Intestinal function and transit associate with gut microbiota dysbiosis in cystic fibrosis

Background: Most people with cystic fibrosis (pwCF) suffer from gastrointestinal symptoms and are at risk of gut complications. Gut microbiota dysbiosis is apparent within the CF population across all age groups, with evidence linking dysbiosis to intestinal inflammation and other markers of health. This pilot study aimed to investigate the potential relationships between the gut microbiota and gastrointestinal physiology, transit, and health. Study Design: Faecal samples from 10 pwCF and matched controls were subject to 16S rRNA sequencing. Results were combined with clinical metadata and MRI metrics of gut function to investigate relationships. Results: pwCF had significantly reduced microbiota diversity compared to controls. Microbiota compositions were significantly different, suggesting remodelling of core and rarer satellite taxa in CF. Dissimilarity between groups was driven by a variety of taxa, including Escherichia coli, Bacteroides spp., Clostridium spp., and Faecalibacterium prausnitzii. The core taxa were explained primarily by CF disease, whilst the satellite taxa were associated with pulmonary antibiotic usage, CF disease, and gut function metrics. Species-specific ordination biplots revealed relationships between taxa and the clinical or MRI-based variables observed. Conclusions: Alterations in gut function and transit resultant of CF disease are associated with the gut microbiota composition, notably the satellite taxa. Delayed transit in the small intestine might allow for the expansion of satellite taxa resulting in potential downstream consequences for core community function in the colon.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021.

45
Cystic fibrosis (CF) associated respiratory infections are the major cause of disease morbidity and mortality.

46
However, a number of gastrointestinal (GI) problems may also arise, limiting the quality of life, including 47 meconium ileus at birth, distal intestinal obstruction syndrome, small intestinal bacterial overgrowth (SIBO), 48 increased risk of malignancy, and intestinal inflammation [1,2]. It is therefore unsurprising that people with CF 49 experience persistent GI symptoms [3,4] with "how can we relieve gastrointestinal symptoms in people with 50 CF?" a top priority question for research [5].

52
Microbial dysbiosis at the site of the GI tract in CF patients has been described, with changes evident from 53 birth through to adulthood [6][7][8]. Moreover, the extent of this divergence from healthy microbiota, initially due 54 to loss of cystic fibrosis transmembrane conductance regulator (CFTR) function [9], is further compounded by 55 routine treatment with broad spectrum antibiotics [10]. The reshaping of the gut microbiota may have functional 56 consequences that could further impact on patients. These include the reduction of taxa associated with the 57 production of short-chain fatty acids (SCFAs) which play key roles in modulating local inflammatory responses 58 and promoting gut epithelial barrier integrity [11][12][13]. Furthermore, studies of microbiota dysbiosis in CF have 59 demonstrated its relationship with intestinal inflammation [14], intestinal lesions [15], and increased gene 60 expression relating to intestinal cancers [16]. Whilst many of these clinical parameters have ties to gut 61 microbiota changes, they remain understudied exclusively past childhood despite advances in less invasive 62 approaches to investigate CF gut physiology and function [17]. Our group has recently published on the use 63 of magnetic resonance imaging (MRI) to assess gut transit time, along with other parameters, in adolescents 64 and adults [18].

66
In this pilot study, we linked those MRI physiology metrics and clinical metadata directly to high-throughput 67 amplicon sequencing data identifying constituent members of the gut microbiota, to explore the relationships 68 between microbial dysbiosis, intestinal function and clinical state.

69
. CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint      is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Subjects marked with an asterisk* indicate those who failed to produce a stool sample for subsequent metagenomic and metabolomic analysis and thus were excluded from downstream analyses. All

126
Sulfonamide. Asterisks denote participants who did not provide any stool samples upon visitation, and thus were excluded from downstream microbiota analysis. 7

128
For both healthy control and CF groups, bacterial taxa were partitioned into core or satellite based on their 129 prevalence and relative abundance as depicted in (Fig. 1). Within the healthy control group, 30 taxa were core 130 constituting 60.5 % of the total abundance, with the remainder accounted for by 386 satellite taxa. In the CF 131 group, 22 core taxa represented 34.7 % of the abundance, with 323 satellite taxa constituting the remainder.

132
Core taxa are listed in Table S7. The whole, core, and satellite microbiota demonstrated similar patterns in 133 diversity, whereby there was significantly reduced diversity in the CF group ( Fig. 2A, Table S8).  Table S7.

143
Within-group core microbiota similarity was higher within the healthy control group, with a mean similarity (± 144 SD) of 0.60 ± 0.08 compared to 0.40 ± 0.11 for the CF group (Fig. 2B). As expected, satellite taxa similarity 145 within groups was much lower than for the core but was also significantly reduced in CF compared to controls, 146 at 0.35 ± 0.08 and 0.21 ± 0.09 for the healthy control and CF group respectively. ANOSIM testing determined 147 the whole microbiota, core, and satellite taxa of the CF group were significantly different in composition 148 compared to healthy controls (Fig. 2B, Table S9). SIMPER analysis was implemented to reveal which taxa 149 . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021. ; https://doi.org/10.1101/2021.08.24.21262265 doi: medRxiv preprint 8 were responsible for driving this dissimilarity ( Table 2). Of the taxa contributing to > 50% of the differences 150 between healthy control and CF groups, those within the genus Bacteroides were represented most.

