Introduction

Probiotics have received a wide scientific and commercial interest. In addition to increasing demand of consumers for organic and naturally produced livestock, the main reason was a widespread use in animal feed of antibiotics as prophylactic and growth-promoting compounds which led to the development of antibiotic-resistant pathogens, failure of antibiotic treatments, decreased immune function, production losses, and the risk of zoonotic infections because of antibiotic accumulation in animal products and transference resistance genes from animal to human microbiota [1]. Thus, Gilchrist et al. [2] indicated that in 1989–1990 no ciprofloxacin-resistant Campylobacter strains were detected whereas, in 2001, 19% of human isolates were resistant to the antibiotic. Consequently, the use of antibiotics as growth promoters was prohibited within the European Union and in many developed countries. It is worth noting that the antibiotic withdrawal had various negative consequences. In particular, the swine and broiler morbidity and mortality have increased and overall animal productivity has decreased [3]. Therefore, an urgent need emerged to find alternative means and strategies that can, like antibiotics, effectively control pathogens and modulate the gut microbiota playing a critical role in maintaining host health [4]. A great number of publications prove that the application of probiotic preparations, i.e., cultures of the microorganisms, capable to create a natural protective barrier between animals and causative agents of infectious diseases became the most real natural alternative to traditional antibiotic therapies [5,6,7,8].

The World Health Organization defines probiotics as “live microorganisms which when administered in adequate amounts confer a health benefit on the host” [9]. Moreover, probiotic organisms used in food must be safe and effective, able of surviving passage through the gut and to proliferate and colonize the digestive tract. Probiotics provide a variety of beneficial effects on the host by regulation of intestinal microbial homeostasis, stabilization of the gastrointestinal barrier function, expression of bacteriocins, enzymatic activity inducing absorption and nutrition, immunomodulatory effects, inhibition of procarcinogenic enzymes and interference with the ability of pathogens to colonize and infect the mucosa [10,11,12].

Probiotics become quite common as therapeutic, prophylactic and growth supplements in animal production and human health. The worldwide probiotic market was valued at 35.9 billion dollars in 2016 and expected to account for 52 billion by 2020 while in the animal feed market is expected to attain 4.71 billion dollars by 2021 [13]. Traditionally, strains of Lactobacillus and Bifidobacterium are used as probiotic products; nevertheless, spore-forming bacteria of the genus Bacillus and related genera have also been studied and commercialized as probiotics in livestock applications, especially in the poultry industry [14,15,16]. It is well known that one of the requirements for probiotics is that the microorganisms should be viable at the end of the product’s shelf life. Therefore, the use of spore-forming bacteria has been proposed perspective.

Spore probiotics are manufactured and used extensively in humans as dietary supplements (for example, Bactisubtil®, France; Nature’s First Food, USA), in various animals as growth promoters and competitive exclusion agents (for example, BioGrow®, UK; Toyocerin®, Japan) and in aquaculture for enhancing the growth and disease-resistance (for example, Biostart®, USA; Promarine®, Belgium). However, analysis of literature data shows that only few Bacillus spp. have been extensively studied thus far and mass production is one of the important aspects of the commercial development of a biocontrol product. Little is known about the physiological peculiarities of bacilli growth and spore production during lignocellulose fermentation, in particular, in the solid-state conditions. Additionally, information on hydrolytic enzyme production by probiotic bacilli during lignocellulose fermentation is limited, although polysaccharides are typically the main resource for bacterial growth and cellulases play a decisive role in steadily supplying a carbon and energy source to the bacteria. Therefore, development of the low-cost and highly efficient technologies of spore-forming probiotic production through the acquisition of new and deeper fundamental knowledge of the bacilli physiology and sporulation efficiency as well as by means of fermentation of a renewable and inexpensive agro-industrial lignocellulosic biomass became necessary.

Recently, spore-forming probiotics selection criteria, their efficacy, safety, and health benefits, mechanisms of action, an application as alternative biocontrol agents have been comprehensively reviewed [1, 10, 11, 14, 17,18,19,20,21,22]. However, to the best of our knowledge, an exhaustive overview of the basic aspects of the microbiology of Bacillus spp. cultivation and spore production in defined and complex media is still lacking in the literature. Therefore, this review summarizes the recent literature and our own data on the current state of knowledge about the physiology of bacilli growth and spore production in the submerged and solid-state fermentation conditions, focusing on the common characteristics and unique properties of individual bacteria as well as on several approaches providing enhanced spore formation.

Bacillus spp. as Probiotics and Their Application

It is widely accepted that good probiotic microorganism, besides exerting a beneficial effect on the host, should have some essential functional and technologic properties, in particular, it has to be safe for host consumption, not conferring any pathogenic or toxic effects; it should be originated from the intestinal tract of healthy organisms; it should be tolerant of gastric and bile acids as well as sufficiently resistant to digestive enzymes to enable its survival after passing the tract; it should manifest a positive influence on the intestinal flora and a capability of reproduction in the host intestine; and lastly, it could be produced with high yield at industrial scale and should be stable and viable throughout the manufacturing process and storage [6, 23, 24]. Recent studies evidence that the microorganisms’ adaptation to a specific ecosystem could play a significant role in the selection of probiotic candidates and that the probiotic efficacy of selected isolates might depend to some extent on the original host [22]. Thus, comparison of the probiotic characteristics among isolates from dairy products and animal rumen revealed that the latter were more tolerant to bile salts and exhibited higher inhibition against pathogens [25]. Probiotics and competitive exclusion agents are thought to enhance the gut microflora by preventing the colonization of the gastrointestinal tract by pathogenic bacteria [26]. There are four basic ways in which this might be achieved: (i) immune exclusion of pathogenic bacteria, (ii) exclusion of a pathogen by competitive adhesion, (iii) synthesis of antimicrobial substances that impair colonization of the gastrointestinal tract by pathogens, and (iv) depletion of or competition for essential nutrients [27, 28].

The majority of available probiotics are species from genera Lactobacillus and Bifidobacterium, bacteria that can ferment carbohydrates to lactic acid that inhibits the growth of some pathogenic bacteria. Strains of Enterococcus, Streptococcus, Saccharomyces, and Aspergillus are also used either as single probiotic species or in combination with other species [8, 23]. Bacillus spp. have received growing attention because of their industrial potential in the development of probiotics, biopesticides, and biofertilizers. The genus Bacillus includes 77 recognized species of Gram-positive, spore-forming, aerobic or facultative anaerobic, rod-shaped, catalase-positive bacteria. The number of species allocated to this genus increased to 318 in the “List of prokaryotic names with standing in nomenclature” (http://www.bacterio.net/bacillus.html).

The Bacillus spp. bacteria form spores under environmental stress such as nutrient deprivation [29]. Spores are a differentiated cell type consisting of metabolically dormant cells, able to resist chemical and physical stresses such as air-drying, high temperature, high pressure, UV light, and acidity. This resistance is due to the presence of several specific layers and the high dehydration level of the spore core. Bacillus spp. initiate its sporulation process at the end of the exponential-stationary growth phases, when nutrients are exhausted, and the production of heat-resistant spores takes approximately 8 h to be completed. Under appropriate conditions, bacilli cell density and sporulation efficiencies are in the range of 1 × 108 – 1.5 × 1010 spores/mL and 30–80%, respectively [10, 30,31,32,33]. Sporulation of Bacillus spp. can be induced not only by nutrient exhaustion but also by subjecting the cells to some adverse environmental conditions including extremes of pH and temperature [34].

