DOI QR코드

DOI QR Code

Analysis of Microbial Community Change in Ganjang According to the Size of Meju

메주의 크기에 따른 간장의 미생물 군집 변화 양상 분석

  • Ho Jin Jeong (Microbial Institute for Fermentation Industry (MIFI)) ;
  • Gwangsu Ha (Microbial Institute for Fermentation Industry (MIFI)) ;
  • Ranhee Lee (Microbial Institute for Fermentation Industry (MIFI)) ;
  • Do-Youn Jeong (Microbial Institute for Fermentation Industry (MIFI)) ;
  • Hee-Jong Yang (Microbial Institute for Fermentation Industry (MIFI))
  • 정호진 ((재)발효미생물산업진흥원) ;
  • 하광수 ((재)발효미생물산업진흥원) ;
  • 이란희 ((재)발효미생물산업진흥원) ;
  • 정도연 ((재)발효미생물산업진흥원) ;
  • 양희종 ((재)발효미생물산업진흥원)
  • Received : 2024.05.08
  • Accepted : 2024.06.17
  • Published : 2024.07.30

Abstract

The fermentation of ganjang is known to be greatly influenced by the microbial communities derived from its primary ingredients, meju and sea salt. This study investigated the effects of changes in meju size on the distribution and correlation of microbial communities in ganjang fermentation, to enhance its fermentation process. Ganjang was prepared using whole meju and meju divided into thirds, and samples were collected at 7-day intervals over a period of 28 days for microbial community analysis based on 16S rRNA gene sequencing. At the genus level, during fermentation, ganjang made with whole meju exhibited a dominance of Chromohalobacter (day 7), Pediococcus (day 14), Bacillus (day 21), and Pediococcus (day 28), whereas ganjang made with meju divided into thirds consistently showed a Pediococcus predominance over the 28 days. Beta-diversity analysis of microbial communities in ganjang with different meju sizes revealed significant separation of microbial communities at fermentation days 7 and 14 but not at days 21 and 28 across all experimental groups. The linear discriminant analysis effect size (LEfSe) was determined to identify biomarkers contributing to microbial community differences at days 7 and 14, showing that on day 7, potentially halophilic microbes such as Gammaproteobacteria, Firmicutes, Oceanospirillales, Halomonadaceae, Bacilli, and Chromohalobacter were prominent, whereas on day 14, lactic acid bacteria such as Pediococcus acidilactici, Lactobacillaceae, Pediococcus, Bacilli, Leuconostocaceae, and Weissella were predominant. Furthermore, correlation analysis of microbial communities at the genus and species levels revealed differences in correlation patterns between meju sizes, suggesting that meju size may influence microbial interactions within ganjang.

간장의 발효는 원재료인 메주와 천일염에서 천이된 미생물 군집에 큰 영향을 받는 것으로 알려져 있다. 본 연구에서는 간장의 발효 고도화를 위하여 메주의 크기 변화가 간장의 미생물 군집에서 분포 변화와 분포 상관관계에 대하여 미치는 영향을 조사하였다. 메주를 전체, 삼등분하여 간장을 제조하였으며, 28일간 발효를 진행하면서 7일 간격으로 간장 시료를 수집하여 16S rRNA 유전자 기반의 미생물 군집 분석을 진행하였다. 속 수준에서 발효가 진행되면서 메주를 전체 사용하여 제조한 간장은 Chromohalobacter (7일), Pediococcus (14일), Bacillus (21일), Pediococcus (28일) 순으로, 메주를 삼등분하여 제조한 간장은 28일 연속으로 Pediococcus가 가장 우점하였다. 메주 크기를 달리한 간장들의 미생물 군집의 beta-diversity 분석 결과, 발효 7일 차와 14일 차에서 메주 크기를 달리하여 제조한 간장의 미생물 군집 분리가 유의미한 차이를 가진다고 보여주었으나 발효 21일 차와 28일 차에는 모든 실험구의 미생물 군집은 분리되지 않았다. 미생물 군집의 분리가 시각적으로 나타난 발효 7일 차와 14일 차에서 미생물 군집 차이를 부여하는 바이오마커를 조사하기 위해 선형 판별 효과 크기 분석을 수행하여 LDA size 별로 정렬하였으며, 발효 7일차에서는 Gammaproteobacteria, Firmicutes, Oceanospirillales, Halomonadaceae, Bacilli, Chromohalobacter 순으로 호염성 미생물군이, 발효 14일 차에서는 Pedioccoccus acidilactici, Lactobacillaceae, Pediococcus, Bacilli, Leuconostocaceae, Weissella 순으로 유산균군이 미생물 군집 차이를 나타내는 것으로 조사되었다. 또한 속과 종에 속하는 바이오마커 미생물군을 주요 미생물로 간주하여 상관관계를 분석한 결과, 메주 크기 별로 미생물군집 간 상관관계가 차이가 있으며, 이에 따라 메주 크기가 간장 내 미생물 간 상호작용에 영향을 미칠 수 있는 것을 확인하였다.

