Introduction

Almost all animal feeds are potentially susceptible to toxigenic molds at any stage during storage, processing, packaging, and transportation. These types of molds produce mycotoxins that are highly toxic secondary metabolites (Rather et al. 2014; Quiles et al. 2016). Aflatoxins are one of the six major classes of mycotoxins produced mainly by Aspergillus species such as Aspergillus flavus and Aspergillus niger (Khanafari et al. 2007). Aflatoxin B1 (AFB1) is the most potent teratogenic, cancerogenic, and oestrogenic agent and a major risk factor for liver cancer (Kabak et al. 2009).

Consumption of contaminated feeds with aflatoxigenic Aspergillus spp. could have serious problem to public health. Therefore, effort to reduce of aflatoxigenic Aspergillus growth is serious concern around the world especially prevalent in developing countries. There are different physical, chemical, and biological methods for reduction of Aspergillus growth in contaminated animal feeds, but only a few of these methods have been accepted for practical use (Kohl et al. 2011). One of mentioned methods is inhibition of Aspergillus spp. growth by lactic acid bacteria (LAB) (Sangmanee and Hongpattarakere 2014). The use of LAB as a natural biopreservative is highly regarded because of their GRAS (generally recognized as safe) status (Belguesmia et al. 2013). In several studies, oral safety of LAB strains has been confirmed. Accordingly, these microorganisms can be used as safe biosupplements in food and pharmaceutical applications (Lara-Villoslada et al. 2007; Tsai et al. 2014; Damodharan et al. 2015).

LAB are a group of microaerophilic bacteria, Gram-positive, non-respiring, non-spore forming, catalase-negative, cocci, or rod cells (Sangmanee and Hongpattarakere 2014). In recent years, various researchers have studied antifungal activity of LAB isolated from traditional fermented foods. In these studies, antifungal potential of Lactobacillus spp. (Sangmanee and Hongpattarakere 2014; Delavenne et al. 2015; Bian et al. 2016; Russo et al. 2016; Dong et al. 2017), Enterococcus spp. (Belguesmia et al. 2013), Lactococcus spp. (Roy et al. 1996; Varsha et al. 2016), and Pediococcus spp. (Dalie et al. 2010; Sadeghi et al. 2016; Sellamani et al. 2016) has also been described.

LAB isolates produce antifungal metabolites including 4-hydroxy-phenyllactic acid and phenyllactic acid (Lavermicocca et al. 2000), 3,6-bis(2-methylpropyl)-2,5-piperazinedion (Yang and Chang 2010), certain organic acids such as acetic acid, propionic acid, and lactic acid (Özcelik et al. 2016), hydroxy fatty acids (Sjogren et al. 2003), and cyclic dipeptides (Strom et al. 2002). These metabolites usually inhibit fungal spoilage by controlling the growth of toxigenic molds (Özcelik et al. 2016).

Due to the presence of LAB, traditional fermented products are usually less exposed to fungal spoilage. Therefore, microbial ecology of these products, especially non-aseptic fermented product ecosystems may be lead to identification of LAB isolates with unique antifungal capabilities. The objective of present study was to molecular identification and investigating the antifungal properties of native LAB isolated from traditional fermented products.

Material and methods

Indicator fungi

Aflatoxin producing A. flavus (PTCC 5004) and A. niger (PTCC 5154) were purchased from Iranian Research Organization for Science and Technology (IROST). These fungi were grown on yeast extract glucose chloramphenicol agar (YGC) (Merck, Germany) at 26 °C for 7 days. Fungal spores were harvested by addition of sterile Tween 80 solution (Merck, Germany) at 0.05% (v/v). Spore concentrations were adjusted to 104 spores/ml by a hemocytometer chamber (Wang et al. 2012).

