Isolation of Streptococcus thermophilus Strains from Plants in Japan and Their Application to Milk Fermentation

Abstract

Streptococcus thermophilus is a lactic acid bacterium used along with Lactobacillus delbrueckii subsp. bulgaricus as a starter culture for yogurt production. Thus, there are many reports on the isolation and characterization of S. thermophilus from dairy products. On the other hand, literature on the isolation of S. thermophilus from plants is limited. Here, we investigate the isolation of S. thermophilus from 688 samples of twigs, leaves, stems, flowers, and berries collected from healthy trees (66 species) and grasses (40 species) found in Japan. Eight S. thermophilus strains were obtained from Acer palmatum var. matsumurae, Cornus controversa, C. officinalis, Fagus crenata, Lagerstroemia indica, Magnolia denudata, and Quercus crispula var. crispula. Then, we evaluated the characteristics of milk fermentation by using these eight strains. All eight strains fermented milk, and six of them coagulated 10% (w/v) skim milk within 6 h at 40 °C. Finally, we characterized susceptibility to syneresis by centrifugation. Milk fermented by the plant strains was more resistant to syneresis than milk fermented by three S. thermophilus strains isolated from commercial yogurts. We found that S. thermophilus can be isolated from some plants found in Japan, and these plants are useful sources of S. thermophilus for milk fermentation.

Introduction

Streptococcus thermophilus is a well-known lactic acid bacterium used along with Lactobacillus delbrueckii subsp. bulgaricus as a starter culture to produce yogurt. S. thermophilus is also used as a starter culture for producing fermented milk products other than yogurt (e.g., labneh and skyr) and cheese (e.g., parmesan cheese and Swiss cheese) (Iyer et al., 2010). As a result, the isolation and characterization of S. thermophilus strains from dairy products has been widely researched (Al-Mariri et al., 2013; Delorme et al., 2017; El-Baradei et al., 2008; Mora et al., 2002). In contrast, little research has been done on the isolation of S. thermophilus from non-dairy products (Michaylova et al., 2007; Muyanja et al., 2003; Nakata et al., 2010; Umamaheswari et al., 2014).

Although industrially produced yogurts and cheeses use well-characterized strains of lactic acid bacteria, traditional, home-made fermented products are made by inoculating milk with a starter culture from previous products (Oberman, 1985). If a previous product is not available, folklore describes plants that can be used to produce some fermented milks (Fondén et al., 2007; Michaylova et al., 2007). For example, Norwegian tettemelk and Swedish tätmjölk, known as Scandinavian ropy fermented milks, have been prepared with leaves from the butterwort plant (Pinguicula vulgaris) or sundew (Drosera rotundifolia); these plants are called tettegras in Norway and tätgräs in Sweden, meaning (milk) thickening grass (Alm, 2005; Alm and Larsson, 1983). Another example is yogurt, defined as milk fermented with symbiotic cultures of L. delbrueckii subsp. bulgaricus and S. thermophilus (i). In Bulgaria, there is a tradition of preparing yogurt at home using plants (Michaylova et al., 2007). Branches of some plant (e.g., Cornus mas) are added to sheep's milk that has been boiled and cooled to about 45 °C. Maintaining the milk at this temperature results in coagulation, and part of this coagulum is used as a starter for yogurt. However, there were no reports on the isolation of these starter bacteria from plants until Michaylova et al. (2007) attempted to isolate them from hundreds of plant samples collected from four regions in Bulgaria. They detected L. delbrueckii subsp. bulgaricus and/or S. thermophilus in 24 of 665 samples and concluded that starter bacteria now used worldwide for commercial yogurt production could have originated from plants in Bulgaria. Besides, Umamaheswari et al. (2014) isolated S. thermophilus strains from vegetables and fodder crops in India.

Here, we report the isolation of S. thermophilus from some plant species found in Japan (excluding vegetables and crop species). We found that S. thermophilus can be isolated not only from plants in Bulgaria and India, but also from plants in Japan, including plants not originally identified as sources. Thus, plants are a useful source of S. thermophilus for fermenting milk, in addition to currently used sources such as raw milk, cheese, and other dairy products.

Materials and Methods

Culture media    CaCO₃-LYP agar (Table S1) plates were used to isolate S. thermophilus from plants and commercial yogurts. Reference strains and the eight strains isolated from plants, which are referred to as the “plant strains” in this paper, were cultured in YGLP broth (Table S2).