Keywords

서론

간장은 메주와 염수를 활용한 대표적인 2차 발효 장류 식품으로서, 대부분의 한식에서 필수 아미노산 섭취와 더불어 감칠맛과 특유의 향미를 부여하고 항산화, 항암, 면역 증진 등 다양한 기능성을 가지고 있는 것으로 보고되었다[15, 17]. 하지만 원재료인 메주의 제조에 약 2개월, 간장의 발효 및 숙성에 약 6개월이 소요되며, 간장의 발효에는 pH, 온도, 습도 같은 환경 조건 외로 메주 및 염수에서 천이된 미생물이 품질 및 식품 안전성에 영향을 가하는 것으로 알려져 있어 이를 관리하기 위한 저장 및 유지비용 등 초기 자본 부담과 불안정한 수익 회수 같은 문제로 인해 산업 성장에 애로사항이 많다[7, 10, 22]. 또한 간장을 포함한 여러 발효식품에서 histamine, tyramine, tryptamine, phenylethylamine 등의 biogenic amine과 aflatoxin을 비롯한 곰팡이 독소인 mycotoxin이 검출되어 식품 안전성에 의문이 제기되고 있다[18, 21, 25].

발효식품의 생산 구조 및 식품 안전성 문제 등 다양한 산업 애로사항을 대응하기 위해 간장의 생산 고도화가 필요하나 최근 국내 간장의 연구 동향으로는 메주 크기 및 다양성에 따른 이화학적 품질특성[4, 19], 약용버섯, 양파추출물, 고수 등의 다양한 첨가물을 이용한 기능성 강화[15, 17, 21], 미생물 군집 탐색[7, 14], 식품 안전성 조사[18, 25] 등으로, 간장 발효의 주요한 요인은 미생물이지만 대부분 이화학적 품질특성과 기능성 평가에 관한 연구가 주를 이루어 간장의 발효 고도화를 위한 미생물 군집 연구는 심화연구를 위한 기반 데이터 구축이 더욱 필요한 실정이다. 또한 간장 제조에 사용되는 원재료인 메주의 변화가 간장에 미치는 이화학적 영향은 분석이 되어있으나[4, 19] 미생물 군집에 미치는 영향에 대한 조사는 부족하다.

본 연구에서는 간장 발효의 원재료인 메주의 크기에 따른 세균 미생물 군집을 MiSeq 플랫폼 기반의 차세대염기서열분석(Next generation sequencing, NGS)을 통해 분석하여 메주의 물리 표면적 변화가 간장의 발효에 미치는 영향을 조사하여 간장 생산성 향상의 미생물학적 자료를 마련하고자 한다.

재료 및 방법

간장 시료의 제조 및 total genomic DNA 추출

미생물 군집 분석을 위한 간장 시료는 원재료인 메주를 성형한 후 전체 15×10×10(L×W×H, cm), 삼등분 5×10×10(L×W×H, cm)으로 메주 크기 조건만 변경하여 이전의 연구방법[14]과 동일하게 제조하였으며, 담금 과정 이후부터 7일 간격으로 액상 부분인 간장을 1 ml 씩 회수하여 균질화 시킨 뒤, DNA 추출 및 미생물 군집 분석을 위한 시료로 사용하였다. 간장 시료를 회수하는 발효기간 동안 고상부분인 메주는 분리하지 않았다. 해당 간장 시료로부터 미생물 군집을 분석하기 위한 DNA 추출은 QIAGEN사의 PowerFood Microbial Kit (QIAGEN, Hilden, Germany)를 이용하여 해당 제품의 이용 가이드에 따라 추출하였으며, 추출한 DNA는 0.8% agarose gel 전기영동 수행을 통해 DNA의 분절 정도를 확인하였다. 또한 추출한 각 DNA가 16S rRNA 유전자 library를 제작하기에 적합한지 확인하기 위하여 품질검사를 수행하였으며, 추출한 각 DNA의 농도를 정밀하게 측정하기 위해 Qubit 4 (Invitrogen, Waltham, Massachusetts, USA)를 사용하였고 각 DNA의 순도측정을 위해서 Nanodrop One 장비를 이용하였다.