Bacilli are widespread in nature, they are associated with soil, water, dust, air, as well as to vegetables and fermented products, such as natto (Japan), gari (Africa), douchi (China), and other [11]. Moreover, they can be found in the normal intestinal flora of humans and animals and are capable to germinate and re-sporulate in the gastrointestinal tract [10]. Bacillus spp. are characterized by high adaptability to environmental conditions and high growth rate on biomaterial, i.e., they can be incorporated into everyday foods. Spores being heat-stable allow long-term storage of preparations without refrigeration or need for encapsulation. Moreover, the spores are capable of surviving the low pH of the gastric barrier [35] and unlike Lactobacillus spp. the entire dose of ingested bacteria reaches the small intestine intact [4]. Furthermore, the high antagonistic activity of bacilli owing to the secretion of antimicrobials (coagulin, amicoumacin, and subtilisin) may also further provide a probiotic effect by suppressing the growth of competing microbes as well as enteric pathogens. Finally, in their vegetative form, Bacillus spp. (especially, Bacillus subtilis, Bacillus amyloliquefaciens, and Bacillus licheniformis) are well known for their ability to produce extracellular amylases, glucoamylases, proteases, cellulases, xylanases, pectinases, and lipases with high product yields that may enhance digestibility and absorption of nutrients in addition to an overall immune function of the gut [11, 36,37,38,39,40,41]. Different species of Bacillus has also been used for the production of vitamins (e.g., riboflavin, cobalamin, inositol) and carotenoids for the synthesis of several health supplements for human consumption [11, 42].

Of the species that have been most extensively examined are B. subtilis, B. cereus, B. coagulans, B. licheniformis, and Bacillus velezensis [10, 11, 17, 43]. The widely known probiotic product, Japanese Natto contains a biofilm of B. subtilis var. Natto, grown on the surface of soybeans. Natto carries as many as 108 viable spores per gram of product and health benefits have been associated with consumption of Natto including stimulation of the immune system [44]. Bacillus amyloliquefaciens is closely related to B. subtilis bacteria. It has been studied extensively as a producer of industrial enzymes, such as α-amylase, subtilisin, barnase (a ribonuclease), and antibacterial and antifungal peptide antibiotics [45]. In addition, it has been reported that several strains have a potential for the biological control of several plant diseases [46, 47]. Moreover, effective formulations having a suspension (1 × 105 spores/mL) of B. amyloliquefaciens for the control of lily grey mold in the field has been established [46]. All these attributes make B. amyloliquefaciens an ideal candidate for the large-scale production of spores for field applications. Bacillus spp. ability to grow in the form of a biofilm is very important since biofilms generate nutritionally functional products in which spore-forming probiotic microorganisms possess a technological plasticity, strong antagonistic activity to pathogens and high enzymatic activity.

Bacillus spp. possess pathogen exclusion, anti-oxidant, antimicrobial, immuno-modulatory abilities [11]. It has been suggested that the health effects of probiotics are genera, species and strain specific. Thus, a high throughput approach was used for preliminary screening of 245 endospore-forming bacterial strains in order to identify Bacillus spp. isolates for use as additives in the feed industry [48]. The spectrum of characteristics, such as antibiotic resistance, pathogen inhibition, sporulation, production of glycosyl hydrolases and biofilms showed that strains belonging to B. amyloliquefaciens, B. subtilis and B. mojavensis exhibited the best probiotic potential compared to B. licheniformis, B. megaterium and B. pumilus. In observing the antagonistic activity of 117 bacilli isolates, 10 of these isolates exhibited distinct growth inhibition of foodborne pathogens, including Salmonella enteritidis, Salmonella typhimurium, Escherichia coli, Staphylococcus aureus, Listeria monocytogenes, and Vibrio cholera [49]. Antimicrobial activities were found connected to the bacilli ability to produce antimicrobial peptides, small extracellular effector molecules and their ability to interact with a host with the help of adhesion and attachment features [50]. The bacilli properties to secrete antimicrobials such as coagulin, amicoumacin, and subtilisin provide a probiotic effect by suppressing the growth of competing microbes as well as enteric pathogens. Bacteriocin-producing strains of probiotic Bacillus spp. have the potential to be introduced as food bio-preservative and as an antimicrobial in human and animal infections.

Already early studies with animal hosts suggested that bacterial spores are able to germinate in the gastrointestinal tracts of dogs, rabbits, mice, pigs, and chicken [10, 51]. Administration of probiotic strains separately and in combination significantly improved feed intake, feed conversion rate (FCR), daily weight gains and total body weight in chicken, pig, sheep, goat, cattle and equine [12, 18, 52, 53] had a beneficial effect on subsequent milk yields, fat and protein content [54]. Especially, there is a noticeable increase in the use of bacilli-based probiotic formulations in poultry farming [55,56,57,58,59]. Several reports indicate favorable results with broilers, layers, and turkey using various strains of Bacillus [60, 61]. Probiotic supplementation has been shown to improve body weight gain, egg production, feed conversion ratio, some egg characteristics, and mortality in poultry [16, 62, 63]. Recently, Mazanko et al. [59] studied the effect of bacilli probiotic preparations on the physiology of laying hens and roosters. Probiotic formulations were prepared as soybean products fermented by B. subtilis KATMIRA1933 and B. amyloliquefaciens B-1895. The introduction of probiotic bacteria into the diet of birds led to an increase in sperm production, egg production, egg quality, and hatchability. The authors speculated that these qualities resulted from the bacilli-produced proteases, amylases, and cellulases which contributed to the better digestion of the feed.

Physiology of Bacillus spp. Growth and Spore Production

The use of Bacillus species as probiotic formulations is expanding rapidly since the spore-based products can be produced comparatively easily and incorporated into everyday foods/feeds with high stability retaining their viability. It is evident that the wide applications of Bacillus probiotics require their production in huge amounts at a low cost. The production process of probiotics consists of several different steps including fermentation and post-fermentation processing [19]. The spore production with a good sporulation efficiency is a key step in bio-products development. For the fermentation process, appropriate medium ingredients need to be selected and the growth conditions for the probiotic have to be optimized with the right pH, temperature, and oxygen tension [10, 32, 64]. Not only growth conditions but other factors, such as the time of harvesting may also influence the exerted functional properties [65]. Unfortunately, the majority of performed studies did not evaluate to which extent growth conditions of bacilli probiotics affect their functional properties and product efficacy.

To develop an efficient technology for the production of spore-forming bacteria, there is a need to elucidate the physiological mechanisms regulating (either enhancing or inhibiting) bacilli growth and sporulation, and to understand the optimal nutrient requirements for both processes. Obviously, a reasonable strategy to increase probiotic production is to create cultivation conditions providing high cell density and subsequently allow sporulation to occur. However, the analysis of literature data shows that only few Bacillus spp. have been extensively studied so far and a current knowledge on their physiology is still limited in order to effectively realize their biotechnological potential on an industrial scale. Moreover, different optimum culture media and culture conditions for bacilli sporulation have been reported, where each particular strain has its own requirements and optimum conditions.

Effect of Carbon Source

Sporulation media commonly used in laboratories are complex, with common features such as a high level of nutrients with a combination of peptones, yeast extract, casamino acids and minerals such as iron, magnesium, calcium, copper, manganese and zinc. In these media, spore-forming bacteria grow until depletion of the nutrients and then sporulate spontaneously. However, these media usually lead to the heterogeneous and poorly reproducible production of spores [66]. Precise regulation of growth and sporulation parameters through the development of a chemically defined media is of great importance for obtaining reproducible and homogeneous spore batches. Currently, chemically defined synthetic media for growth and sporulation are frequently used but all factors which can influence bacterial biomass and spore accumulation have to be optimized.

