Ebola: Symptoms, treatment, and causes
Security Guard Accused Of Holly Willoughby Murder Plot Searched 'how To Meet People Who Plan To Kidnap Celebrities' More Than A Decade Before His Arrest, Court Hears
The man accused of plotting to rape and murder Holly Willoughby had searched online about abducting famous people more than a decade before his arrest, a court heard yesterday(TUES).
Gavin Plumb, 37, allegedly looked up 'how to meet people who plan to kidnap celebrities' in September 2011.
Over the next few years he looked up a range of phrases including 'what does it feel like to be raped' and 'where (sic) Jewish women raped in the war', jurors were told.
The first known reference to Miss Willoughby in electronic devices seized from the security guard's flat in Harlow, Essex, was in 2021 when he is said to have looked up fake images on a free porn website.
However, a police officer giving evidence said there were 'millions' of images of her and other celebrities and police had only managed to go through about ten per cent of them.
A court artist's drawing of Gavin Plumb (right) at Chelmsford Crown Court in Essex yesterday
Plumb also described tying up married mother-of-three Miss Willoughby's family during the raid at her home, it is claimed.
He is accused of planning to enter the former This Morning presenter's London property with other 'obsessive' fans, who would then move her to a 'dungeon-style room' elsewhere in the country where no one could hear her screams and 'repeatedly rape' her before murdering her and disposing of her body in a lake.
Plumb has previous convictions for attempting to kidnap two women on trains in separate incidents in 2006, Chelmsford Crown Court has heard.
He was jailed for false imprisonment following another incident in 2008 when he threatened two 16-year-old girls with a knife and tried to tie them up at the Woolworths branch near his home where they all worked.
The jury was taken through a sequence of events yesterday by prosecutor Alison Morgan KC and Essex Police officer Detective Constable William Belsham, who helped sift through the defendant's electronic devices.
The document chronologically details the prosecution's case against Plumb.
The father-of-one began chatting online with an individual called Marc in late 2021, with the man allegedly writing: 'We do a home invastion, tie the whole family up and take her.
'Could even… burn the house down or shoot the rest of her family.'
Plumb is said to have discussed Miss Willoughby's security, saying: 'I know she doesn't have security unless she's going to a major TV event.'
Holly Willoughby said in October last year that she was stepping down from This Morning
The balding defendant allegedly also discussed restraining the entire family, saying to Marc: 'We'll wait for him [husband TV producer Dan Baldwin, 49) to leave then strike… or we'll tie them up as well before taking her… pick out some of her outfits… we'll be masked up.'
He added: 'I'm actually looking forward to doing it. I'm at the point where I don't care about the risks of consequences.'
The men discussed access routes to the Dancing on Ice presenter's home, the jury heard, with Plumb referring to an 'ambush point'.
Among the 'numerous' images of her that Plumb shared with Marc were deep fake porn photos, as well as graphic images showing women 'tied up and/or gagged'.
In January 2022, the defendant allegedly talked about staking out Miss Willoughby's home, adding: 'We'll give her a birthday present she'll never forget.'
Marc said: 'If I ever do get caught, I'll be f****** admired by f****** people.'
The pair exchanged messages about other woman they wanted to abuse, it is claimed, and sent each other images of two female celebrities.
Marc allegedly commented: 'Celebrities they think they're full of themselves, they need to be f****** put into place.
Gavin Plumb is accused of hatching a plot to kidnap, rape and solicit the murder of the star
In another exchange, Marc reportedly messaged: 'Her (Miss Willoughby) in yellow looks hot.'
Plumb responded: 'There isn't anything she doesn't look hot in that's the point.'
The defendant also discussed cutting off his victim's hair or pulling her by it.
And he allegedly wrote: 'Consensual non consent – we'll make her say we're aloud (sic) to do whatever we want to her and record it.'
The defendant, who the Crown say was in contact with a number of people about the plot, was arrested in October last year after striking up a relationship with a man called David Nelson who turned out to be a US-based undercover police officer.
He is said to have sent him a video of a 'restraint kit' he had assembled which included handcuffs, a gag ball, chloroform and cable ties to prove he was serious about the plot.
Opening the case on Monday, Ms Morgan said: 'The defendant's online discussions reveal his real intention to carry out a plot to kidnap Holly Willoughby from her family home – to take her to a location where she would be raped repeatedly, before the defendant then intended to kill her.
'It was not just the ramblings of a fantasist. The defendant had carefully planned what he would do and how he would do it.'
Miss Willoughby pulled out of an appearance on This Morning following the arrest in October last year and quit the show a few days later.
Plumb denies soliciting kidnap, rape and murder.The trial continues.
9 Sports Journalists Headed Into Cincinnati Journalism Hall Of Fame
They were the faces of Cincinnati television sports for a generation of local viewers: Dennis Janson, Ken Broo, John Popovich and George Vogel.
And they'll be inducted together into the Greater Cincinnati Society of Professional Journalism Hall of Fame at a Sept. 16 dinner in the Bally Sports Club at Great American Ball Park.
A total of nine local sports journalists — from print, broadcast and new media — will be honored in the largest induction class since the Hall of Fame was established in 1990.
