There are 13 vitamins that must be obtained exogenously due to the inability of humans to synthesize them; thereby they are essential nutrients in the human diet, and although in small amounts, a daily requirement is necessary to prevent deficiencies. Although most vitamins are present in a variety of foods, human vitamin deficiencies still occur in many countries, mainly because of malnutrition, not only as a result of insufficient food intake but also because of unbalanced diets (LeBlanc et al., 2011).

Although milk contains many vitamins fermentation by LAB and bifidobacteria constitute an effective way to increase vitamin levels in milk (Laiño et al., 2013). Some bacterial strains included in the genera Lactobacillus and Bifidobacterium can provide an additional source of B vitamins (thiamine, riboflavin, cobalamin, folate, and biotin) during dairy fermentation. Deficiencies in riboflavin (vitamin B₂) or thiamine (vitamin B₁) can lead to both liver and skin disorders and alterations in brain glucose metabolism, respectively (Russo et al., 2014). In this regard, L. casei KNE-1 has been shown to produce thiamine and riboflavin in fermented milk drinks (Drywien et al., 2015). B. infantis CCRC14633 and B. longum B6 strains have been reported to produce riboflavin and thiamine during soymilk fermentation (Tamime, 2006). It was recently indicated that soymilk fermented by the riboflavin-producing strain L. plantarum CRL2130 was able to prevent ariboflavinosis and experimental colitis in a murine model (Juarez del Valle et al., 2016; Levit et al., 2016). Some propionibacteria can produce cobalamin, folic acid, and biotin (Hugenholtz et al., 2002).

Folate (vitamin B₉) is involved in several vital processes and its deficiency is generally linked to neural tube defects, certain forms of cancer, poor cognitive performance and coronary heart diseases. Even though vitamins are widely present in foods, the prevalence of folate deficiency -especially among women of child bearing age- is a growing concern and thereby folate fortification programs have been implemented in various countries (Divya and Nampoothiri, 2015). Rather than incorporating synthetic folate, foods can be naturally fortified with folate synthesized by LAB and bifidobacteria during manufacture of fermented foods (Lin and Young, 2000; Saubade et al., 2016). The strains Streptococcus thermophilus CRL803/CRL415 and L. bulgaricus CRL871 were reported to be suitable for the elaboration of yogurt naturally bio-enriched in this vitamin (Laiño et al., 2013). High folate concentration (up to 150 μg/l) can be reached in yogurt as a result of the ability of S. thermophilus to produce this vitamin (Hugenholtz et al., 2002). Among bifidobacteria, B. catenulatum ATCC 27539 was shown to produce high levels of folate in vitro (D’Aimmo et al., 2012), and B. lactis CSCC5127, B. infantis CSCC5187, and B. breve CSCC5181 strains increased folate concentration during fermentation of reconstituted skim milk (Crittenden et al., 2003). Similarly, L. amylovorus CRL887 can be used for natural folate bio-enrichment of fermented milk (Laiño et al., 2014).

The deficiency of cobalamin (vitamin B₁₂) can be common, particularly in vegetarians who avoid ingestion of animal protein and use soymilk as an alternative to dairy milk (Gu et al., 2015). Animals, plants and fungi are incapable of producing this vitamin, and hence, it is exclusively produced by microorganisms (LeBlanc et al., 2011). It has been demonstrated that cobalamin can be synthesized by some bacteria such as L. reuteri ZJ03, Propionibacterium freudenreichii, B. animalis Bb12 in soy-yogurt, kefir and fermented milk, respectively (Van Wyk et al., 2011; Patel et al., 2013, Gu et al., 2015; Moslemi et al., 2016). Microorganisms can biosynthesize two different isoforms, the vitamin and the pseudovitamin. For example, in a recent work, the production of vitamin and pseudovitamin B₁₂ by P. freudenreichii was quantified specifically and shows that at the initial phase of the fermentation both isoforms are biosynthesized at similar levels; however, by the end of the fermentation the pseudovitamin is not detected, most likely because it is converted to the vitamin form (Deptula et al., 2017). It seems crucial to differentiate between the two isoforms of this vitamin, as the transporter protein in the human GIT has very low affinity for the pseudovitamin, making it un-available to humans (Varmanen et al., 2016).

Biotin (vitamin B₇) deficiency can be caused by inadequate dietary intake or some inborn genetic disorders that affect its metabolism. Subclinical deficiency can cause mild symptoms, such as hair thinning or skin rash typically on the face. Biotin can be synthesized by some LAB in dairy products, e.g., L. helveticus MTCC 5463 increased biotin content in fermented milks (Patel et al., 2013). Some propionibacteria can also produce biotin (Hugenholtz et al., 2002).