Especially, a carbon source and its concentration in the nutrient medium plays a crucial role in the bacilli growth and sporulation process. Thus, Warriner and Waites [67] tested several carbohydrates (the final total carbohydrate concentration was 5.7 mmol/L in all cases) to determine which types of carbon source enhance sporulation in the B. subtilis PS 346 cultivation. As compared with the glucose-containing medium, carbohydrates such as raffinose, sucrose, and fructose gave lower spore yields and appeared to actually suppress sporulation even when used in combination with glucose. The authors established that this phenomenon is not associated with growth rate or bacterial biomass at the end of the exponential phase since it was found that at the end of the 4 d incubation period the optical densities were equal to, or greater than, those cultures supplied with glucose. This suggests that, following growth on substrates such as raffinose, sucrose, and fructose, a significant proportion of cells do not initiate sporogenesis following the onset of stationary phase. In contrast, substitution of glucose for pentose, ribose or arabinose, two-fold increased the relative spore yield in the B. subtilis cultivation. Moreover, usage of gluconate as a carbon source increased five-fold the bacilli relative spore yield. Finally, when the dual substrates of gluconate, ribose or arabinose were supplied with glucose, the 11-, 14-, 19-fold increase in spore yield was obtained. Enhanced sporulation was also observed by supplementing glucose-containing media with pyruvate but not citrate or malate (the TCA intermediates). From the results obtained, the authors assumed that the sporulation enhancement effect caused by supplying dual substrates could be attributed to maintaining, either directly or indirectly, of a high pyruvate pool from which derive metabolites that play a critical role in sporogenesis regulation.

In our previous study [68], the highest yield of spores (2.3 × 109 spores/mL) was attained after 72 h of submerged cultivation of B. subtilis KATMIRA1933 in the 5 g/L glucose-containing medium closely followed by sucrose- and xylose-based media. Glycerol appeared to be the poorest carbon source providing two-fold lower spore yield as compared with glucose. Similarly, glucose at the concentration of 0.5% stimulated both bacterial growth and spore production by B. amyloliquefaciens B-1895, xylose as well served as an appropriate carbon source for spore formation [69]. However, for this strain sucrose appeared to be the poorest carbon source ensuring the only 1.5-fold increase of spore production as compared with the control medium. Apparently, unlike B. subtilis KATMIRA1933, B. amyloliquefaciens B-1895 expressed a low sucrase activity to provide the bacterial culture with a carbon source.

In the evaluation of the effect of a number carbon sources (11 g/L) with different structure and composition, the polysaccharides appeared to be better carbon sources than monosaccharides and disaccharides for the spore production during the submerged cultivation of B. subtilis WHK-Z12 [70]. Cornstarch provided the highest spore yields (2.73 × 109 spores/mL), followed by potato starch (1.95 × 109 spores/mL). This enhancement in yields may be due to cornstarch ability to be slowly hydrolyzed to glucose in liquid cultures, leading to alleviation of catabolite repression on sporulation and growth caused by glucose. Among monosaccharides and disaccharides, d-maltose and d-xylose provided high spore yields (1.60–1.83 × 109 spores/mL), other carbohydrates, including glucose (8.55 × 108 spores/mL) yielded significantly lower spore number. Especially, d-lactose was the poorest carbon source for spore formation by B. subtilis WHK-Z12 (2.65 × 108 spores/mL). Unlike this strain, lactose appeared to be the best carbon source as compared with glucose and maltose for the spore production by B. amyloliquefaciens B128 [71]. Although the cultivations were performed using a comparatively high concentration (20 g/L) of carbon sources bacilli produced 4.28 × 108, 0.88 × 108, and 0.67 × 108 spores/mL at the presence of lactose, glucose, and maltose, respectively, in medium containing soy peptone as the nitrogen source.

Undoubtedly, the carbon source concentration may play a decisive role in the sporogenesis process by individual bacilli. Thus, Monteiro et al. [32] evaluated the effect of glucose on B. subtilis strain MB24 growth and sporulation in 2-L batch bioreactors by varying the initial glucose concentration between 3.5 and 20 g/L. The maximum vegetative cell concentration increased with the increase in glucose concentration up to 5 g/L achieving 5.6 × 109 spores/mL. However, initial glucose concentrations higher than 5 g/L inhibited sporulation. It turned out that no glucose was detected by the end of the exponential growth phase when it initially was added to the medium at 5 g/L, while at higher glucose concentrations there was still glucose consumption during the stationary phase. Later, the same authors [72] tested another B. subtilis strain 210 to determine the most effective medium for spore production and compared their yield with that in the standard complex medium used for spore production (Difco Sporulation Media, DSM). Media consisted of salts with glucose, glycerol, citric acid or a combination of glucose and glycerol. The results indicated that in all the experiments carried out using chemically defined media spore production was lower than that in the experiment using DSM. It is interesting that unlike the earlier experiments with B. subtilis strain MB24, the best results were obtained using media formulation with higher carbon and nitrogen concentration. Notably that an augmentation of glucose concentration from 4 to 20 g/L resulted in the five-fold increase of the B. subtilis strain 210 spore number to 2.4 × 109 spores/mL, but the further increase of glucose concentration inhibited the spore production.

Posada-Uribe et al. [64] tested 15 media by the Plackett and Burman design and only 4 (M1, M4, M11, and M15) produced spores after 96 h of B. subtilis EA-CB0575 cultivation. These media had low glucose content in common (2.0 g/L) and provided spore cell densities 0.51–1.87 × 109 spores/mL and sporulation efficiencies between 50.7–93.2%. It is important to note that the media with high quantities of glucose (11 g/L) did not produce spores. The authors assumed that this phenomenon may be related to sporulation being controlled by catabolite repression, as glucose may be involved in inhibition of induction of several enzymes at least partially responsible for sporulation. In addition, the negative effect of glucose on sporulation suggests that an excess of glucose inhibits sporulation by repressing the transcription of the spo0A gen. This gene is responsible for encoding Spo0A, a response regulator activated by phosphorylation in response to several internal and external stimuli and is the master regulator to entry into sporulation [31].

B. subtilis KATMIRA1933 also appeared to be sensitive to glucose concentration [68]. In fact, the maximum concentration of spores (6 × 109 spores/mL) was achieved using 0.2% glucose, while a further gradual increase in the concentration of the carbon source was accompanied by a three-fold decrease in spore yield even at 0.5% glucose-containing medium. Study of B. amyloliquefaciens B-1895 spore formation in the dependence of glucose concentration showed that even the lowest concentration (0.2%) of carbon source five-fold increased the spore yield as compared with the control medium [69]. The maximum concentration of spores (7.1 × 109) was achieved when the medium contained 0.4% glucose and a further increase of the carbon source concentration caused a decrease in spore yield. It is important to note that in these experiments the complete glucose depletion at the end of the cultivation was observed in all media with the exclusion of 0.6% glucose-containing medium where 0.2–0.3 mg/mL of the sugar was detected after 72 h growth of bacterial cultures.

