Probiotic-based functional beverages: a comprehensive review of strain selection, formulation strategies, viability challenges, and health-promoting potential

1 Introduction

The link between diet and health has been recognized since antiquity. It was Elie Metchnikoff’s early twentieth-century observation, however, that Bulgarian peasants who regularly consumed fermented milk enjoyed unusual longevity, an insight that first framed live dietary microorganisms as a determinant of host health. This observation seeded the modern concept of “probiotics,” subsequently formalized by the FAO/WHO expert consultation of 2001 as live microorganisms that, when administered in adequate amounts, confer a health benefit on the host (Koirala and Anal, 2021). Over the past two decades, this definition has anchored a substantial body of research demonstrating that probiotic bacteria can modulate immune responses, attenuate inflammation, and reshape the intestinal microbiota (Palomar et al., 2014); these effects extend further to lowering serum cholesterol, alleviating lactose intolerance, and exerting anti-hypertensive and anti-carcinogenic activity (Latif et al., 2023). These findings have driven the incorporation of probiotic microorganisms into an increasingly diverse range of foods, pharmaceuticals, and dietary supplements, making probiotics a commercially significant and scientifically active subdivision of the functional food industry (Jalgaonkar et al., 2019; Nazhand et al., 2020; Tamang et al., 2016).

Functional foods, broadly defined as products consumed as part of a normal diet that additionally confer a physiological benefit beyond basic nutrition, have grown into a substantial global market as health-conscious consumers increasingly seek dietary tools for disease prevention and wellbeing (Jalgaonkar et al., 2019; Nazhand et al., 2020). Within this category, functional beverages occupy a distinctive niche: their liquid format allows for convenient, rapid delivery of bioactive compounds, hydration, and sensory pleasure simultaneously, spanning sports drinks, energy drinks, dairy-based drinks, fruit- and vegetable-based drinks, and nutraceutical formulations (Nazhand et al., 2020). The convenience and palatability of the beverage format have made it a preferred vehicle for probiotic delivery, and food industry research has consequently intensified around probiotic-containing beverages built on Lactobacillus, Bifidobacterium, and related genera (Latif et al., 2023; Tamang et al., 2016). For example, pomegranate juice fermented with Lactobacillus plantarum ATCC 14917 has been shown to develop a more favourable volatile compound profile, higher antioxidant activity, and greater total phenolic content relative to unfermented juice (Mantzourani et al., 2018), illustrating how fermentation can simultaneously deliver probiotic organisms and enhance the intrinsic functional value of the base matrix. The successful survival of these cultures within a beverage system, however, depends on a complex interplay of pH, acidity, redox potential, dissolved oxygen, and hydrogen peroxide concentration, all of which must be actively managed during formulation (Nigam, 2018).

Dairy products, particularly fermented milk and yogurt, have historically served as the dominant probiotic delivery matrix. Yet lactose intolerance, veganism, dietary cholesterol concerns, and milk-protein allergenicity have collectively increased consumer demand for non-dairy probiotic alternatives (Aspri et al., 2020; Vasudha and Mishra, 2013). Fruit and vegetable juices are particularly attractive substrates in this regard: their intrinsic carbohydrate, mineral, and vitamin content provides a favourable fermentation substrate, while the lactic acid generated during fermentation confers antibacterial activity, extends shelf life, and imparts a characteristic tangy flavour profile (Butu and Rodino, 2019; Mojikon et al., 2022). Cereal-, legume-, and herbal-based matrices further widen the substrate base available for probiotic delivery, each carrying its own technological opportunities and constraints. Taken together, this diversification of matrices, strains, and processing approaches has produced a large and rapidly expanding literature—but one that remains fragmented across disciplinary silos (microbiology, food engineering, sensory science, and clinical nutrition) and across matrix-specific research communities that rarely cross-reference one another.

Several narrative reviews have previously addressed individual facets of this field, including probiotic mechanisms of action in the food industry generally (Latif et al., 2023), non-dairy probiotic beverages specifically (Aspri et al., 2020; Vasudha and Mishra, 2013), and the broader relationship between probiotics and human health (Sarita et al., 2025). However, existing summaries tend to be either mechanism-centred (focused on host physiology) or matrix-centred (focused on a single substrate category, such as dairy or fruit juice), with comparatively little synthesis that integrates strain selection criteria, cross-matrix formulation science, physicochemical stabilization, and viability engineering within a single, comparative framework specific to the beverage format. Building on our own earlier comprehensive review of probiotics and human health (Sarita et al., 2025), which examined host-level physiological outcomes, the present review is deliberately application- and technology-focused: it is organized around the practical formulation science of probiotic beverages themselves rather than around host health outcomes in isolation.

