A study by Liu et al.⁵⁴ noted a decrease in the levels of butyrate producers and increase in the levels of Bifidobacterium after administration of FOS and GOS in a healthy population, which may be caused by high levels of lactic acid being produced, therefore making the environment inhospitable for butyrate producers. It is also worth noting that this intervention was conducted for a period of 14 days; longer intervention studies are needed to better evaluate the effects of prebiotic administration. Nevertheless, these results prove intriguing and emphasize the need for more human studies.

Table 2. Human studies conducted in relation to prebiotics and newly identified health-promoting bacteria.

Prebiotic Administered	Study Design	Cohort	Delivery	Effect on Microbiota	Reference
Extensively studied prebiotics
FOS vs. GOS	A randomized, double-blind, cross-over study	35 healthy participants (10 males, 25 females)	16 g/d for 14 d	FOS: ↑ Bifidobacterium; and ↓ in Phascolarctobacterium, Enterobacter, Turicibacter, Coprococcus and Salmonella GOS: ↑ in Bifidobacterium; ↓in Ruminococcus, Dehalobacterium, Synergistes & Holdemania	Liu et al., 2017^Citation54
GOS	A randomized, double-blind, parallel-group, multisite placebo-controlled study	62 lactose intolerant subjects	GOS or placebo was escalated in 5-d increments from 1.5 g to 15 g once a day. Taken for 35 d	In response to GOS administration ↑ in Bifidobacterium, Faecalibacterium & Lactobacillus was observed. Subsequent dairy consumption resulted in ↑ Roseburia levels	MA. Azcarte-Peril et al., 2017^Citation55
Inulin- oligofructose	Balanced cross-over study	12 healthy adults split into 2 groups (control and prebiotic)	10 g prebiotic or control per day over a 16-d period	↑ F. prausnitzii across all volunteers & ↑ in different Bifidobacterium species dependent on individual as a result of prebiotics	C. Ramirez-Farias et al., 2009^Citation56
Inulin/oligofructose mix (50:50)	A double blind, placebo controlled, intervention study	30 obese women	16 g prebiotic or control per day for 3 months	Prebiotics led to ↑ in bifidobacteria & F. prausnitzii while ↓ in Bacteroides intestinalis, Propionibacterium & Bacteroides vulgatus was observed	EM. Dewulf et al., 2013^Citation53
Inulin-type fructans	A randomized, double-blind, placebo-controlled, cross-over trial	42 healthy adults	12 g chicory-derived Orafti inulin or control per day for 4 weeks	Inulin consumption led to ↑ Bifidobacterium & Anaerostipes abundances while Bilophila numbers ↓	Vandeputte et al., 2017^Citation57
Potential prebiotics
Red wine polyphenols	A randomized, crossover-controlled intervention study	10 obese subjects with metabolic syndrome (MetS) and 10 healthy controls	Initial wash-out followed by two intervention periods where participants drank red wine (272 mL/d) or de-alcoholised red wine (272 mL/d) separated with a wash-out phase (15 d) in between cross-over	In healthy individuals ↑ levels in F. prausnitzii & Roseburia after red wine and de-alcoholised red wine consumption in comparison to baseline levels. MetS patients also had ↑ levels in these bacteria while also in Blautia coccoides-E. rectale group and Lactobacillus. Also, differences in microbiota between both groups were significantly ↓ after interventions	I. Moreno-Indias et al., 2016^Citation58
Red wine polyphenols	A randomized, crossover, controlled, intervention study	10 healthy males	After a 15-d wash-out period, each participant completed 3 consecutive 20-d periods in which they drank de-alcoholised red wine (272ml/d), red wine (272ml/d), or gin (100ml/d)	Red wine polyphenols ↑ Enterococcus, Prevotella, Bacteroides, Bifidobacterium, Bacteroides uniformis, Eggerthella lenta & Blautia coccoides – Eubacterium rectale groups	M. I. Queipo-Ortuño et al., 2012^Citation59
Cocoa flavanols	A randomized, double-blind, crossover, controlled intervention study	22 healthy volunteers	Subjects either consumed high cocoa flavanol (HCF – 494 mg) or low cocoa flavanol (LCF – 29 mg) drink per day for 4 wks followed by 4 wk washout period before switching to alternate drink	HCF ↑ Bifidobacterium & Lactobacilli levels while ↓ levels of C. histolyticum group. Both HCF & LCF led to slight ↑ in E. rectale-C. coccoides group but no significant differences between the two	X. Tzounis et al., 2011^Citation60
XOS	A double-blind, randomized, placebo-controlled study	32 healthy subjects	1.4 g XOS, 2.8 g XOS or placebo taken daily	Both XOS doses ↑ bifidobacteria, no change in lactobacilli, ↑ in Faecalibacterium sp. & Akkermansia sp. in those supplemented with the higher dose	S. Finegold et al., 2014^Citation61
Resistant starch (RS)	A randomized, crossover dietary study	39 subjects with reduced insulin sensitivity	Participants either consumed a high (HC) or low carbohydrate (LC) diet followed by a baseline diet. Then the HC subjects consumed either a high RS (HRS – 66 g/d) or low RS (LRS – 4 g/d). Subjects which consumed LC diet consumed either 48 g for HRS or 3 g for LRS	HRS led to ↑ in the ratio of Firmicutes to Bacteroidetes in comparison to LRS. In particular there were ↑ levels of F. prausnitzii, Ruminococcus, Roseburia, E. rectale & A. muciniphila	TV. Maier et al., 2017^Citation62
Resistant Starch (RS) type 2	A balanced study	20 healthy young adults (10 male & 10 female)	48 g of potato starch (24 g twice per day) for 7 d after a 3-d acclimatization period	Individuals with high or enhanced levels of butyrate concentrations showed ↑ in B. adolescentis or R. bromii after RS consumption. In 5 individuals an ↑ in E. rectale was observed. Huge inter-individual variation was evident	Venkataraman et al., 2016^Citation26