Also to be inducted are former ESPN SportsCenter anchor Game Day Communications founder Betsy Ross; "The Morning Line" online blogger and former newspaper columnist Paul Daugherty; Bengals Radio Network host Wayne "Box" Miller; WKRC-TV executive sports producer Kevin Barnett; and the late Enquirer baseball beat writer John Fay.
The ceremony will be held in conjunction with the chapter's 2024 Excellence In Journalism contest awards presentation. Tickets for the dinner and programs are $75 each.
Here are the nine honorees:
Courtesy WCPO-TV
John PopovichJOHN POPOVICH: "Popo," as everyone calls him, set the gold standard for sports storytelling during his 40 years at WCPO-TV (1979-2019). He oversaw daily sports coverage, created the Sunday night Sports Of All Sorts magazine show (as a live, one-hour program) in 1980, and for nearly 30 years, was half of the city's best sports TV team with Denny Janson.
Popovich took over the main sports anchor chair when Dennis Janson retired in 2013.
Courtesy WCPO-TV
Dennis JansonDENNIS JANSON: The Price Hill native and 1968 Elder High School graduate bounced around with a number of jobs at WSAI-AM, WKRC-AM and Dayton's WONE-AM and Channel 2 before quitting the business to manage the Chapter 13 bar in Mount Adams. That's what he was doing in 1977 when WKRC-TV needed someone to help ABC edit film of the Beverly Hills Supper Club fire in May 1977, which killed 165 people.
Soon he was doing sports for WKRC-TV's popular Nick Clooney-Ira Joe Fisher team. He jumped to WCPO-TV in 1984 and anchored sports until retiring in 2013.
Courtesy Betsy Ross
BETSY ROSS: The 1968 Connersville High School graduate made her Cincinnati TV debut as a WCPO-TV news reporter in 1981, after working in South Bend, Ind. Her love of sports prompted her to volunteer for Channel 9 sports assignments. After a stop in Indianapolis, she moved to New York to anchor SportsChannel America's nightly sportscast. She returned to Cincinnati to anchor at WLW-TV for six years, then made the big leap to ESPN in 1997 to anchor for ESPN News and SportsCenter.
Ross came back to Cincinnati again in 2000 to create Game Day Communications. She also has worked part-time as a WXIX-TV sports anchor; written a book in 2010, "Playing Ball with the Boys: The Rise of Women in Men's Sports"; and done public address announcing for University of Cincinnati women's basketball, soccer and lacrosse.
WAYNE "BOX" MILLER: Lifelong Bengals fan Wayne "Box" Miller got his dream job in 2018, when named host of the Bengals' Countdown To Kickoff pregame show, half-time show and post shows on the Bengals Radio Network, which he juggles around his job as St. Xavier High School's director of diversity, equity and inclusion.
Courtesy Wayne "Box" Miller
Wayne "Box" MillerThe 1973 Woodward High School graduate started as a print salesman for the Cincinnati Enquirer, New York Times, Family Circle and Golf Digest in the late 1970s until joining new WIZF-FM (The WIZ) in the mid-1980s. Miller, who got the nickname "Box" as a kid growing up in Avondale, started a sports marketing firm in 1988, but retuned to radio in 2000-08 when WDBZ-AM (1230 The Buzz) was launched. He's also appeared on WLWT-TV's Sports Rock Sunday night show.
Courtesy WKRC-TV
Kevin BarnettKEVIN BARNETT: First hired by WKRC-TV sports anchor Ken Broo in 1991, Kevin Barnett has served as executive sports anchor assisting sportscasters Broo, Brad Johansen, Ken Anderson, Walt Maher and Gary Miller.
Barnett created the "Friday Night Final" extended high school sports highlights show, the Sunday night Sports Authority show and Bengals Nation weekly program taped before a live audience (which won a Bronze Telly Award this year).
The former St. Bernard/Elmwood Place high school sports star has produced Channel 12's coverage of Reds Opening Day; Bengals preseason, playoff and Super Bowl specials; Luke Fickell's UC coach show; a FC Cincinnati preseason special; and Jim Beam Turfway Park specials.
Courtesy Paul Daugherty
Paul DaughertyPAUL DAUGHERTY: By his count "Doc," as he's known, has written more than 10,000 columns — many of those for the Cincinnati Post (1987-1994) and Cincinnati Enquirer (1994-2022). The Maryland high school graduate wrote for the Virginian-Pilot, Dallas Times-Herald and Newsday before coming to Cincinnati.
Although never shy at giving his opinion, which continues with his "The Morning Line" blog, Daugherty also liked to tell personal stories about his annual trips to North Carolina with his son Kelly, "the 'Kid Down The Hall,' " and his 2015 book "An Uncomplicated Life," about his daughter Jillian, who was born with Down syndrome.
KEN BROO: Ken Broo holds the distinction of anchoring sports for three TV stations in town — WLWT, WKRC-TV and WCPO-TV — in a broadcasting career which continues on WLW-AM as his retirement from local television in 2018.