Table S1. CaCO₃-LYP agar

Ingredients	Weight or volume
Lactose·H2O	10.0 g
Yeast extract (Oxoid)	10.0 g
Trypticase peptone (BD)	5.0 g
Lab-Lemco powder (Oxoid)	2.0 g
Sodium acetate·5H2O	2.0 g
MgSO4·7H2O	200 mg
MnSO4·5H2O	10 mg
FeSO4·7H2O	10 mg
NaCl	10 mg
Tween	80 500µL
CaCO3	3.0 g
Cycloheximide	10µg
Polymixin B	5µg
Sodium azide	10µg
Agar	12.0 g
Water	1,000 mL

pH 6.8

Table S2. Composition of YGLP broth

Ingredients	Weight or volume
Trypticase peptone (BD)	5.0 g
Phytone peptone (BD)	5.0 g
Lab-Lemco powder (Oxoid)	8.0 g
Yeast extract (Oxoid)	3.0 g
KH2PO4	2.5 g
K2HPO4	2.5 g
MgSO4·7H2O	200 mg
MnSO4·5H2O	50 mg
Glucose	5.0 g
Lactose·H2O	5.0 g
Water	1,000 mL

pH 6.8

Reference strains S. thermophilus    JCM 17834^T and S. salivarius JCM 5707^T were obtained from the Japan Collection of Microorganisms (Wako, Saitama, Japan) and grown in YGLP broth (Table S2) at 45 °C overnight. S. thermophilus YSa, YSb, and YSc were isolated with CaCO₃-LYP agar (Table S1) plates from commercial yogurts obtained in Hirosaki, Japan. These three strains are referred to as the “yogurt strains” in this paper.

Sampling of plants and isolation of lactic acid bacteria    Between May 2015 to July 2016, twigs, leaves, stems, flowers, and berries were collected from healthy trees (66 species) and grasses (40 species) growing in Aomori, Iwate, and Saitama Prefectures in Japan (Fig. S1 and Table S3). The plant samples were collected using disinfected scissors and put into 50-mL sterilized plastic tubes by hand (with disinfected gloves). The tubes were capped securely and returned to the laboratory under refrigeration. Within the same day, the tubes were aseptically filled with 40 mL sterilized litmus milk (100 g/L skim milk powder, 0.4 g/L litmus) and incubated statically overnight at 45 °C. When the litmus milk coagulated and changed from a purple to pink color without gas formation, an aliquot of culture was Gram-stained and observed under a light microscope. After detecting Gram-positive streptococci through microscopy, an aliquot of culture was streaked onto CaCO₃-LYP agar (Table S1) plates and incubated at 45 °C overnight under anaerobic conditions with an AnaeroPack (Mitsubishi Gas Chemical, Tokyo, Japan) to isolate these organisms. Colonies surrounded by clear zones produced by acid formation were picked for Gram-staining. When Gram-positive streptococci were observed, the isolation step was repeated to confirm successful isolation.

Fig. S1.

The map shows the northern part of the main island of Japan.

(A) Shiriya fishing port in Higashidohri village. (B) Natsudomari peninsula in Hiranai town. (C) Eboshidake Mountain in Noheji town. (D) Oinonagane park in Goshogawara city. (E) Bonjyusan Mountain in Aomori city. (F) Shirakamidake Mountain in Fukaura town. (G) Futatsumori Mountain in Ajigasawa town. (H) Anmon waterfalls in Nishimeya village. (I) Shirakami Natural Science Park in Nishimeya village. (J) Hirosaki city. (K) Ninohe city. (L) Saitama city.