Targeting newly identified health-associated microbes through emerging prebiotics

The prebiotic paradigm has been shifting in recent years as a result of the identification of newly recognized putatively beneficial members of the gut microbiota to target for enrichment. These studies have been facilitated through new developments in cultivation techniques and, in particular, high-throughput sequencing, culture-independent approaches. These approaches have been used to assess the impact of specific fibers as well as polyphenols, other oligosaccharides and the less explored marine derived foods including seaweed-based products, which represents a relatively untapped source of food bioactives, on the gut microbiota of animals and humans.

Polyphenols

Polyphenols are naturally occurring secondary metabolites of plants that consist of a wide category of compounds and are classified primarily based on structure.⁶³ Plant-based foods and beverages, including fruits, vegetables, coffee, tea and wine, are rich sources of dietary polyphenols. Consumption of plant foods is associated with lower risk of chronic diseases including cancer,⁶⁴ heart diseases,⁶⁵ type 2 diabetes⁶⁶ and other inflammatory diseases. A large number of these biological effects are attributed to phytochemical components of plant foods.^67,68 Moreover, it has been suggested that many of the beneficial impacts of these secondary metabolites on overall health is mediated through the manipulation of gut bacteria in the colon and their transformation therein by the gut bacteria present.⁶⁹ One such example is cranberry extract, which is rich in polyphenols, and has been shown to ameliorate diet-induced obesity and several features of metabolic syndrome (MetS) in mice, while increasing the abundance of A. muciniphila.⁷⁰ Cranberry administration has also been shown to enhance mucus secretion, which could possibly create favorable conditions for A. muciniphila.⁷¹ Similarly, the administration of Concord grape and California table grape extracts, rich in a class of polyphenol called proanthocyanidin, also increases the abundance of Akkermansia.^72,73 Other polyphenols present in red wine, tea and cocoa flavonols have also been investigated. Cocoa flavanols increase the levels of lactobacilli and bifidobacteria in addition to enhancing levels of the Clostridium coccoides-Eubacterium rectale group subtly.⁷⁴ From a red wine perspective, intake of 272 mL per day over a 30-d period altered the composition of the gut microbiota in patients with MetS and resulted in increased levels of Bifidobacterium, Lactobacillus, F. prausnitzii and Roseburia sp.⁵⁸ The study also suggests that the increase in F. prausnitzii levels was associated with the decrease of blood glucose levels in MetS patients possibly due to the production of SCFAs interacting with gut hormones which in turn may have an impact on carbohydrate metabolism.⁵⁸ In the same study, beneficial modulatory impacts in healthy volunteers were attributed to red wine polyphenols as levels of Blautia coccoides – Eubacterium rectale group, bifidobacteria, Bacteroides uniformis and Eggerthella lenta were enhanced. Consequently, there was a significant decrease in the levels of the potentially harmful Clostridium histolyticum group, which encompasses some pathogens such as Clostridium perfringens. Although the exact mechanisms involved are not fully understood, these in vitro and human studies form the basis of utilizing certain polyphenols, alone or in combination, to alter the composition of the gut microbiota.