After graduating from Ohio University in 1974, Broo worked at radio stations in Wilmington, Ohio, and New Castle, Pa., before coming to Cincinnati's WSAI-AM in 1976 to do sports with morning host Jim Scott. He did TV in Tulsa and Tampa before returning here to do his "Boos & Bravos" as sports anchor for WLWT-TV's top-rated news anchors Jerry Springer and Norma Rashid. He jumped to WKRC-TV in 1990 to anchor sports and do Bengals radio play-by-play until 1996.
He spent four years at WUSA-TV in Washington, D.C., before returning to WLWT-TV in 2000. He joined WCPO-TV when Janson retired in 2013 and worked with Popovich until retiring from TV in 2018. Broo continues to host WLW-AM's Sunday Morning Sports Talk and fill in for Bill Cunningham and Scott Sloan.
JOHN FAY: In his career spanning six decades at the Enquirer, most of John Fay's stories were as the paper's baseball beat writer while traveling with the Reds. Fay also was one of the rotating baseball writers who regularly chatted with Marty Brennaman during games on the Reds radio network.
Fay was a gifted, versatile writer who also wrote about the Cincinnati Bengals; UC, Xavier and Miami sports; high schools games; Olympic participants; the Thanksgiving Day race and many other events.
Fay left the paper for a couple years about 10 years ago and joined WCPO-TV, where he wrote for the station's website and hosted a podcast, then returned to the Enquirer. Fay died in August 2023, at age 66.
GEORGE VOGEL: The 1975 Georgetown High School graduate was an important part in launching WLWT-TV's high school and weekend sports coverage in 1987 with Thom Brennaman. He started at Channel 5 as a newsroom assignment editor in 1982 straight out of the University of Cincinnati. He worked with Channel 5 sportscasters Steve Physioc, Zip Rzeppa, Bill Brown, Todd Donoho, Tom Varrato, Steve Shapiro, Greg Hoard, Brandon Saho and Broo.
In 1990, when Brennaman and Broo quit, he was main sports anchor until Hoard was hired and Broo returned. He resumed the main gig when Broo left in 2013.
Courtesy George Vogel
George VogelYou can read profiles of each nominee here written by Tom McKee, retired WCPO-TV reporter and long-time Cincinnati SPJ board member, on the chapter's website.
Doors will open at 5:30 p.M. For the event. The Excellence In Journalism Awards presentations starts at 6 p.M., followed by a buffet dinner at 7 p.M. The induction ceremony begins at 8 p.M.
Tickets for the dinner and programs are $75 apiece. Reservations can be made here.
Sponsorship opportunities are available by contacting McKee at tmckee9@yahoo.Com or (513) 608-1782.
Gut Microbiome Signatures Associated With Type 2 Diabetes In Obesity In Mongolia
1 IntroductionIn Asia, lifestyle-related diseases are known to be associated with social problems. In particular, East Asia, where there have been rapid changes in diet owing to the introduction of an urban lifestyle, is facing an epidemic of metabolic diseases. In Asian people, however, metabolic diseases arise not only from diet but also from phenotypic and genotypic factors (Ramachandran et al., 2012; Mathis et al., 2022). Notably, East Asians tend to have higher abdominal obesity than other ethnic groups (Ramachandran et al., 2012); this is associated with their susceptibility to glucose homeostasis (Hu, 2011; Kodama et al., 2013) and lower pancreatic Ī²-cell mass, which predisposes them to impaired insulin secretion even in relatively lean subjects. In fact, the type 2 diabetes (T2D) risk is higher in lower-body mass index (BMI) Asian populations than in lower-BMI Western populations (Chan et al., 2009).
The gut microbiota plays a key role in metabolic control and homeostasis (Pinart et al., 2022; Zhou et al., 2022). It is reported that the gut microbiome exhibits lower alpha-diversity and a different community structure in obesity than non-obesity people (Therdtatha et al., 2021; Pinart et al., 2022). Gut microbiota dysbiosis, in which reduced gut barrier function enables gut microbes and their components to enter the bloodstream, is associated with chronic diseases, including T2D (Scheithauer et al., 2020; Yang et al., 2021; Mishra et al., 2023). In contrast, microbiota-secreted short-chain fatty acids (SCFAs), specifically acetate, propionate, and butyrate, play important roles in preventing metabolic disorders by maintaining both metabolic homeostasis and the gut barrier, and via their anti-inflammatory activity (Portincasa et al., 2022). Levels of Roseburia and Faecalibacterium, SCFA-producing bacteria, are reduced in patients with T2D (MartĆnez-LĆ³pez et al., 2022).
Mongolia, a landlocked East Asian country, exhibits a unique dietary habit characterized by high consumption of animal products and low consumption of plant matter, established traditionally under a nomadic lifestyle adapted to the dry climate (Bromage et al., 2020; Delgermaa et al., 2023). This leads to insufficient dietary intake of particular components, including fiber, folate, and vitamin D, and can lead to obesity (Bromage et al., 2020). In 2016, 20.4% of the population of Mongolia suffered from obesity (BMI > 30) (NCD Risk Factor Collaboration, 2020), the highest prevalence among East Asian countries. However, the incidence of lifestyle-related diseases such as diabetes and coronary heart disease is lower than expected; this is recognized as the "Mongolian paradox" (Torii et al., 2012). This is possibly explained by their unique dietary habit; extremely high consumption of dairy products, particularly fermented animal milk with probiotic effect, and also supplementary consumption of medical plants and herbs with antioxidant effects (Torii et al., 2012).