Vitamin K is an important promoter of bone and cardiovascular health. It has been associated with the inhibition of arterial calcification and stiffening, and the reduction of vascular risk. This vitamin is nearly non-existent in junk food, with little being consumed even in a healthy Western diet (Maresz, 2015). Its deficiency has been implicated in several clinical ailments such as intracranial hemorrhage in newborn infants and possible bone fracture resulting from osteoporosis (LeBlanc et al., 2011). Vitamin K occurs in two forms: firstly, phylloquinone (vitamin K₁), which is present in green plants, and secondly, menaquinone (vitamin K₂), which is produced by some intestinal bacteria (LeBlanc et al., 2011). Menaquinone can be biosynthesized by some LAB species (mainly Lactococcus lactis) commonly used in industrial fermentation of cheese, buttermilk, sour cream, cottage cheese, and kefir (Walther et al., 2013). Other LAB have been screened for the ability to produce menaquinone; these included strains from the genera Lactococcus, Bifidobacterium, Leuconostoc, and Streptococcus (Morishita et al., 1999). Although the MK forms are ubiquitous in bacteria, it should be noted that some genera such as Lactobacillus have lost the functional ability to produce them (Lechardeur et al., 2011; Walther et al., 2013).

Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter of the central nervous system (CNS). Several important physiological functions of GABA have been characterized, such as neurotransmission, induction of hypotension, diuretic effects, antidiabetic, relaxing and tranquilizer effects (Inoue et al., 2003; Marques et al., 2016). In fact, some GABA_A-receptor agonist drugs (e.g., benzodiazepines) are important pharmacological agents used for clinical treatment of anxiety (Foster and Kemp, 2006).

Gamma-aminobutyric acid is biosynthesized through α-decarboxylation of glutamate, an enzymatic conversion which is catalyzed by glutamate decarboxylase (GAD) (Tajabadi et al., 2015). Several food-grade LAB have been reported to exhibit GABA-producing ability. Among them, most of the GABA-producing strains are lactobacilli (L. brevis, L. paracasei, L. delbrueckii, L. buchneri, L. plantarum, L. helveticus), Streptococcus thermophilus, and Lactococcus lactis (Li and Cao, 2010; Dhakal et al., 2012). Some, Bifidobacterium spp. were also reported to produce GABA, although with lower capacity than LAB (Park et al., 2005; Barrett et al., 2012).

Some fermented dairy products enriched in GABA using GABA-producing LAB as starters have been developed. The strains L. casei Shirota, S. salivarius fmb5 and L. plantarum NDC75017 were utilized to ferment and enrich milk in GABA (Inoue et al., 2003; Shan et al., 2015; Chen et al., 2016). More recently, yogurt enriched with 2 mg GABA/ml was produced using the strain S. thermophilus APC151 (Linares et al., 2016a, 2017). Also, fermented soya milk (using L. brevis OPY-1 as source of GABA) (Park and Oh, 2007), or cheese (Lactococcus lactis as source of GABA) (Nomura et al., 1998; Pouliot-Mathieu et al., 2013) have been produced. Thus, GABA has potential as a health-promoting bioactive component in foods (Li and Cao, 2010).

During milk fermentation, LAB, making use of their proteolytic system can transform milk proteins into biologically active peptides. These peptides can exert a wide range of effects, such as antimicrobial, antihypertensive, antithrombotic, immunomodulatory, and antioxidative (LeBlanc et al., 2002; Nongonierma and FitzGerald, 2015). The most studied mechanism of bioactive peptides is the antihypertensive action displayed by the inhibition of the angiotensin-I-converting enzyme (ACE; peptidyldipeptide hydrolase, EC 3.4.15.1) which regulates blood pressure (Fernandez et al., 2015). ACE inhibitory peptides have been isolated from a variety of fermented dairy products including cheese, fermented milks and yogurt (Fitzgerald and Murray, 2006; Pritchard et al., 2010). The best known ACE-inhibitory biopeptides, Val-Pro-Pro (VPP) and Ile-Pro-Pro (IPP), have been identified in milk fermented by L. helveticus (Slattery et al., 2010). In addition, other dairy starter cultures industrially used in the manufacture of fermented dairy products (e.g., L. helveticus, L. delbrueckii ssp. bulgaricus, L. plantarum, L. rhamnosus, L. acidophilus, Lactococcus Lactis, or S. thermophilus) can generate bioactive peptides (Hajirostamloo, 2010; Hafeez et al., 2014). Other ACE-inhibitory peptides such as β-casein f(72-81), Ser-Lys-Val-Tyr-Pro-Phe-Pro-Gly-Pro-Ile (SLVYPFPGPI) have been produced by L. delbrueckii ssp. bulgaricus LB340 in fermented milk (Qian et al., 2011).