The novelty and contribution of this review are therefore threefold. First, it consolidates dairy-based, non-dairy (fruit, vegetable, and cereal-based), and herbal probiotic beverage literature within a single comparative structure, allowing matrix-specific advantages and constraints to be evaluated side by side rather than in isolation. Second, it foregrounds the physicochemical and engineering determinants of product viability and quality—titratable acidity, viscosity, total soluble solids, colour, flavour, and encapsulation strategy—as a unifying analytical lens across matrices, rather than treating these as an afterthought to strain-level microbiology. Third, and most importantly, it moves beyond a descriptive catalogue of individual studies to a critical, gap-oriented appraisal that identifies recurring methodological inconsistencies in the literature and proposes a concrete research agenda for the field. Accordingly, this review aims to (i) classify and compare the major categories of probiotic functional beverages; (ii) synthesize current strain selection, characterization, and safety-screening practices; (iii) evaluate matrix-specific formulation and processing strategies; (iv) critically examine the physicochemical determinants of beverage quality and probiotic viability, including stabilization and encapsulation approaches; (v) summarize the health-promoting mechanisms, safety considerations, and regulatory landscape relevant to probiotic beverages; and (vi) identify the principal knowledge gaps and priority directions for future formulation research.

2 Review methodology and scope

This article is structured as a comprehensive, matrix-comparative narrative review rather than a formal systematic review with meta-analytic pooling of quantitative outcomes; given the heterogeneity of substrates, strains, processing conditions, and outcome measures reported across the underlying literature, a narrative synthesis with explicit comparative and gap-analysis structure was judged the most appropriate and transparent format for this body of evidence, consistent with the scope of a comprehensive review article in Frontiers in Food Science and Technology. The review draws on peer-reviewed research articles, book chapters, and authoritative technical reports addressing probiotic-based dairy, non-dairy (fruit, vegetable, and cereal-based), and herbal beverages, together with foundational regulatory and definitional documents from the FAO/WHO. Sources were organized thematically around beverage classification, strain selection and characterization, formulation and process development, physicochemical stabilization, probiotic viability engineering, and health/safety/regulatory considerations, with cross-matrix comparison used throughout to surface convergent findings, contradictions, and gaps in the literature. Studies reporting original strain-beverage combinations, fermentation or fortification conditions, and quantitative viability or physicochemical outcomes were prioritized for inclusion in the comparative tables. Where the literature is explicitly noted as limited, inconsistent, or absent for a given sub-topic (for example, standardized viability endpoints or comparative shelf-life protocols across matrices), this is flagged directly in the relevant section and revisited in the critical appraisal (Section 8) rather than left implicit. Literature was primarily identified through structured searches of PubMed, Scopus, and Web of Science, supplemented by Google Scholar and publisher-indexed sources, covering studies published up to 2026 with particular emphasis on evidence from approximately the last 15 years; earlier foundational sources (for example, the 2001 FAO/WHO probiotic consultation) were retained where they remain the primary regulatory or definitional reference for the field. For consistency with the majority of the primary literature synthesized in this review, bacterial strain names throughout the manuscript follow the historical Lactobacillus genus nomenclature rather than the emended 2020 taxonomic reclassification of the family Lactobacillaceae into 23 novel genera (e.g., Lacticaseibacillus, Lactiplantibacillus, Limosilactobacillus) (Zheng et al., 2020); readers cross-referencing strain names against current taxonomic databases should note this convention.

3 Classification of probiotic functional beverages

Probiotic functional beverages can be broadly classified into three matrix categories that differ substantially in their intrinsic composition, processing requirements, and target consumer segments: dairy-based beverages, non-dairy beverages derived from fruits, vegetables, and cereals, and herbal or botanical beverages. Although these categories are frequently discussed in isolation within the literature, they share common underlying technological challenges—principally the preservation of strain viability against a background of matrix-specific chemical stressors—that justify a comparative treatment (Figure 1).

3.1 Dairy-based probiotic beverages

Probiotic, prebiotic, and synbiotic dairy beverages were the first functional beverage category to achieve commercial scale and continue to dominate the global functional beverage market. Based on the primary dairy ingredient employed, these beverages are conventionally divided into whey-based, buttermilk-based, and milk-based formulations, drawing respectively on liquid whey (a cheese/casein industry by-product), liquid buttermilk (a butter-manufacturing by-product), and liquid milk as the principal carrier (Mudgil and Barak, 2019). Fermentation of whey in particular yields protein hydrolysates that are more readily absorbed than intact proteins, supports gastrointestinal health, may reduce allergy- and intolerance-associated symptoms, and can strengthen immune function (Kaur et al., 2022). In a representative example, whey fermented with the probiotic strain Lactobacillus rhamnosus MTCC 5897 demonstrated anti-inflammatory activity against Escherichia coli-induced inflammation in Caco-2 intestinal epithelial cells (Gupta et al., 2021), illustrating that the dairy matrix itself can act as a functional carrier rather than a passive delivery vehicle. This is consistent with a broader pattern reported across dairy-based probiotic-fermented products, in which whey- and milk-derived bioactive peptides, exopolysaccharides, and other fermentation metabolites contribute functional properties independent of the probiotic strain itself, reinforcing that dairy matrix selection is not a neutral formulation choice but an active contributor to the health-promoting profile of the finished beverage (Butu and Rodino, 2019; Kaur et al., 2022; Mudgil and Barak, 2019).