Gut microbiota of Mongolians has attracted interest in relation to their unique dietary habit. The first study on the gut microbiota of Mongolian people (Zhang et al., 2014) examined regional and seasonal differences, finding that the core intestinal microbiota comprised Prevotella, Bacteroides, and Faecalibacterium. A comparative study with Han and European populations found that Faecalibacterium prausnitzii and Coprococcus comes are enriched in Mongolian people which may contribute to their health through the production of SCFA, notably butyrate with anti-inflammatory activity (Liu et al., 2016). Our earlier study (Shinoda et al., 2021) showed that the general Mongolian population has Prevotella-dominant microbiome in their gut and that lactic acid bacteria are enriched in Mongolian relative to other Asian populations.
The gut microbiota is now recognized as an interface between foods and host health, notably energy and metabolic homeostasis. In this context, the gut microbiome has to be investigated to address the Mongolian paradox. Therefore, we herein investigated gut microbial factors involved in the prevention of the co-occurrence of T2D with obesity in Mongolians, by comparing the gut microbiome of Mongolian individuals with obesity with and without T2D; first, the difference in the bacterial composition was investigated by the 16S rRNA amplicon sequencing, second, the difference in the metabolites was investigated by targeting short-chain fatty acids (SCFAs) and bile acids (BAs), third, the difference of their community functional potential was assessed by the shotgun metagenomics, and finally we address the alteration in the community function correlated with the development of T2D in Mongolian with obesity.
2 Materials and methods 2.1 Ethics declarationThis study was approved by the Ethics Committees of the Faculty of Agriculture in Kyushu University (Approval No. 107). Ethical clearance was obtained from the Ethics Committee of the Ministry of Health of Mongolia (Approval No. 78). All methods were performed in accordance with the relevant guidelines and regulations. Written informed consent was obtained from all participants, and the samples and data were analyzed, entered, and published anonymously.
2.2 Study designParticipants were Mongolian adults living in Ulaanbaatar (n = 39) and Bulgan (n = 27). Subjects were screened according to the inclusion and exclusion criteria for AMP phase IV study (published in the Supplementary Material of Therdtatha et al., 2021), except that some of the subjects had received antidiabetic therapy in the past 2 months. Supplementary Table S1 presents details about the subjects.
Samples were collected from the 66 participants with obesity (BMI > 30), who were classified into two groups, with T2D (DO: HbA1c ≥ 6.5) or without T2D (NDO: HbA1c < 6.5), based on the National Glycohemoglobin Standardization Program (NGSP) criteria. Table 1 summarizes the demographic and clinical characteristics of each group. There were significant differences in age, sex, and BMI between the DO and NDO groups; these were subsequently treated as non-microbial confounding factors. The NDO group was further divided into a prediabetes group (PDO, HbA1c > 5.7) and a healthy obese group (HO, HbA1c ≤ 5.7) to examine differences in microbiome community structure.
Table 1. Demographic and clinical characteristics of the 66 Mongolian subjects.
2.3 Sample collectionAll subjects provided two fresh stool samples for metagenome and metabolome analyses, in separate tubes (76 × 20 mm, Sarstedt, Munich, Germany), each containing 2 mL RNAlater (Invitrogen, Thermo Fisher Scientific, Waltham, MA) or MeOH, and five YTZ 2.5 mm zirconia balls (Nikkato, Sakai, Japan). For each tube, two spoons of fecal sample were collected from different parts in the middle section of feces and immediately suspended in the solution by shaking the tubes vigorously.
The collected feces were transferred to the laboratory, vortexed to completely suspend the feces, and stored at −80°C in Mongolian University of Life Sciences. The samples were transported to Kyushu University, Japan, while kept below 4°C and were immediately stored at −80°C until DNA or metabolite extraction.
2.4 DNA extraction, 16S rRNA gene amplicon sequencing, and data processingTotal bacterial DNA was extracted from stool samples using the bead-beating method, as previously described (Matsuki et al., 2004). Briefly, DNA was extracted by first vigorously shaking the feces with the zirconium beads; the DNA was then purified by phenol–chloroform extraction followed by isopropanol precipitation (Supplementary Material of Therdtatha et al., 2021). High-throughput 16S rRNA gene sequencing was performed as previously described (Kisuse et al., 2018). Briefly, the V3 to V4 region of the bacterial 16S rRNA gene was amplified via two-step PCR, in which the bacterial universal primer set, Bakt_341F (5′-CGCTCTTCCGATCTCTGCCTACGGGNGGCWGC AG-3′) and Bakt_805R (5′-TGCTCTTCCGATCTGACGACTA CHVGGGTATCTAATCC-3′), was used in the first step; in step two, the same primers plus index-sequences were used. Amplicons with index sequences were subjected to sequencing using the MiSeq Reagent Kit v3 (MS-102-3003; Illumina, San Diego, CA, United States).