Table 1 summarizes surprisingly limited available data on spore yields produced by different Bacillus spp. strains in dependence on carbon source. Only a small number of carbon compounds was tested to evaluate their suitability for the bacilli spore formation as compared with that of easily metabolizable glucose. Exactly glucose was the best carbon source (even in the media with starch or lactose) for the Bacillus spp. spore production ensuring accumulation of 1.87–7.1 × 109 spores/mL. The data received indicate that the concentration of glucose (carbon source) in the culture medium should be reduced to increase the sporulation efficiency and spore yield. Moreover, it is evident that depletion of the carbon source is the main stimulus for the sporulation by Bacillus spp.

Table 1 Bacillus spp. spore yields in media with different carbon sources and lignocellulosic substrates
Full size table

Effect of the Lignocellulosic Growth Substrate

The production cost can be reduced by use of agro-industrial wastes/by-products as a substrate for the growth of probiotic microbes. However, in the literature a scarce information describing the effect of lignocellulosic substrates on bacilli spore formation is available. In particular, a combination of tapioca (16.7 g/L) with lactose (12.7 g/L) in a nutrient medium for submerged cultivation of B. amyloliquefaciens B128 resulted in spore yield of 5.92 × 108 spores/mL [71]. Maximum spore production (7 × 108 spores/mL) was achieved in the cultivation of B. subtilis in a statistically optimized medium containing corn flour (1.2%) in combination with 4.54% (NH4)2SO4 [73].

In the recent study, corn meal and soybean meal positively influenced the spore production by B. amyloliquefaciens BS-20 [74]. However, no significant effect was found from wheat bran and molasses. From the verification experiments, the optimized medium (8 g/L glucose, 9.0 g/L corn meal, 9.5 g/L soybean meal, 7.2 g/L beef extract, 1.0 mM Mn2+, 3.0 mM Fe2+ and 2.1 mM Ca2+) gave an 8.8-fold increase in the spore yield compared with the control (8 g/L glucose, 7.2 g/L beef extract, 10 g/L NaCl). In this medium, the bacilli produced as high as 8.05 × 109 spores/mL.

An interesting approach was exploited by Wangka-Orm et al. [75]. These researchers obtained the water extracts from sweet potato, cassava root, rice and sticky rice which were then supplemented to medium with 20 g dextrose/L and used for submerged fermentation by Bacillus KKU02 and Bacillus KKU03 strains. Only cassava root and sweet potato supplementation gave an equal or better spore concentration as compared with the nutrient broth. Supplementation of cooked cassava medium with 0.1 g/L MgSO4 and 2 g/L (NH4)2SO4 resulted in the 1.75-fold increase of spore production in both strains. Nevertheless, the highest concentrations of spores were comparatively low–1.62 × 108 and 6.61 × 107 spores/mL for Bacillus KKU02 and Bacillus KKU03, respectively. Cassava concentrations from 50 to 200 g/L were tested for spore production and the highest yields were obtained at the substrate concentration of 100 g/L (1.78 × 108 and 1.48 × 108 spores/mL, respectively).

A wide range of lignocellulosic materials with different chemical compositions was for the first time tested at a concentration of 40 g/L to evaluate B. subtilis KATMIRA1933 spore production after 3 days of submerged fermentation [68]. Among the tested growth substrates, milled soybean and the sunflower processing by-products resulted in a good growth of bacilli and accumulation of vegetative cells but failed to promote mass sporulation as compared to the control medium. On the contrary, mandarin peels followed by ethanol production residue (EPR) from corn grains provided an especially high yield of spores (5.7 × 1010 and 2.9 × 1010 spores/mL, respectively, 71- and 37-fold higher as compared with that in the control medium without a carbon source). The authors established that the spore number (4 × 109 spores/mL) in the medium containing 10 g mandarin peels/L increased 7- and 10-fold with augmentation of the growth substrate concentration to 30 and 40 g/L, respectively. Further increase in mandarin peel concentration did not favor spore formation. It is worth noting that the authors for the first time showed a suitability of the cheese whey and cottage cheese whey usage instead of distilled water for the preparation of nutrient medium containing mandarin peels, EPR or a mixture of the two growth substrates. The use of these ingredients together in the optimized medium provided a rapid initial growth and a strong synergistic effect on the bacilli cellulase and xylanase activities and spore production increasing their yield to 5.8–7.4 × 1010 spores/mL.

B. amyloliquefaciens B-1895 appeared to be an efficient spore-forming bacterium producing 8.2–10.8 × 109 spores/mL in submerged fermentation of majority tested materials (40 g/L of milled corn cobs, ethanol production residue from wheat grain, wheat bran, sunflower extraction cake, mandarin peels) [69]. Like in the cultivation of B. subtilis KATMIRA1933, the spore yield gradually increased only with an elevation of the growth substrates concentration from 1 to 4%. Comparative analysis of the data received shows that in the submerged fermentation of lignocellulosic materials both bacilli produced higher yields of spores as compared with those in the glucose-containing medium (Table 1). One may assume that these materials contain some components necessary for both bacterial growth and efficient sporulation. Moreover, B. subtilis KATMIRA1933 [68] and B. amyloliquefaciens B-1895 [69] expressed comparatively low endoglucanase activity (0.2–1.0 U/mL) in the fermentation of the tested materials, although it was higher as compared with B. subtilis strain MU S1 [76]. Obviously, in combination with other carbohydrases, this cellulase activity provided a slow but continuous hydrolysis of lignocellulose polysaccharides to metabolizable sugars to ensure the bacterial cultures with a required carbon source. Consequently, only traces of reducing sugars were detected even at the end of submerged fermentation, when the bacterial metabolism and proliferation had significantly declined. These circumstances may lead to the prevention of sporulation inhibition caused by elevated concentrations of sugars. Thus, these results indicate that various lignocellulosic materials may be successfully exploited as growth substrates for the cultivation of spore-forming bacteria.

Effect of Nitrogen Source

Only a few research groups investigated the effect of nitrogen sources on the bacilli growth and spore production. These studies have proved that both the nature and concentration of nitrogen sources are crucial nutritional factors affecting bacilli growth and spore production. The effect of different nitrogen-based nutrients on the spore production by B. subtilis WHK-Z12 was tested by maintaining glucose as the constant variable for carbon nutrition in the bacilli submerged cultivation [70]. Corn steep liquor appeared to be the best nitrogen source for spore production (1.73 × 109 spores/mL), followed by soybean flour and yeast extract. The authors assumed that the corn steep liquor and soybean flour improved spore yields as they are rich in nitrogen nutrients and provide amino acids required for growth (19.18 and 43.83% amino acids, respectively) as well as provide some of the growth factors required for the bacterial growth and sporulation. At the same time, among the complex organic nitrogen sources tested, low spore yields were obtained when soy peptone or meat peptone were used as a single nitrogen source (0.40 and 1.56 × 108 spores/mL, respectively). Moreover, amino acids and inorganic ammonium phosphates were poor nitrogen sources yielding 0.2–0.36 × 109 spores/mL. More importantly, in this study, the researchers utilized a unique combination of nutritional ingredients for B. subtilis WHKZ12 spore production in a 30-L fermenter. Using cornstarch, wheat bran, corn flour, corn steep liquor, soybean flour, and yeast extract at optimal concentrations they achieved maximal spore yield 1.56 × 1010 spores/mL after 40 h fermentation.

Rao et al. [71] compared the effect of several nitrogen sources on the spore production by B. amyloliquefaciens B128 and established that during the bacilli cultivation in the lactose-based medium using peptone or soy powder as a nitrogen source their concomitant supplementation with (NH4)2SO4 two-fold increased the spore yield. Applying the response surface methodology for further optimization, the authors established that the mixture of 1.8 g/L of ammonium sulfate and 8.0 g/L of peptone in the medium containing 12.7 g/L of lactose and 16.7 g/L of tapioca provides the optimum cultivation parameters for the spore production (5.93 × 108 spores/mL) by B. amyloliquefaciens B128.