간장 시료의 16S metagenomic library 제작

간장 시료의 미생물 군집분석을 수행하기 위해 16S rRNA library를 제작하였으며, Illumina사에서 배포한 16S metagenomic sequencing library preparation guide [11]에 의거하여 16S rRNA 유전자의 V3-V4 구역을 증폭하기 위하여 1차 PCR을 진행하였다. 1차 PCR 수행을 위한 PCR premix는 universal primer set (forward : 5'-TCG TCG GCA GCG TCA GAT GTG TAT AAG AGA CAG CCT ACG GGN GGC WGC AG-3', reverse : 5'-GTC TCG TGG GCT CGG AGA TGT GTA TAA GAG ACA GGA CTA CHV GGG TAT CT A ATC C-3', N; A or C or G or T, W; A or T, H; A or C or T, V; A or C or G)와 2× KAPA HiFi HotStart Ready Mix (Roche, Basel, Switzerland)를 혼합하여 이전의 연구방법[7]과 동일하게 25회 PCR을 진행하여 1차 PCR 산물을 획득하였으며, 불순물을 제거하기 위해 AMpure XP bead(Beckman Coulter, Brea, California, USA)를 이용하였다. 각 sample에서 증폭된 raw read들을 구분하기 위해 index primer로 Nextera XT Index kit v2 (Illumina)를 사용하였으며, 증폭된 library에 index를 결합시키기 위한 2차 PCR은 이전의 연구방법[7]과 동일하게 8회 PCR을 진행하였다. 2차 PCR 종료 후, AMpure XP bead를 이용하여 불순물을 제거하여 최종 증폭산물만 남겨 미생물 군집분석을 위한 16S rRNA 유전자 library를 제작하였다. 최종 16S rRNA 유전자 library의 품질검사를 위해 1.5% agarose gel에 전기영동을 진행하여 library의 크기를 조사하였으며, Nanodrop One 장비와 Qubit 4 장비를 사용하여 각 library의 순도와 농도를 분석하여 염기서열 분석을 진행하였다.

MiSeq를 통한 16S metagenomic sequencing

각 library를 10 mM Tris-HCl (pH 8.5) buffer로 4 nM 수준의 농도로 희석한 뒤 동일 부피로 혼합하여 정규화를 수행하였으며, HT1 buffer (Illumina)를 사용하여 최종적으로 7 pM이 되도록 library를 제작하였다. 염기서열 신호 중첩 오류 보정 및 분석 품질 확인을 위하여 PhiX (Illumina)를 대조군으로 활용하여 7 pM의 library와 혼합하였으며[11], PhiX가 30% 포함되도록 만들어진 최종 combined library를 MiSeq reagent kit V3 (Illumina)의 cartridge에 주입하여 Illumina사의 MiSeq 플랫폼을 통해 2×301회의 paired-end sequencing을 수행하였다.

통계분석

MiSeq sequencing 이후 확보한 FASTQ 파일의 각 간장시료로부터 얻은 read에 대하여 생물 다양성 정보(alpha-diversity, beta-diversity), 미생물 분포 및 biomarker 분석은 EzBiocloud 16S-based microbiome taxonomic profiling (MTP) pipeline [30]을 이용하였다. 군집분석에 사용될 valid reads는 분석과 무관한 index 서열 및 chimeric, non-bacterial, low quality amplicon을 제거하였으며, 각 valid reads에서 분류학적 수준에서 OTUs를 기반으로 미생물을 분리하기 위하여 PKSSSU 4.0 database [6]를 활용하여 97% cut off 기준으로 분석하였다. Alpha-diversity 지표와 미생물 분포, Good’s coverage of library 수치는 Single MTP browser를사용하여 산출하였으며, 각 간장 시료간의 미생물 군집 차이를 확인하기 위한 beta-diversity 분석에는 UniFrac distance metric [20]을 기반으로 comparative MTP analyzer를 사용하여 principal coordinates analysis (PcoA), UPGMA-clustering, permutational multivariate analysis of variance(PERMANOVA) 분석[26]을 진행하였다. 또한 같은 도구를 이용하여 각 간장 시료 간 유의미한 차이를 나타내는 미생물학적 biomarker 탐색을 위하여 linear discriminant analysis effect size (LefSe) 분석[23]을 수행하였다. 각 간장 군집 내 주요 미생물간의 분포 상관관계를 조사하기 위하여Pearson correlation coefficient [1]를 산출하여 분석하였다.