Table S3. Plants used in this study

Name	Sampling regions in Fig S1 (No. of sampling times)	No. of samples tested	No. of individuals collected samples
Woody plants
Acer japonicum	F(1), I(1)	6	2
Acer palmatum var. matsumurae	J(9)	22	16
Acer pictum	I(1), J(1)	8	2
Actinidia arguta	J(1)	1	1
Aesculus turbinata	E(1)	2	1
Aronia melanocarpa	J(1)	1	1
Aucuba japonica var. borealis	I(1)	1	1
Berchemia racemosa	D(1)	2	1
Betula maximowicziana	F(1)	5	1
Cerasus × yedoensis. ‘Somei-yoshino’	J(7)	51	12
Cerasus sargentii	I(1)	1	1
Chamaecyparis pisifera	J(5)	5	5
Clerodendrum trichotomum	I(1)	5	1
Cornus brachypoda	B(1)	4	2
Cornus controversa	B(1), C(1), D(1), G(1), I(1), J(8)	111	23
Cornus florida	J(1)	3	1
Cornus kousa	D(1), J(1)	15	2
Cornus officinalis	J(13)	83	15
Daphniphyllum macropodum	J(1)	2	1
Diosoyros lotus	J(1)	1	1
Diospyros kaki Thunb.	J(5)	5	5
Dryopteris crassirhizoma	E(1)	2	1
Elaeagnus umbellata	A(1)	2	1
Euonymus alatus form. ciliatodentatus	J(1)	1	1
Fagus crenata	E(1), F(1), H(1), I(4), J(3)	74	17
Fraxinus lanuginosa	E(1)	2	1
Hydrangea macrophylla	J(1)	2	1
Hydrangea serrata var. megacarpa	E(1)	2	1
Ilex leucoclada	(E)(1)	2	1
Ilex serrata	J(1)	3	1
Lagerstroemia indica	J(3), L(1)	12	6
Lindrera umbellata	E(1), I(2)	4	3
Magnolia denudata	J(7)	11	7
Magnolia kobus	J(2)	4	2
Malus pumila	J(3)	7	3
Morus bombycus	E(1)	4	1
Nandina domestica	J(1)	3	1
Padus ssiori	I(1)	1	1
Picea glehnii	J(1)	1	1
Pinus thunbergii	A(1)	2	1
Prunus mume	J(7)	11	7
Prunus salicin	J(7)	7	7
Pterocarya rhoifolia	E(1)	2	1
Quercus crispula var. crispula	H(1), I(3)	18	7
Quercus dentana	A(1), J(1)	3	2
Quercus glauca	J(2)	2	2
Rhododendron quinquefolium	I(1)	1	1
Rosa rugosa	A(1)	4	2
Rubus palmatus var. coptophyllus	J(1)	2	1
Sambucus racemosa ssp. sieboldiana	E(1)	2	1
Sasa kurilensis	E(1), I(1)	3	2
Sorbus commixta	F(1), J(5), K(1)	17	7
Taxus cuspidata var. cuspidata	J(3)	5	3
Taxus cuspidata var. nana	J(1)	1	1
Thujopsis dolabrata var. hondae	E(1)	2	1
Tilia japonica	J(1)	5	1
Ulmus japonica	E(1)	2	1
Vaccinium japonicum var. japonicum	J(1)	1	1
Vaccinium oldhamii	J(1)	2	1
Vibrunum dilatatum	E(1), J(1)	2	2
Vibrunum furcatum	E(1), I(2)	4	3
Viburnum opulus var. sargentii	J(1)	1	1
Vitis coignetiae	J(1)	1	1
Zelkova serrata	J(2)	3	2
Herb
Anemone pseudoaltaica	I(1)	1	1
Angelica ursina	A(1)	2	1
Aralia cordata	K(1)	1	1
Artemisia indica var. maximowiczii	J(1)	1	1
Artemisia stelleriana	A(1)	2	1
Boehmeria japonica var. longispica	A(1)	2	1
Carex kobomugi	A(1)	3	1
Cephalotaxus harringtonia var. nana	E(1), H(1)	4	2
Chloranthus japonicus	I(1)	1	1
Convallaria majalis	J(1)	1	1
Dactylis glomerate	J(1)	5	1
Disporum smilacinum	I(1)	1	1
Elymus mollis	A(1)	2	1
Epimedium koreanum	I(1)	1	1
Eragrostis cilianensis	J(1)	2	2
Erythronium japonicum	I(2)	1	1
Festuca arundinacea	J(2)	10	2
Galeola septentrionalis	I(1)	2	1
Glaucidium palmatum	I(1)	1	1
Glehnia littoralis	A(1)	2	1
Hemarthria sibrica	J(1)	5	1
Ixeris repens	A(1)	2	1
Lathyrus japonicas	A(1)	2	1
Miscanthus sinensis	J(1)	5	1
Monotropastrum humile	H(1)	1	1
Patasites japonicus	I(2)	2	2
Pennisetum alopecuroides	J(1)	5	1
Phalaris arundinacea	J(1)	5	1
Phragmites australis	J(1)	5	1
Plantago asiatica	J(1)	1	1
Plantago lanceolate	A(1), J(2)	8	3
Plantago major	I(1)	2	1
Poa pratensis	A(1)	2	1
Setaria viridis	J(1)	5	1
Smilacina japonica	A(1), I(1)	8	3
Smilay riparia var. ussuriensis	J(1)	2	1
Sonchus asper	A(1)	2	1
Trifolium pretense	A(1)	2	1
Trillium apetalon	I(1)	1	1
Viola cilianensis	I(2)	2	2
Zoysia macrostachya	A(1)	2	1

DNA extraction    To characterize the plant strains by polymerase chain reaction (PCR), the plant strains and the reference strains were propagated statically in YGLP broth (Table S2) at 45 °C overnight. One milliliter of culture was centrifuged at 9 057 g at 4 °C for 10 min. After removing the supernatant, the pellet was suspended in 100 µL 25 mmol/L NaOH, heated at 100 °C for 10 min, and neutralized with a 100 µL 80 mmol/L Tris-HCl buffer (pH 7.5) (Yamazaki-Matsune et al., 2007). After centrifugation, the supernatant was used as a template for PCR (see Tentative identification by PCR).

For 16S rRNA gene sequencing and DNA-DNA hybridization experiments, total DNA was extracted as follows. The plant strains and reference strains were propagated statically in YGLP broth at 40 °C for 6 h. Ten milliliters of culture was centrifuged at 5 796 g at 4 °C for 20 min. After removing the supernatant, the pellet was suspended in saline sodium citrate (SSC) buffer (0.15 mol/L NaCl, 15 mmol/L sodium citrate, pH 7.0) and centrifuged at 9 057 g at 4 °C for 20 min. After removing the supernatant, the pellets were stored at −30 °C overnight; they were thawed the next day by leaving them at room temperature for approximately 1 h. Then, the pellets were resuspended in a 170 µL lysis enzyme solution containing 5 mg lysozyme (Nacalai Tesque, Kyoto, Japan), 0.7 mg labiase (Seikagaku Biobusiness, Tokyo, Japan), and 100 units mutanolysin (Sigma-Aldrich, St. Louis, USA) in SSC and incubated for 2.5 h at 40 °C. The cell lysate was frozen at −30 °C for 1 h then thawed by placing on ice for approximately 30 min. Ten microliters of RNase A solution (Sigma-Aldrich) was added to the cell lysate, and the mixture was incubated at 37 °C for 1 h. DNA was then extracted from the cell lysate with the QuickGene SP Kit DNA Tissue system (Kurabo Industries Ltd., Osaka, Japan), according to the manufacturer's instructions. Absorbance of the isolated DNA was measured with a Nanodrop (Thermo Fisher Scientific, Waltham, USA). DNA with both an A₂₆₀/A₂₈₀ ratio of 1.8–2.0 and an A₂₆₀/A₂₃₀ ratio of > 2.0 was used for DNA-DNA hybridization experiments.