On an industrial scale, two fermented milk products with antihypertensive claims, Calpis^TM and Evolus^®, have been tested extensively in rats and in clinical trials, and are commercialized as functional foods (Dziuba and Dziuba, 2014). Evolus^® is available in the market as a L. helveticus fermented milk -produced in Finland- proven to decrease the systolic blood pressure in hypertensive subjects due to the actions of L. helveticus bioactive peptides (EFSA, 2008). Calpis^TM is defined as a milk product marketed in Japan (Calpis Co. Ltd.) with antihypertensive properties. It is prepared by fermenting skimmed milk with L. helveticus and Saccharomyces cerevisiae, which produce VPP and IPP peptides from β-casein and κ-casein (Dziuba and Dziuba, 2014).

Bacteriocins are ribosomally synthesized antimicrobial peptides produced by a particular bacterium that are active against other competitor bacteria; thereby they constitute an important part of the microbial defense system (Nes et al., 2007). Such bacteriocin-producing strains may offer potential as an alternative to antibiotics, and may be useful as a means of controlling pathogen carriage, therefore being highly suitable as microbial food additives (Cotter et al., 2013) (Table 2). Bacteriocins from LAB have attracted much interest because they are frequently produced by commercially useful strains that are generally regarded as safe (GRAS) for human consumption (Nes et al., 2007). These antimicrobial molecules are among the beneficial peptides intrinsically synthesized by some LAB during milk fermentation and they have been traditionally used as naturally produced food biopreservatives. In addition, they may function in the GIT as potential natural biotherapeutic agents facilitating the competition of probiotic strains and/or inhibition of pathogens; thereby they are potential contributors to the microbiota balance and human health (Dobson et al., 2012).

Nisin is the most well-known bacteriocin used as food preservative due to its antibacterial effect against Listeria, clostridia spores and LAB associated to spoilage. Nisin has been approved as a food additive (E234) in the European Union according to Directive 95/2/EC (EC, 1995) in the following products: semolina and tapioca puddings (3 mg/kg); ripened and processed cheese (12.5 mg/kg), clotted cream (10 mg/kg), and Mascarpone cheese (10 mg/kg). It is also permitted in over 40 countries world-wide including USA, Australia, South Africa, Russia, and India for use as an antimicrobial agent in a variety of food products (EFSA, 2006). Nisin-containing Camembert and semihard cheeses with prolonged shelf-life were made using Lactococcus lactis (strains CNRZ150 or TAB50, respectively) as nisin producers (Arques et al., 2015). Apart from nisin, plantaricins are very well-known bacteriocins. For example, plantaricin C is a broad spectrum bacteriocin produced by L. plantarum, isolated from ripening cheese (Gonzalez et al., 1994). Plantaricins have been reported to produce an immunomodulatory effect on dendritic cells (Meijerink et al., 2010). However, bacteriocins other than nisin have so far only few and limited authorized uses in foods (Yang et al., 2015). Consequently, the use of bacteriocin-producing bacteria as starter culture for in situ biosynthesis during milk fermentation becomes an effective alternative strategy to incorporate bacteriocins in dairy foods. Similarly, the lacticin 3147-producing strain Lactococcus lactis DPC3147 used as a protective culture in Cheddar cheese reduced numbers of Listeria monocytogenes to <10 cfu/g within 5 days at 4°C (Ross et al., 1999; Chen and Hoover, 2006). Other bacterial species such as L. acidophilus can be added as an adjunct in many food fermentation processes to contribute to bacteriocin production and food preservation (Anjum et al., 2014). Other LAB strains such as L. plantarum WHE92 used as adjunct to the starter culture reduced Listeria monocytogenes, Listeria innocua, and Escherichia coli O157:H7 counts in cheese as a consequence of the production of plantaricin (Arques et al., 2015). Using a similar concept, Danisco developed a freeze-dried culture of Pediococcus acidilactici (marketed as CHOOZIT Flav 43) for use as a bacteriocin-producer adjunct in Cheddar and semihard cheeses (Mills et al., 2011).

Studies of the direct impact of dairy foods containing bacteriocins on human health and microbiome are still limited. In vivo antimicrobial activity of nisin and lacticin 3147 has been recently demonstrated in a murine infection model. A nisin-producing Lactococcus lactis CHCC5826 modified the microbiota composition of human microbiota-associated rats increasing bifidobacteria levels and decreasing Enterococcus/Streptococcus populations. Lacticin 3147 has the potential to be employed in the treatment of Clostridium difficile diarrhea and to eliminate the pathogen when added to an anaerobic fecal fermentation (Arques et al., 2015).

