Central to Hooks et al.’s analysis is their critique of unwarranted causal claims about the impact of the gut microbiota on psychobiological functioning following probiotic interventions, given the inconsistent evidence on the capacity of probiotics to modulate gut microbiota composition and function. Although we agree, evidence on the effects of dietary fiber (DF) and prebiotics on microbiota composition and function and subsequent psychobiological changes are not discussed. We argue that such findings illustrate the potential of prebiotic interventions in supporting causal claims about the impact of gut microbiota on psychobiological functioning. Nonetheless, before meriting such claims, direct investigation into the mechanisms that mediate DF/prebiotic effects on brain function are needed.

Critical to warranting causal claims in microbiota-gut-brain (MGB) axis research is the availability of tools to steer the microbial ecosystem into a desired composition/function that subsequently improves brain function. DF/prebiotics may constitute such a tool. DF is defined as carbohydrate polymers with three or more monomeric units, which are not hydrolyzed by the endogenous small intestinal human enzymes, are naturally occurring or isolated from foods, and demonstrate a physiological health benefit (Jones Reference Jones2014). Fermentable fibers provide metabolic substrates for most gut bacteria, influence their diversity and richness, and increase levels of fermentation products such as short chain fatty acids (SCFAs) (den Besten et al. Reference den Besten, van Eunen, Groen, Venema, Reijngoud and Bakker2013). Some fibers can be classified as “prebiotic” if they are “selectively utilized by host microorganisms conferring a health benefit” (Gibson et al. Reference Gibson, Hutkins, Sanders, Prescott, Reimer, Salminen, Scott, Swanson, Cani, Verbeke and Reid2017, p. 493).

A systematic review and meta-analysis of 64 intervention studies in healthy adults found that DF resulted in consistently higher abundance of Bifidobacterium spp. and Lactobacillus spp. (So et al. Reference So, Whelan, Rossi, Morrison, Holtmann, Kelly, Shanahan, Staudacher and Campbell2018). However, the application of next-generation sequencing techniques that allow microbiota-wide assessment of relative abundance shifts suggests that modification of microbiota composition by means of DF/prebiotics is not limited to these specific taxa (Davis et al. Reference Davis, Martínez, Walter, Goin and Hutkins2011; Everard et al. Reference Everard, Lazarevic, Derrien, Girard, Muccioli, Neyrinck, Possemiers, Van Holle, François, de Vos, Delzenne, Schrenzel and Cani2011; Reference Everard, Lazarevic, Gaïa, Johansson, Ståhlman, Backhed, Delzenne, Schrenzel, François and Cani2014; Holscher et al. Reference Holscher, Caporaso, Hooda, Brulc, Fahey and Swanson2015; Martínez et al. Reference Martínez, Kim, Duffy, Schlegel and Walter2010; Vandeputte et al. Reference Vandeputte, Falony, Vieira-Silva, Wang, Sailer, Theis, Verbeke and Raes2017a; Walker et al. Reference Walker, Ince, Duncan, Webster, Holtrop, Ze, Brown, Stares, Scott, Bergerat, Louis, McIntosh, Johnstone, Lobley, Parkhill and Flint2011). Notably, these studies indicate that changes in gut microbiota composition are reversible, maintained only with continued consumption of DF/prebiotics, and exhibit inter-individual variation probably dependent on baseline microbiota profile, including presence of keystone species or variation in enzymatic capacity of certain strains (Falony et al. Reference Falony, Joossens, Vieira-Silva, Wang, Darzi, Faust, Kurilshikov, Bonder, Valles-Colomer, Vandeputte, Tito, Chaffron, Rymenans, Verspecht, De Sutter, Lima-Mendez, D'Hoe, Jonckheere, Homola, Garcia, Tigchelaar, Eeckhaudt, Fu, Henckaerts, Zhernakova, Wijmenga and Raes2016; Ze et al. Reference Ze, Le Mougen, Duncan, Louis and Flint2013).