Tentative identification by PCR    PCR was performed using the S. thermophilus species-specific primers Th I (5′-ACGGAATGTACTTGAGTTTC-3′) and Th II (5′-TTTGGCCTTTCGACCTAAC-3′) under the conditions reported by Tilsala-Timisjärvi and Alatossava (1997). Specifically 25 µL of PCR reaction mixture containing 5 µL template DNA solution, 5 pmol of each primer, and 12.5 µL 2 × Quick Taq^® HS DyeMix (Toyobo Co., Ltd., Osaka, Japan) was subjected to the following thermal cycles: 94 °C for 2 min followed by 30 cycles at 94 °C for 30 sec, 55 °C for 1 min, and 68 °C for 1 min. Then, the PCR products were separated on a 1.5% agarose S gel (Nippon Gene, Tokyo, Japan; 52 or 106 mm wide × 60 mm long) in Tris-acetate EDTA (TAE) buffer (40 mM Tris-acetate, 1 mM EDTA; pH 8.2) at 100 V for approximately 30 min. The Gene Ladder 100 (Nippon Gene) was used as the molecular weight marker.

Phylogenetic analysis    The nearly full-length 16S rRNA genes of the plant strains were amplified by PCR from the purified genomic DNA using the bacteria-specific primers 27f (Lane, 1991) and 1492r (Turner et al., 1999) and were sequenced, as described previously (Horino et al., 2015). The sequences were deposited in in GenBank/EMBL/DDBJ under the accession numbers LC485977 for YS1, LC485978 for YS2, LC485979 for YS3, LC485980 for YS4, LC485981 for YS5, LC485982 for YS6, LC485983 for YS7, and LC485984 for RO1. Phylogenetic analysis was performed using the MEGA7 software package (Kumar et al., 2016). The 16S rRNA gene sequences of the plant strains and their phylogenetic relatives, whose 16S rRNA gene sequences were retrieved from GenBank/EMBL/DDBJ, were aligned using MUSCLE (Edgar 2004). A phylogenetic tree was reconstructed using the maximum-likelihood (ML) method (Gu et al., 1995). Evolutionary distances were calculated according to the Jukes– Cantor model (Jukes and Cantor, 1969) selected as the best model on the process of model selection by MEGA7. The bootstrap analysis was performed to assess the reliability of reconstructed phylogenetic tree with 1,000 resamplings.

DNA-DNA hybridization    DNA-DNA hybridization was conducted according to the microtitration plate method (Ezaki et al., 1989) modified by Dianou et al. (2001). Values are expressed as means of four to five replicates.

Preparation of fermented milk    Reconstituted skim milk (10% w/v) was autoclaved at 115 °C for 20 min and cooled to 40 °C. The milk was inoculated with 2.5% (v/v) of overnight culture in skim milk of a S. thermophilus strain and incubated at 40 °C for 24 h. The pH was measured with a pH meter (Model D-71, Horiba Ltd., Tokyo, Japan) at 0, 2, 4, 6, 8, 10, and 24 h.

Measurement of syneresis by centrifugation    Forty milliliters of fermented milk was prepared in a 50-mL conical tube (3181-345, Labcon, Petaluma, USA), as described in “Preparation of fermented milk” above, except that the milk inoculated with S. thermophilus JCM 17834^T, strains YSa and YSc were incubated for 24 h, and the milk inoculated with the plant strains was incubated for 10 h. The resulting fermented milk was refrigerated at 4 °C for 24 h. It was then centrifuged at 4 °C for 10 min under different centrifugal forces ranging from 200 × g to 1 000 × g with an MX-307 centrifuge (Tomy Seiko Co. Ltd., Tokyo, Japan) and an AR510-04 rotor. The separated clear supernatant (Ws) was weighed, and syneresis (%) was calculated as Ws (gram)/weight of fermented milk in the conical tube (gram) × 100. Tests were carried out in triplicate.

Results

Isolation of thermophilic streptococci strains from plants    We attempted to isolate S. thermophilus from 688 samples of twigs, leaves, stems, flowers, and berries collected from healthy trees (66 species) and grasses (40 species) in Aomori, Iwate, and Saitama Prefectures in Japan (Fig. S1 and Table S3). When the individual samples were incubated in litmus milk at 45 °C, only 15 of the 688 samples coagulated the litmus milk and changed the color purple to pink without gas formation, and Gram-positive streptococci were observed in eight of these 15 samples. Then, eight strains of acid-producing Gram-positive bacteria, the “plant strains”, were isolated from the eight coagulated litmus milk samples on CaCO₃-LYP agar plates incubated at 45 °C.