Lactic acid bacteria associated to dairy fermentations possess enzymes which can be produced in situ during fermentation of dairy foods and have bioactive potential on the consumer. Examples are hydrolytic enzymes that may exert potential synergistic effects on digestion and alleviate symptoms of intestinal malabsorption (Patel et al., 2013). A well-known example is the β-galactosidase activity, which can achieve lactose degradation and thereby improve health and reduce symptoms of lactose intolerant consumers. Yogurt and other conventional starter cultures and probiotic bacteria in fermented and unfermented milk products improve lactose digestion and alleviate symptoms of intolerance in lactose malabsorbers. These beneficial effects are due to microbial β-galactosidase (de Vrese et al., 2001).

Conjugated linoleic acid (CLA) is a polyunsaturated fatty acid (PUFA) that can be biosynthesized by LAB and bifidobacteria through bioconversion of linoleic acid (LA; cis-9,cis-12 C18:2). The two isomers that have been shown to have bioactive potential are cis-9,trans-11 (c9,t11) and trans-10,cis-12 (t10,c12). The health-promoting properties of CLA include anticarcinogenic, antiatherogenic, anti-inflammatory, and antidiabetic activity, as well as the ability to reduce body fat (Sosa-Castañeda et al., 2015). Although it is a native component of milk, the amount consumed in foods is far from that required in order to obtain desired beneficial effects. Thus, increasing the CLA content in dairy foods through milk fermentation with specific LAB offers a promising alternative. An effective way to increase CLA uptake in humans is to increase CLA levels in dairy products by using strains with high production potential (Lee et al., 2007). A number of food-grade LAB and bifidobacteria were reported to produce CLA in milk products (Sosa-Castañeda et al., 2015; Yang et al., 2015), as is the case of Lactococcus lactis LMG, L. rhamnosus C14, L. casei CRL431, L. acidophilus Lac1, L. plantarum-2, B. bifidum CRL1399 and B. animalis Bb12 (Van Nieuwenhove et al., 2007a; Florence et al., 2009). Some of these strains were also used as adjunct cultures for the manufacture of high CLA-content buffalo cheese (Van Nieuwenhove et al., 2007b). The CLA-producing starter culture of Lactococcus lactis CI4b enhanced levels of total CLA in Cheddar cheese (Mohan et al., 2013). Similarly, L. bulgaricus LB430 and S. thermophilus TA040 are suitable for production of CLA-enriched yogurt (Florence et al., 2009).

In addition, it has been shown that specific microorganisms such as L. plantarum PL60 or B. breve NCIMB702258, are able to produce CLA following dietary administration in animal models (Wall et al., 2009, 2012) and following the administration as a freeze-dried product in humans (Lee and Lee, 2009). Thus, intestinal CLA production by bacteria may contribute to enhance CLA supply in addition to the CLA provided by the producing strain in fermented milks during the manufacture (Teran et al., 2015).

Exopolysaccharides (EPS) are complex extracellular carbohydrate polymers that can be produced by some LAB in situ during dairy fermentations. Some of them promote selective growth of bifidobacteria, thus playing a role on the host microbiota and immune system (Fernandez et al., 2015; Salazar et al., 2016). In this regard, EPS derived from yogurt fermented with L. delbrueckii ssp. bulgaricus OLL1073R-1 exerted immune-stimulatory effects in mice (Makino et al., 2016). Yogurt, Swiss-type, and Cheddar cheeses represent suitable food matrices for the delivery of the hypocholesterolemic EPS-producer strain L. mucosae DPC 6426 (Ryan et al., 2015). Other microorganisms with potential to produce EPS in cheese are P. freudenreichii KG15/KG6, Lactococcus lactis SMQ-461 or S. thermophilus MR-1C (Dabour et al., 2005; Darilmaz and Gumustekin, 2012). Significant levels of EPS can also be produced in kefir by L. plantarum YW11 (Wang et al., 2015). Recently, EPS produced by bifidobacteria have attracted the attention due to their immune modulation capability (Hidalgo-Cantabrana et al., 2012).

Exopolysaccharides can also improve the quality, sensory and rheological properties of the final food product, which in many cases results in a reduction of the amount of chemical stabilizers required, thus favoring a more natural product. For example, strains of B. longum subsp. infantis CCUG 52486 and S. thermophilus were suitable to produce yogurt and fermented ice-cream with improved viscosity and texture and reduced syneresis as a consequence of their high EPS production (Prasannaa et al., 2013; Han et al., 2016; Dertli et al., 2016).