결과 및 고찰

발효 기간에 따른 메주 크기 별 간장의 alpha-diversity 분석

간장 시료 제조에 사용된 메주 크기 별 조건과 간장시료로부터 얻은 raw read의 trimming 후의 결과에 대한 요약을 Table 1에 정리하였다. 메주 전체를 사용한 시료를 WM, 메주를 삼등분하여 사용한 시료를 TM으로 실험구를 명명한 뒤, 각각 7일 간격으로 sampling 하여 총 8개의 실험구를 미생물 군집분석을 위한 시료로 사용하였다. 시료당 평균 91,088.09개의 valid reads를 획득하였으며, 모든 시료에서 99.81% 이상의 Good’s coverage of library 수치를 확보하여 각 시료로부터 획득한 vaild reads가 미생물 군집분석을 하기 위해 충분함을 확인하였다. 각 기간 별 alpha-diversity 지표를 분석한 결과(Table 2), 종 풍부도를 나타내는 Chao1 지표와 종 추정치 인 OTUs는 발효 7일 차와 14일 차에서 WM과 TM은 통계적으로 유의한 차이를 가졌으나, Chao1과 유사하게 종 풍부도를 나타내는 또 다른 지표인 ACE는 p-value가 0.05보다 높은 수치를 가져 통계적으로 유의미한 차이를 나타내지 않았다. Chao1, ACE 지표 모두 희귀 종을 고려하나 ACE의 산출에 풍부 종의 표본 수가 강조되어 나타난 결과로 판단되며[2], 이에 따라 종 추정치 및 풍부도는 발효 7일 차에서는 TM이 조금 더우세하고 발효 14일 차에서는 WM이 조금 더 우세한 것으로 나타났다. 종 균등성 및 다양성을 나타내는 Shannon, Simpson과 phylogenetic diversity 지표 또한 발효 7일 차와 14일 차에서 WM과 TM은 통계적으로 유의한 차이가 있는 것으로 나타났다. Shannon 지표는 종 수의 상대적 풍부도를 고려한 지표로 값이 클수록 다양성이 높고 Simpson 지표는 하나의 sample에서 두 read를 임의로 선택하였을 때 같은 종일 확률을 측정한 값으로 0과 1 사이 값을 가지고 값이 작을수록 다양성이 높으며, phylogenetic diversity는 종 간 계통학적 차이를 수치화한 지표로 계통도의 node간 최단거리를 합산하여 계산되며, 값이 클수록 다양성이 높은 것을 의미한다[27]. 이에 따라 WM과 TM 모두 종균등성 및 다양성은 발효 21일 차까지 점차 상승하나, 발효 28일 차에서 감소하는 경향이 나타났다.

Table 1. Information of samples used in this study and record of valid reads left after trimming

1)Ganjang made from the whole meju

2)Ganjang mdae from the meju divided into thirds

Table 2. Summary of alpha-diversity indices indicating species richness, diverstiy and evenness

1)Ganjang made from the whole meju

2)Ganjang mdae from the meju divided into thirds

3)Significance on each of alpha-diversity indices between WM and TM, *p<0.05

4)The abundance-based coverage estimator

발효 기간에 따른 메주 크기 별 간장의 미생물 분포 변화 분석

메주 크기를 달리하여 제조한 간장의 미생물 분포 변화를 분석하기 위해 28일의 발효기간 동안 7일 간격으로 4번 sampling하여 미생물 군집을 분석하였으며, 각 군집의 미생물의 상대적 분포를 생물학적 분류 수준에 따라 조사한 결과를 Table 3에 정리하였고 Fig. 1과 2에 시각화 하였다. 문(phylum) 수준에서, WM과 TM group에서 우점한 문은 Firmicutes와 Proteobateria 였으며, WM에서 Firmicutes 문은 9.50%(7일), 94.00%(14일), 71.52%(21일), 82.45%(28일)로, Proteobacteria 문은 90.16%(7일), 5.24%(14일), 22.99%(21일), 15.60%(28일)로 발효기간에 따라 상대적 분포 비율이 변화하였다. TM의 경우, Firmicutes 문은 71.61%(7일), 96.79%(14일), 87.96%(21일), 90.23%(28일)로, Proteobacteria 문은 26.79%(7일), 2.75%(14일), 7.22%(21일), 3.96%(28일)로 발효기간에 따라 상대적 분포 비율이 변화하였다. 속(genus) 수준에서, WM과 TM group에서 우점한 속은 Chromohalobacter, Pediococcus, Bacillus, Enterococcus, Weissella로 대부분 유산균 속으로 나타났다. WM에서 Chromohalobacter 속은 88.90%(7일), 3.29%(14일), 10.44%(21일), 8.27%(28일)로, Pediococcus 속은 3.92%(7일), 42.96%(14일), 27.66%(21일), 35.41%(28일)로, Bacillus 속은 2.83%(7일), 27.49%(14일), 27.83%(21일), 28.73%(28일)로 발효기간에 따라 상대적 분포 비율이 변화하였다. TM의 경우, Chromohalobacter 속은 25.06%(7일), 2.24%(14일), 5.35%(21일), 2.96%(28일)로, Pediococcus 속은 35.78%(7일), 51.56%(14일), 40.21%(21일), 48.82%(28일)로, Bacillus 속은 20.20%(7일), 26.20%(14일), 28.51%(21일), 32.93%(28일)로 발효기간에 따라 상대적 분포 비율이 변화하였다. WM에서 발효 14일 차에 급격한 미생물 군집구조 변화가 관찰되었으나 TM은 발효기간 전반에 걸쳐서 Pediococcus 속, Bacillus 속, Enterococcus 속이 빠르게 군집에 우점되어 군집구조가 정착되는 경향을 보여주었다. 종 수준에서 미생물 군집구조는 속 수준과 유사한 구조 변화를 보여주었으며, WM이 TM보다 비교적 같은 속에 포함되는 미생물종이 더 다양한 것으로 나타나 alpha-diversity 지표와 동일한 결과를 확인할 수 있었다. 과거 연구 결과[19]에 따르면 메주 크기가 작아질수록 간장의 총 질소에는 큰 영향을 미치지 않으나 유리당은 증가하는 경향성이 있다고 보고하였으며, 메주 크기에 따라 간장의 pH와 염도 등 다양한 환경 조건이 변화하여 미생물 생육에 영향을 미치는 것으로 판단된다. 추가적인 이화학적 분석 연구가 미생물 군집 분석과 연계하여 수행된다면 메주 크기 조건이 간장의 미생물 군집구조 차이에 미치는 영향을 면밀히 규명할 수 있을 것으로 생각된다.