Identification of the plant strains    To identify whether the eight plant strains belong to the species S. thermophilus, we conducted a PCR with the S. thermophilus species-specific primers Th I and Th II, as reported by Tilsala-Timisjärvi and Alatossava (1997). For this, S. salivarius JCM 5707^T was used as the reference strain because S. thermophilus was previously temporarily misclassified as a subspecies of S. salivarius (S. salivarius subsp. thermophilus) (Schleifer et al., 1991) due to their very similar phenotypic features. Furthermore, S. salivarius is known as an opportunistic pathogen found in the upper respiratory tract, mouth, and etc. (Ireland, 2010). Thus, it is very important to differentiate the eight plant strains from S. salivarius. As shown in Fig. 1, the eight plant strains all produced a single fragment with an expected size of 259-bp. However, unfortunately, S. salivarius also produced a PCR product of the same size (Fig. 1), presumably because the primer pair Th I and Th II used are not specific only to S. thermophilus but also to S. salivarius; actually, the primers were not tested for S. salivarius in the article by Tilsala-Timisjärvi and Alatossava (1997) and it appeared in this study that they are not suitable to differentiate S. thermophilus and S. salivarius. Hence, phylogenetic analysis based on the 16S rRNA gene sequences was performed to confirm the assignment of the eight plant strains to S. thermophilus species. The phylogenetic analysis revealed that S. thermophilus ATCC 19258T was the closest relative of the eight plant strains with sequence similarities between 99.93 and 100% (Table S4). S. salivarius ATCC 7073^T was the second best relative with sequence similarities between 99.66% and 99.73%. As shown in the maximum likelihood (ML) phylogenetic tree (Fig. 2), the eight plant strains formed a clade with S. thermophilus ATCC 19258^T with a high bootstrap value of 98% and were clearly discriminated from S. salivarius ATCC 7073^T. Furthermore, we performed DNA-DNA hybridization to confirm the species' identification. As shown in Table 1, DNA from the eight plant strains was 74%–90% similar to the S. thermophilus type strain, exceeding the threshold value of 70% for species delineation (Stackebrandt et al., 2002). Furthermore, the low similarity values of 30%–41% between the eight plant strains and S. salivarius also supported their differentiation at the species level.

Fig. 1.

PCR products amplified from DNA from the plant strains using S. thermophilus species-specific primers Th I and Th II.

M, 100-bp DNA ladder; lane 1, S. thermophilus JCM 17834^T; lane 2, S. salivarius JCM 5707^T; lane 3, YS1; lane 4, YS2; lane 5, YS3; lane 6, YS4; lane 7, YS5; lane 8, YS6; lane 9, YS7; and lane 10, RO1.

Table S4. Similarity matrix based on 16S rRNA sequences of the 8 plant strains and the type strains of S. thermophilus and S. salivarius

Strain*	% Pairwise nucleotide similarity to:
	1	2	3	4	5	6	7	8	9
1.YS1 (LC485977)
2.YS2 (LC485978)	99.93
3.YS3 (LC485979)	99.93	99.93
4.YS4 (LC485980)	99.93	99.93	100
5.YS5 (LC485981)	99.86	99.86	99.93	99.93
6.YS6 (LC485982)	99.93	99.93	100	100	99.93
7.YS7 (LC485983)	99.93	99.93	100	100	99.93	100
8.RO1 (LC485984)	99.93	99.93	100	100	99.93	100	100
9. S. thermophilus ATCC 19258T (AY188354)	99.93	99.93	100	100	99.93	100	100	100
10. S. salivarius ATCC 7073T (AY188352)	99.66	99.66	99.73	99.73	99.66	99.73	99.73	99.73	99.73

*  The numbers in parentheses are the GenBank/EMBL/DDBJ accession numbers for the sequences used in the analysis.

Fig. 2.

ML phylogenetic tree based on 16S rRNA gene sequences, showing the close relationship of the plant strains (YS1, YS2, YS3, YS4, YS5, YS6, YS7, and RO1) with S. thermophilus ATCC 19258T. The sequence of Streptococcus saliviloxodontae NUM 6306T was used as the outgroup. The values at each node represent the percentage of bootstrap support from 1000 resamplings. Bar, 0.005 substitutions per site. All sequences were retrieved from GenBank/EMBL/DDBJ and those accession numbers are shown in parentheses.

Table 1. DNA-DNA hybridization between plant isolates and the Streptococcus thermophilus type strain.

Strains	% similarity to
	S. thermophilus JCM 17834T	S. salivarius JCM 5707T
S. thermophilus JCM 17834T	100	41
S. salivarius JCM 5707T	41	100
YS1	84	30
YS2	75	33
YS3	78	20
YS4	75	31
YS5	90	40
YS6	81	38
YS7	82	32
RO1	74	41

% similarity values are expressed as means of four to five replicates

Consequently based on the phylogenetic analysis and the results of DNA-DNA hybridization, all eight plant strains were identified as S. thermophilus (Table 2).

Table 2. S. thermophilus strains isolated from Japanese plants.