Oligosaccharides

In addition to the well-studied oligosaccharides, FOS and GOS, more recently there have been many studies focusing on the prebiotic effect of other oligosaccharides, such as xylo-oligosaccharides (XOS) and arabinoxylan (AX), pectin-oligosaccharides (POS), isomatlooligosaccharides (IMO), soybean oligosaccharides and human milk oligosaccharides (HMOs).

XOS is the product of the hydrolysis of xylan and can act as a substrate for gut bacteria, resulting in the production of SCFAs,⁷⁵ with associated health benefits. The selectivity of XOS was noted in a study where 6 out of 11 Firmicutes examined, including R. intestinalis, R. faecis, E. rectale and A. caccae, were able to utilize this substrate, along with some Bifidobacterium sp. such as B. adolescentis.⁷⁶ A number of studies have highlighted that XOS can have a bifidogenic effect. In one instance, a human study evaluated the impact of XOS alone and XOS in conjunction with inulin. XOS alone increased fecal Bifidobacterium and butyrate, with a decrease in acetate and p-cresol and the combination of substrates appeared to reduce the inflammatory repercussions of a Western diet.⁷⁷ XOS supplementation was also determined to increase Faecalibacterium sp. and Akkermansia sp. as well as bifidobacteria but did not have a significant impact on levels of lactobacilli.⁶¹ This study demonstrated that XOS promoted intestinal health through modulation of the microbial community; although there was no notable net increase in SCFA production detected, due consideration must be taken when examining dose effects.

Arabinoxylan (AX), a non-digestible carbohydrate often found in the cell walls of plants, can be selectively fermented in the colon by fibrolytic gut bacteria possessing AX-degrading enzymes.⁷⁸ AX can be hydrolyzed to arabinoxylan-oligosaccharides (AXOS). Both AX and AXOS have demonstrated an ability to increase desirable bacteria and butyrate producers in the colon.^79–81 Indeed, in one instance when a high-fat (HF) diet was administered to mice, the addition of AX subsequently increased levels of bifidobacteria.⁸⁰ AX also restored bacterial levels to that of the initial control level before HF diet induced obesity, in particular enhancing populations of the Roseburia species and Bacteroides-Prevotella species that were reduced upon HF feeding. These microbial shifts as a result of AX supplementation highlight its prebiotic potential.

Pectin and pectic oligosaccharides (POS) have been identified as emerging prebiotics as evidenced through their selective utilization by certain members of the colonic microbiota such as E. eligens and F. prausnitzii.^12,82 The most frequent source of pectin is from citrus fruits and apple pulp, but it is also abundant in agricultural by-products such as sugar beet pulp. POS can be obtained through depolymerization of pectin and both pectin and POS escape host digestion and reach the distal colon when consumed.⁸³ In an in vitro assay of F. prausnitzii growth in the presence of different substrates, it was ascertained that growth was enhanced with apple pectin in most cases.^12,84 Although pectin is extensively fermented in the gut, pectin utilization has not been reported for many of the bacterial groups residing in the gut. It has been demonstrated that one of the main fermentation products from pectin utilization by F. prausnitzii was butyrate. The ability of F. prausnitzii to compete for pectin when cultured with B. thetaiotaomicron and E. eligens was also noted.¹² Growth on some uronic acids such as galactouronic acid was also observed in F. prausnitzii. This is important as galactouronic acid is a major component of pectin and it is not reported to be utilized by many other bacterial groups other than Bacteroides sp.¹² The impact of sugar beet pulp and lemon peel wastes on gut microbial communities have also been assessed using in vitro fermentation assays.⁸⁵ Results suggested that POS of both lemon peel and sugar beet possessed better prebiotic potential than the corresponding pectin, and similar or even better than FOS (which was used as a control). Lemon peel waste oligosaccharides (LPOS) in particular resulted in an increased level of Faecalibacterium and Roseburia sp. as well as lactobacilli. Both LP and LPOS increased the level of E. rectale, although larger effects were observed with pectin. Ultimately, the study indicated that POS and pectin brought about a beneficial impact on the microbial population, presenting an intriguing opportunity for further investigation.