Monteiro et al. [72] investigated the effect of two mineral salts in a chemically defined medium to determine the most effective medium for the spore production by B. subtilis strain 210 in shake flask cultivations. Varying (NH4)2SO4 concentration from 0.2 to 0.6 g/L, they established that the salt content of 0.4 g/L provides the best sporulation process. However, (NH4)2HPO4 at the same concentration appeared to be more appropriate nitrogen source for spore formation by the bacterial culture. More interesting, simultaneous addition of both compounds at the indicated concentration resulted in the formation of 2.4 × 109 spores/mL after 48 h of B. subtilis cultivation.

It is known that in the plant raw materials, nitrogen is usually available at suboptimal concentrations for the cultivation of microorganisms, making it necessary to include an additional nitrogen source for optimal growth. To provide the bacilli with adequate conditions for abundant growth and sporulation, several inorganic salts and organic compounds were tested as nitrogen sources for the B. subtilis KATMIRA1933 growth and spore production in addition to the nitrogen already available in the mandarin peels [68]. The mandarin peels without an additional nitrogen source represented an excellent growth substrate providing an accumulation of 2.3 × 1010spores/mL. The yield of spores largely varied when bacilli were grown using different nitrogen sources. Among inorganic compounds, KNO3 appeared to be the best nitrogen source, ensuring a two-fold increase in spore number when compared with the control medium. At the same time, supplementation of the control medium with ammonium sulfate sharply inhibited the sporulation process, maybe due to an acidification of the medium. By contrast, all organic nitrogen sources favored spore production in the submerged fermentation of mandarin peels by B. subtilis KATMIRA1933. Among them, peptone ensured accumulation as high as 6.6 × 1010 spores/mL, almost a three-fold increase in the spore count as compared with the control medium. Moreover, the culture needed a comparatively high concentration of peptone for a maximum sporulation because the number of produced spores changed from 2.0 × 1010 to 8.3 × 1010 spores/mL when the nitrogen concentration in the nutrient medium was gradually increased from 0 to 40 mM. The positive effect of peptone may be attributed to the higher production of bacterial biomass and increased sporulation efficiency. However, the further elevation of nitrogen source concentration delayed the development of the bacterial culture and the sporulation process. It is worth noting that the higher was the peptone concentration, the higher was the medium pH at the end of submerged cultivation obviously favoring the spore formation.

Analogically, several inorganic salts and organic compounds at the concentration of 20 mM as nitrogen were tested in the cultivation of B. amyloliquefaciens B-1895 as an additional nitrogen source to available in the corn cobs nitrogen [69]. The results revealed several general features. Firstly, the substrate fermentation in the presence of peptone, potassium nitrate, and ammonium nitrate accompanied with an increase of the medium pH while the presence of physiologically acid ammonium sulfate caused acidification of the medium. Secondly, the corn cobs without additional nitrogen source represent an excellent growth substrate for this strain providing accumulation of 7.2 × 109 spores/mL. Thirdly, the yields of spores were increased by supplementation of the medium with organic nitrogen sources. Among them, casein hydrolysate provided the three-fold increase of spore number as compared with that in the control medium. Inorganic salts, ammonium sulfate, and ammonium nitrate rather inhibited the sporulation process whereas KNO3 favored more than three-fold increase of spore yield. Overall, the data received indicate that an establishment of optimal nitrogen source and its concentration is necessary for each Bacillus species with considering of individual physiological peculiarities and carbon source in the nutrient medium.

Effect of Medium pH

Several authors described the influence of the medium pH on the bacilli growth and sporulation. For example, Monteiro et al. [32] observed that the cultivation of B. subtilis in 2-L bioreactor without control pH accompanied with decrease of pH from 6.7 to 6.5 during the exponential growth phase, then a sharp increase to 8.1 was observed until the end of the exponential growth phase, and a slow increase to pH 9.0 occurred during the sporulation process. In this conditions, the maximum vegetative cell concentration achieved 2.6 × 109 cells/mL but the sporulation efficiency was low (approximately 16%), leading to a final spore concentration of 4.2 × 108 spores/mL. Butch cultivation of B. subtilis at a controlled pH revealed that within the pH range of 6.0–9.0, the sporulation efficiency (approximately 50%) did not depend on the medium pH value, whereas the medium pH 5.0 sharply reduced the sporulation efficiency to 6%. Bacilli cultivation at pH 7.5 increased the maximum vegetative cells concentration up to 7.5 × 109 cells/mL and gave a sporulation efficiency of approximately 50%. This led to a final spore concentration of 3.6 × 109 spores/mL, which corresponds to a nine-fold increase when compared to the batch performed without pH control.

The effect of medium pH on B. subtilis EA-CB00575 spore production was studied by Posada-Uribe et al. [63] at bioreactor level. The authors established that maintaining the medium pH at a constant value (6.5 or 7.0) or fermentation under non-controlled pH did not affect significantly total cell biomass, spore cell density, and sporulation efficiency. Under non-controlled conditions, when pH variation was between 5.5 and 7.0 during 48 h of fermentation, spore cell densities and sporulation efficiencies of 2.27 × 109 spores/mL and 92.9% were achieved, respectively.

Tzeng et al. [47] in the shake-flasks experiments with B. amyloliquefaciens B-128 showed that the initial alkaline pH value of 8.0 favored the spore production which increased to 1.03 × 109/mL compared to 7.96 × 108/mL in control cultivations. In contrast to this, the acidic pH (6.0 and 5.0) values significantly suppressed maximal spore production. It is interesting that with a starting pH of 9.0 and 8.0, after 48 h cultivation the medium pH ended at 7.34 and 7.08, respectively, while no significant reduction of medium pH was noted in the starting pH of 7.0 or lower. Another picture was revealed when B. amyloliquefaciens B-128 was cultivated in a 20-L airlift bioreactor at 1.5 vvm aeration rate. Spore production process was controlled at two different pH operations (7.0 and 8.0) and compared with uncontrolled pH operation. Among the three pH values studied, no significant difference was found in spore yield at the end of cultivations. However, in uncontrolled pH experiments, at the end of the run, the vegetative cells concentration was increased to 8.6 × 109/mL, which corresponded to a six-fold increase when compared to the batch performed with pH controlled cultivations.

Summarizing the literature data and our own observations during the evaluation of carbon and nitrogen sources effects on the spore formation by B. amyloliquefaciens B-1895 and B. subtilis KATMIRA1933, we suppose that the medium pH close to neutral value favors Bacillus spp. growth and maximum accumulation of vegetative cells while the medium acidification delays/suppress the bacilli growth and decreases the sporulation efficiency, while a slightly alkaline pH promotes sporogenesis.

Effect of Agitation and Aeration

Several studies indicate that medium agitation and aeration rates may affect biomass and metabolites production by different Bacillus species [77]. In the batch cultivation of B. megaterium BA-019 in the molasses-based medium at pH 7.0 the highest cell biomass (8.78 g L−1) was attained at 60% DO followed by at 80% DO (8.26 g L−1) and 40% DO (6.38 g L−1) [78]. The researchers underlined that at DO level of 40%, the cultivation time was prolonged and productivity was decreased dramatically due to insufficient oxygen in the culture broth to meet aerobic anabolic demands. However, higher DO levels (80%) reduced the bacterial growth due to oxidative and shear stresses that resulted from the high agitation speeds. The authors concluded that the concentration of dissolved oxygen is important for efficient spore production.