Isolation and molecular identification of LAB

Ten different samples of traditional fermented products including Chal: fermented camel milk (four samples), whole barley sourdough (three samples), and whole wheat sourdough (three samples) were collected from rural area in north of Iran (Golestan Province). The samples were placed in a cooling box and transported to the Department of Food Science, Gorgan University of Agricultural Sciences and Natural Resources. Ten grams of each sample was stomachered in 90 ml of sterile ringer solution (Merck, Germany) with a Stomacher 400 (Seward, England) at 260 rpm for 3 min. An aliquot (1 ml) of the homogenate sample was spread plated onto MRS (de Man, Rogosa, and Sharpe) agar (Merck, Germany) and plates were then incubated at 37 °C in 10% CO2 for 48 h. After incubation, colonies were streaked on MRS agar plates for obtaining single colonies. Then colonies were tested for Gram-stain morphology, catalase activity, and microscopic cell morphology (Yang et al. 2012).

For molecular identification, isolated strains were cultured in MRS broth at 37 °C in 10% CO2 for 24 h, and DNA was extracted using the purification kit based on the manufacturer’s protocol (GeneAll, Korea). 16S rRNA gene of isolates were subjected to PCR amplification with primers F44 (5′-RGTTYGATYMTGGCTCAG-3′) and R1543 (5′-GNNTACCTTKTTACG ACTT-3′) by a thermo-cycler (Corbett N15128, Australia). Optimized PCR reaction was carried out according to Yang et al. (2012) in 20-μl final volume including 10-μl Taq DNA polymerase master mix red (Ampliqon, Denmark), 1.5 μl of each primer at concentration of 0.5 mM, 5 μl ddH2O, and 2 μl DNA with concentration of 100 ng/μl. Subsequently, PCR products (1500 bp amplicon) were electrophoresed in 1.5% (w/v) agarose gel in TBE buffer for 40 min at a constant voltage of 90 V and were visualized by UV transilluminator (Kiagen, Iran). Finally, the PCR products were sequenced by MWG Co. (Germany) and the sequencing results were checked by BLASTn procedure with available data in gene bank.

Antifungal activity of LAB isolates

Overlay method was used to investigate the antifungal activity of LAB isolates with slight modifications (Magnusson and Schnurer 2001). This method was performed using MRS agar plates (6 cm) on which LAB were inoculated as two 3-cm-long lines and incubated at 37 °C for 72 h in 10% CO2. The plates were then overlaid with 15 ml of YGC agar containing 104 fungal spores/ml and incubated at 26 °C until the fungi growth was almost complete in the control plates. Inhibition zones in the pictures taken by a digital camera were measured by the ImageJ software (version 1.41).

Determination of logarithmic and stationary phases of LAB isolate

Overnight culture of the most effective LAB isolate was diluted to 103 CFU/ml in MRS broth and then incubated at 37 °C in 10% CO2. Quantities of optical density (OD) of the isolate culture were monitored by spectrophotometer (LTD-T80T, England) at 600 nm at 1-h intervals, until receiving the growth curve to stationary phase (Gulahmadov et al. 2009).

Preparation of cell-free supernatants (CFSs) from LAB isolate

After twice activation of each LAB isolate in MRS broth, CFSs from logarithmic and stationary phases were respectively prepared from culture collected 2 h before and after the end of the logarithmic phase. The supernatant was then centrifuged (Sigma 2-16 KL, United States) at 4 °C, 14,000 g for 5 min and subsequently filtered through sterile 0.22-μm syringe microfilters (Biofil, China) to obtain the untreated CFS (Wang et al. 2012).

Antifungal activity of CFSs

Antifungal activity of CFSs was tested as reported by Wang et al. (2012) with some modifications. Briefly, CFSs were mixed with YGC (45 °C) to achieve a final concentration of 1–10%. Fungal spores (3 μl of Aspergillus spore suspensions containing 104 spore/ml) were then spotted at the center of plates and incubated at 26 °C. The control YGC plate was CFS-free and containing 10% (v/v) of sterile distilled water that was inoculated with 3 μl of Aspergillus spores at the center. Antifungal activity of CFS was measured by the ImageJ software, until the fungi growth in the control plates was almost complete. The lowest CFS concentration that inhibited visible growth of Aspergillus was also defined as minimum inhibitory concentration (MIC).

Preparation of activated Pediococus lolii for oral gavage in rats

For this purpose, the method of Tsai et al. (2014) was used. Briefly, bacteria grown in MRS broth were harvested by centrifugation (10,000 rpm for 5 min at 4 °C), then washed twice with phosphate-buffered saline (PBS), and resuspended in PBS to final OD of 0.8 (108 CFU cells/ml) at 600 nm, as measured by a spectrophotometer (LTD, England).

