Intestinal dysbiosis in carbapenem-resistant Enterobacteriaceae carriers

Infection with Carbapenem-Resistant Enterobacteriaceae (CRE) became an important challenge in health-care settings and a growing concern worldwide. Since infection is preceded by colonization, an understanding of the latter may reduce CRE-infections. We aimed to characterize the gut microbiota after colonization by CRE, assuming that an imbalanced gastrointestinal tract (GIT)-associated microbiota precedes CRE-colonization. We evaluated the GIT-microbiota using 16S rRNA genes sequencing extracted of fecal samples, collected from hospitalized CRE-carriers, and two control groups of hospitalized non-carriers and healthy adults. The microbiota diversity and composition in CRE-colonized patients differed from that of the control groups participants. These CRE-carriers displayed lower phylogenetic diversity and dysbiotic microbiota, enriched with members of the Enterobacteriaceae family. Concurrent with the bloom in Enterobacteriaceae, a depletion of anaerobic commensals was observed. Additionally, changes in several predicted metabolic pathways were observed for the CRE-carriers. Concomitant, we found higher prevalence of bacteremia in the CRE-carriers. Several clinical factors that might induce change in the microbiota were examined and found as insignificant between the groups. CRE-colonized patients have dysbiotic gut microbiota in terms of diversity and community membership, associated with increased risk for systemic infection. Our study results provides justification for attempts to restore the dysbiotic microbiota with probiotics or fecal transplantation.

approximately 9% of carriers [4]. Moreover, CRE-carriers serve as a major reservoir for its 50 dissemination in healthcare facilities [4,5]. 51 The complex commensal microbiota that normally colonizes mucosal surfaces in healthy 52 individuals allows colonization resistance and inhibits expansion and domination by antibiotic-53 resistant exogenous bacteria such as Enterobacteriaceae [6,7].Microbial dysbiosis may lead to an 54 overgrowth of antibiotic resistant pathogens [8], which can be calamitous for susceptible 55 patients, resulting in bacteremia and sepsis [9] and is associated with increased risk for 56 transmission due to increased shedding to the environment [10,11]. 57 Fresh fecal samples were collected from hospitalized participants by the research cadre (CRE-88 carriers and non-carrier groups). Following collection, swabs were stored at -80 º C. Samples from 89 healthy participants were self-collected by participants, and transported in a freezer pack to the 90 laboratory within 24 hours of collection, and then stored at -80 º C. 91

Microbiota sequencing and taxonomy assignment: 92
Total DNA was extracted from the samples using QIAamp Fast DNA Stool Mini Kit (QIAGEN 93 Inc., Hilden, Germany), according to manufacturer`s instructions. Genomic

141
Page 8 To study the microbiota profile in CRE-carriage we analyzed the clinical parameters and 142 microbial composition of three groups: hospitalized CRE-carriers, hospitalized non-carriers, and 143 healthy controls. 144

Study cohort clinical characteristics 145
The demographic and clinical characteristics of the study cohort for all groups are presented in 146 Tables 1 and 2. There were no significant differences between the groups regarding confounding 147 factors such as gender, ethnic origin, gastrointestinal disease, radiotherapy, chemotherapy, and 148 diabetes mellitus. Moreover, the comparison between these factors and the microbial profile 149 described below was insignificant. 150 Hospitalized patients were older than the healthy individuals (average ages 68.3 and 42.2, 151 respectively). In general, antibiotic usage (broad vs. narrow spectrum), and positive culture 152 prevalence were similar in both hospitalized groups. However, the rate of bacteremia was twice 153 as high in the CRE-carriers. 154 Regarding antibiotics usage, it was noted that Vancomycin and Tazocin treatments were used 155 more in the CRE-carriers, while Amikacin and Rocephin treatments were used more in the 156 hospitalized non-carriers (Kruskal-Wallis p<0.05). Notably, most comparative analyses 157 conducted between the two hospitalized groups (excluding the healthy group), because of the 158 different age average, the strict exclusion criteria and the lack of antibiotic treatment, which 159 effect the microbiota. 160 The prevalence of positive urine, sputum, and blood cultures (not specifically by CRE) was 161 around 80% and similar between both hospitalized groups. However, a higher bacteremia rate Page 9 was found in the in the CRE-carriers as compared to the non-carriers (53% vs 24%, respectively, 163 Taxonomic classification revealed that the dominant bacterial phyla were Bacteroidetes (56-172 62%), Firmicutes (19-35%), and Proteobacteria (6-21%) (Fig 1A). Firmicutes prevalence was 173 significantly lower in the CRE-carriers compared to non-carriers and healthy controls (p<0.005); 174 Proteobacteria were significantly higher in the CRE-carriers (p<0.005). The ratio between 175 Firmicutes to Bacteroidetes, considered highly relevant in human gut microbiota composition, 176 was the lowest in the CRE-carriers (0.35±0.05), higher for hospitalized non-carriers (0.41±0.05), 177 and highest in the healthy group (0.63±0.05). 178 179

Microbial diversity and composition 180
Microbial richness assessment determined using the Shannon index, revealed that CRE-carriers 181 had significantly lower richness as compared to the other groups (p<0.005) (Fig 1B). 182 Interestingly, the healthy and non-carriers groups did not differ in the richness measure.