To determine the effect of the agitation and aeration rate on B. subtilis EA-CB0575 spore production under non-controlled pH conditions in a fermenter, Posada-Uribe et al. [64] evaluated operating conditions between 300 and 500 rpm of agitation rate and 8–16 L/min of aeration rate. At these conditions, the spore yield and sporulation efficiency were between 1.24 × 109 and 9.33 × 109 spores/mL and 78.9–96.2%, respectively, which represented a 6.8 increase in spore number with respect to the shake flask conditions. The highest yield of spore (9.33 × 109 spores/mL) was achieved at 400 rpm of agitation and 12 L/min of aeration although the sporulation efficiency was rather lower - 91.2%.

Monteiro et al. [32] investigated the effect of dissolved oxygen concentration on B. subtilis growth and sporulation in a batch reactor with the dissolved oxygen concentration of 10%, 30%, and 50% of air saturation. The researchers demonstrated that this parameter did not influence the microorganism growth, although a slightly higher spore concentration was reached controlling the dissolved oxygen concentration above 30% of air saturation. Study of the effect of agitation on spore production by B. amyloliquefaciens B-128 in the 5-l stirred-tank bioreactor with the 0.5 vvm aeration rate showed that the spore formation was started after 30 h and reached its maximal value of 4.65 × 108/mL in 48 h of inoculation with a relatively mild agitation rate of 200 rpm [47]. From this time onwards there was no substantial increase in spore production. In contrast to this, in an agitation speed of 300 rpm, the spore production was started after 50 h and their concentration reached 3.56 × 108/mL after 100 h. The residual sugars concentration profiles were different for the two levels of agitation tested, where the assimilation of sugars decreased with agitation speeds shifting from 200 to 300 rpm. According to the authors, this could be due to higher shear stress at the increased agitation rate, which may cause a decrease in the biomass of shear-sensitive bacilli and affect the spore yield. The same researchers optimized the aeration rate in 20-L airlift reactor carrying out cultivations in a range of aeration rate from 1.5 up to 3.0 vvm. The aeration rate of 2.5 vvm provided maximal spore concentration of 3.82 × 109/mL after 29 h of inoculation, which was five-fold higher than that at an aeration rate of 1.5 vvm.

It is worth noting that the bacilli growth may take place under anaerobic conditions but sporulation efficiency is highly reduced. Moreover, spore formation depends on the nutrient medium composition. For example, in anaerobiosis, the MOD medium with glucose concentration at 5.4 g/L and without microelements supported the anaerobic growth of the B. cereus strains but no spore formation was observed [79]. The same bacilli manifested the sporulation capacity (average of 5 × 104 spores/mL) in the modified MODS medium containing 1.8 g/L glucose and supplemented with Ca, Mn, Zn, Fe, although the sporulation was lower than that in aerobiosis with an average of 8 × 108 spores/mL.

Effect of Microelements

Microelements such as Ca, Mn, Mg, Fe and Zn in appropriate concentrations are essential for the sporulation process since they are present in the spore layers and enable it to resist high temperatures [30, 47]. As mentioned above, the B. cereus strains had the sporulation capacity in anaerobiosis when the modified MODS medium containing 1.8 g/L glucose was supplemented with Ca, Mn, Zn, Fe [79]. O’Hara and Hageman [80] showed that in chemically defined sporulation medium, B. subtilis cells are strongly dependent on the addition of calcium ions to the medium to achieve maximal sporulation. Monteiro et al. [32] also revealed that cells of B. subtilis strain 210 are strongly dependent on the addition of calcium ion to achieve maximal sporulation. The authors studied the effect of calcium ion concentration within the range of 0.4 to 1.2 g/L and observed that spore production increased with an increase of the calcium concentration up to 0.6 g/L. It is worth noting that the Ca2+ chelate of dipicolinic acid is a major constituent of the dormant spore core, accounting for approximately 10% of the total spore dry weight [81]. Recently, Posada-Uribe et al. [64] established that B. subtilis produced spores in media with and without MnCl2, but those that had this salt at its highest concentration (0.5 g/L) manifested the highest sporulation efficiencies. Finally, of the six metal ions, four ions including Mn2+, Fe2+, Ca2+, and Mg2+ showed significant positive influence on the enhancement of sporulation by B. amyloliquefaciens BS-20 compared with the control, but the inclusion of zinc had a negative effect on sporulation [74]. The optimum concentrations of metal ions were 1.0 mM of Mn2+, 3.0 mM of Fe2+, 2.0 mM of Ca2+, and 3.0 mM of Mg2+; their combined use at optimum concentrations in culture media produced a 3.4-fold increase in spore yields., The authors suggested that different strains had a different response to metal ions in the medium and a thorough screening procedure is important before optimizing the concentration of metal ions.

Effect of the Cultivation Method

Mass production is one of the important aspects of the commercial development of a probiotic product. The cost of probiotic production significantly depends upon the method of fermentation. At present, submerged cultivations are often preferred in industry, as they are shorter and easier to automate. However, several studies have reported that probiotic produced by the solid-substrate fermentation (SSF) is cost-effective and environment-friendly [82, 83]. Among advantages of the SSF over the submerged fermentation process are less elaborate equipment, relatively low investment, ease of handling, higher fermentation productivity and end-product concentration, low waste water output, opportunity to cultivate microorganisms adapted for fermentation of lignocellulosic substrates. Moreover, after the SSF, the products can be lyophilized directly without centrifugation. However, cultivation of microbes using the SSF is related with several technological issues, such as oxygen supply for aerobic metabolism, removal of heat, CO2, and volatile components produced during the metabolism, maintenance of suitable for growth moisture contents [84, 85].

The above-reviewed literature date evidence that the submerged fermentations have been basically used for the production of spore-forming probiotics. The SSF is widely applied for the cultivation using filamentous fungi; however, the results of Gangadharan et al. [86] study proved that a bacterial culture such as B. amyloliquefaciens can be successfully used for the SSF of wheat bran and groundnut oil cake mixture (in a mass ratio of 1:1) with relatively high moisture level (85%) and production of alpha amylase with very high yield during three days of fermentation. Nevertheless, only a few studies exploited the SSF of plant raw materials for the Bacillus probiotic production, although these studies lack a comparative information on the production of Bacillus spp. probiotics under submerged and SSF conditions.

Combination of wheat bran with other lignocellulosic materials for the SSF by B. licheniformis B36 revealed the highest yield of spores (1 × 1011 spores/g) when a mixture of 15 g wheat bran and 5 g rice straw powder was used as a growth substrate [87]. Addition of an extra carbon source, glucose or sucrose, increased spore production by 35% and 25%, respectively, while peptone and yeast extract used as additional nitrogen sources increased the spore yield by16% and 24%, respectively. The optimization of the contents of four nutrients using the orthogonal scheme L9 (34) gave a maximum of 1.7 × 1011 spores/g dry substrate. It is important to note that unlike B. amyloliquefaciens [86] B. licheniformis B36 produced the highest spore yield when an initial moisture of the substrate was 65%.