Isomaltooligosaccharides (IMO) are often considered potential prebiotics. Their impact on newly identified health beneficial microbes is limited and necessitates further investigation. IMOs are naturally found in foods such as honey as well as fermented foods such as miso and soy. In addition, various commercial preparations are made enzymatically by processing an assortment of starches and are readily available on the market today. Commercially available IMOs are composed of a mixture of α(1–6) and α(1–4)-linked glucosyl oligosaccharides.⁸⁶ One of the glucose oligomers identified in IMOs is isomaltose, which is a major constituent of honey thereby giving IMOs a distinctive sweet honey taste. In a recent study evaluating the impact of co-supplementation of cranberry extract with IMO in HF diet fed mice, Roseburia, Faecalibacterium, and Eubacterium were enhanced with an associated increase in butyrate.⁸⁷

Unsurprisingly, HMOs have also been the focus of much attention, with strong effects being observed on Bifidobacterium levels in particular.⁸⁸ Breast milk is the natural first nutritional source for newborns and provides oligosaccharides that promote the growth of desirable Bifidobacterium sp. in the infant gut. HMOs and other nutrients in breast milk can act as substrates for bacteria in the gut, thereby stimulating the growth of beneficial bacteria located here. Two compounds in HMOs are 2ʹ-O-fucosyllactose and lacto-N-neotetraose, which was the subject of one human study evaluating their putative prebiotic effect on the human gut microbiota in 100 healthy, adult volunteers.⁸⁹ This HMO supplementation was shown to modify the gut microbiota with a resulting increase in the relative abundance of Actinobacteria and, in particular Bifidobacterium, while reducing the relative abundance of Firmicutes and Proteobacteria. While there are much data available relating the impact of HMOs on the infant gut, and of Bifidobacterium strains found therein,⁹⁰ their full effect on the overall infant gut microbial population, and resulting adult microbiota, is the subject of many on-going investigations.

Other oligosaccharides, including those from soybean and mannan, have putative prebiotic effects. Soybean oligosaccharides seem to have a positive influence on Bifidobacterium levels and lactate production.⁹¹ Soybean forms a substantial part of diets in Asia and is now more frequently seen in Western diets, and benefits are attributed partly to fiber and the oligosaccharides present. Mannan-oligosaccharides (MOS) are commonly used as additives in animal feeds. MOS are not absorbed by the animal so can reach the gut microbiota where they are utilized beneficial bacteria such as bifidobacteria and lactobacilli which can out-compete potential pathogenic bacteria. The effect on animal intestinal bacteria and mechanisms involved is just beginning to be evaluated.⁹² Although their effect on human gut intestinal inhabitants has yet to be assessed, there may be considerable merit in doing so. Combinatorial benefits of various oligosaccharides are also beginning to be assessed, which may lead to the identification of optimal levels and combinations of prebiotics for the promotion of health-associated bacteria in the gut.⁹³