The sequence data were processed using QIIME2 pipeline 2021.2 (Bolyen et al., 2019). Briefly, after the total reads were demultiplexed using the index sequences, paired-end sequences were merged and denoised, and primer sequences were removed using DADA2 (Callahan et al., 2016). The forward and reverse primers were eliminated using the 3′ and 5′ end trim parameters "--p-trim-left-f 17" and "--p-trim-left-r 21," respectively. The parameters "p-trunc-len-f 290" and "p-trunc-len-r 210" were used to trim the forward and reverse reads to lengths of 290 and 210 bases, respectively. The sequence of each amplicon sequence variant (ASV) with 97% sequence identity was taxonomically classified via the QIIME "classify-sklearn" feature using the SILVA 138 database.
2.5 Beta-diversity analysisBeta diversity was assessed using Bray Curtis, Jaccard, weighted UniFrac, and unweighted UniFrac distances, and permutational multivariate analysis of variance (PERMANOVA) was used to investigate statistical differences between the PDO and HO groups and between the DO and NDO groups.
2.6 Whole shotgun metagenomicsFor whole shotgun metagenomic analysis, nine samples were selected from both the DO and NDO groups to be evenly distributed in the weighted and unweighted UniFrac-PCoA ordinations of both groups (Supplementary Figure S1). The DNA extract prepared for the 16S rRNA amplicon sequencing was treated using RNase (NIPPON GENE Co., Ltd., Tokyo, Japan) and was purified using a Gel/PCR extraction kit (NIPPON GENE Co., Ltd., Tokyo, Japan) to remove RNA. DNA concentration was quantified using the Quant-iT PicoGreen dsDNA Assay Kits with a dsDNA standard (Invitrogen, Thermo Fisher Scientific, Waltham, MA). More than 1 Ī¼g of DNA in TE buffer (>12.5 ng/Ī¼L) was subjected to 150 bp paired-end sequencing on a DNBSEQ system (BGI, Shenzhen, China).
The obtained raw sequences were filtered using the SOAPnuke tool (Chen et al., 2018) to remove (i) reads matching ≥25.0% of the adapter sequences (up to three base mismatches allowed); (ii) short reads (<150 bp); (iii) reads in which undetermined nucleotides account for ≥0.1% of the entire read; and (iv) low-quality reads in which bases with quality <20 accounts for ≥40% of the entire read. Then, the cleaned pair-ended reads were merged by using BBMerge (Bushnell et al., 2017). Human DNA sequences were then removed using the KneadData tool and the database Homo_sapiens_hg37_and_human_contamination_Bowtie2_v0.1. The sequences were further processed using the KneadData tool (version 0.12.0) with the default settings and the human DNA sequences were removed by Bowtie2 (version 2.4.2) on very-sensitive-local mode, phred33 parameter, and the database Homo_sapiens_hg37_and_human_contamination_Bowtie2_v0.1.
These quality-filtered sequences were functionally profiled using the HUMAnN pipeline (v. 3.6.1) using MetaPhlAn 4.0; the mpa_vJan21_CHOCOPhlAnSGB_202103 nucleotide database was used as an intermediate step in the taxonomic classification of each DNA fragment. Each DNA fragment was functionally annotated to UniRef90 gene families, and then categorized into Kyoto Encyclopedia of Genes and Genomes (KEGG) categories at the enzyme, orthology, and pathway levels. Gene abundance in each sample (and similarly in each category) was calculated as reads per kilobase (RPK) and was normalized to obtain the count per million (CPM).
2.7 Quantification of SCFAs and bile acidsSCFAs and BAs in the feces were quantified as previously described (Tanaka et al., 2020; Therdtatha et al., 2021). Briefly, feces collected in MeOH were placed in a SpeedVac concentrator (Thermo Fisher Scientific, Waltham, MA) to evaporate the methanol, and were resuspended in PBS in deuterated water (100 mM, pH 7.4) containing 3-(trimethylsilyl) propionate-2.2.3.3-d4 as an internal standard. The supernatant was used for quantitative 1-H nuclear magnetic resonance (NMR) spectroscopy to quantify acetate, propionate, and butyrate, using a 400 MHz spectrometer (JNM-ECZ400, JEOL, Tokyo, Japan). The SCFA data of each sample was deposited in Supplementary Table S3.
BAs were extracted from the fecal pellet by hot ethanol with an internal standard of nor-deoxycholic acid at 60°C for 30 min and subsequently at 100°C for 3 min. After cleaning using an Oasis HLB cartridge column (WAT094225, Waters, Milford, MA), the extract was subjected to liquid chromatography with a triple quadrupole mass spectrometry system, in which 15 major human fecal BAs were quantified (LCMS-8050, Shimadzu, Japan). These procedures are described in detail by Tanaka et al. (2020). The BA data of each sample was deposited in Supplementary Table S4.
