Subscriber access provided by UNIVERSITY OF CONNECTICUT
Communication
Selective imaging of Gram-negative and Gram-positive microbiotas in the mouse gut Wei Wang, Yuntao Zhu, and Xing Chen Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.7b00539 • Publication Date (Web): 06 Jul 2017 Downloaded from http://pubs.acs.org on July 7, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
Selective imaging of Gram-negative and Gram-positive microbiotas in the mouse gut Wei Wang1, Yuntao Zhu1, Xing Chen1,2,3,4* 1
College of Chemistry and Molecular Engineering, 2Peking-Tsinghua Center for Life Sciences, 3Synthetic and Functional Biomolecules Center, and 4Key Laboratory of Bioorganic Chemistry and Molecular Engineering of Ministry of Education, Peking University, Beijing 100871, China KEYWORDS. Gut microbiota imaging, Gram-stain level selectivity, metabolic labeling, LPS, vancomycin probe.
ABSTRACT: The diverse gut microbial communities are crucial for host health. How the interactions between microbial communities and between host and microbes influence host, however, has not been well understood. To facilitate gut microbiota research, selective imaging of specific groups of microbiotas in the gut is of great utility, but remains technically challenging. Here we present a chemical approach that enables selective imaging of Gramnegative and Gram-positive microbiotas in the mouse gut, by exploiting their distinctive cell wall components. Cell-selective labeling is achieved by combined use of metabolic labeling of Gram-negative bacterial lipopolysaccharides with a clickable azidosugar, and direct Gram-positive bacteria labeling with a vancomycin-derivatized fluorescent probe. We demonstrated this strategy by two-color fluorescence imaging of Gram-negative and Gram-positive gut microbiotas in the mouse intestines. This chemical method should be broadly applicable to different gut microbiota research fields, and other bacterial communities studied in microbiology.
The bacterial communities inhabiting the gut of human and other animals consist of hundreds to thousands of species.1 Accumulating evidence indicates that different bacterial groups heterogeneously colonize among niches in the gut, and their spatial distribution affects host health and diseases.2 Furthermore, the eubiosis and dysbiosis of gut microbiota are often characterized by the relative abundance of two dominating phyla, Bacteroidetes and Firmicutes, which are Gram-negative and Gram-positive bacteria, respectively.3 How the microbial eubiosis/dysbiosis and hostmicrobiota communication influence host physiology, however, has not been completely understood.4 To explore these issues, selective imaging of Gram-negative and Gram-positive microbiotas in the gut is of great utility, but remains a technical challenge. Because many microbes in the gut microbiota are not yet amenable to genetic manipulations, tagging the microbiota with fluorescent proteins has only met limited success to date.5 Alternatively, gut microbiotas can be visualized by using fluorescence in situ hybridization (FISH) probes targeting bacterial 16S rRNA.6 However, FISH suffers from its incompatibility with living cells, and the uneven labeling with different bacterial taxa.6, 7 A new method reported recently combined the use of DNA staining with computational analysis of large image datasets to quantify the spatial organization of the gut microbiota.8 It offers a facile way of analyzing the distributions of gut bacteria, but the method alone
does not provide taxonomic information. In addition, the substantial DNA staining of the host tissue and plant materials in the digesta compromises the imaging of gut microbes.7 Another approach reported recently exploited metabolic glycan labeling to incorporate N-azidoacetylgalactosamine (GalNAz) into the polysaccharides of selected bacterial species in culture.9 Subsequent click-labeling with an alkyne-fluorophore conjugate enabled fluorescence imaging of the labeled living bacteria after inoculated to mice. However, this method cannot be used to selectively label gut microbiotas in vivo, because GalNAz is also taken up by host cells and metabolically incorporated into their glycoproteins.10 In addition, not all bacterial species in the gut can be cultured in vitro, even though cultivation procedures are under rapid improvement.11, 12 Therefore, given the