Strains	Source plant	Part	Time	Regiona
YS1	Quercus crispula var. crispula	Twigs	Jun-15	I
YS2	Fagus crenata	Leaves	Jun-15	I
YS3	Cornus officinalis	Leaves	Jul-15	J
YS4	Cornus controversa	Leaves	Jul-15	D
YS5	Lagerstroemia indica	Twigs	Aug-15	L
YS6	Fagus crenata	Twigs	Oct-15	I
YS7	Acer palmatum var. matsumurae	Leaves	Oct-15	J
RO1	Magnolia denudata	Twigs	Dec-15	J

^a  Abbreviations are same as in Fig. S1

Fermentation of milk by the plant strains    To evaluate the ability of the S. thermophilus isolated from plants (the plant strains) to ferment milk, we assayed acid production and milk coagulation. The plant strains were monitored in 10% (w/v) skim milk at 40 °C alongside the yogurt strains (Fig. 3). The rate at which the pH decreased varied across strains. After 6 h incubation, the pH values ranged from 4.60 to 5.23 in milk fermented by the plant strains (Fig. 3A) and from 5.00 to 5.91 in milk fermented by the yogurt strains (Fig. 3B). In milk fermented by the plant strains, we observed coagulation by strain YS1 after 4 h, whereas coagulation by five plant strains (YS2, YS3, YS4, YS5, and YS6) took at least 6 h. In milk fermented by the three yogurt strains, only strain YSb produced coagulation after 6 h. Thus, like the yogurt strains, the plant strains were capable of producing acid and coagulating milk.

Fig. 3.

Growth of S. thermophilus strains isolated from plants (A) or commercial yogurt (B) in skim milk. Values are expressed as the means of triplicate measurements.

(A) YS1 (○), YS2 (●), YS3 (△), YS4 (▴), YS5 (□), YS6 (■), YS7 (◊), RO1 (◆), JCM 17834^T (); (B) YSa (), YSb (), YSc ()

Coagulation observed at 4 h (YS1), 6 h (YS2, YS3, YS4, YS5, YS6, and YSb), 8 h (RO1), 10 h (YS7), and 24 h (JCM 17834^T, YSa, and YSc).

Next, we used centrifugation to evaluate the susceptibility of the fermented milks to syneresis. As shown in Fig. 4, the degree of syneresis (%) increased in proportion to the strength of the centrifugal force in all cases, although susceptibility varied depending on the strain. At 200 g, syneresis (%) was 5.0%–18.0% in milks fermented by the plant strains (Fig. 4A), compared to 8.8%–21.5% for the yogurt strains (Fig. 4B). At 1,000 × g, syneresis (%) was 22.4%–43.3% in milks fermented by the plant strains, compared to 35.1%–51.9% for the yogurt strains. Likewise, at 500 and 700 g, milk fermented by the plant strains showed less syneresis (%) than milk fermented by the yogurt strains. Notably, the plant strain RO1 produced fermented milk that was the most resistant to syneresis among 12 S. thermophilus strains, including those from the three yogurt strains.

Fig. 4.

Susceptibility to syneresis of fermented milk produced by S. thermophilus strains isolated from plants (A) or commercial yogurt (B). Values are expressed as the means of triplicate measurements.

(A) YS1 (○), YS2 (●), YS3 (△), YS4 (▴), YS5 (□), YS6 (■), YS7 (◊), RO1 (◆), JCM 17834^T (); (B) YSa (), YSb (), YSc ()

Discussion

We isolated eight S. thermophilus strains from seven plant species (Table 2) among 688 samples of 106 plant species growing in Japan (Fig. S1 and Table S3). Two strains were isolated from Fagus crenata, but only one strain each was isolated from Acer palmatum var. matsumurae, Cornus controversa, C. officinalis, Lagerstroemia indica, Magnolia denudata, and Quercus crispula var. crispula. Several researchers reported isolating S. thermophilus from non-dairy sources, specifically plants (Michaylova et al., 2007; Umamaheswari et al., 2014), bushera (a Ugandan non-alcoholic fermented beverage made from sorghum) (Muyanja et al., 2003), and a sugar factory (Nakata et al., 2010). Only two of these studies isolated S. thermophilus from a plant source. Umamaheswari et al. (2014) isolated S. thermophilus strains from vegetables and fodder crops, but the scope of their study does not overlap with ours. Further, there are no reports on the isolation of S. thermophilus from the plant species mentioned above.

Michaylova et al. (2007) isolated 20 S. thermophilus strains from Calendula officinalis, Capsella bursa-pastoris, Chrysanthemum, Cichorium intybus, Colchicum, Cornus mas, Dianthus sp., Hedera sp., Nerium oleander, Plantago lanceolata, Prunus spinosa, Rosa sp., Tropaeolum sp., and other unidentified plant species. Among them, four strains were from C. mas, and two strains were from C. bursa-pastoris. Michaylova et al. (2007) isolated L. delbrueckii subsp. bulgaricus and/or S. thermophilus (data on S. thermophilus alone are not available) from C. mas samples at a rate of 14%, which is much higher than the average rate of 3.6% for all plant samples in the literature. C. mas is used to prepare yogurt starters in the Bulgarian tradition. As C. mas is not found in Japan, we attempted to isolate S. thermophilus from five species in the genus Cornus (Table S3). Among the five species tested, only the 111 C. controversa and 83 C. officinalis samples yielded S. thermophilus, with isolation rates of 0.9% and 1.2%, respectively. Of the other plant species from which Michaylova et al. (2007) isolated S. thermophilus, we obtained P. lanceolata and Rosa. However, we could not isolate S. thermophilus despite analyzing 11 Plantago spp. (P. asiatica, P. laceolata and P. major) and four Rosa rugosa samples (Table S3).

