Anaerobic Growth of Listeria monocytogenes on Rhamnose Is Stimulated by Vitamin B12 and Bacterial Microcompartment-Dependent 1,2-Propanediol Utilization

ABSTRACT The foodborne pathogen Listeria monocytogenes can form proteinaceous organelles called bacterial microcompartments (BMCs) that optimize the utilization of substrates, such as 1,2-propanediol, and confer an anaerobic growth advantage. Rhamnose is a deoxyhexose sugar abundant in a range of environments, including the human intestine, and can be degraded in anaerobic conditions into 1,2-propanediol, next to acetate and lactate. Rhamnose-derived 1,2-propanediol was found to link with BMCs in some human pathogens such as Salmonella enterica, but the involvement of BMCs in rhamnose metabolism and potential physiological effects on L. monocytogenes are still unknown. In this study, we first test the effect of rhamnose uptake and utilization on anaerobic growth of L. monocytogenes EGDe without or with added vitamin B12, followed by metabolic analysis. We show that vitamin B12-dependent activation of pdu stimulates metabolism and anaerobic growth of L. monocytogenes EGDe on rhamnose via 1,2-propanediol degradation into 1-propanol and propionate. Transmission electron microscopy of pdu-induced cells shows that BMCs are formed, and additional proteomics experiments confirm expression of pdu BMC shell proteins and enzymes. Finally, we discuss the physiological effects and energy efficiency of L. monocytogenes pdu BMC-driven anaerobic rhamnose metabolism and the impact on competitive fitness in environments such as the human intestine. IMPORTANCEListeria monocytogenes is a foodborne pathogen causing severe illness and, as such, it is crucial to understand the molecular mechanisms contributing to its survival strategy and pathogenicity. Rhamnose is a deoxyhexose sugar abundant in a range of environments, including the human intestine, and can be degraded in anaerobic conditions into 1,2-propanediol. In our previous study, the utilization of 1,2-propanediol (pdu) in L. monocytogenes was proved to be metabolized in bacterial microcompartments (BMCs), which are self-assembling subcellular proteinaceous structures and analogs of eukaryotic organelles. Here, we show that the vitamin B12-dependent activation of pdu stimulates metabolism and anaerobic growth of L. monocytogenes EGDe on rhamnose via BMC-dependent 1,2-propanediol utilization. Combined with metabolic and proteomics analysis, our discussion on the physiological effects and energy efficiency of BMC-driven rhamnose metabolism shed new light to understand the impact on L. monocytogenes competitive fitness in ecosystems such as the human intestine.

ranging from 20 to 30% (1,3). L. monocytogenes is found ubiquitously in natural environments, and it can survive a variety of stress conditions leading to the colonization of different niches, including a range of food processing environments (1,3,4). To survive in such a variety of niches, L. monocytogenes should be able to adapt to environmental stresses and to use a range of nutrients for growth in aerobic and anaerobic conditions (1,5,6).
Recent studies on anaerobic growth of L. monocytogenes have provided evidence that it has the capacity to form proteinaceous organelles so-called bacterial microcompartments (BMCs) that enable extension of its metabolic repertoire by supporting the utilization of 1,2-propanediol and ethanolamine (7)(8)(9). BMCs are self-assembling organelles that consist of an enzymatic core that is encapsulated by a semipermeable protein shell (7,10,11). The separation of the encapsulated enzymes from the cytosol is thought to protect the cell from toxic metabolic intermediates such as aldehydes, and prevent unwanted side reactions (7,10,11). In our previous studies, we showed that the L. monocytogenes 1,2-propanediol utilization gene cluster (pdu) is activated in the presence of 1,2-propanediol and vitamin B 12 , resulting in stimulation of growth in anaerobic conditions (8). Vitamin B 12 is required for activation of the pdu cluster in L. monocytogenes (8,12) and to act as a cofactor of 1,2-propoanediol reductase (13). Activation of BMC-dependent pdu supports degradation of 1,2-propanediol via the toxic intermediate propionaldehyde into 1-propanol and propionate via respective reductive and oxidative branches, with the latter resulting in extra ATP generation leading to enhanced anaerobic growth of L. monocytogenes (8). Notably, 1,2-propanediol is a major end product from the anaerobic degradation of mucus-derived rhamnose by human intestinal microbiota, and it is thought to be an important energy source supporting the intestinal growth of selected pathogens such as Salmonella spp. and L. monocytogenes (7,(14)(15)(16).