limitations of existing strategies, improved methods for selectively labeling gut microbiotas are still needed. Herein, we report the development of a cell-selective labeling method for two-color fluorescence imaging of Gram-negative and Gram-positive microbiotas in the mouse gut, by exploiting their distinctive cell wall components (Figure 1A). Gram-negative bacteria possess the characteristic lipopolysaccharides (LPS) in their outer membrane. LPS in almost all Gram-negative bacteria contains a monosaccharide component, 3-deoxy-D-mannooctulosonic acid (Kdo),13 which is absent in Gram-positive bacteria or animals. An azido analog of Kdo, 8-azido-8-deoxy-Kdo (8AzKdo), has been shown to be able to metabolically label LPS in several Gram-negative species including Escherichia coli, Salmonella typhimurium, Legionella pneumophila, etc. 14, 15 We therefore envisioned that 8AzKdo might be used to selectively label Gram-negative bacteria in the complex gut microbiotas. To evaluate this probe, we first adopted a recently optimized cultivation procedure that allowed for recovery of a representative fraction of human gut microbiota in vitro to assess the 8AzKdo’s labeling specificity within complex microbial samples.11 Microbiotas collected from the large intestine of a specific pathogen free C57BL/6 mouse were anaerobically cultured with 2 mM 8AzKdo for 6 d, followed by reaction with alkynetetramethylrhodamine (TAMRA) via copper(I)-catalyzed azidealkyne cycloaddition (CuAAC, also termed click chemistry).16 Confocal fluorescence microscopy showed strong labeling in a subpopulation of the microbiota with various sizes and shapes (Figure 1B). The unlabeled population was presumably the Grampositive bacteria. Super-resolution imaging by structured illumination microscopy (SIM, Figure 1C) and stochastic optical reconstruction microscopy (STORM, Figure S1) revealed
ACS Paragon Plus Environment
Biochemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 6
Figure 1. Cell-selective labeling of Gram-negative and Gram-positive microbiotas. (A) 8AzKdo enters the biosynthetic pathway of LPS and is metabolically incorporated into LPS in Gram-negative bacteria. Click chemistry is then used to conjugate a fluorescent probe. Gram-positive bacteria and mice do not utilize Kdo and therefore cannot be labeled. Fluorophore-conjugated vancomycin specifically binds to Gram-positive bacteria by recognizing the stem peptide of their exposed peptidoglycan. (B) Confocal fluorescence images of mouse gut microbiotas cultured with vehicle (left) or 8AzKdo (right) and reacted with alkyne-TAMRA. Shown are merged images of DIC and fluorescence. Scale bar, 10 µm. (C) SIM fluorescence images of representative TAMRA-labeled microbiotas. Bright-field (BF) images are shown to the right. Scale bars, 5 µm. (D) Relative phylum (top) and genus (bottom) abundance, as determined by 16S rRNA gene sequencing, of the 8AzKdo-treated and alkyne-TAMRA-labeled microbiotas before and after FACS. Striped bars represent Grampositive bacteria. Bacterial taxa represented by less than 1% are not displayed. (E) Two-color confocal fluorescence images of dually labeled microbiotas. The gut microbiotas were cultured with 8AzKdo and reacted with alkyne-TAMRA, followed by labeling with VancoBODIPY. Scale bar, 20 µm. fluorescence signals on cell surfaces, supporting the metabolic of sorting, the 8AzKdo-labeled bacteria were analyzed by 16S incorporation of 8AzKdo into LPS. We also optimized the click rDNA sequencing, which indicated that they consisted of mostly chemistry protocol to minimize the copper cytotoxicity, thus enaGram-negative bacteria dominated by the Bacteroidetes and Probling its use in labeling live bacteria. The experimental conditeobacteria phyla (Figure 1D, top panel). The dominating Firmictions including premixed 100 µM CuSO4 and the ligand tris[(1utes phylum in the Gram-positive microbiota was not significantly benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA) at a 1:1.5 sorted through the 8AzKdo-based labeling. As shown by the ratio and sodium ascorbate at 400 µM were found to result in the analyses at the genus level, most Gram-negative genera were metabolically labeled with 8AzKdo and sorted with minimal conhighest bacterial survival rate during the reaction (Figure S2). tamination from the Gram-positive genera (Figure 1D, bottom To further validate 8AzKdo’s specificity for Gram-negative panel). These results demonstrate that 8AzKdo can be bacteria, the click-labeled microbiotas were subjected to fluorescence-activated cell sorting (FACS, Figure S3). After two rounds