Page 10
The bacterial communities of the three groups were compared using weighted UniFrac based on 184 PCoA (Fig 1C). The samples from healthy individuals clustered separately from hospitalized 185 participants. The PC1 and PC2 vectors significantly discriminated between the groups (Kruskal-186 Wallis p<0.001). 187 LefSe was used to identify bacterial taxa associated with CRE-carriage, by comparing the 188 microbiota of CRE-carriers and non-carriers (Fig 2A)  In this study, we used 16S rRNA genes to study microbial communities in  However, this marker gene cannot directly identify metabolic or other functional capabilities of 199 the microorganisms. Nonetheless, PICRUSt is a technique that uses evolutionary modeling to 200 predict metagenomes from 16S data and a reference genome database. We used this nd their 201 correlation with relative abundances of selected bacteria identified by LefSe. 202 We found that CRE-carriers were enriched in functional categories associated with xenobiotics 203 biodegradation and metabolism (L2), and amino benzoate degradation (L3) (LDA=3.03, 2.14 204 respectively; p<0.02; Fig S2). Moreover, Enterobacteriaceae family positively correlated with Page 11 xenobiotics biodegradation and metabolism (R=0.534; FDR p<0.003). Other functional 206 categories changed in CRE-carriers included reduction in histidine metabolism (L3), elevation in 207 ubiquinone and other terpenoid-quinone biosyntheses (L3), and tryptophan metabolism (L3) (Fig  208   S1). 209