3.2 Non-dairy probiotic beverages (fruit, vegetable and cereal-based)

A substantial body of traditional, non-alcoholic, grain-based fermented beverages exists across global food cultures, and contemporary research has systematically extended probiotic fortification to cereals, millets, legumes, fruits, and vegetables. These non-dairy matrices are attractive for several converging reasons: they are comparatively low-cost, rich in phytochemical and dietary-fibre content, free of the cholesterol burden associated with dairy, and inherently suited to lactose-intolerant and vegan consumer segments (Butu and Rodino, 2019). Grain-, fruit-, and vegetable-derived juices additionally supply minerals, dietary fibre, and cancer-protective phytochemicals that reinforce their positioning within the broader healthy-food category (Mishra et al., 2021). Table 1 summarizes representative non-dairy probiotic beverages reported in the literature, organized by raw material source and probiotic strain employed, and illustrates the breadth of substrates—from tropical fruits (pineapple, papaya) to root vegetables (beetroot) to fermented cereal beverages (Bushera, Boza) — that have been explored to date.

TABLE 1. Representative non-dairy probiotic beverages, source substrates, and probiotic strains reported in the literature.

Considered together, the studies summarized in Table 1 indicate that fruit- and vegetable-based matrices are most frequently paired with acid-tolerant Lactobacillus species (notably L. plantarum, L. casei, and L. rhamnosus), while cereal-based traditional beverages such as Boza and Bushera support more complex, spontaneously fermenting microbial consortia involving multiple lactic acid bacteria genera and co-occurring yeasts. This distinction has direct formulation implications: single-strain fruit and vegetable juice systems are more amenable to standardized, reproducible industrial fermentation, whereas traditional multi-species cereal beverages present greater strain-identification and quality-control complexity but may offer broader metabolite diversity.

3.3 Herbal and botanical probiotic beverages

Herbal beverages, more precisely termed tisanes, herbal infusions, or botanical infusions to distinguish them from true tea (Camellia sinensis), are prepared from fresh or dried flowers, immature fruits, leaves, seeds, and roots through infusion or boiling (Chandrasekara and Shahidi, 2018). As a probiotic delivery matrix, herbal beverages occupy a distinctive dual-functionality niche: the botanical substrate itself frequently contributes non-digestible carbohydrates and phenolic compounds that act as prebiotics, supporting probiotic proliferation and enhancing intestinal colonization capacity, while Lactobacillus species in particular have been shown to inhibit colonization by invasive pathogens and help restore gut microbial balance (Wang et al., 2021). In an illustrative formulation study, a probiotic herbal beverage combining Hemidesmus indicus (Indian sarsaparilla, locally “Nannari”) roots, Ocimum tenuiflorum (holy basil), and Aloe vera, flavoured with lemon, pineapple, or ginger and fermented using Enterococcus durans isolated from curd, exhibited improved nutritional and antioxidant properties relative to the unfermented herbal base (Sundararajan and Priya, 2021). Beyond individual formulation studies, biotransformation of medicine-food homology botanicals by probiotics and intestinal microbiota has been shown more broadly to enhance bioactivity, improve flavour, and support antimicrobial, antioxidant, anti-inflammatory, and gut-modulating outcomes, positioning herbal probiotic beverages as a promising but comparatively under-industrialized frontier relative to dairy and fruit/vegetable systems (Cui et al., 2024; Shi et al., 2024).

4 Selection, characterization and safety screening of probiotic strains

The probiotic strain selected for beverage incorporation must, above all, be non-pathogenic and safe for the intended host population; in practice, the overwhelming majority of strains employed in food and beverage applications belong to the genera Lactobacillus and Bifidobacterium, both generally regarded as safe (Fijan, 2014; Koirala and Anal, 2021). Strain development for beverage applications follows a broadly conserved workflow. Isolation and identification typically begin with 16S rRNA gene sequencing, confirmed through complementary phenotypic and genotypic characterization. Candidate strains are then screened against a defined functional panel encompassing bile acid tolerance, gastric juice resistance, adherence to human epithelial cell lines, capacity to reduce pathogen burden in the gut, antimicrobial activity, and bile salt hydrolase activity (Zielińska et al., 2018). Strains that pass this functional screen must additionally undergo safety assessment for undesirable metabolite production, antimicrobial resistance gene carriage, and haemolytic activity, with confirmatory evidence increasingly sought from human clinical trials before commercial deployment (Zielińska et al., 2018).