ACS Paragon Plus Environment
Page 3 of 6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biochemistry
Figure 2. In vivo metabolic labeling and visualization of Gram-negative microbiotas in the mouse gut. (A) In vivo metabolic labeling of Gram-negative microbiota with 8AzKdo. The microbiotas collected from vehicle- (left) or 8AzKdo-gavaged mice (right) were reacted with alkyne-TAMRA and visualized by confocal fluorescence microscopy. Scale bar, 10 µm. (B) The gut microbiotas from vehicle- (left) or 8AzKdo-gavaged (right) mice were reacted with alkyne-biotin and stained with streptavidin-Alexa Fluor 647 (AF647), followed by intragastrical administration into another group of living mice. The intestine sections were then prepared and visualized by fluorescence microscopy. Scale bar, 200 µm. (C) In vivo visualization of previously labeled Gram-negative bacteria in the gut by intravital two-photon microscopy. Representative micrographs of the small intestine with (right) or without (left) administration of TAMRA-linked microbiota are shown. Scale bar, 100 µm. Hoechst staining was used to label the nucleus (blue). (D) Images using in vivo imaging system on dissected intestines after gavage of Cy7-labeled microbiota showed the temporal and spatial distributions of the Gram-negative bacteria. Bar graph demonstrates fluorescence quantification as the average radiant efficiency at each indicated time point. used to selectively label a broad spectrum of Gram-negative bacTAMRA, fluorescent labeling of the Gram-negative microbiota teria in the microbiota culture. was observed (Figure 2A, right). In contrast, no labeling was observed for vehicle-treated mice (Figure 2A, left). As shown by To selectively label the Gram-positive microbiota, we chose to the flow cytometry analysis, the incorporation of 8AzKdo was target their peptidoglycan exposed on the surface (Figure 1A). dose-dependent (Figure S6). Notably, the in vivo labeling effiVancomycin (Vanco) is a widely used Gram-positive-specific ciency of 8AzKdo for microbiota in the mouse gut was signifiantibiotic, functioning by binding the D-Ala-D-Ala motif of pepticantly lower than that performed in the cultured microbiota. This doglycan and inhibiting its construction. The LPS in Gramwas probably due to the relatively lower uptake of 8AzKdo by the negative bacteria is impermeable to Vanco, thus causing their gut microbiota in the living mice. Nevertheless, this in vivo metaintrinsic resistance. Inspired by the successful use of fluorescent bolic labeling of microbiota with 8AzKdo demonstrated its potenVanco probe for labeling peptidoglycan of individual Gramtial as a LPS expression probe for different gut Gram-negative positive species such as Bacillus subtilis and Staphylococcus aubacteria in situ. reus,17, 18 we reasoned that Vanco-based probes might be used to The fluorescently labeled microbiotas were then intragastrically selectively label Gram-positive bacteria in a complex bacterial administered to another group of living mice, and the distribution sample like the gut microbiota. We used the commercially availof the labeled Gram-negative bacteria was visualized by fluoresable Vanco-BODIPY and also synthesized Cyanine 3-derivatized cence microscopy in the frozen intestinal sections (Figure 2B). Vanco (Vanco-Cy3) for microbiota labeling (Scheme S1). We first applied Vanco-BODIPY to the cultured gut microbiotas that Furthermore, employing intravital two-photon microscopy, the administered Gram-negative bacteria could be directly imaged in had already been metabolically incorporated with 8AzKdo and the small intestines of living mice, especially in the villus regions click-labeled with alkyne-TAMRA. As expected, Vancowhere mammalian tissues were delineated by nuclei staining BODIPY fluorescently labeled a subpopulation of the microbiota (Figure 1E). And remarkably, dual labeling with the combination (Figure 2C). In addition, the distribution of 8AzKdo-labeled miof Vanco-BODIPY and 8AzKdo covered a majority (~85%) of crobiota on excised mouse digestive tracts were visualized by an in vivo imaging system on excised mouse digestive tracts (Figure the microbiota with minimal (