2.8 Statistical analysisStatistical analyses and data visualization were performed using RStudio 2022.07.2, R 4.2.1 (R Core Team, 2023), Stata/SE 12.0 (Stata Technical Support, 2023), and EZR 4.0.3 (Kanda, 2013).
To evaluate statistical differences between two groups, the Wilcoxon rank-sum test was applied using "wilcox.Test" in R. To evaluate statistical differences in qualitative variables between groups, Fisher's exact test was used in EZR. Spearman's correlation analysis was performed to calculate correlations between indices, using Stata/SE12. Multiple logistic regression analysis was performed using EZR to assess the risk of each factor to the occurrence of T2D. Linear discriminant analysis of effect size (LEfSe) was performed using the online Galaxy platform (Segata et al., 2011).
2.9 Accessions of sequence dataThe raw sequence data obtained was deposited in the DNA Data Bank of Japan (DDBJ) (Fukuda et al., 2021) under BioProject no. PRJDB5860. The Sequence read archives of 16S rRNA gene amplicon sequences (DRA017379) and whole shotgun metagenomic sequence data (DRA017451) were deposited with fecal sampling data in the BioSample database (BioSample, 2023) (accessions SAMD00656652-717 and SAMD00657900-17, respectively).
3 Results 3.1 Differences in the microbial community composition between the DO and NDO groupsThe gut microbial community composition was profiled by barcode sequencing of 16S rRNA V3-V4 region. Eventually, 812,093 high quality sequences corresponding to 12,304 ± 2,774 reads (minimum 6,231) per sample were obtained from 17,273 ± 3,757 raw reads per sample and were clustered into 1,722 ASVs; these were assigned to 2 domains, 13 phyla, 19 classes, 40 orders, 72 families, 218 genera, and 502 species. The counts of each ASV for each sample were tabulated in Supplementary Table S2 with taxonomic information for each ASV.
Differences in gut microbial composition were calculated using four distance matrices, and discreteness between the different diabetic status groups was examined using permutation analysis (Table 2). The DO and NDO groups differed significantly, based on the Jaccard, Bray-Curtis, and unweighted UniFrac distances (Supplementary Figure S1 for the ordination plot). When the NDO group was further divided into PDO and HO groups only the Jaccard distance showed a significant difference between DO and PDO, while there were no significant differences between the PDO and HO groups. This indicates that the DO group harbored a significantly different gut microbial community from the NDO group, whereas the community did not differ significantly between the PDO and HO groups. Therefore, the grouping of DO and NDO including PDO and HO subjects was henceforth used in this study.
Table 2. Pairwise differences in microbial community composition, based on Jaccard, Bray-Curtis, weighted UniFrac, and unweighted UniFrac distances.
The relative abundances of the 20 most abundant genera are illustrated separately for the DO and NDO groups (Figure 1A). In both groups, the most abundant genus was Prevotella_9, followed by Faecalibacterium and Blautia. Although there was no significant difference in the abundances of the top 20 genera between the DO and NDO groups, some trends were observed, including elevated Faecalibacterium (p = 0.066) levels and lower Holdemanella (p = 0.089) levels in the NDO group. Correlations between the abundance of the most or second most dominant genera (Prevotella_9 and Faecalibacterium, respectively) and BMI or HbA1c were calculated using the Spearman's rank sum test (Figure 1B); Prevotella_9 was significantly positively correlated with BMI.
Figure 1. Composition and abundance of the dominant fecal bacterial genera in fecal samples from Mongolian subjects with obesity. (A) Comparison of the top 20 genera in the diabetic obese (DO) and nondiabetic obese (NDO) groups. The abundance of each genus is shown in the bar graph. *, higher values in the relevant group, approaching significance (p < 0.1). (B) Correlations between the abundance of Prevotella_9 or Faecalibacterium and BMI or HbA1c levels. Correlations were analyzed using Spearman's rank sum test.
LEfSe analysis was performed to examine differences in gut microbiota composition at different taxonomic levels between the DO and NDO groups (Figure 2A): the genera Anaerostipes, Limosilactobacillus and Parasutterella were significantly abundant in the NDO group, whereas in the DO group, a wide variety of taxonomic groups belonging to domain Archaea, six bacterial families, and nine bacterial genera, including Methanobrevibacter, Desulfovibrio, and Solobacterium, were significantly abundant (Figure 2B). Further, correlation analysis (Supplementary Table S5) indicated that Methanobrevibacter and Solobacterium positively correlated with HbA1c.
Figure 2. Comparison of fecal bacterial composition between the diabetic obese (DO; green) and nondiabetic obese (NDO; red) groups. (A) Linear discriminant analysis of effect size (LEfSe) analysis to compare the fecal bacterial composition at each taxonomic rank between the DO and NDO groups. The Wilcoxon rank-sum test was used to calculate the linear discriminant analysis (LDA) scores. Taxonomic groups with LDA > 2.0 and p < 0.05 are highlighted by the indicated color on the cladogram. (B) Box plots of the abundance of genera that differed significantly between the DO and NDO groups in the LEfSe analysis.