Safety assessment

Eighteen 4-week female Wistar rats were purchased from Pasteur Institute of Iran. Eighteen rats were then divided into two different groups for feeding and each rat was placed in a separate cage. For 28 days, the rats received orally/daily 100 μl of PBS containing 108 CFU of P. lolii, while control group received 100 μl PBS only. The rats were also given free access to food and drinking water and the housing conditions were 22 ± 3 °C with relative humidity of 60 ± 5% and a 12-h dark and 12-h light cycle. These animals were observed daily, and weight, behavior, activity, and hair luster were recorded. Specific growth rate (SGR) or average weekly weight gain (g) was calculated based on the following formula: (W − W 0 / W 0) × 100. W and W 0 indicate the rat weight on the defined feeding days and on day 0, respectively. On day 28, all rats were sacrificed and blood samples were collected for laboratory analysis. Platelet counts, red blood cell counts (RBC), white blood cells counts (WBC), mean corpuscular hemoglobin concentration (MCHC), mean corpuscular hemoglobin (MCH), and mean corpuscular volume (MCV) were determined by Kavosh Laboratory (Gorgan, Iran). Furthermore, plasma levels of liver enzymes including alanine transaminase (ALT), aspartate transaminase (AST), and alkaline phosphatase (ALP) were also determined (Tsai et al. 2014).

Statistical analysis

All experiments were performed in triplicate, and data were reported as mean ± standard error. LSD multiple comparison was also applied to obtain significance level (P < 0.05). All tests were analyzed using one-way analysis of variance (ANOVA) design with SPSS software (version 20).

Results

Molecular identification of LAB isolates

A total of 40 potential LAB strains were randomly isolated from fermented products on MRS agar; 28 of these isolates (12 isolates from Chal, nine isolates from whole wheat sourdough, and seven isolates from whole barley sourdough) were Gram positive, catalase negative, cocci or rod, and non-spore forming bacteria. Molecular identification of the isolates was done by specific PCR as described above. Fig. 1 shows the agarose gel electrophoresis of PCR products with a target gene of about 1500 bp. These 28 isolates were identified by sequencing of the 16S rRNA gene that showed 99% similarity with the corresponding species in the NCBI database (Table 1).

Fig. 1
figure 1

Agarose gel electrophoresis of PCR products for the LAB isolates. Lane 1 negative control, lanes 2–14 extracted DNA from the LAB isolates (1500 bp), lane 15 100 bp DNA ladder

Full size image
Table 1 Molecular identification of the LAB isolates from “traditional fermented products” by 16S rRNA sequence analysis
Full size table

Antifungal activity of LAB isolates

Antifungal activity of LAB isolates is shown in Table 2 (against A. flavus) and in Table 3 (against A. niger). As it is indicated in Table 2, A. flavus filled the entire surface of the control plate at the end of the fourth day, while LAB isolates showed different inhibitory effects ranging from 0–53.68% (Chal), 4.26–65.9% (whole wheat sourdough), and 27.61–70.32% (whole barley sourdough). After 4 days, among the 28 isolates of tested LAB, eight isolates from Chal and three isolates from whole wheat sourdough had no inhibitory effect against A. flavus, while all isolated LAB from whole barley sourdough had significant effect on A. flavus growth (P < 0.05) in comparison to control plate.

Table 2 The percentage of A. flavus growth in the presence of the LAB isolates during 4 days
Full size table
Table 3 The percentage of A. niger growth in the presence of the LAB isolates
Full size table

Among the entire LAB isolates, P. lolii isolated from whole barley sourdough had the highest inhibitory effect (70.32% inhibition) against A. flavus (Fig. 2). In this study, we also observed that LAB isolated from traditional fermented products had proper inhibitory effect against A. niger growth. A. niger filled the entire surface of the control plate at the end of the fifth day, but growth of the mold in the presence of LAB isolated from Chal, whole wheat, and whole barley sourdoughs ranged from 28.49–95.06, 26.96–100, and 1.2–33.15% after 5 days, respectively (Table 3). According to our finding, LAB isolated from whole barley sourdough showed a broad spectrum of antifungal activity against A. niger and among the isolates, P. lolii (Fig. 2) and Pediococus pentaseous had respectively the highest inhibitory effects against A. niger growth with 98.8 and 74.21% inhibition at the end of the fifth day.