151
Escherichia coli contributed most towards the differences between groups, despite satellite status, followed by

155
Whole microbiota (black plots) and partitioned data into core (orange plots) and satellite taxa (grey plots) are 156 given. (A) Differences in Fisher's alpha index of diversity between healthy controls and cystic fibrosis samples.

157
Black circles indicate individual patient data. Error bars represent 1.5 times inter-quartile range (IQR). Asterisks 158 between groups denote a significant difference in diversity following use of Kruskal-Wallis tests (P < 0.001).

159
Summary statistics are provided in Table S8. (B) Microbiome variation measured within and between sampling 160 groups, utilising the Bray-Curtis index of similarity. Error bars represent standard deviation of the mean.

161
Asterisks indicate significant differences between sampling groups following the use of one-way ANOSIM 162 testing (P < 0.001). Summary statistics are provided in Table S9.

163
. CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint    is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021. ; https://doi.org/10.1101/2021.08.24.21262265 doi: medRxiv preprint Table 3 Redundancy analysis to explain percent variation in whole microbiota, core taxa and satellite taxa between all subjects from significant clinical variables 179 measured.

180
Var. Exp (%) represents the percentage of the microbiota variation explained by a given parameter within the redundancy analysis model. P (adj) is the adjusted significance value following false discovery 181 rate correction. Antibiotics is the presence/absence of recurrent antibiotic regimes for a given patient. BMI -Body mass index, Colon Fasting Vol -Colon volume at baseline corrected for body surface area,

11
A species redundancy analysis biplot (RDA) was constructed to investigate how significant clinical variables 184 from the whole microbiota direct ordination approach explained the relative abundance of taxa from the 185 SIMPER analysis (Fig. 3)  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021. ; https://doi.org/10.1101/2021.08.24.21262265 doi: medRxiv preprint

203
In this pilot study, we investigated the relationships between clinical factors, MRI markers of GI function and 204 the composition of faecal bacterial microbiota. We have shown that it is possible to partition the gut microbiota 205 into core and satellite taxa to investigate potential community functions and relationships, with the notion that 206 the core constituents contribute to the majority of functionality exhibited by the community [20,24]. As to be 207 expected, the core taxa made up most of the abundance within the healthy control group. Whilst many taxa 208 were also commonly represented in the CF group, the latter was dominated in abundance by the satellite taxa.

209
Our findings of reduced diversity across the whole, core, and satellite microbiota are in agreement with  (Table S7), and of which are implicated in both CF lung and gut microbiomes 222 [8,9,24,[26][27][28]. The concept of the "gut-lung axis" in CF arises from the direct translocation of the respiratory 223 microbiota from sputum swallowing to the gut [29], but also the emergence of species in the gut prior to the 224 respiratory environment [27]. This apparent bidirectionality is further supported by the administration of oral 225 probiotics to decrease pulmonary exacerbations in CF [30]. Aside from sputum swallowing, the increase in

230
E. coli contributed most to the dissimilarity between healthy and CF groups despite maintaining satellite status 231 throughout both the healthy and CF groups, seemingly resultant of the wide age range of our study participants, 232 of which the higher relative abundances were observed in the younger adolescent patients (Table 2). In

233
. CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021. ; https://doi.org/10.1101/2021.08.24.21262265 doi: medRxiv preprint 13 childhood studies, a significantly higher relative abundance of Proteobacteria is often reported in relation to 234 dysbiosis, with E. coli abundance associating with poor growth outcomes and intestinal inflammation [32][33][34].

235
Other notable taxa contributing to the dissimilarity observed between groups encompassed a variety of key 236 species associated with SCFA production in the colon. This included F. prausnitizii and E. rectale, both of 237 which were significantly decreased in abundance within the CF group, but also R. bromii and B. luti. These

247
Variance across the whole microbiota and satellite taxa was significantly explained by the use of antibiotics 248 (Table 3) is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021. ; https://doi.org/10.1101/2021.08.24.21262265 doi: medRxiv preprint similarly pro-inflammatory intestinal environments, such as Crohn's disease or ulcerative colitis , including

276
Although dietary profiles were similar between groups (Tables S3-6) and did not contribute to significant 277 variation in the microbiota, increased fat intake to meet energy requirements is a staple of the CF diet [46].

278
The infant gut metagenome demonstrates enrichment of fatty acid degradation genes [32] whilst CF-derived  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

295
This cross-sectional pilot study has identified relationships between markers of clinical status, gastrointestinal 296 function and bacterial dysbiosis in the CF population. By partitioning the community into core and satellite taxa, 297 we were able to reveal the relative contributions of CF-associated lifestyle factors and elements of intestinal 298 function to these subcommunity compositions, and how specific taxa were affected by these clinical factors.

299
Further, as the first study to combine high-throughput gene amplicon sequencing with non-invasive MRI to 300 assess underlying gut pathologies, we demonstrate the potential for future collaborations between 301 gastroenterology and microbiology with larger cohort recruitment to investigate these relationships between 302 gut function and the microbiome further.     is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021.  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 26, 2021. ; https://doi.org/10.1101/2021.08.24.21262265 doi: medRxiv preprint