Although DF/prebiotics consistently modify gut microbiota composition, only a few studies have explored their effect on neurobehavioral functioning. Human studies, although notably scarce, showed positive effects on hypothalamic-pituitary-adrenal axis activity, emotional attention, and anxiety and depression symptomology (Azpiroz et al. Reference Azpiroz, Dubray, Bernalier-Donadille, Cardot, Accarino, Serra, Wagner, Respondek and Dapoigny2017; Farhangi et al. Reference Farhangi, Javid, Sarmadi, Karimi and Dehghan2017; Schmidt et al. Reference Schmidt, Cowen, Harmer, Tzortzis, Errington and Burnet2015). Animal studies revealed effects on stress response (Forsatkar et al. Reference Forsatkar, Nematollahi, Rafiee, Farahmand and Lawrence2017), anxiety- and depressive-like behavior (Mika et al. Reference Mika, Day, Martinez, Rumian, Greenwood, Chichlowski, Berg and Fleshner2017; Reference Mika, Gaffney, Roller, Hills, Bouchet, Hulen, Thompson, Chichlowski, Berg, Fleshner and Fleshner2018; Savignac et al. Reference Savignac, Couch, Stratford, Bannerman, Tzortzis, Anthony and Burnet2016), stress-induced sleep alterations (Thompson et al. Reference Thompson, Roller, Mika, Greenwood, Knight, Chichlowski, Berg and Fleshner2016), cognition (Gronier et al. Reference Gronier, Savignac, Di Miceli, Idriss, Tzortzis, Anthony and Burnet2018), and related neurobiological mechanisms such as gamma aminobutyric acid (GABA) and serotonin receptor gene expression (Burokas et al. Reference Burokas, Arboleya, Moloney, Peterson, Murphy, Clarke, Stanton, Dinan and Cryan2017) and brain-derived neurotrophic factor (BDNF) and N-methyl-d-aspartate receptor subunit levels (Savignac et al. Reference Savignac, Corona, Mills, Chen, Spencer, Tzortzis and Burnet2013; Williams et al. Reference Williams, Chen, Savignac, Tzortzis, Anthony and Burnet2016). None of the human studies, and only some of these animal studies, concurrently quantified microbiota composition and found increased abundance in fecal Bifidobacterium and Lactobacillus spp. using selective bacterial culture or quantitative polymerase chain reaction (Azpiroz et al. Reference Azpiroz, Dubray, Bernalier-Donadille, Cardot, Accarino, Serra, Wagner, Respondek and Dapoigny2017; Gronier et al. Reference Gronier, Savignac, Di Miceli, Idriss, Tzortzis, Anthony and Burnet2018; Kao et al. Reference Kao, Spitzer, Anthony, Lennox and Burnet2018; Mika et al. Reference Mika, Day, Martinez, Rumian, Greenwood, Chichlowski, Berg and Fleshner2017; Reference Mika, Gaffney, Roller, Hills, Bouchet, Hulen, Thompson, Chichlowski, Berg, Fleshner and Fleshner2018; Savignac et al. Reference Savignac, Corona, Mills, Chen, Spencer, Tzortzis and Burnet2013; Reference Savignac, Couch, Stratford, Bannerman, Tzortzis, Anthony and Burnet2016; Thompson et al. Reference Thompson, Roller, Mika, Greenwood, Knight, Chichlowski, Berg and Fleshner2016). One study (Burokas et al. Reference Burokas, Arboleya, Moloney, Peterson, Murphy, Clarke, Stanton, Dinan and Cryan2017) used 16S rRNA sequencing and showed changes in β-diversity and shifts at different taxonomic levels. A limited subset of these studies, exclusively in rats, correlated prebiotic-induced changes in microbiota composition and relative abundance with changes in brain function. Prebiotic-induced increases in fecal Lactobaccillus spp. positively correlated with altered cfos and serotonin receptor gene expression in multiple brain regions (Mika et al. Reference Mika, Gaffney, Roller, Hills, Bouchet, Hulen, Thompson, Chichlowski, Berg, Fleshner and Fleshner2018) and predicted stress-protective alternations in mRNA expression in serotonergic dorsal raphe nucleus neurons during inescapable stress (Mika et al. Reference Mika, Day, Martinez, Rumian, Greenwood, Chichlowski, Berg and Fleshner2017). Furthermore, lower levels of Deferribacteres following a prebiotic diet correlated with longer non–rapid eye movement episode durations (Thompson et al. Reference Thompson, Roller, Mika, Greenwood, Knight, Chichlowski, Berg and Fleshner2016). Following ingestion of fructo- or galacto-oligosaccarides, the number of fecal bifidobacteria correlated positively with frontal cortex NR1 protein (Savignac et al. Reference Savignac, Corona, Mills, Chen, Spencer, Tzortzis and Burnet2013).

Bacterial fermentation of DF leads to the production of SCFAs (den Besten et al. Reference den Besten, van Eunen, Groen, Venema, Reijngoud and Bakker2013). SCFAs – predominantly acetate, propionate, and butyrate – constitute the major anions in the colon and serve as an energy source for colonocytes. Furthermore, they inhibit histone deacetylation and activate G-protein coupled receptors, thereby acting as signaling molecules linking diet, gut microbiota, and host (Tan et al. Reference Tan, McKenzie, Potamitis, Thorburn, Mackay and Macia2014) and interacting with gut-brain signaling pathways (Dalile et al. Reference Dalile, Van Oudenhove, Vervliet and Verbeke2019). Few studies have explored whether the effects of DF/prebiotics on brain function are mediated by SCFAs. Fructo- and galacto-oligosaccaride-induced increases in cecal SCFAs correlated with effects on depressive and anxious behavior and stress responses, as well as changes in gene expression in mice (Burokas et al. Reference Burokas, Arboleya, Moloney, Peterson, Murphy, Clarke, Stanton, Dinan and Cryan2017). Prebiotic Bimuno galacto-oligosaccharides increased plasma acetate levels (Gronier et al. Reference Gronier, Savignac, Di Miceli, Idriss, Tzortzis, Anthony and Burnet2018; Kao et al. Reference Kao, Spitzer, Anthony, Lennox and Burnet2018), cortical GluN2B subunits (involved in glutamate neurotransmission), and acetyl-CoA carboxylase mRNA, all of which also increased following direct administration of acetate (Gronier et al. Reference Gronier, Savignac, Di Miceli, Idriss, Tzortzis, Anthony and Burnet2018), suggesting that acetate may play a mechanistic role in the observed effects. Other studies that explored the effects of SCFA administration on brain function have been reviewed elsewhere (Dalile et al. Reference Dalile, Van Oudenhove, Vervliet and Verbeke2019).

Although current evidence does not convincingly support a causal role of gut microbiota in modulating neurobehavioral functioning, we believe that prebiotics have a higher potential than probiotics to warrant such causal claims. However, before maintaining such claims, high-quality, adequately powered (human) prebiotic intervention studies, measuring both changes in microbiota composition or function (Bindels et al. Reference Bindels, Delzenne, Cani and Walter2015) and psychobiological functioning using state-of-the-art methodology, are needed. Such interventions should use mediation analysis to estimate the contribution of microbiota composition/function in the observed psychobiological effects. Claims that causally implicate the role of gut microbiota in the MGB axis should be based on studies that isolate products of microbial activity and directly demonstrate their causal effects on the brain.