Fig. 2
figure 2

Inhibitory effect of P. lolii isolated from whole barley sourdough after 4 days against A. flavus (a) and at the end of the fifth day (the marked circle show the fungus growth) against A. niger (c) compared with controls (b and d, respectively)

Full size image

Antifungal activity of CFSs

Based on the results of overlay tests in the present study, P. lolii showed the highest antifungal activity against A. flavus and A. niger and so on. We focused on investigating the antifungal activity of P. lolii CFS. Stationary phase from P. lolii was started 10 h after incubation (data not shown). Therefore, CFSs from logarithmic and stationary phases were respectively prepared after 8 and 12 h after incubation. Table 4 shows the antifungal activity of CFSs obtained from P. lolii against A. flavus. According to Table 4 data, the MIC values of P. lolii CFSs from logarithmic and stationary phases against A. flavus were 2 and 1% (v/v), respectively. The media containing 9% of logarithmic phase CFS and 7% of stationary phase CFS completely inhibited the growth of A. flavus mycelia.

Table 4 The percentage of A. flavus growth in the presence of P. lolii CFSs
Full size table

The results indicated that antifungal activity of P. lolii stationary phase CFS was significantly (P < 0.05) higher than the effect of the logarithmic phase CFS. In addition, the media containing ≥4% of the stationary phase CFS (v/v) had totally inhibited from germination of A. flavus spores (Fig. 3).

Fig. 3
figure 3

Growth of A. flavus in the presence of P. lolii CFS

Full size image

MIC values of P. lolii CFSs from logarithmic and stationary phases against A. niger were shown in Table 4, too. MIC value of the stationary phase CFS was 1%, while for the logarithmic phase was 4%. Complete inhibition from A. niger growth was also occurred at 8% of logarithmic phase CFS and 6% of stationary phase CFS, respectively. Fig. 4 indicates that germination of A. niger spores was absolutely inhibited in the presence of 4% or more of the stationary phase CFS (v/v).

Fig. 4
figure 4

Growth of A. niger in the presence of P. lolii CFS

Full size image

SGR, hematological, and clinical chemistry parameters

SGR of rats is shown in Table 5. Based on these results, during 28-day oral administration of P. lolii, the SGR increased in both control and treated groups but there was no significant difference (P > 0.05) between control and treated groups on the defined feeding days. Furthermore, any noticeable difference in the activity, behavior, and hair luster of the rats was observed in comparison to control group.

Table 5 Specific growth rate of rats that were fed with P. lolii for 28 days
Full size table

Clinical chemistry and hematological analysis data were also shown in Table 6. As it is indicated, in treated group, no significant change was observed in platelets, WBC, RBC, MCH, MCHC, MCV, and liver enzymes and all values were within normal physiological ranges.

Table 6 Clinical chemistry and hematological finding in rats that were fed with P. lolii for 28 days (mean ± SD)
Full size table

Discussion

Modern agriculture and animal production systems need to reliable and safe methods to prevent and minimize Aspergillus growth rate. Inhibition of the mold growth by natural biological antagonists is a good example of these safe and efficient ways (Sangmanee and Hongpattarakere 2014). Furthermore, consumer demand for reduced use of antibiotics and chemical preservatives has led to more interest in biological preservation of foods and animal feeds in recent years. Among natural preservatives, LAB are special interest groups and their application has a long history in foods and animal feeds (Russo et al. 2016; Khanafari et al. 2007). Fermented foods such as sourdough are the most common sources for LAB isolation. Based on competition ability and adaptability of predominant sourdough LAB to non-aseptic conditions of this fermented ecosystem, these LAB usually show strong antimicrobial activity (Sadeghi et al. 2016).