We also isolated a single S. thermophilus strain each from 22 Acer palmatum var. matsumurae, 12 L. indica, 11 M. denudata, and 18 Q. crispula var. crispula samples, as well as two strains from 74 F. crenata samples (Table 2). To our knowledge, S. thermophilus has never been isolated from these five plant species until now. The isolation rate for these five plant species was 2.7%–9.1%. We were unable to isolate S. thermophilus strains from the same plants from which the eight plant strains had been isolated previously. Isolation rate is as low as only 2.7%, even in the case of F. crenata, which is the only plant species from where the two S. thermophilus strains were isolated. This low isolation frequency, and lack of isolation reproducibility, suggests that S. thermophilus is not indigenous to either plant, but is transferred to the plants by wild animals, birds, and/or insects. Michaylova et al. (2007) also proposed that L. delbrueckii subsp. bulgaricus and S. thermophilus could be transferred to plants from other materials by insects, such as ants.

The characteristics of milk fermentation by plant strains have been studied by Michaylova et al. (2007) and Umamaheswari et al. (2014). Michaylova et al. (2007) observed that the characteristics of plant strains (e.g., regarding acid production, urease activity, proteolytic activity, and kinematic viscosity) are by no means inferior to those of industrial strains currently used to produce yogurt. Further, Umamaheswari et al. (2014) concluded that the plant strains and reference strains obtained from Indian dairy culture collections did not demonstrate any considerable differences in acidification ability, proteolytic activity, urease activity, or acetaldehyde production. We also showed that the rate of pH decrease, the time required to coagulate milk, and the resistance to syneresis for the plant strains was equivalent or superior to yogurt strains. Therefore, our results, together with those of Michaylova et al. (2007) and Umamaheswari et al. (2014), show that plants are useful sources of S. thermophilus for milk fermentation.

Acknowledgements    This work received financial support from the Hirosaki University research project foundation. Yuri Saito thanks Chisaki Kudo and members of the Shirakami Fungus Group for their technical advice.