Resistant starch

Resistant starch (RS) is a form of starch that cannot be degraded in the small intestine and, thus, subsequently reaches the large intestine where it is fermented by the colonic bacteria. There are four types of RS (type 1–4), each with different properties that contribute to resisting digestion. RS1 is physically inaccessible due to its compact structure, RS2 has a granular structure, which prevents digestive enzymes from hydrolyzing it, RS3 are retrograded starches and RS4 is modified with chemicals to enhance resistance. RS2 has been the focus of many investigations, both ex vivo and in humans. Interestingly, in a co-culture experiment, R. bromii demonstrated a unique ability to stimulate RS degradation by other bacteria.²⁵ When cultured with E. rectale, B. thetaiotaomicron and B. adolescemtis on YCFA (yeast extract-casitone-fatty acid) medium, a semi-defined medium, utilization of RS was enhanced in comparison to combinations without R. bromii. This is intriguing as it was determined that YCFA medium did not support the growth of R. bromii very well. This indicates that R. bromii has a superior ability to ferment RS and even possibly produce metabolites that are available for cross-feeding interactions with co-inhabitants. Many studies have particularly emphasized the importance of R. bromii as a keystone species with regard to degrading RS within the human gut.^25,26,94 The R. bromii population substantially increased in fecal samples collected from those consuming a high RS diet.²⁵

Venkataraman et al.²⁶ demonstrated that RS2, when used to supplement the normal diet of healthy volunteers, increased fecal butyrate concentrations albeit with considerable inter-individual differences. This high amylose starch enhanced SCFA production, although those with low initial levels of SCFAs did not increase substantially, indicating that possibly the baseline level of required bacteria was too low for the full benefits of supplementation to be achieved. In those where an effect was evident, there was a substantial increase in RS-degrading bacteria such as B. adolescentis and R. bromii with some individuals displaying an additional enhanced level of E. rectale.

Similarly, in a study evaluating the effect of RS2 on subjects with insulin resistance using a multi-omics approach, a high-RS diet resulted in an increased Firmicutes to Bacteroidetes ratio, with a particular enhancement of F. prausnitzii, R. bromii and Roseburia species.⁶² The Firmicutes to Bacteroidetes ratio has been used, albeit somewhat controversially, as a marker of gut homeostasis with increased proportions Bacteroidetes being proposed to reflect a beneficial modulation of gut microbiota. There is evidence to suggest that different types of RS may lead to distinct compositional modulations. In a study examining the effect of RS2 or RS4 on the composition of the fecal microbiota in 10 human subjects, in line with the aforementioned studies RS2 enriched B. adolescentis, R. bromii and E. rectale.⁹⁵ On the other hand, RS4 increased the levels of Parabacteroides distasonis and R. bromii and decreased the levels of E. rectale. Both R. bromii and E. rectale were also enriched in fecal samples of obese human subjects during consumption of RS3 indicating some similarity to RS2.⁹⁶ This highlights the importance of the chemical structure of RS in dictating its impact on the colonic bacteria. A recent dietary intervention study looked at the impact of two different RS, one from potatoes and the other from maize, as well as inulin on the gut microbiota of healthy volunteers.⁹⁷ Both types of RS were RS2 although the resulting changes in the gut microbiota differed with the RS from potatoes yielding a greater increase in SCFA production than either RS from maize or inulin. Again, the presence of R. bromii was significant. It was noted that if either R. bromii or Clostridium charatabidum were present after consuming RS it was more likely for butyrate to be produced, especially when the microbial population included the butyrate-producer E. rectale. This indicates that there may be primary degraders of RS such as R. bromii and the breakdown products may subsequently be utilized by certain butyrate-producers. Bifidobacterium was also enhanced in most of the gut microbiotas after consuming any of the three fermentable fibers. The resulting effects of RS degradation on the microbial population has yet to be fully explored, however, it is evident that R. bromii has an integral role in initiating the breakdown of this resistant but fermentable starch. This selective metabolism is the driving force behind the ever-increasing interest in the use of RS as a prebiotic ingredient in functional foods.⁹⁸ Indeed, several commercialized products containing RS are already available on the market and there has been a continuing effort to incorporate the RS as a functional food component.