3.2 Comparison of fecal SCFA levels between the DO and NDO groupsWe compared fecal SCFA concentrations between the DO and NDO groups (Figure 3A). The levels of the three major SCFAs–acetate, propionate, and butyrate–tended to be higher in the NDO group, and acetate levels were significantly higher in the NDO group. Figure 3B illustrates the abundance of each SCFA stacked in the bar graph and ordered according to total SCFA concentration, together with a heat map of HbA1c levels (Figure 3B). In most samples, the ratio of acetate, propionate, and butyrate was approximately 6:3:2. The Spearman's rank sum test revealed an inverse correlation between total SCFA concentration and HbA1c with approached significance (r = −0.23, p = 0.06).
Figure 3. Distribution of the abundances of short-chain fatty acids (SCFAs) and bile acids (BA) in the feces of the diabetic obese (DO) and nondiabetic obese (NDO) groups. (A) Comparison of fecal concentrations of acetate, propionate, butyrate and of total SCFA (sum of acetate, propionate, and butyrate). The box plot illustrates these levels in the DO and NDO groups; statistical significance was evaluated using the Wilcoxon rank-sum test. *, 0.05 < p < 0.1, **, p ≤ 0.05. (B) Stacked bar plot showing the levels of the three SCFAs, order by total abundance, in each of the 66 Mongolian subjects. The HbA1c level of each subject is shown in the heatmap below the bar graph. Correlations between the levels of total SCFA and HbA1c were analyzed using Spearman's rank sum test. (C) Comparison of fecal concentrations of tauroursodeoxycholic acid and ursodeoxycholic acid. Statistical differences between groups were calculated as in (A). *, 0.05 < p < 0.1, **, p ≤ 0.05.
3.3 Comparison of fecal BA levels between the DO and NDO groupsWe measured the concentrations of 15 major BAs in fecal samples (Figure 3C). Ursodeoxycholic acid and its taurine conjugated forms, tauroursodeoxycholic acid (TUDCA), tended to be more abundant in the NDO group. In particular, levels of TUDCA were significantly higher in the NDO group than in the DO group (p = 0.049).
3.4 Comparison of metagenomic profile between the DO and NDO groupsTo address differences in the functional potential of the microbial communities in the DO and NDO groups, we performed a whole shotgun metagenomic analysis by using 9 DO and 9 NDO samples. Ultimately, per sample, 24.01 ± 2.48 million high-quality non-human reads, corresponding to 5.60 ± 0.48 Gb sequences, were obtained from 24.30 ± 2.50 million raw reads corresponding to 5.64 ± 0.48 Gb sequences. In the KEGG pathway categories, only the phosphotransferase system (PTS) exhibited significantly differential enrichment (Figures 4A,B). The enrichment of PTS genes in the NDO group can be attributed to the marked enrichment of Anaerostipes hadrus (Figure 4C). Further, genes involved in SCFA biosynthesis tended to be more abundant (although not significantly) in the NDO group, with A. Hadrus contributing to this enrichment (Figure 5).
Figure 4. KEGG pathway differential enrichment in the diabetic obese (DO) and nondiabetic obese (NDO) groups. (A) Volcano plot illustrating differences in the abundance of each KEGG pathway between the DO and NDO groups. The vertical axis displays –log10p and the horizontal axis displays the linear fold change (NDO/DO). (B) Box plot of ko02060 (PTS) expression (in count per million, CPM) in the DO and NDO groups. (C) Stacked bar plot of ko02060 (PTS) expression (in count per million, CPM) in each species. **p < 0.05.
Figure 5. Abundances of KEGG pathway/orthology involved in short-chain fatty acid (SCFA) biosynthesis in the diabetic obese (DO) and nondiabetic obese (NDO) groups. The three stacked bar plots refer, respectively, to the KEGG pathways/orthologies involved in acetogenesis, propanogenesis (ko00640), and butanogenesis (ko00650), and reflect count per million (CPM). For acetogenesis, the counts of KEGG orthologies involved in the Wood–Ljungdahl pathway (17 KOs) and of the K00158 (pyruvate oxidase) and K01512 (acyl phosphatase) KOs, were summed. **p < 0.05 comparing the DO and NDO groups.
A series of genes encoding enzymes involved in acetogenesis, pyruvate oxidase (EC 1.2.3.3), phosphate acetyltransferase (EC 2.3.1.8), and acylphosphatase (EC 3.6.1.7), were higher in the NDO group than in the DO group (Figures 6A,B), although nonsignificantly. Figure 6B illustrates the composition of the bacteria encoding the enzymes (based on KEGG database) in each group. The enrichment of pyruvate oxidase (EC 1.2.3.3), which metabolizes acetyl phosphate from pyruvate, was attributed to the higher abundance of oral lactic acid bacteria (Lactobacillus and Streptococcus species) in the NDO group. The higher abundance of A. Hadrus was responsible for the enrichment in the NDO group of phosphate acetyltransferase (EC 2.3.1.8) and acylphosphatase (EC 3.6.1.7) which consecutively metabolize acetyl-CoA to acetate.