Beyond Lactobacillus and Bifidobacterium, genera including Streptococcus, Propionibacterium, Enterococcus, Pediococcus, and Saccharomyces have also been characterized as suitable for non-dairy probiotic applications (Fijan, 2014). Screening studies isolating candidate strains from traditional fermented vegetables illustrate the rigor required at this stage: in one such study, only 14 of 38 isolates from fermented cucumber and cabbage were confirmed as Lactobacillus spp., and although most surviving strains tolerated pH 2.5–3.5 with survival exceeding 90%, none survived pH 1.5 except four strains that persisted for 90 min at survival rates below 39%. Bile salt tolerance ranged from 81% to 94% at 0.2% concentration to 59%–94% at higher (2%–4%) concentrations, all strains resisted 0.4% phenol, and four candidates were ultimately excluded due to antibiotic resistance or undesirable metabolic activity (Zielińska et al., 2015). This example underscores that probiotic functionality is not an intrinsic, matrix-independent property of a species but rather a strain- and context-specific outcome that must be re-verified for each intended application; the same conclusion is reinforced by juice-matrix comparison studies showing that Lactobacillus and Bifidobacterium strains exhibit markedly different acid resistance in orange, pineapple, and cranberry juice, with L. casei, L. rhamnosus, and Lactobacillus paracasei sustaining viable counts above 7.0 log CFU/mL in orange juice and above 6.0 log CFU/mL in pineapple juice for at least 12 weeks, yet failing to survive standard pasteurization treatments (76 °C/30 s or 90 °C/1 min) or high-pressure processing (400 MPa/5 min) at commercially relevant levels (Sheehan et al., 2007). Strain selection, in other words, cannot be decoupled from the intended matrix and process; a strain validated in one juice or dairy system cannot be assumed functionally equivalent in another.

Table 2 summarizes the principal starter culture categories used across dairy and non-dairy probiotic beverages together with their dominant metabolic products, spanning traditional lactic acid bacteria (Lactococcus, Leuconostoc, Pediococcus acidilactici, Streptococcus thermophilus, Lactobacillus delbrueckii spp.), non-traditional lactic acid bacteria (L. acidophilus, L. helveticus, L. johnsonii, L. casei, L. reuteri, L. plantarum, L. rhamnosus, Bifidobacterium spp., Enterococcus faecium/faecalis), and yeasts (Candida, Saccharomyces, Kluyveromyces, Debaryomyces spp.) (Rasguido and Luis, 1997; Tamime and Marshall, 1997). The most widely applied species across dairy fermentation include L. acidophilus, L. gasseri, L. helveticus, L. delbrueckii subsp. bulgaricus, L. casei, L. paracasei, L. rhamnosus, L. fermentum, L. reuteri, L. plantarum, and L. lactis, alongside Bifidobacterium bifidum, Bifidobacterium longum, Bifidobacterium breve, and Bifidobacterium animalis subsp. lactis (Li et al., 2020). In vitro investigations remain the primary source of evidence on strain characterization, genomic analysis, DNA-based identification, and viability, and are conventionally benchmarked against the criteria recommended by the FAO/WHO working group (Koirala and Anal, 2021; Soto et al., 2019).

TABLE 2. Starter cultures used in probiotic beverage fermentation and their principal metabolic products (data compiled and modified from Rasguido and Luis, 1997; Tamime and Marshall, 1997).

5 Formulation and process development of probiotic beverages

Beverages are conventionally defined as liquids intended for human consumption that provide nutrition, refreshment, and palatability, occasionally with a stimulating physiological effect; functional beverages extend this definition by delivering additional health benefits through the presence or deliberate fortification of bioactive components (Playne et al., 2003). Formulation strategy differs substantially across the three matrix categories reviewed here, reflecting differences in intrinsic composition, microbial compatibility, and processing tolerance.

5.1 Dairy matrix development

Beyond their fundamental nutritional role, milk-based beverages carry an intrinsic complement of peptides, oligosaccharides, enzymes, vitamins, and minerals—including colostrum-derived factors, glycomacropeptides, lactoferrin, lactoperoxidase, casein and whey protein hydrolysates, milk minerals, and α-lactalbumin—that can themselves function as bioactive, health-promoting ingredients (Nagpal et al., 2012). Protein, lactose, fat, and mineral content collectively shape both the technological behaviour and functional value of the resulting beverage. Commercial fermented dairy products such as Yakult (L. casei Shirota), LAJOIE (Lactobacillus paracasei LPC-37), and Cherita (L. rhamnosus GG ATCC 53103) are typically built around a single probiotic strain, but single-strain fermentation characteristically requires extended fermentation times (Soto et al., 2019). Nutrient supplementation studies illustrate one route to overcoming this limitation: fortifying milk with a defined nutrient package (asparagine, aspartate, cysteine, leucine, methionine, riboflavin, guanine, uracil, and Mn²⁺) shortened L. acidophilus LA-5 fermentation time by 9 h while maintaining product quality and, in mixed-culture systems with Streptococcus thermophilus and L. delbrueckii subsp. bulgaricus, increased viable L. acidophilus LA-5 counts 1.3-fold (Meng et al., 2021).