Algae and seaweed

The marine environment is often considered a vast, untapped reservoir with regards to the microorganisms and compounds present, the metabolites produced and interactions that occur in this ecosystem. New bioactives and potential prebiotic substrates are being harnessed from this rich environment. Seaweeds, also known as marine macroalgae, are rich in polysaccharides and bioactive compounds, which could be applied to improve human and animal health.^99,100 Seaweeds can be subdivided into three categories based on their pigmentation; Rhodophyta (red seaweed), Chlorophyta (green seaweed) and Phaeophyta (brown seaweed).⁹⁹ Differences in composition and structure of the polysaccharides present in these seaweeds are also evident. Brown seaweed is of particular interest as it possesses polysaccharides such as fucoidan, laminarin and alginate.¹⁰¹ Fucoidans comprise a class of fucose-rich sulfated polysaccharides often located in the cell walls of brown macroalgae. It is suggested that fucoidans with differing structures may impact the gut in a variety of ways. For example, a study carried out by Shang et al.,¹⁰² established that fucoidans isolated from Ascophyllum nodosum (FuA) and Laminaria japonica (FuL) had positive but slightly differing effects on the murine gut microbiota due to their structural differences. FuA has a type I back-bone structure whereas FuL has a type II structure. FuL markedly increased the levels of Ruminococcaceae while FuA enhanced Lactobacillus, Anaeroplasma and Thalassospira. In a follow-up study on HF diet-fed mice, these fucoidans improved MetS induced by a HF diet.¹⁰³ Dietary supplementation with fucoidan decreased body weight in HF diet-fed mice and also improved glucose intolerance and insulin resistance. Both fucoidans separately ameliorated intestinal dysbiosis caused by HF diet and significantly increased Akkermansia abundance. An evolving body of evidence has linked A. muciniphila with a lean phenotype and often levels are lower in those with obesity and its related metabolic alterations. Additionally, SCFA producers such as Blautia and Alloprevotella were increased. This contributes to the improvement of the health status of the gut microbial community. Remarkably there did not seem to be any major differences between FuL and FuA in this instance as both influenced the gut bacterial community correspondingly. Thus, ingredients from marine resources, particularly fucoidan, have potential as mediators of gut microbes.

Other components influencing growth

Gut metabolites as modulators

Apart from dietary and environmental factors, the host-derived as well as microbial-derived metabolites in the gut can act as modulators of the gut ecosystem. Lactate is one of the many studied metabolites. Interest in lactate as a substrate to impact on the gut microbiota has increased in recent years due to the fact that some lactate-utilizing bacteria have the ability to produce butyrate.¹⁴ Notably, E. hallii and A. caccae have demonstrated an ability to use both D- and L-lactate.¹⁴ Evidence of lactate being utilized in cross-feeding studies is particularly intriguing as it may explain the butyrogenic effect of certain dietary bioactives, including resistant starch.^14,104 Indeed, Bifidobacterium and Bacteroides who are known degraders of RS do not produce butyrate. It is hypothesized that these active starch degraders may produce lactate initially and that, subsequently, lactate utilizers such as E. hallii and A. caccae may produce butyrate using lactate as a substrate.¹⁴ The synergistic effect of this cross-feeding mechanism highlights the intricate and elaborate relationships of co-inhabitants within the gut. Interestingly, synbiotic administration of A. caccae and GOS improved beneficial organic acid production in comparison with GOS alone.¹⁰⁵ Also, in a separate study, Roseburia sp. strain A2-183 was unable to consume lactate or grow on potato starch or FOS in pure culture, but when co-cultured with B. adolescentis L2–32 in the presence of starch or FOS, the bacterium produced butyrate.¹⁰⁴

As noted earlier, acetate is the most abundant SCFA in the colon and while it is normally an end-product of anaerobic fermentation, it can also be utilized by some butyrate-producing species within the gut.¹⁰⁶ For example, F. prausnitzii, one of the most abundant bacteria in the colon, is also a prominent acetate-consumer¹⁴ and grows poorly on media deficient in acetate.¹⁰⁷ Along with F. prausnitzii, R. intestinalis and E. rectale are also known acetate consumers.^106,108 Although they are unable to use acetate as a sole source of energy, they do show net utilization of acetate possibly to produce butyrate via acetyl-CoA formation. Strains of F. prausnitzii and Roseburia sp. possess butyryl-CoA:acetate-CoA transferase which is involved in catalyzing butyryl-CoA to butyrate.¹⁰⁶ This reaction may also be carried out by butyrate kinase instead of the transferase.¹⁰⁹ Other members of the intestinal microbiota may also play a role in making acetate available for utilization by other microbes present in the gut. A. muciniphila has the ability to degrade and utilize host mucus and consequently can produce 1,2-propanediol, acetate and propionate, thereby stimulating the nearby butyrate producers as a result of this specialized activity within this niche environment.¹³ Barcenilla et al., working with human fecal samples, demonstrated that 95% of strains isolated that were net acetate utilizers were butyrate producers.¹⁰⁸ Half of butyrate-producing isolates examined in this study, however, exhibited net acetate consumption. Not all acetate, of course, is re-directed for butyrate synthesis. However, there appears to be a strong link between acetate consumption and butyrate production by colonic bacteria. Further exploration to establish to what degree acetate is required for colonic butyrate production is necessary in developing our understanding of this relationship.