Figure 6. Abundances of KEGG enzymes involved in acetogenesis in the diabetic obese (DO) and nondiabetic obese (NDO) groups. (A) KEGG enzyme map for the acetogenesis pathway from pyruvate to acetate. (B) Box plot showing the counts of each acetogenesis KEGG enzyme (EC1.2.3.3, pyruvate oxidase; EC2.3.1.8, phosphate acetyltransferase; and EC3.6.1.7, acylphosphatase), and stacked bar plots showing the counts of KEGG enzymes per species per subject. **p < 0.05 comparing the DO and NDO groups.
3.5 Multiple logistic regression of factors associated with the occurrence of T2D in obesityTo examine the effect of fecal SCFA levels and host demographic factors on the occurrence of T2D in obesity in this population, we performed multiple logistic regression analyses. Model 1A and Model 1B were obtained with all sets of possible host demographic confounding factors and the fecal concentration of either acetate or total SCFA (acetate + propionate + butyrate), respectively, as microbial factors (Table 3). This model predicted the occurrence of T2D with statistical significance; acetate and total SCFA contributed the most respectively, and age and BMI exhibited significant correlations, whereas sampling region, sex, and metformin intake did not. Regarding sampling region, we further investigated the differences in the levels of acetate and total SCFA in Ulaanbaatar and Bulgan, separately, and confirmed their higher levels in both regions, although not statistically significant (Supplementary Figure S2). Regarding metformin, it is known that metformin intake increases the level of SCFA (Mueller et al., 2021). However, we observed an opposite trend that the acetate and total SCFA levels decreased in the metformin-prescribed subjects in the DO group (DO-Met (+)) (Supplementary Figure S3). The DO-Met (+) group showed significantly higher BMI, HbA1c and fasting blood glucose levels than the non-metformin-prescribed subjects in the DO group (DO-Met (−)). The fact that the DO-Met (+) group had significantly more sever diabetic symptom appears to cause the decrease of the SCFA level rather than the direct effect by metformin.
Table 3. Model 1 of multiple logistic regression analysis of factors potentially associated with the occurrence of T2D in obesity in the Mongolian subjects.
Following Model 1, we did the multiple logistic regression analysis excluding these insignificant valuables, resulting in Model 2 with enhanced significance and reasonably high prediction strength (p < 0.0001 and pseudo R2 = 0.3285 for the model including acetate and p < 0.0001 and pseudo R2 = 0.3091 for the model including total SCFA). As shown in Figure 7, the odds ratios of acetate and total SCFA were −0.42 and −0.65 per 100 Ī¼mol/g feces, respectively. By using this model, we further examined the odds ratios of the abundances of the certain bacteria genera and TUDCA. Prevotella_9, Faecalibacterium, Anaerostipes, and TUDCA showed the tendency of the negative risk associations for the occurrence of T2D, although nonsignificantly.
Figure 7. Multiple logistic regression of gut microbial factors for the occurrence of T2D in obesity. The multiple logistic regression analysis of each gut microbial factor was performed with age and BMI (Model 2) and the estimated 95% confidence interval for odds ratio of each factor are shown in the forest plot. Asterisks represent statistically significance (p < 0.05).
4 DiscussionThe main aim of this study was to address the gap of the lower-than-expected frequency of diabetes in obesity in the Mongolian population in terms of intestinal microbiology. First of all, we overviewed the difference in the fecal microbiome structure between the DO and NDO subjects and also between PDO and HO subjects to clarify how the intestinal microbial community differ by the level of diabetes. Community structure differed between the DO and NDO groups, but not between the PDO and HO groups. Various studies have reported significant alterations in the gut microbiota of patients with prediabetes (Letchumanan et al., 2022; Rathi et al., 2023). Our current analysis may have missed small-scale alterations unique to the PDO group, owing to the small sample size. However, the community composition of the NDO group including PDO subjects was clearly different from that of the DO group (Table 2). These differences might be involved in the prevention of T2D development in obesity in Mongolia and were further analyzed.
Prevotella_9 was the most dominant genus in both DO and NDO groups. Although Prevotella_9 did not differ in abundance between the DO and NDO groups, it showed positive correlation with BMI (Figure 1B) and also showed marginally lower risk associations with the occurrence of T2D in the multiple logistic regression analysis (Figure 7). This is consistent with prior findings showing a link between obesity and the abundance of Prevotella, which exhibits substantial potential to produce energy via high levels of sugar fermentation activity (Precup and Vodnar, 2019). Prevotella is also known to exert an antidiabetic function (Kovatcheva-Datchary et al., 2015). Therefore, high levels of Prevotella may be involved in the low frequency of T2D in obesity in the Mongolian population.
The second dominant genus, Faecalibacterium, was less abundant in the DO group with approached significance (Figure 1A). Lower Faecalibacterium abundance is commonly observed in individuals with diabetes (Gurung et al., 2020). A reduction in Faecalibacterium prausnitzii abundance was observed even at the prediabetic stage (Wu et al., 2020). Since Faecalibacterium prausnitzii is recognized as beneficial, notably as a butyrate producer, its reduction represents dysbiosis and this might be the case in the DO group.
Dysbiotic trend in the DO subjects was also observe
Comments
Post a Comment