Table 3. Compositonal variation of bacterial community of ganjang depending to the size of meju by fermentation period

1)Taxon excluding anything that exceeds 1% in either of the two groups

2)Ganjang made from the whole meju

3)Ganjang mdae from the meju divided into thirds

Fig. 1. Microbial community composition and relative abundance on genus level in ganjang made from the whole meju (WM) and meju divided into thirds (TM).

Fig. 2. Microbial community composition and relative abundance on species level in ganjang made from the whole meju (WM) and meju divided into thirds (TM).

메주 크기 별 발효 기간에 따른 간장의 미생물 군집 beta-diversity 분석

메주 크기를 달리 하여 제조한 간장들의 미생물 군집 구조가 발효의 진행에 따라 차이가 나타나는 지 조사하기 위해 속 수준과 종 수준에서 UPGMA-clustering 분석과 UniFrac distance metric 기반 principal coordinate analysis(PCoA) 분석을 수행하였다(Fig. 3, 4). UPGMA-clustering 및 PcoA 분석 결과, 속 수준과 종 수준 모두 동일하게 발효 7일 차와 14일 차에서 WM과 TM 그룹 간의 명확한 분리를 확인할 수 있었으나, 발효 21일 차와 28일 차에서는 미생물 군집의 분리가 이루어지지 않은 것으로 판단된다. 또한 속 수준 및 종 수준 모두 TM의 산포 범위는 WM의 산포 범위보다 좁은 것으로 확인되었으며, TM이 WM보다 비교적 미생물 다양성이 적고 특정 군집구조에 정착되어있음을 시사하여 composition 결과에 유사하게 분석되었다. 시각적으로 확인된 발효 기간에 따른 미생물 군집 구조에 메주 크기 변화가 미치는 양상이 통계적으로 분석하기 위하여 PERMANOVA을 진행하여 Table 4와 5에 정리하였다. 발효 진행에 따른 메주 크기 별 미생물 군집간 차이에 대한 p-value를 확인한 결과, 발효 7일 차와 14일 차에서 WM과 TM의 미생물 군집의 산포 차이는 통계적으로 유의미하였으나 발효 21일 차와 28일 차에서는 산포 차이가 통계적으로 유의미하지 않았다(Table 4). 각 group에서 발효기간 별로 미생물 군집 간 차이에 대한 통계분석을 확인한 결과, WM은 발효 21일 차와 28일 차 간의 산포 차이가 통계적으로 유의미하지 않았으며, TM의 경우 발효 14일 차와 28일 차, 발효 21일 차와 28일 차간의 산포 차이가 통계적으로 유의미하지 않았다(Table 5). Pseudo F는 산포 차이를 수치화하며, 값이 작을수록 산포 차이가 적다는 것을 의미한다[3, 29]. Pseudo-F 값과 p-value 값을 종합적으로 고려하였을 때, WM은 발효 21일 차 이후로 미생물 군집 구조가 안정화된 것으로 나타나고 TM은 발효 14일 차 이후로 미생물 군집 구조가 안정화되는 것으로 보이며, 비교적 발효기간 동안 Pseudo F 값의 격차가 적었던 TM이 빠르고 안정적으로 미생물 군집구조가 구축되는 것으로 판단된다.

Fig. 3. UPGMA clustering and Principal coordinate analysis (PCoA) result based on UniFrac distance metric of microbial communities on genus level (red, ganjang made from the whole meju, WM; green, ganjang made from the meju divided into thirds, TM).

Fig. 4. UPGMA clustering and Principal coordinate analysis (PCoA) result based on UniFrac distance metric of microbial communities on species level (red, ganjang made from the whole meju, WM; green, ganjang made from the meju divided into thirds, TM).