Ying et al. [88] co-cultivated B. subtilis MA139, Lactobacillus fermentum and S. cerevisae in the unsterilized substrate (feed-grade soybean meal and wheat bran). In this study, S. cerevisae was co-inoculated to consume the oxygen inside the fermenting bag enabling the growth of lactobacillus. The growth of enterobacteria was effectively inhibited, and these strains were not detected in the substrate after 4 days of fermentation. The counts of the starter strains increased rapidly during the first 2 days of fermentation. The maximum number of yeast cells (7.71 log spores/g) was observed after 3–4 days of fermentation and decreased slightly during days 5–10. The number of bacilli reached its maximum (8.68 log spores/g) on day 3. The maximum number of L. fermentum detected in the substrate had a value of 9.04 log spores/g) and their growth reached a stable level starting on day 5. The authors established that owing to the production of antimicrobial substances by B. subtilis MA139, probiotic microorganisms successfully controlled the growth of enterobacteria (since Escherichia coli K88 and Salmonella typhimurium were not detected) and provided a facile and low-cost way to produce solid-state fermentation feed.

In order to develop a multi-microbe probiotic preparation of B. subtilis MA139 and Lactobacillus reuteri G8–5 in the SSF, a wide range of parameters were optimized sequentially in shake flask culture containing feed grade soybean meal, corn flour, and wheat bran with the proportion of 1:1:2 (w/w/w) [89]. The optimized process was as follows: a comparatively low water content, 50%; initial pH, 6.5; inocula volume, 2%; flask dry contents, 30 ∼ 35 g/250 g without sterilization; fermentation time, 2 days. The multi-microbial preparations finally provided the maximum concentration of Lactobacillus of about 9.01 log spores/g and spores of Bacillus of about 10.30 log spores/g. Moreover, the viability of L. reuteri G8–5 significantly enhanced in the presence of B. subtilis MA139, which favored the production of probiotics for animal use.

In our studies, we for the first time compared the spore production during the submerged and SSF of plant raw materials by B. amyloliquefaciens B-1895 [69, 90] and by B. subtilis KATMIRA1933 [68, unpublished work]. The data obtained revealed several general features (Table 2). In particular, a distinctive peculiarity of both bacilli is their capability to utilize various inexpensive lignocellulosic wastes/by-products as growth substrates for the high-yield spore production. Like the submerged fermentation (SF), the SSF is a suitable method for the bacilli cultivation; moreover, in the most media, it favored a significant increase of spore number as compared with that produced during the same time in the SF. However, the tested species to some extent differently utilized the lignocellulosic materials for spore formation in their SF and SSF. For example, as compared with the SSF, the sunflower oil cake appeared to be the worst growth substrate in the SF by B. subtilis KATMIRA1933 sporulation, but it was a preferable source of nutrients in the SF by B. amyloliquefaciens B-1895.

Table 2 Bacillus species spore production (CFU/g) in the SSF of lignocellulosic materials
Full size table

Among the growth substrates, SSF of wheat bran followed by mandarin peels provided especially high yields of B. subtilis KATMIRA1933 spores whereas corn cobs and the ethanol production residue (EPR) from the wheat grains promoted spore formation in the SSF by B. amyloliquefaciens B-1895. In the SF, mandarin peels appeared to be the superior growth substrate for B. subtilis KATMIRA1933 spore production whereas B. amyloliquefaciens B-1895 was capable to efficiently sporulate fermenting majority of tested materials. The observed enhancement in spore yield in presence of these materials suggests that both bacilli possess potent enzymatic systems to deconstruct plant raw materials and provide all the components necessary for an abundant bacterial growth, whereas the materials chemical composition, particles structure, and adhesive properties favored biofilm formation, when nutrients and oxygen transfer processes were sufficient for the promotion of the microbial growth and efficient sporulation.

Scaled up Bacillus spp. Probiotics Production

After optimization of nutrient medium in shake flask experiments, a technical feasibility of large-scale spore production by Bacillus spp. should be confirmed and the optimal cultivation conditions which stimulate the active growth and sporulation of bacilli should be determined. Fermentation scale-up is a final and important step in the probiotic production process development.

Sen et al. [91] performed a 24 full factorial central composite design followed by a multistage Monte Carlo process optimization for Bacillus coagulans RK-02 cultivation and sporulation. They established the optimal process conditions for maximum biomass production: pH–6.65; temperature–38.3 °C; agitation–247 rpm and aeration–1.05 vvm and those for the maximum sporulation: pH–6.27; temperature–41.4 °C; agitation–115 rpm and aeration–0.33 vvm. A two-stage strategy was developed with biomass production in exponential phase under the optimal growth conditions in the first stage followed by the second stage in stationary phase under the optimal conditions for sporulation to obtain a maximum probiotic biomass yield of 4.3 g/L and spore yield of 9 × 1011 spores/g of dry biomass for the formulation of effective nutraceuticals.

It is interesting that the cultivation of B. amyloliquefaciens B128 for the spore production was carried out in 5 L stirred tank bioreactor at a constant temperature of 30 °C and aeration rate of 0.5 vvm but with different agitation speeds [47]. At a relatively mild agitation rate of 200 rpm, the spore production was started after 30 h and reached its maximal value of 4.65 × 108 spores/mL in 48 h of inoculation. When an agitation speed was increased to 300 rpm, the spore production was started only after 50 h and reached to 3.56 × 108 spores/mL in 100 h after inoculation. By contrast, Monteiro et al. [72] cultivated B. subtilis in the 2 L bioreactor using the optimized chemically defined medium and during the exponential growth phase increased the agitation rate from 100 to 1200 rpm to compensate for the oxygen consumption rate. The maximum vegetative cell concentration (1.3 × 1010 cells/mL) was obtained at the end of the exponential growth phase but after that, a high cell lysis was observed and only 48% of the vegetative cells gave rise to heat-resistant spores, the final spore concentration being 6.3 × 109 spores/mL.

We performed B. subtilis KATMIRA1933 cultivation in a 7-L fermenter filled with an optimized medium containing mandarin peels as a growth substrate [68]. The fermenter agitation speed (300 rpm) and the aeration rate (1.0 L/L/min) were constant during the entire period of the fermentation. Application of these conditions prevented the sedimentation of bacterial cells and provided sufficient dispersion of air. Nevertheless, after 24 h, the dissolved oxygen concentration decreased to 5%, while after 3 days of fermentation it increased to 13%. The data collected showed that bacilli multiplication occurred rapidly and the number of vegetative cells increased from 3 × 106 spores/mL to 2.4 × 1010 spores/mL after 24 h of fermentation. However, the spore number observed by this time was low, only 3 × 108 spores/mL. During the second day of fermentation, the vegetative cells and spore number increased to 8.1 × 1010/mL and 9.3 × 109 spores/mL, respectively. In the subsequent cultivation, the B. subtilis KATMIRA1933 cell number increased to 10.2 × 1010 spores/mL after 72 h and then to 10.4 × 1010 spores/mL after 96 h fermentation. By contrast, the spore number increased more rapidly, with a maximum yield of 6.5 × 1010 spores/mL after 96 h. It is interesting that during bacilli cultivation, the reducing sugar content in the culture liquid gradually decreased from 7.7 mg/mL on day 1 to 0.6 mg/mL on day 4 although the bacterial culture expressed comparatively high endoglucanase and xylanase activities with maximum activity (3 and 5 U/mL, respectively) after 3 days of the mandarin peel fermentation.