A parallel and increasingly prominent strategy is the development of non-fermented probiotic dairy products, motivated by the recognition that antagonistic interactions between probiotics and conventional starter cultures—together with the naturally acidic environment of fermented products—can compromise probiotic viability (Jang et al., 2024). In one representative non-fermented formulation, sterilized milk enriched with L. plantarum Ln1 or L. rhamnosus GG (10⁹ CFU/mL) and refrigerated for 14 days showed the expected decline in pH and viable counts alongside rising titratable acidity, but no significant compositional or sensory differences relative to a non-probiotic control; notably, the L. plantarum-enriched formulation exhibited elevated antioxidant activity by DPPH and ABTS⁺ assays despite the absence of fermentation-driven metabolite generation (Jang et al., 2022). This finding is instructive: it demonstrates that functional value in probiotic dairy beverages can arise from live-cell presence and strain-specific metabolic activity independent of extensive fermentation, widening the formulation design space beyond traditional fermented product formats.

5.2 Non-dairy matrix development

Unlike dairy systems, fruit- and vegetable-based matrices are inherently free of lactose, milk allergens, and cholesterol, broadening their suitability across sensitive consumer populations while retaining an appealing, naturally refreshing sensory profile. Probiotic enrichment of these matrices proceeds via two principal routes: direct inoculation of probiotic strains into unfermented juice, or fermentation of the juice by the probiotic culture itself. Fermentation is generally the preferred route, since it yields a lower-sugar product, improves microbial adaptation and survival within the matrix, and can generate protective and quality-enhancing metabolites such as bacteriocins that extend shelf life (Aspri et al., 2020). The chemical consequences of fermentation can be substantial even beyond viability outcomes: apple juice fermented with L. plantarum ATCC14917 showed enhanced DPPH and ABTS radical-scavenging activity and cellular antioxidant capacity despite a reduction in total phenolic and flavonoid content, an apparent paradox explained by LC-MS/MS analysis showing a marked increase in specific high-activity phenolic compounds (5-O-caffeoylquinic acid, quercetin, and phloretin) generated by fermentative biotransformation (Li et al., 2018). This illustrates a broader principle relevant to non-dairy formulation: fermentation reshapes rather than simply preserves the phytochemical profile of the base matrix, and net antioxidant outcomes cannot be predicted from total phenolic content alone.

Cereal grains provide a further important substrate category, contributing protein, carbohydrate, vitamins, minerals, and dietary fibre, while their non-digestible carbohydrate fraction can selectively promote Lactobacillus and Bifidobacterium growth in the colon, conferring a secondary prebiotic function (Prado et al., 2008). Traditional cereal-based probiotic products illustrate the geographic and microbial diversity of this category: Mahewu, a South African sour maize/sorghum/millet/malt/wheat product, undergoes spontaneous fermentation dominated by Lactococcus lactis subsp. lactis (Blandino et al., 2003), while Togwa, prepared from fermented multi-grain flours in East African and Asian traditions, relies on L. plantarum and Streptococcus species (Panghal et al., 2018). Table 3 consolidates representative process parameters (culture, temperature, duration, and inoculum density) reported for non-dairy probiotic beverages spanning cereal, root vegetable, legume, and fruit substrates, and highlights the wide variation in fermentation temperature (30 °C–37 °C), duration (3.2 h–48 h), and target inoculum density (10³–10⁹ CFU/mL) currently reported without a clearly converged industry standard.

TABLE 3. Representative probiotic beverages with reported fermentation conditions.