References

•  Alm,  T. (2005). Pinguicula vulgaris (Lentibulariaceae) and its uses in Norway. SIDA, Contrib. Bot., 21, 2249-2274.
•  Alm,  L. and  Larsson,  I. (1983). From antiquity into the future. The Nordic ropy milk-a product with a long history. Nordisk Mejeriindustri (English edition), 10, 398-400.
•  Al-Mariri,  A.,  Abou-Younes,  A., and  Sharabi,  N.E. (2013). Characterization of thermophilic streptococci isolated from rustic white cheese. J. Gen. Appl. Microbiol., 59, 97-103.
•  Delorme,  C.,  Legravet,  N.,  Jamet,  E.,  Hoarau,  C.,  Alexandre,  B.,  El-Sharoud,  W.M.,  Darwish,  M.S., and  Renault,  P. (2017). Study of Streptococcus thermophilus population on a world-wide and historical collection by a new MLST scheme. Int. J. Food Microbiol., 242, 70-81.
•  Dianou,  D.,  Miyaki,  T.,  Asakawa,  S.,  Morii,  H.,  Nagaoka,  K.,  Oyaizu,  H., and  Matsumoto,  S. (2001). Methanoculleus chikugoensis sp. nov., a novel methanogenic archaeon isolated from paddy field soil in Japan, and DNA-DNA hybridization among Methanoculleus species. Int. J. Syst. Evol. Microbiol., 51, 1663-1669.
•  Edgar,  R. C. (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res., 32, 1792-1797.
•  El-Baradei,  G.,  Delacroix-Buchet,  A., and  Ogier,  J.C. (2008). Bacterial biodiversity of traditional Zabady fermented milk. Int. J. Food Microbiol., 121, 295-301.
•  Ezaki,  T.,  Hashimoto,  Y., and  Yabuuchi,  E. (1989). Fluorometric deoxyribonucleic acid-deoxyribonucleic acid hybridization in microdilution wells as an alternative to membrane filter hybridization in which radioisotopes are used to determine genetic relatedness among bacterial strains. Int. J. Syst. Bacteriol., 39, 224-229.
•  Fondén,  R.,  Leporanta,  K., and  Svensson,  U. (2007). Nordic/Scandinavian Fermented Milk Products. In “Fermented Milks,” ed. by  A.Y.,  Tamime. Blackwell, Oxford, pp. 156-173.
•  Gu,  X.,  Fu,  Y. X., and  Li,  W. H. (1995) Maximum likelihood estimation of the heterogeneity of substitution rate among nucleotide sites. Mol. Biol. Evol., 12, 546-557.
•  Horino,  H.,  Ito,  M., and  Tonouchi,  A. (2015) Clostridium oryzae sp. nov., from soil of a Japanese rice field. Int. J. Syst. Evol. Microbiol., 65, 943-951.
•  Ireland,  R. (2010) Streptococcus. in “A Dictionary of Dentistry” ed. by  R.  Ireland. Oxford University Press, Oxford, p 329.
•  Iyer,  R.,  Tomar,  S.K.,  Maheswari,  T.U., and  Singh,  R. (2010). Streptococcus thermophilus strains: Multifunctional lactic acid bacteria. Int. Dairy J., 20, 133-141.
•  Jukes,  T.H. and  Cantor,  C.R. (1969) Evolution of Protein Molecules. In “Mammalian Protein Metabolism” ed. by  H. N.  Munro. Academic Press, New York, pp. 21-132.
•  Kumar,  S.,  Stecher,  G., and  Tamura,  K. (2016) MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol., 33, 1870-1874.
•  Lane,  D. J. (1991). 16S/23S rRNA sequencing. In “Nucleic Acid Techniques in Bacterial Systematics” ed. by  E.  Stackebrandt, and  M.  Goodfellow. John Wiley & Sons, Inc, New York, pp. 115-175.
•  Michaylova,  M.,  Minkova,  S.,  Kimura,  K.,  Sasaki,  T., and  Isawa,  K. (2007). Isolation and characterization of Lactobacillus delbrueckii ssp. bulgaricus and Streptococcus thermophilus from plants in Bulgaria. FEMS Microbiol. Lett., 269, 160-169.
•  Mora,  D.,  Fortina,  M.G.,  Parini,  C.,  Ricci,  G.,  Gatti,  M.,  Giraffa,  G., and  Manachini,  P.L. (2002). Genetic diversity and technological properties of Streptococcus thermophilus strains isolated from dairy products. J. Appl. Microbiol., 93, 278-287.
•  Muyanja,  C.M.,  Narvhus,  J.A.,  Treimo,  J., and  Langsrud,  T. (2003). Isolation, characterisation and identification of lactic acid bacteria from bushera: a Ugandan traditional fermented beverage. Int. J. Food Microbiol., 80, 201-210.
•  Nakata,  H.,  Hasegawa,  H.,  Sakurai,  H., and  Tamura,  M. (2010). Distinctive flavor and strong antifungal activity in a sourdough bread made using unique lactic acid bacteria obtained from a sugar factory. Nippon Shokuhin Kagaku Kogaku Kaishi, 57, 85-90 (in Japanese).
•  Oberman,  H. (1985). Fermented milks. In “Microbiology of fermented foods, vol 1,” ed. by  B.J.B.  Wood. Elsevier Applied Science, London, pp. 167-195.
•  Schleifer,  K.-H.,  Ehrmann,  M.,  Krusch,  U., and  Neve,  H. (1991). Revival of the species Streptococcus thermophilus (ex Orla-Jensen, 1919) nom. rev. System. Appl. Microbiol., 14, 386-388.
•  Stackebrandt,  E.,  Frederiksen,  W.,  Garrity,  G.M.,  Grimont,  P.A.D.,  Kämpfer,  P.,  Maiden,  M.C.J.,  Nesme,  X.,  Rosselló-Mora,  R.,  Swings,  J.,  Trüper,  H.G.,  Vauterin,  L.,  Ward,  A.C., and  Whitman,  W.B. (2002). Report of the ad hoc committee for the re-evaluation of the species definition in bacteriology. Int. J. Syst. Evol. Microbiol., 52, 1043-1047.
•  Tilsala-Timisjärvi,  A. and  Alatossava,  T. (1997). Development of oligonucleotide primers from the 16S-23S rRNA intergenic sequences for identifying different dairy and probiotic lactic acid bacteria by PCR. Int. J. Food Microbiol., 35, 49-56.
•  Turner,  S.,  Pryer,  K. M.,  Miao,  V. P. W., and  Palmer,  J. D. (1999) Investigating deep phylogenetic relationships among cyanobacteria and plastids by small submit rRNA sequence analysis. J. Eukaryot. Microbiol., 46, 327-338.
•  Umamaheswari,  T.,  Anbukkarasi,  K.,  Singh,  P.,  Tomar,  S.K., and  Singh,  R. (2014). Streptococcus thermophilus strains of plant origin as dairy starters: Isolation and characterisation. Int. J. Dairy Technol., 67, 117-122.
•  Yamazaki-Matsune,  W.,  Taguchi,  M.,  Seto,  K.,  Kawahara,  R.,  Kawatsu,  K.,  Kumeda,  Y.,  Kitazato,  M.,  Nukina,  M.,  Misawa,  N., and  Tsukamoto,  T. (2007). Development of a multiplex PCR assay for identification of Campylobacter coli, Campylobacter fetus, Campylobacter hyointestinalis subsp. hyointestinalis, Campylobacter jejuni, Campylobacter lari and Campylobacter upsaliensis. J. Med. Microbiol., 56, 1467-1473.
• i)  Codex (2003) Codex standard for fermented milks (Codex Stan 243-2003). Codex Alimentarius Commission. http://www.fao.org/input/download/standards/400/CXS_243e.pdf. (Accessed 25 Nov 2018)