Rhamnose is a naturally occurring deoxyhexose sugar abundant in glycans on surfaces of mammalian and bacterial cells and in the cell walls of many plant and insect species (14,17). Anaerobic metabolism of rhamnose has been studied previously in a range of bacteria, including Escherichia coli, and rhamnose is parallelly metabolized into lactaldehyde and dihydroxyacetone phosphate (DHAP) (18,19). DHAP is converted in the glycolytic pathway, leading to a variety of fermentation products, while lactaldehyde is converted to 1,2-propanediol that is subsequently secreted (18,19). Notably, for example in Salmonella spp. and Clostridium phytofermentans, rhamnose-derived 1,2-propanediol can be converted to 1-propanol and propionate via BMC-dependent pdu (14,16). Although rhamnose-derived 1,2-propanediol was found to be metabolized via a pduD-dependent pathway in Listeria innocua (20), the possible activation and contribution of BMC-dependent pdu to anaerobic metabolism and growth of L. monocytogenes on rhamnose remains to be investigated.
In this study, we first quantified the effect of rhamnose as sole carbon source on anaerobic growth and metabolism of L. monocytogenes in absence or presence of vitamin B 12 (cobalamin), an essential cofactor of 1,2-propanediol reductase, the signature enzyme of BMC-dependent pdu (13). Next, we analyzed rhamnose utilization and end product formation and, combined with transmission electron microscopy (TEM) and proteomics, we provide evidence for a B 12 -dependent pdu-induced metabolic shift. We summarize our findings in a model integrating BMC-dependent pdu with rhamnose metabolism and discuss impact on growth and survival of L. monocytogenes in anaerobic environments such as the human intestine.

RESULTS
Activation of pdu stimulates anaerobic growth of L. monocytogenes EGDe on rhamnose. We first examined whether rhamnose can function as a sole carbon source to support anaerobic growth of L. monocytogenes EGDe in MWB defined medium without or with added vitamin B 12 (cobalamin) (Fig. 1). In MWB (modified Welshimer's broth) defined medium supplied with 20 mM rhamnose, the optical density at 600 nm (OD 600 ) reaches a maximum of about 0.37 after 48 h, while in MWB supplied with 20 mM rhamnose and 20 nM B 12 OD 600 continues to increase after 48 h, reaching a significantly higher OD 600 of 0.51 at 72 h. Enhanced growth on MWB supplied with rhamnose and B 12 compared to MWB plus rhamnose is also evident from plate counts, which increase from 6.5 to 8.2 log 10 CFU/ml and from 6.5 to 7.2 log 10 CFU/ml, respectively (Fig. 1B). There is no significant difference in growth performance of L. monocytogenes EGDe on MWB supplied with 20 mM glucose and MWB supplied with 20 mM glucose and 20 nM B 12 , and at 48 h final levels of 8.8 log 10 CFU/ml were reached (see Fig. S1 in the supplemental material). These results suggest that B 12 -stimulated anaerobic growth of L. monocytogenes EGDe on MWB medium with rhamnose as the sole carbon source is linked to the activation of pdu.
Activation of pdu supports 1,2-propanediol degradation and stimulates rhamnose metabolism. To confirm possible activation of pdu, metabolic analysis via high-pressure liquid chromatography (HPLC) was conducted to quantify substrate consumption and product formation after anaerobic growth of L. monocytogenes EGDe on MWB plus 20 mM rhamnose and on MWB plus 20 mM rhamnose and 20 nM B 12 . As shown in Fig. 2A, at 72 h, the initial 20 mM rhamnose is completely consumed under a pdu-induced condition, whereas 3.5 mM rhamnose is retained under a pdu-noninduced condition. Additional end product analysis at 72 h shows the accumulation of ;6.7 mM 1,2-propanediol under a pdu-noninduced condition and nearly zero produc-  Visualization of BMCs and expression analysis of BMC shell proteins. To determine whether BMCs are formed to support the utilization of rhamnose-derived 1,2-propanediol, TEM was performed to observe BMCs structures, and proteomics was applied to measure the expression of BMC shell proteins (Fig. 3A). The pdu-induced cells clearly contain BMC-like structures (60 to 70% of 300 BMC-positive cells) with an approximate diameter of 50 to 80 nm, while similar structures were not observed in pdu-noninduced cells. Notably, the identified structures strongly resemble TEM pictures of previously reported pdu BMCs in L. monocytogenes (8,9) and in S. enterica and E. coli (13,21). Compared to pdu-noninduced cells, pdu-induced cells show significant upregulation of 21 measurable Pdu proteins (Fig. 3B), including seven proteins annotated as BMCs shell proteins, PduTUABKJN. Notably, pdu-induced and pdu-noninduced rhamnose-grown cells show similar expression of proteins in the rhamnose metabolism cluster (lmo2850, rhaA, rhaB, and rhaM) (Fig. 3B), which indicates that the activation of pdu BMC does not affect the expression of these enzymes.
Proteomics-based pathway visualization of propanoate metabolism and vitamin B 12 metabolism. To visualize the metabolism from 1,2-propanediol to propanoate (propionate) and 1-propanol, the identified proteins and expression levels presented in Table S1, are mapped to propanoate metabolic pathways of L. monocytogenes EGDe. As shown in Fig. 4A, the enzymes involved in degradation of rhamnose-derived 1,2propanediol into propanoate (propionate) and 1-propanol are all significantly upregulated under pdu-induced conditions compared to pdu-noninduced conditions. The  Table S1.