5.3 Herbal matrix development

Herbal medicines and botanical materials frequently contain bioactive compounds with anti-aging, anti-cancer, and immune-enhancing properties, but these compounds can be chemically unstable and, in some cases, associated with toxicity risk at high concentrations. Biotransformation via microbial or enzymatic fermentation has therefore gained prominence as a strategy to improve both the stability and bioactivity of herbal substrates, aligning with rising consumer interest in natural, disease-preventive “medicine-food homology” products. Probiotic-fermented herbal substrates have been shown to improve flavour, elevate concentrations of bioactive compounds, and support the growth and metabolite output of the probiotic culture itself, generating antimicrobial, antioxidant, anti-inflammatory, skin-protective, and gut-modulating benefits, alongside a smaller body of evidence flagging associated safety considerations that warrant further characterization (Cui et al., 2024). Lactic acid bacteria applied to herb-derived functional health food products further generate bioactive constituents through hydrolysis, deglycosylation, and deacylation reactions, with the resulting secondary metabolites acting synergistically to enhance bioavailability and amplify functional benefit relative to the unfermented botanical alone (Shi et al., 2024). Relative to dairy and fruit/vegetable systems, herbal matrix development remains comparatively early-stage, with fewer standardized process parameters reported in the literature—a gap discussed further in Section 8.

6 Physicochemical behaviour and stabilization strategies

Maintaining defined physicochemical characteristics throughout shelf life is central to consumer acceptance of probiotic beverages and is frequently the limiting factor determining commercial viability, independent of probiotic functionality per se.

6.1 Titratable acidity and pH

Careful control of pH is essential to probiotic strain survival within a beverage matrix. Titratable acidity quantifies total acid content via NaOH titration with phenolphthalein indicator, and is conventionally expressed as percentage citric acid in unfermented fruit and vegetable matrices (reflecting citric acid as the dominant native organic acid) or as percentage lactic acid following lactic acid bacterial fermentation, since lactic acid becomes the principal fermentation by-product (Mojikon et al., 2022). Beyond its direct relevance to microbial survival, pH governs the structural and colloidal stability of dairy beverage components: low pH and elevated acidity can drive casein and whey protein insolubility, precipitating sedimentation defects that are particularly consequential in fermented functional dairy products (Mudgil and Barak, 2019).

6.2 Viscosity and texture

Viscosity is a principal determinant of mouthfeel and overall consumer acceptability (Tripathi and Giri, 2014). Lactic acid bacteria generate exopolysaccharides during fermentation that intrinsically increase viscosity and confer a creamy texture, an effect that can be reinforced through hydrocolloid stabilizers such as guar gum or pectin (Duboc and Mollet, 2001). Yogurt-style probiotic beverages rely substantially on the combined metabolic activity of Streptococcus thermophilus and Lactobacillus delbrueckii for their characteristic creamy consistency (Amara and Shibl, 2015), and optimal viscosity is closely associated with improved phase stability during storage, with hydrocolloid selection playing a central formulation role (Mudgil and Barak, 2019).

6.3 Total soluble solids and sugar utilization

Total soluble solids (TSS), measured refractometrically, track sugar content and decline over fermentation as lactic acid bacteria metabolize available sugars for growth and reproduction; the corresponding reduction in sweetness is a defining sensory signature distinguishing fermented from unfermented probiotic beverages (Gangwar et al., 2018). Ingredient water solubility and dispersibility—themselves pH-dependent—are critical formulation parameters, since poor solubility and dispersibility are directly linked to phase instability during storage, a consideration that stabilizer and emulsifier selection must specifically address.

6.4 Colour and appearance

Colour stability is a key driver of consumer acceptance and is influenced by both native pigment chemistry and fermentation-associated biochemical change. Fermentation of fruit-based probiotic beverages frequently enhances colour stability through antioxidant protection against discoloration (Maia et al., 2023), and beetroot-based probiotic beverages retain their characteristic vibrant colour owing to the fermentation-stable nature of betalain pigments (Janiszewska-Turak et al., 2022). However, colour outcomes are not uniform across matrices: no significant colour change was observed across formulation or microbial treatment in one juice system (de Oliveira et al., 2021), whereas fermentation of Chinese jujube juice produced lighter, less red samples (Li et al., 2021), and fermented pineapple and juçara juices instead trended toward stronger red and yellow coloration, respectively (de Almeida Bianchini Campos et al., 2019). This heterogeneity indicates that colour behaviour under fermentation is strongly substrate-specific and cannot be generalized across fruit matrices without direct empirical testing. Representative colour and viability outcomes reported for specific matrix-microorganism combinations are summarized in Table 4.

TABLE 4. Representative physicochemical and viability outcomes reported for specific matrix–microorganism combinations in probiotic beverages.

6.5 Flavour and aroma

Flavour stability throughout storage is essential to consumer acceptance, and bioactive or functional ingredients incorporated into beverage formulations can occasionally introduce unpalatable off-notes that require masking agents to neutralize (Mudgil and Barak, 2019). Aroma formation is strain-specific and reflects the underlying metabolic diversity of the fermenting culture; single-species versus mixed-species starter systems consequently play distinct roles in substrate transformation, generating measurably different aroma and flavour profiles even from an identical base matrix (Cui et al., 2019).