Table 4. Results of PERMANOVA showing the statistical significance of ganjang in ecah day on bacterial community structure of genus level

1)Significance on bacterial distribution between WM and TM, *p<0.05

2)Ganjang made from the whole meju

3)Ganjang mdae from the meju divided into thirds

Table 5. Results of PERMANOVA showing the statistical significance of fermentation period in ecah ganjang group on bacterial community structure of genus level

1)Significance on bacterial distribution between WM and TM, *p<0.05

2)Ganjang made from the whole meju

3)Ganjang mdae from the meju divided into thirds

메주 크기 별 발효 기간에 따른 간장의 taxonomic biomarker 및 상관관계 분석

메주 크기를 달리하여 제작한 간장에서 미생물 군집 차이가 유의미하였던 발효 7일 차와 14일 차에서 미생물 군집 구조의 상대적 분포비율을 기반으로 통계적으로 유의미한 차이를 부여하는 미생물을 조사하기 위해 LEfSe 분석을 수행하였으며, taxonomic biomarker를 선형 판별 분석 점수(LDA effect size, Linear discriminant analysis effect size)가 높은 순으로 상위 20종을 나타내었다(Table 6, 7). 발효 7일 차의 경우, Gammaproteobacteria, Firmicutes, Oceanospirillales, Halomonadaceae, Bacilli, Chromohalobacter 순으로 나열되어, 호염성 미생물이 속하는 분류군이 각 메주 크기 별 간장의 미생물 군집의 상대적 분포 비율 차이가 비교적 크게 나타나 발효 초기에 영향을 줄 것으로 예상된다. 발효 14일 차의 경우, Pedioccoccus acidilactici, Lactobacillaceae, Pediococcus, Bacilli, Leuconostocaceae, Weissella 순으로 나열되어, 발효 7일 차와 달리 유산균에 속하는 분류군이 각 메주 크기 별 간장의 미생물 군집에서 상대적 분포 비율이 분명하게 차이가 나타나 발효가 진행됨에 따라 유산균이 미생물 군집 구조 형성에 중요한 역할을 할 것으로 판단된다. 발효 7일 차와 14일 차의 상위 20종의 taxonomic biomarker에서 속과 종에 포함된 미생물군을 주요한 미생물로 간주하여 해당 미생물군 간의 상호작용을 분석하기 위하여 발효 기간 전체에 대하여 분포 상관관계(Pearson correlation coefficient)를 분석하였다(Fig. 5). 속 수준에서 WM과 TM 모두 공통적으로 Chromohalobacter-Pediococcus, Chromohalobacter-Bacillus, Chromohalobacter-Enterococcus간의 관계가 강한 음의 상관관계를 나타내었으며, WM보다 비교적 TM에서 미생물 간 상관관계가 더 복잡하게 구축되는 것으로 나타났다. TM에서 Weissella 속은 유산균에 속하는 미생물이지만 Chromohalobacter 속과 약한 양의 상관관계를 나타내었는데 이는 TM의 미생물 군집이 Pediococcus, Bacillus, Enterococcus 속이 빠르게 우점하면서 생긴 결과로 판단되며, 이러한 경향성은 종 수준에서도 유사하게 분석되었다. 간장의 미생물 군집 속 Bacillus, Enterococcus, Pediococcus, Weissella 속은 메주에서, Chromohalobacter, Halomonas 속은 천일염에서 기인된 것으로 보고되었으며[8], 메주의 발효에 Bacillus 속이 주요한 미생물인 반면, 간장에서는 상대적으로 Bacillus 속의 분포 비율이 낮고 호염성 또는 내염성 미생물 군집들이 더 우점한다는 기존 연구[5, 8, 12]와 유사한 결과를 얻었다. 간장의 발효과정 중 젖산 발효가 주요한 역할을 하는 것으로 알려져 있어[5] 이와 관련된 호염성 또는 내염성 유산균인 Pediococcus, Enterococcus, Weissella 속은 간장의 발효에 주요한 역할을 하는 미생물 군집으로 판단된다. 또한 Pediococcus, Enterococcus, Weissella 속은 이형젖산발효를 하는 것으로 알려져 있어[9, 24] 젖산뿐만 아니라 아세트산, 알코올류 등의 다른 대사물질도 만들어 간장 풍미에 영향을 줄 것으로 여겨진다. Chromohalobacter, Halomonas 속은 호염성 미생물로, Halomonas 속은 대체로 비병원성을 띤다고 알려져 있지만[28], 일부 Chromohalobacter 속은 특정 환경에서 biogenic amines을 생성한다고 보고되어[13, 16, 21] 식품 안전성 관리를 위해 환경 조건 및 미생물 간 상관관계를 활용하여 유익군과 유해군의 분포를 발효과정 동안 지속적으로 관리해야 할 필요성이 있다고 판단된다. 본 연구는 메주 크기를 달리 하여 간장을 제조하였으며, 각 간장의 미생물 군집 분석을 통하여 미생물 분포 변화와 분포 상관관계 분석을 수행하여 메주의 물리적 크기 변화가 간장 발효에 미치는 영향에 대하여 미생물학적 기초자료를 확보하였다. 해당 연구를 기반으로 메주의 크기, 형태 등의 다양한 조건에서 미생물 군집 분석뿐만 아니라 추가적인 연구로 pH, total nitrogen, aflatoxin, biogenic amine 등의 간장의 품질 및 안전성 특성 분석이 진행된다면 식품 안전성이 확보되면서 발효기간을 단축시킨 간장의 고도화와 제품 개발에 기여할 수 있을 것으로 기대된다.