210
This study has demonstrated that CRE-colonized patients have dysbiotic microbiota in terms of 211 community membership with different functional metabolic microbiota profiles. 212 The intestinal microbiota can protect itself against colonization with new bacteria (colonization 213 resistance), while dysbiosis is exploited, apparently, by CRE for colonization. On the other hand, 214 it is also possible that the established CRE-colonization induce significant perturbations to the 215 microbiota, which in turn may act as a pathogenic community to perpetuate host pathology [13]. 216 We observed that healthy individuals has higher microbial diversity while CRE-carriers has the 217 lowest diversity (Fig 1B) [14,15]. Moreover, we observed three clusters indicating different 218 microbial community structure for each of the experimental groups ( Fig 1C). This is in 219 agreement with numerous studies that have shown reduced bacterial diversity in obesity, 220 inflammatory bowel disease, irritable bowel syndrome, and type-2 diabetes mellitus [14,15]. 221 Few specific taxa of the microbiota were different between the CRE-carriers and non-carriers. 222 First, in addition to the CRE-itself, we observed increased Enterobacteriaceae abundance 223 (Enterobacter, Erwinia, Pantoea, Klebsiella), among which were resident species with virulence 224 potential that are normally kept at low levels. This consequently predisposes the host to Second, concurrently with the bloom in Enterobacteriaceae, a depletion of anaerobic-228 commensals was observed (Fig 2), among which were Coprococcus and Faecalibacterium 229 abundance, an important short-chain fatty acids (SCFA) producing commensal bacteria [12]. 230 These SCFAs are physiological byproducts of carbohydrate fermentation by the microbiota, and 231 serve to salvage energy for the host, enhance the mucosal barrier, inhibit intestinal inflammation, 232 and oxidative stress [16]. Among the functional consequences of reduction in anaerobic-bacteria, 233 is a reduced metabolic capacity, often exemplified by a decline in SCFAs production. Dysbiosis 234 caused by broad-spectrum antibiotics (e.g. clindamycin, cephalosporins), which in our case can 235 presumably enable CRE-colonization, is commonly associated with low intestinal SCFAs levels 236 [16]. 237 Colonization resistance depends on microbiota diversity, as well as microbial composition. The 238 intestinal microbiota can protect efficiently against colonization by many enteric pathogens. 239 Therefore, it is not surprising that during dysbiosis, intestinal colonization resistance is impaired. 240 Interestingly, it was shown that Barnesiella spp., which in our study was less abundant in CRE-241 carriers as compared to non-carriers, has the ability to restrict the growth of intestinal pathogens, 242 limit colonization with highly antibiotic-resistant bacteria, and is required to prevent expansion 243 of oxygen-tolerant bacteria such as Enterobacteriaceae [6,7]. 244 As an outcome of dysbiosis, predictions of metabolic function also indicated a profile shift. In 245 CRE-carriers we found changes in abundance of several pathways including increased histidine 246 metabolism, and decreased ubiquinone and other terpenoid-quinone biosynthesis, tryptophan Page 13 metabolism, xenobiotics biodegradation and metabolism, and amino benzoate degradation (Fig  248   S1). These metabolic alterations have been previously linked to modulation of the immune 249 system response to pathogens and the adaptive immune system activation [17,18], and 250 compromised intestinal epithelial barrier and function, which allow bacterial translocation [6,19]. 251 Xenobiotics biodegradation and metabolism category, and specifically the aminobenzoate 252 degradation pathway, generate catechol (1,2-dihydroxybenzene), which promotes 253 Enterobacteriaceae growth and virulence [20]. This can explain the enrichment of this pathway 254 (Fig S1A, S1B) and its positive correlation with Enterobacteriaceae, and may suggest a causative 255 scenario: the change in the microbiota leading to Enterobacteriaceae enrichment. Taken together, 256 the functional prediction of the microbiota leading to enrichment in Enterobacteriaceae, immune 257 system modulation and intestinal epithelial damage, can explain the higher rate of blood-stream 258 infections in the CRE-carrierssince colonization with Enterobacteriaciae has been associated 259 with increased risk for bacteremia [21]. 260 These compositional and functional changes predispose the host to invasive infection and death. 261 We found higher rates of bacteremia (not caused only by CRE) in the CRE-carriers group 262 compared to the non-carriers (Table 1). Interestingly, only one patient had bacteremia with K. 263 pneumoniae KPC, which is consistent with a previous study showing that K. pneumoniae isolates 264 from blood samples were less likely to harbor KPC [3]. This can be explained by the fitness costs 265 of resistance, typically observed as a reduced bacterial growth rate [22]. 266 It can be assumed that the dysbiotic microbiota and high rate of bacteremia in CRE-carriers is 267 linked by low levels of SCFAs, shown to interact with innate mechanisms of defense against Page 14 infection (regulation of immune cell function by SCFA), and low levels of "defensive bacteria" 269 such as Barnesiella [6,16]. 270 Once established, the gut microbiota composition is relatively stable throughout adult life, but 271 can alter as a result of the action of several vectors. In our study, a trend towards a statistically 272 significant difference between the experimental groups was found in the following factors: 273 treatment with carbapenem, chemotherapy treatment, and gastrointestinal disease/disorder. 274 However, we cannot point to the exact determinants influencing the microbiota composition in 275 CRE-carriers. Whatever the predominant factors that modify the microbiota are, the end result is 276 an "unhealthy microbiota" which lost "key species" required for shaping a "healthy microbiota." 277 Indeed, gut microbiota has been previously shown to effect susceptibility to infections caused by 278 other pathogens such as Vibrio cholerae [23], and C. difficile [24]. 279 One major limitation of our study is the strong impact of antibiotic treatment on the gut 280 microbiota [25]. To weaken the effect, one of our control groups was comprised of hospitalized 281 non-CRE-carriers. Both hospitalized groups were hospitalized (for at least seven days) at the 282 same healthcare facility, and were treated with antibiotics profile similar to that of the 283 hospitalized CRE-carriers. As mentioned, no significant confounding factors differed between 284 the CRE-carriers and non-carriers, except for the CRE-carriage itself. There may be other 285 unexamined factors that may cause the microbial differences we found. Moreover, all taxa-286 associated analyses were conducted by excluding the healthy controls (not hospitalized), since 287 the lack of antibiotic treatment does not allow proper comparison between the groups. 288 In addition, on the basis of this study it is not possible to determine causality between dysbiosis 289 and CRE-colonization, as dysbiosis can be both a cause and a result of CRE-colonization. 290 Page 15

291
Overall the results in our cohort, indicate that the interrelation between dysbiotic microbiota, its 292 pool of bacterial genes (microbiome), and their expressing functions might weakens the 293 protection and resistance against CRE-colonization and infection and other pathobionts. 294 Therefore, our study supports the possibility of fecal transplantation as a therapeutic strategy for 295 CRE-carriage, a strategy already efficiently used to treat C. difficile recurrent infection [16,26]   Values are expressed as number (%) of patients, mean value ± SEM. 304 *As compared to healthy group; whereas, the difference between the two other study groups was 305 not significant. 306 Values are expressed as number (%) of patients, mean value ± SEM. 309 *All hospitalized participants were under antibiotic treatment during the fecal sampling; 310 **Bacteremia caused by any bacteria, and not specifically by CRE.