6.6 Probiotic viability, encapsulation and shelf-life

Ultimately, the health relevance of a probiotic beverage depends on delivering a sufficient viable-cell dose—conventionally cited as a minimum of 10⁶–10⁷ CFU/mL—through gastric transit to the intestinal site of action, and viability loss during processing, storage, and digestion remains one of the most consistently reported challenges across the literature (Iravani et al., 2015; Tripathi and Giri, 2014). Cell counts characteristically rise before declining over storage, a pattern attributed to residual oxygen or trace nitrogenous compounds; declining pH increases the metabolic energy cost of maintaining intracellular pH homeostasis in lactic acid bacteria, depleting ATP reserves and ultimately triggering cell wall disintegration and death, while the characteristic absence of catalase and a functional electron transport chain in most probiotic species renders them intrinsically vulnerable to oxidative damage as dissolved oxygen accumulates. Beyond production and storage, gastric acidity, bile salts, intestinal lysozyme, digestion-derived phenolic toxins, bacteriophages, antibiotics, and anaerobic transit conditions further constrain probiotic survival in vivo (Bhat et al., 2015).

Multiple complementary strategies have been reported to mitigate viability loss in fruit-juice-based systems specifically, including prebiotic fortification (dietary fibre, cellulose, or other protective ingredients), refrigerated storage under carbon-dioxide-enriched atmospheres, antioxidant addition, and microencapsulation (Aspri et al., 2020; da Costa et al., 2017;Tripathi and Giri, 2014) (Figure 2). Microencapsulation in particular has emerged as the most extensively studied viability-protection strategy, shielding probiotic cells from low pH, oxygen exposure, and thermal stress during both processing and gastrointestinal transit through a protective coating matrix that must nonetheless permit bidirectional nutrient exchange. Alginate, carrageenan, whey protein, gelatin, chitosan, cellulose acetate phthalate, and locust bean gum are among the most commonly used encapsulation materials, with alginate particularly favoured for its thermal stability (60 °C–80 °C) and capacity to maintain cell viability under acidic conditions; extrusion, emulsion, spray drying, and freeze drying represent the principal encapsulation production methods (Martín et al., 2015).

Empirical comparisons of free versus encapsulated probiotics in low-pH fruit juices illustrate both the promise and matrix-dependence of encapsulation: alginate-microencapsulated L. casei (1:1 ratio of 5 g/L lyophilized culture at 10¹⁰ CFU/g to 2% sodium alginate, produced via a 1000 Hz encapsulator at 20 mL/min flow with a 450 μm nozzle, yielding 500–700 μm microcapsules) maintained 100% viability in pineapple juice despite partial capsule breakage and over 91% viability in orange juice across 28 days of refrigerated storage at 4 °C, yet viability declined rapidly in raspberry juice owing to anthocyanin absorption within the microcapsules rather than low pH per se (Olivares et al., 2019). Prebiotic fortification offers a complementary, non-encapsulation route: inulin supplementation of mixed carrot–orange juice containing L. plantarum CECT 220 sustained higher viability across 15 days of storage relative to unsupplemented controls, acting both as a carbon source and as a physical protective agent through aggregate formation in the aqueous matrix (Valero-Cases and Frutos, 2017). Storage temperature control is similarly critical for non-fermented probiotic beverages: in a banana–strawberry–juçara blend, Bifidobacterium animalis subsp. lactis viability fell substantially within 30 days, L. acidophilus declined by nearly 2 log CFU/mL over 60 days, L. casei maintained stable counts for 60 days, and L. plantarum sustained viability for up to 75 days with sensory properties comparable to the non-probiotic control at 90 days and less than a 2 log CFU/mL reduction under simulated gastrointestinal conditions after 30 days of storage (de Oliveira Ribeiro et al., 2020). Collectively, these studies demonstrate that no single viability-protection strategy performs uniformly across matrices; strain–matrix–protectant compatibility must be empirically established rather than assumed from single-system data, a point returned to in Section 8.

7 Health implications, regulatory status and safety considerations

Regular consumption of beneficial live bacteria is understood to support gut microbial balance and, by extension, broader host health. Most commercial probiotic formulations draw on lactic acid bacteria genera—Lactobacillus, Bifidobacterium, and Streptococcus—associated with reduced inflammation, alleviation of gastrointestinal discomfort and diarrhoea, and enhanced immune function. A minimum viable-cell threshold of 10⁶–10⁷ CFU/mL is conventionally required for a probiotic beverage to exert a measurable physiological effect, reflecting the practical dose-dependence of probiotic benefit. Beyond gastrointestinal outcomes narrowly defined, probiotic consumption has been investigated across a widening range of conditions, including bacterial vaginosis, depression, obesity, allergic disease, and cancer risk modulation (Zucko et al., 2020). Mechanistically, probiotics are understood to support intestinal homeostasis by inhibiting or displacing harmful microbial populations and by triggering host immune recognition that reinforces a protective mucosal barrier against pathogen colonization (Chandrasekaran et al., 2024).