Table 6. Taxonomic biomarker analysis in 7 day using LEfSe analysis

1)Significance on relative abundance between WM and TM, *p<0.05

2)Ganjang made from the whole meju

3)Ganjang mdae from the meju divided into thirds

Table 7. Taxonomic biomarker analysis in 14 day using LEfSe analysis

1)Significance on relative abundance between WM and TM, *p<0.05

2)Ganjang made from the whole meju

3)Ganjang mdae from the meju divided into thirds

Fig. 5. The Pearson correlation coefficient (r) matrix heatmap among core microorganisms in ganjang produced with different meju size. The p-value is investigated for significance in the linear correlation between core microorganisms, and the data with p-value higher than 0.05 is not indicated. Core genus on ganjang made from the whole meju, WM (A); core species on ganjang made from the whole meju, WM (B); core genus on ganjang made from the meju divided into thirds, TM (C); core species on ganjang made from the meju divided into thirds, TM (D) (*p<0.05).

감사의 글

본 연구는 2023년 농림축산식품부 농축산·식품마이크로바이옴 통합 바이오 뱅크구축사업의 지원에 의해 수행된 것입니다.

The Conflict of Interest Statement

The authors declare that they have no conflicts of interest with the contents of this article.

References

  1. Benesty, J., Chen, J. and Huang, Y. 2008. On the importance of the pearson correlation coefficeint in noise reduction. IEEE Trans. Audio. Speech. Lang. Process. 16, 757-765.
  2. Brose, U. and Martinez, N. D. 2004. Estimatig the richness of species with variable mobility. Oikos. 105, 292-300.
  3. Cho, E. and Hwang, I. 2008. The effective criteria for determining the number of clusters: an application with SPSS and microsoft excel. J. Prod. Res. 26, 1-9.
  4. Choi, J., Kim, M., Shon, M. Y., Park, S. K., Choi, S. D. and Hong, U. 2002. Production and quality properties of capsule type meju prepared with Rhizopus oligosporus. Kor. J. Food Preserv. 9, 315-320.
  5. Chun, B. H., Han, D. M., Kim, H. M., Park, D. B., Jeong, D. M., Kang, H. A. and Jeon, C. O. 2021. Metabolic features of ganjang (a Korean traditional soy sauce) fermentation revealed by genome-centered metatransciptomics. mSystems 6, e00441-21.
  6. Chun, J. S., Lee, J. H., Jung, Y. Y., Kim, M. J., Kim, S. I., Kim, B. K. and Lim, Y. W. 2007. EzTaxon: a web-based tool for the identification of prokaryotes based on 16S ribosomal RNA gene sequences. Int. J. Syst. Evol. Microbiol. 57, 2259-2261.
  7. Ha, G. S., Jeong, H. J., Noh, Y. J., Kim, J. W., Jeong, S. J., Jeong, D. Y. and Yang, H. J. 2022. Comparative micro- biome analysis of and microbial biomarker discovery in two different fermented soy products, doenjang and ganjang, using next-generartion sequencing. J. Life Sci. 32, 803-811.
  8. Han, D. M., Chun, B. H., Feng, T., Kim, H. M. and Jeon, C. O. 2020. Dynamics of microbial communities and metabolites in ganjang, a traditional Korean fermented soy sauce, during fermentation. Food Microbiol. 92, 103591.
  9. Han, D. M., Chun, B. H., Kim, H. M. and Jeon, C. O. 2021. Charaterization and correlation of microbial communities and metabolite and volateile compounds in doenjang fermentation. Food Res. Int. 148, 110645.
  10. Hur, J. Y., Kim, M. J., Hong, S. P. and Yang, H. J. 2020. Anticancer effects of ganjang with different aging periods. J. Kor. Soc. Food Cult. 35, 215-223.
  11. Illumina. 2013. 16S metagenomic sequencing library preparation protocol: Preparing 16S ribosomal RNA gene amplicons for the Illumina MiSeq system. Part no. 15044233 Rev B. Illumina, San Diego, CA.
  12. Jang, K. T., Lee, I. S. and Lee, M. H. 2023. Microbial identification and network analysis of the Korean traditional ganjang. Culin. Sci. Hosp. Res. 29, 59-67.