N-acetylglucosamine (GlcNAc) and N-acetylgalactosamine (GalNAc) are host-derived monosaccharide derivatives of glucose and galactose, respectively, and are found abundantly in the mucus lining of the gut.³⁵ As noted, A. muciniphila is an intestinal bacterium that resides in this ecological niche.^29,110 This beneficial bacterium is adept at colonizing the mucus lining and has the ability to utilize a number of sugars including GlcNAc and GalNAc, although optimal growth in vitro is achieved with media containing mucin.³⁵ Other components of mucin may further enhance its growth.

The more recent generation of culture-independent, sequence-based understanding of gut communities facilitates the use of computational modelling to elucidate metabolic functions possessed by a single species^22,111 or microbial gut communities,¹¹² as well as predicting possible growth substrates based on the pathways present. Based on whole genome sequences, models have been developed to predict the metabolic capabilities of F. prausnitzii²² and A. muciniphila^111,113 in silico. The ability of microbes to utilize a variety of carbohydrates, including the host-derived sugar GlcNAc, and amino acids or other growth promoters can be confirmed with cultivation experiments. In evaluating the predicted conditions, GlcNAc was found to enhance the growth of F. prausnitzii better than glucose.²² This ability was also observed in several tested strains of F. prausnitzii suggesting that multiple strains can grow well on GlcNAc, highlighting its ability to utilize both diet- and host-derived substrates.¹² Although GlcNAc is not commonly found in foods, it is found in mushrooms as GlcNAc is a monomer of chitin¹¹⁴ and GlcNAc supplements are also available on the market.

Minerals

Minerals, such as iron, magnesium, sodium and calcium, are essential elements for living organisms, including most bacteria, and play pivotal roles in many biological processes. They are present abundantly in many food sources and, when these foods are ingested; microbes in the gut have access to these minerals. Although a little is known about the regulation of mineral absorption mediated by gut bacteria and their impact on the gut microbiota, information on the importance of major dietary minerals on health is available. Of these, iron has sparked the interest of many researchers over the last number of years, with a focus on the impact on malnourished children. Iron deficiency is very prevalent among African children with the WHO estimating 62.3% of preschool children are anemic.¹¹⁵ Iron-containing micronutrient powders (MNPs) are added to foods in a bid to reduce iron deficiency and prevent anemia in children but the impact of iron supplementation on gut microbiota composition and functions has yielded conflicting results.^116–119 These differences might be partly due to age-related differences in various cohorts, diverse geographical settings, the variety of iron compounds involved and the doses administered in the different studies, which are summarized below.