In the present study, we identified 28 LAB isolates by PCR method. Among those, 12 and nine LAB isolated from Chal and wheat sourdough belonged to the genera of Lactobacillus and Enterococcus, respectively. Moreover, seven LAB isolated from barely sourdough belonged to the genera of Lactobacillus and Pediococcus. Among mentioned isolates, P. lolii had the highest inhibitory effects against A. flavus and A. niger. According to these results, antifungal activity of P. lolii stationary phase CFS was higher than logarithmic phase CFS significantly (P < 0.05). Few studies have been conducted on P. lolii functional features. For example, Lee et al. (2012) analyzed the profile of organic acid in culture media of P. lolii by GC/MS and recently, Ju et al. (2016) confirmed the protective efficacy of P. lolii against influenza virus but there is no report about the antifungal activity of P. lolii until now. It should be noted that we identified P. lolii with strong antifungal activity from traditional whole barely sourdough for the first time.

Digaitiene et al. (2012) identified five strains of antifungal LAB (Lactobacillus sakei KTU05-06, Pediococcus acidilactici KTU05-7, Pediococcus pentosaceus KTU05-8, KTU05-9, and KTU05-10) with high activity against Aspergillus, Fusarium, Mucor, and Penicillium. Accordingly, antifungal LAB may be used as a suitable biopreservative to control the fungal spoilage in food and feed products. Antifungal activity of L. rhamnosus L60 and L. fermentum L23 was approved against aflatoxigenic Aspergillus spp. by Gerbaldo et al. (2012). Lactobacilli strains fully inhibited the Aspergillus growth of all strains assayed. Mentioned researchers have also concluded that mold contaminants of animal feeds could be significantly controlled by L60 and L23 strains. Rather et al. (2014) isolated Lactobacillus plantarum YML007 from Korean kimchi. L. plantarum YML007 and its supernatant delayed the growth of A. niger, A. flavus, Fusarium oxysporum, Saccharomyces cerevisiae, Candida albicans, and Pichia membranifaciens in maize. Furthermore, Wistar rats fed with supernatant treated maize showed more weight gain, compared with the control group. Based on these studies, antifungal activity of LAB is due to synergistic interactions between their metabolites such as weak organic acids, free fatty acids, cyclic dipeptides, piperidine derivatives, phenyl lactic acid, and phenolic compounds. Furthermore, recently, antifungal activity of cyclic dipeptides (Muhialdin et al. 2016), organic acids (Özcelik et al. 2016; Russo et al. 2016), and phenyl lactic acid (Russo et al. 2016) produced by LAB was approved. According to Strom et al. (2002), cyclic dipeptides are effective on quorum-sensing mechanism. Production of fungistatic bacteriocin-like substance by regulating signal peptides in LAB is a type of this quorum sensing. Free fatty acids, organic acids, and phenyl lactic acid also directly interact with the lipid bi-layer of fungi cell membrane and increase its fluidity. This interaction leads to change in membrane proteins, release of intracellular materials, cytoplasmic disorder, and finally cell destruction (Avis and Belanger 2001).

Some LAB strains produce piperidine derivatives (Li et al. 2012) and phenolic compounds (Annan et al. 2003) by several antifungal activities. Piperidine derivatives inhibit α-glycosidase activity, which plays a crucial role in gene regulation corresponding to carbohydrate metabolism in Aspergillus (Kato et al. 2002). Phenolic compounds can also interact with free radicals and prevent from formation of hydroperoxy molecules (Annan et al. 2003), which are essential for both sexual and asexual body development of filamentous Aspergillus spp. (Calvo et al. 2001).

In the present study, we also observed that oral administration of P. lolii led to significant increase in the SGR of rats. Moreover, hematological and clinical chemistry parameters of the rats did not differ significantly from those of the control group. Over the past years, various researchers have reported safety of LAB isolated from traditional fermented foods. In these studies, safety of Pediococcus spp. (Tsai et al. 2014), Lactobacillus spp. (Sulemankhil et al. 2012; Jones et al. 2012; Szabo et al. 2011), and Enterococcus spp. (Tsai et al. 2004) has also been described. They reported that LAB isolates had no adverse effects regarding the behavior, activity, hematology, clinical chemistry indices, and growth rate of rats in comparison to control group.

Conclusion

In conclusion and based on our results, by considering the proper antifungal activity of P. lolii isolate and its CFS against aflatoxigenic Aspergillus spp. and approval of the isolate safety, it is possible to use from mentioned LAB and its CFS as biopreservative agents in processing of different foods and feeds.