  13. Jeong, D. W., Heo, S. J. and Lee, J. H. 2017. Safety assessmetn of Tetragenococcus halophilus isolates from doenjang, a Korean high-salt-fermented soybean paste. Food Microbiol. 62, 92-98.
  14. Jeong, H. J., Ha, G. S., Lee, J. M., Song, Y. J., Jeong, D. Y. and Yang, H. J. 2023. Diversity and succession of the bacterial community during the initial fermentation period in modernized soy sauce (ganjang). J. Life Sci. 33, 481-489.
  15. Jeong, J. H., Lee, C. M., Kim, S. H., Choi, Y. R., Lee, D. H., Lee, C. Y., Choi, B. K. and Huh, C. K. 2023. Enhancing the quality, characteristics, and antioxidant activity of Korean soy sauce made using lactic acid bacteria fermentation-based onion extract. Kor. J. Food Sci. Tech- nol. 55, 258-265.
  16. Jung, J. Y., Chun, B. H. and Jeon, C. O. 2015. Chromohalobacter is a causing agent for the production of organic acids and putrescine during fermentation of ganjang, a Korean traditional soy sauce. J. Food Sci. 80, M2853.
  17. Jung, Y., Park, S. Y., Park, S, H., Yeom, E. S., Jeong, W. Y., Kim, S. M. and Song, H. N. 2023. Antioxidant and antibiotic acitivity of seasoned soy sauce added with medicinal mushrooms and oriental medicinal plants. Kor. J. Food Nutr. 36, 033-041.
  18. Kim, J., Jeong, J., Jang, M., Kim, J. C. and Lee, H. 2023. Comparative analysis of microbial and mycotoxin contamination in korean traditional soybean paste and soy sauce production with and without starter. Fermentation 9, 621.
  19. Lee, J. G., Kwon, K. I., Choung, M. G., Kwon, O. J., Choi, J. Y. and Im, M. H. 2009. Quality analysis on the size and the preparation method of meju for the preparation of Korean traditional soy sauce (kanjang). J. Appl. Biol. Chem. 52, 205-211.
  20. Lozupone, C. and Knight, R. 2005. UniFrac: a new phylogenetic method for comparing microbial communities. Appl. Environ. Microbiol. 71, 8228-8235.
  21. Mannaa, M., Seo, Y. S. and Park, I. M. 2020. Addition of coriander during fermentation of Korean soy sauce (gangjang) causes significant shift in microbial composition and reduction in biogenic amine levels. Foods 9, 1346.
  22. Na, H. J., Cho, S. H. and Jeong, D. Y. 2020. Current status of the Jangryu industry and future development direction. Food Sci. Nutr. 53, 183-199.
  23. Segata, N., Izard, J., Waldron, L., Gevers, D., Miropolsky, L., Garrett, W. S. and Huttenhower, C. 2011. Metagenomic biomarker discovery and explanation. Genome Biol. 12, R60.
  24. Sharma, R., Garg, P., Kumar, P., Bhatia, S. K. and Kulshrestha, S. 2020. Microbial fermentation and its role in quality improvement of fermented foods. Fermentation 6, 106.
  25. Song, D. H., Chun, B. H., Lee, S., Son, S. Y., Reddy, C. K., Mun, H. I., Jeon, C. O. and Lee, C. H. 2021. Comprehensive metabolite profiling and microbial communities of doenjang (fermented soy paste) and ganjang (fermented soy sauce): a comparative study. Foods 10, 641.
  26. Tang, Z. Z., Chen, G. and Alekseyenko, A. V. 2016. PERMANOVA-S: Association test for microbial community composition that accommodates confounders and multiple distances. Bioinformatics 32, 2618-2625.
  27. Thukral, A. K. 2017. A review on measurement of alpha diversity in biology. Agric. Res. J. 54, 1-10.
  28. Vreeland R. H. 2015. Halomonas. Bergey's manual of systematics of archaea and bacteria 1-19.
  29. Wilkinson, L., Engelman, L., Corter, J. and Coward, M. 1999. Cluster analysis. sas/stat user's guide. 8, 95-135.
  30. Yoon, S. H., Ha, S. M., Kwon, S. J., Lim, J. M., Kim, Y. S., Seo, H. S. and Chun, J. S. 2017. Introducing EzBio Cloud: A taxonomically united database of 16S rRNA and whole genome assemblies. Int. J. Syst. Evol. Microbiol. 67, 1613-1617.