The impact iron exerts on the production of bacterial metabolites, in particular butyrate, has also been the focus of some investigations. Butyrate seems to be strongly affected in response to iron. In one in vitro colonic fermentation model, moderately low levels of iron increased butyrate production whereas very low levels impaired its production.¹¹⁹ The same study found that under high iron conditions, an increase in abundance of propionate-producing Bacteroidaceae and a decrease in butyrate-producing Lachnospiraceae were reflected based on the metabolites produced. Butyrate levels produced by R. intestinalis in batch culture were negatively affected by low iron levels while lactate and formate were enhanced. The results were reversed with high iron levels suggesting that levels of a single mineral can have a big impact on gut microbiota. However, some recent data has indicated that iron supplementation has adverse effects on the gut microbiota.^116,117,120 Notably iron supplementation resulted in decreased levels of desirable bacterial groups such as Bifidobacteriaceae and Lactobacillus and increased levels of Enterobacteriaceae. The production of siderophores in many pathogenic bacteria may have been responsible for the latter phenomenon.¹¹⁶ One intriguing study by Paganini et al.¹²⁰ involving three groups of Kenyan infants demonstrated that the addition of GOS with a MNP containing iron (FeGOS group) mitigated the adverse effects observed in comparison to children that were administered the MNP and iron without the prebiotic (Fe group). The FeGOS group exhibited a similar gut microbiota composition to the control group (MNP without iron or GOS) but had lower levels of virulence and toxin genes (VTGs) and increased production of SCFAs. On the other hand, a shift in microbiota composition to a more adult-like community was observed in the Fe group. Thus, the prebiotic seemed to stabilize the gut microbial community by limiting the negative effects induced by the addition of iron. Further, the addition of GOS did not seem to affect the bioavailability of iron. The MNP used in this study was fortified with 5 g of iron, i.e., much lower than the average amount of 12.5 g used in standard MNP supplements in earlier studies.¹²¹ The lower dose maintained efficacy in that the iron was of high bioavailability and reduced anemia so therefore may be more suitable than the current high dose levels which are between 10 and 12.5 mg of fortified iron according to the WHO.¹²² As noted, the varying doses and iron compounds investigated throughout the literature highlight a need for standardization in supplementation regimes. However, achieving this across different cohorts is difficult as many factors contribute to iron bioavailability and differences in microbiota across different populations makes it difficult to establish the complete impact of iron.

The effect of other minerals on the gut microbiota, in particular bacteria recently identified as beneficial, is currently not well understood. While there are some insights to suggest that other minerals, such as zinc, can have an impact,^123,124 little is currently known.

Synbiotics

Synbiotics have also attracted significant attention in recent years. Synbiotics are combinations of a probiotic and an appropriate prebiotic designed to enhance specifically the growth and survival of the probiotic or other desirable bacteria within the gastrointestinal environment, enabling beneficial effects to be induced more effectively.¹²⁵ This provides a very practical method of overcoming the difficulties that a probiotic may endure when initially introduced into the gut environment. Challenges in designing a synbiotic include selecting the right combination of pro- and prebiotic and ensuring sufficient quantities of each. As bifidobacteria and lactobacilli are well-characterized members of the human intestinal consortium and the benefits of specific strains with respect to human health have been established, microbes from these genera are most frequently included in synbiotics. FOS, GOS, inulin and lactulose are common choices for synbiotic formulations. Some newly identified health-promoting bacteria have been the focus of investigations to assess the consequence of combining their use with some well-established prebiotics. For example, a synbiotic administration of A. caccae L2 with GOS to rats enhanced intestinal organic acid production more effectively than just GOS alone.¹⁰⁵ It was hypothesized that GOS increased levels of bifidobacteria, resulting in greater lactate production, which in turn promoted A. caccae growth, therefore, facilitating the production of butyrate by A. caccae. This demonstrates one way of utilizing the complex cross-feeding mechanisms to selectively manipulate target microbes.

Conclusion

The human gut harbors a diverse collection of microbes whose interactions and functionality have yet to be completely elucidated. However, it is already clear that these microbes play an integral role in human health and well-being. An increasing awareness that various microbes that can influence human health has been a catalyst for an ever-rising number of investigations into benefits of gut microbiota manipulation. Prebiotics, and other nutrients, targeted to specific health-promoting bacteria are establishing themselves as a significant means of improving host health and disease. This has been evident in the considerable body of work that has accumulated in the last few years by demonstrating that species composition within the microbiota can be modified with very few changes in food consumption. The expansion of the prebiotic definition to substrates that go beyond the classical prebiotics and the inclusion of bacteria other than Bifidobacterium and Lactobacillus have enabled a greater examination of this intriguing relationship between diet and the microbiota. It is anticipated that these substances can soon be harnessed to promote populations of newly-identified health promoting bacteria in the gut.

Additional information

Funding

CL was funded by the Teagasc Walsh Fellowship Scheme [Grant no. 2017047] and internal Teagasc RMIS funding. DT was funded by a DAFM grant [Grant no. 15/F/635]. PDC was funded in the form of a center grant [APC Microbiome Ireland Grant Number SFI/12/RC/2273].