The regulatory foundation for this field remains the FAO/WHO expert consultation convened in Córdoba, Argentina, in October 2001, which established the working definition of probiotics and emphasized the necessity of clear regulatory guidelines, systematic evaluation methodologies, and pre-market safety assessment, recommending a combination of in vitro characterization and animal safety studies to support human application decisions (Koirala and Anal, 2021). This framework continues to underpin strain safety and efficacy assessment internationally, although its 2001-era guidance predates much of the matrix-specific formulation science reviewed in Sections 5 and 6, a temporal gap that is itself relevant to the research agenda proposed below.

8 Critical appraisal, research gaps and future directions

Synthesizing the evidence reviewed across Sections 3–7 surfaces several recurring limitations that constrain both scientific comparability and industrial translation within the probiotic beverage field, and which together constitute the principal novelty contribution of this review relative to prior matrix-specific or mechanism-specific summaries.

First, viability and shelf-life outcomes are reported using inconsistent endpoints (log CFU/mL reduction, percentage viability, or survival duration to an undefined threshold) across studies, and storage conditions (temperature, packaging atmosphere, and duration) are rarely standardized even within a single matrix category, making direct cross-study or cross-matrix comparison difficult; a standardized minimum reporting set (baseline and endpoint CFU/mL, storage temperature and duration, and packaging/atmosphere conditions) would substantially improve the comparability of future formulation studies.

Second, strain-matrix compatibility is empirically established on a largely case-by-case basis rather than through a systematic, predictive screening framework: the same strain (for example, L. casei) can behave very differently across pineapple, orange, and raspberry juice systems (Olivares et al., 2019), yet no consolidated matrix-compatibility screening protocol currently exists to guide strain selection prospectively for a new substrate, representing a clear methodological gap.

Third, herbal probiotic beverages remain markedly under-characterized relative to dairy and fruit/vegetable systems, with fewer standardized fermentation parameters, safety data, and viability studies reported; given the dual prebiotic–bioactive potential of many botanical substrates, this represents a comparatively high-opportunity, low-evidence area for future formulation research.

Fourth, most viability and functionality data derive from in vitro or short-duration storage studies, with comparatively limited in vivo or clinical validation of health outcomes specific to the beverage delivery format (as opposed to probiotic supplementation generally); closing this gap would strengthen the evidentiary basis for health claims attached to probiotic beverage products specifically.

Fifth, encapsulation research has focused heavily on viability outcomes with comparatively less attention to consumer-facing constraints, including clean-label formulation expectations, sensory impact of encapsulation materials, and cost-effective scale-up of microencapsulation processes for beverage-format products; integrating food-engineering, sensory-science, and consumer-acceptance research within encapsulation studies is an important direction for translating laboratory viability gains into commercially viable products.

Finally, the regulatory framework underpinning strain safety assessment remains anchored to the 2001 FAO/WHO consultation, predating the matrix diversification (particularly non-dairy and herbal systems) documented in this review; periodic re-evaluation of regulatory guidance to reflect the expanded substrate and processing landscape would benefit both researchers and manufacturers navigating novel formulations.

9 Conclusion

Probiotic-based functional beverages constitute a rapidly diversifying segment of the functional food industry, extending from their traditional dairy foundation into fruit, vegetable, cereal, legume, and herbal matrices in response to consumer demand for lactose-free, allergen-conscious, and plant-based alternatives. Across this diversity, successful commercial formulation depends on the coordinated management of strain selection and safety screening, matrix-specific fermentation or fortification strategy, and the physicochemical determinants of quality—acidity, viscosity, total soluble solids, colour, and flavour—that govern both consumer acceptance and probiotic viability throughout processing, storage, and gastrointestinal transit. Microencapsulation and prebiotic fortification have emerged as the most effective viability-protection strategies identified to date, though their performance is consistently shown to be matrix- and strain-specific rather than universally transferable. While the health-promoting potential of probiotic beverages remains substantial, this review’s critical synthesis indicates that the field’s advancement is now constrained less by a shortage of individual formulation studies than by the absence of standardized viability reporting, systematic strain-matrix compatibility screening, and matrix-specific in vivo validation. Addressing these gaps—alongside expanded characterization of the comparatively under-studied herbal beverage category and periodic modernization of the underlying regulatory framework—represents the clearest path toward evidence-based, reproducible, and consumer-aligned development of next-generation probiotic functional beverages.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Soubhagya Tripathy, Indian Institute of Technology Kharagpur, India

Reviewed by: Kamila Goderska, Poznan University of Life Sciences, Poland

Dulce Del Carmen Velásquez Reyes, Universidad Veracruzana, Mexico