Analogically, the B. amyloliquefaciens B-1895 cultivation was carried out in the same fermenter. However, the agitation speed was constant (250 rpm) while the aeration rate was 0.5 L/L/min during the first day, then it was increased to 1.0 L/L/min [69]. Under these conditions, the vegetative cells number increased from 2 × 106 spores/mL to 1.4 × 1010 spores/mL during the first day of cultivation, but no spore production was observed. During the second day, the vegetative cells number increased to 3.5 × 1010 spores/mL and after 48 h we detected 4 × 109 spores/mL. Maximum of the B. amyloliquefaciens B-1895 vegetative cells number (3.9 × 1010 spores/mL) was achieved after 72 h while the spore number continued to increase with a maximum yield after 4 days (2.5 × 1010 spores/mL). It is worth noting that the sporulation efficiency increased from 1.1% after 48 h cultivation to 64% after 96 h. Apparently, the sporulation rate of B. amyloliquefaciens B-1895 can be enhanced; therefore, further study should be performed in the fermenter to optimize the sporulation process.

Das et al. [92] selected B. coagulans RK-02 for probiotic production because its unique properties and superiority in pharmaceutical industries: it is a spore-forming lactic-acid-producing bacteria, it can be stored at room temperature and handled easily in different stages of processing and packaging, it can produce a number of industrially important extracellular enzymes. The researchers applied an interesting strategy to achieve maxima in the same batch, by maintaining three different conditions for obtaining maximum biomass, lipase and spore concentration, one after another in the same batch. From the beginning, 36.86 °C temperature, 237 rpm agitation, and 150 L/h aeration were maintained 9 h till glucose is consumed. Oil in the concentration of 0.05% was added initially as an antifoaming agent so that bacteria can adapt to this different medium and more oxygen is available for each cell. Since glucose was found to be consumed within first 9 h of cultivation, therefore slightly before end-log phase (i.e., after 6 h) rest 9 ml oil was added, when glucose concentration was just below 6 g/L. Starting from 9th h to 22nd h 35.2 °C temperature, 213 rpm agitation, and 220 L/h aeration were maintained for maximum lipase production. Then, 40.85 °C temperature, 158.4 rpm agitation, 106 L/h aeration were maintained till the end of the batch for maximizing spore formation rate in media. Following this strategy yields of biomass, spore and lipase activity were improved manifold. After 36 h of fermentation, the respective values of biomass concentration, lipase activity, and spore yield were 6.19 g/L, 9.1 IU and 6 × 1012 per gram of biomass, respectively.

The use of fed-batch culture by the fermentation industry takes advantage of the fact that the concentration of the limiting substrate may be maintained at a very low level, thus avoiding the repressive effects of high substrate concentration (in this case glucose). Furthermore, the fed-batch system also gives some control over the organism’s growth rate due to control over the oxygen demand of the cultivation. It is important that all other nutrients are in excess so that the growth is solely controlled by the levels of the carbon source present [93]. This approach allows exponential growth of the culture at a specific rate for a much longer duration thereby generating a constant amount of biomass per amount of carbon [94]. Monteiro et al. [32] developed a fed-batch cultivation process in 2 L bioreactor for B. subtilis spore production with high yield (up to 7.4 × 109 spore/mL). The cells were initially grown (5 h) in batch mode in 1.3 L of Difco sporulation medium containing 3.5 g/L of glucose; then a nutrient feed was started in the middle of the exponential growth phase, before the complete depletion of the nutrients present in the media, i.e. before the beginning of the sporulation process. This feeding strategy permitted to extend the exponential growth phase (10 h after inoculation), leading to a maximum vegetative cell concentration of 3.6 × 1010 cells/mL at the end of the growth phase. During this exponential growth phase, the agitation rate was increased from 100 to approximately 1000 rpm to compensate for the oxygen consumption rate. At the end of the fed-batch phase, glucose was completely depleted from the medium causing a spike in the dissolved oxygen concentration, indicating the onset of the sporulation process. The fed-batch strategy applied had two main objectives: avoid glucose limitation during the vegetative growth phase, as this would induce sporulation, and avoid also concentrations higher than 3.5 g/L to prevent catabolite repression of several enzymes that may be involved in sporulation process and thus to achieve increased spore production. This fed-batch process of B. subtilis cultivation resulted in an increase in spore production with the highest yield of 7.4 × 109 spores/mL.

In the recent study, Pandey and Vakil [94] carried out the high cell density cultivation to obtain a higher yield of the probiotic Bacillus coagulans. Batch cultivation without regulating pH at C/N 35:1 yielded 21 g/L (corresponding to 2.9 × 1011 cells/mL), which was higher than biomass obtained under optimized shake flask conditions (8.0 g/L). Then fed-batch fermentation was performed where glucose was added intermittently as shots (about 50 mL of 250 g glucose/L), pH was not controlled and the C/N ratio was maintained at 35:1. The yield of 25 g/L (corresponding to 3.2 × 1011 cells/mL) was achieved. In the last fed-batch, C/N was further reduced to 30:1. The batch was harvested at the age of 32 h. Finally, batch B4 yielded 30 g/L of biomass corresponding to 3.8 × 1011cells/mL with high spore titer of 1.9 × 1011/mL and the sporulation efficiencies around 81%. High biomass production was achieved by maintaining the DO concentration greater than the critical level (20% DO) to satisfy the organism’s maximum specific oxygen demand. In all the batches, pH initially dropped from 6.5 to 4 and then in the later stages, it shifted to 6–6.5. The initial drop in pH was likely to be due to the conversion of glucose to lactic acid. In the later stage, when almost all the glucose is consumed cells start metabolizing lactate.

Results represented in Table 3 show wide variations in media composition used for different spore-forming probiotics production in the batch and fed-batch fermentations. Luna et al. [95] used a simple molasses-containing medium for cultivation of B. subtilis R14 that yielded 1.0 × 109 spores/mL, Monteiro et al. [32, 72] succeeded through supplementation of the nutrient media by salt mixtures whereas Khardziani et al. [68, 69] exploited the lignocellulosic material potential. Finally, Pandey and Vakil [93] achieved the highest yield of probiotic spores through the creation of cultivation conditions providing especially high biomass density. However, a number of studies employing plant raw materials as a growth substrate for the probiotics production is limited. Moreover, to the best our knowledge, there is no publication describing the scaled-up spore production in the SSF of lignocellulosic substrates. At the same time, this approach has great potential for the efficient production of cheap probiotics. Thus, in our experiments (unpublished results), B. subtilis KATMIRA1933 produced 4.9 × 1011 spores/g and 4.3 × 1011 spores/g when it was cultivated in polypropylene bags filled with 2 kg of wheat bran or milled corn cobs, respectively, moistened by cheese whey instead of water. Similarly, B. amyloliquefaciens B-1895 formed 2.1 × 1011 spores/g in the SSF of the ethanol production residue from wheat grains filled in cuvettes.

Table 3 Probiotic spore production in the batch and fed-batch fermentations
Full size table

Conclusions

Bacillus spp. probiotics are manufactured and used extensively due to their important advantages including ease of production and the stability of finished products in addition to immune stimulation, antimicrobial and competitive exclusion activities. However, an effective realization of their biotechnological potential and the development of competitive technologies of spore-forming bacteria production requires an understanding of physiological mechanisms regulating bacilli growth and sporulation and elucidation of conditions favoring both processes. Provided in this review available literature data clearly evidence that these processes depend on wide range factors, such as medium composition, pH, aeration, temperature etc. Moreover, each particular strain has its own requirements and optimum conditions. Further work is needed to establish physiological mechanisms by which individual factors (and their interrelationships) affect sporogenesis and increase the sporulation efficiency. In addition, more studies employing plant raw materials as the bacteria growth substrates are needed. Undoubtedly, evaluation and understanding to which extent growth conditions of bacilli probiotics affect their functional properties and product efficacy is necessary, i.e. it is important to perform clinical trials of each obtained formulation.