DISCUSSION
The presented model of 1,2-propanediol BMCs in rhamnose metabolism is based on growth phenotypes, metabolic analysis, proteomics, TEM visualization, and our understanding of 1,2-propanediol BMCs in anaerobic growth of L. monocytogenes EGDe. As illustrated in Fig. 5, the rhamnose catabolism gene cluster (rha) in L. monocytogenes EGDe is composed of lmo2846-lmo2851 (25). lmo2850 encodes a secondary transporter which has high similarity with L-rhamnose permease RhaT in E. coli (26)(27)(28) and is conceivably acting as the transporter of a-L-rhamnose. L-Rhamnose mutarotase RhaM mediates the conversion of a-L-rhamnose into b-L-rhamnose (also called L-rhamnopyranose) (25,29). b-L-Rhamnose is converted to L-rhamnulose by L-rhamnose isomerase RhaA (25,30). L-Rhamnose is then phosphorylated to L-rhamnulose 1-phosphate by rhamnulokinase RhaB with one ATP consumption (25,30). L-Rhamnulose 1-phosphate is split into (S)-lactaldehyde and dihydroxyacetone phosphate (DHAP) by rhamnulose-1-phosphate aldolase RhaD (25,30). DHAP can be metabolized to glyceraldehyde 3-phosphate via triosephosphate isomerase 1 TpiA1 and, via the glycolytic pathway (14,31) and the GABA (g-aminobutyric acid) shunt in the incomplete tricarboxylic acid cycle in L. monocytogenes (32), to the end products acetate and lactate, as confirmed in our metabolic analysis. The observed production of 1,2-propanediol in pdu-noninduced conditions confirms the predicted anaerobic conversion of lactaldehyde to 1,2-propanediol in L. monocytogenes EGDe. The activity of lactaldehyde reductase has not been described in L. monocytogenes (33), but protein similarity alignment with lactaldehyde reductase FucO of Escherichia coli (33) suggests four putative candidates annotated as alcohol dehydrogenase in L. monocytogenes EGDe, including lmo1166, lmo1171, lmo1634, and lmo1737, detected in the proteomes of both pdu-noninduced and pdu-induced cells (for details see Text S1 in the supplemental material). Since the discovery of the role of pdu BMCs dehydratase in rhamnose (and fucose) utilization, two pathway scenarios have been proposed, one with and one without lactaldehyde reductase encapsulated inside BMCs (25,29). In line with previously reported comparative genomic analysis (25,29), our data now provide evidence for the latter model to be active in L. monocytogenes since rhamnose is converted via lactaldehyde to 1,2-propanediol in the absence of BMCs under the pdu-noninduced condition, while with added B 12 the metabolism of 1,2-propanediol proceeds via pdu BMCs.
The activation of pdu BMCs enhances anaerobic rhamnose metabolism in L. monocytogenes and conceivably generates additional energy via the ATP-producing propionate branch in pdu and via enhanced flux into the glycolytic pathway resulting in a significant stimulation of growth. At 72 h, 20 mM rhamnose is metabolized into 7.6 mM acetate, 5.1 mM lactate, 1.4 mM 1,2-propanediol, 3.4 mM propionate, and 3.6 mM 1-propanol under the pdu-induced condition, whereas 16.5 mM rhamnose is metabolized into 4.1 mM acetate, 2.3 mM lactate, and 6.7 mM 1,2-propanediol under the pdu-noninduced condition. The theoretical ATP yield from rhamnose conversion to lactate, acetate, and propionate includes the production of 1.5 ATP per 1 lactate, 2.5 ATP per 1 acetate, and 0.5 ATP per 1 propionate produced (for details of the reactions, see Table S3). Based on the concentrations of end products at 72 h, pdu-induced cells theoretically generate 1.425 ATP per 1 rhamnose, while pdu-noninduced cells generate 0.830 ATP per 1 rhamnose (for details on the calculations, see Table S3). The theoretical energy gain of L. monocytogenes EGDe from anaerobic rhamnose metabolism with the activation of 1,2-propanediol BMCs could offer an explanation for the 10-fold-higher number of CFU reached (8.2 log 10 CFU/ml) compared to pdu-noninduced conditions (7.2 log 10 /ml).
Our data provide evidence for another extension of the BMC-dependent metabolic repertoire of L. monocytogenes under anaerobic conditions that now includes BMC-dependent ethanolamine utilization (eut) (9), BMC pdu (8), and BMC pdu-stimulated rhamnose metabolism. The indicated substrates can be found in a wide range of environments, including foods and the human gastrointestinal tract. Substrates for microcompartment metabolism such as ethanolamine and 1,2-propanediol are constantly produced in the human intestine by bacterial metabolism of food or host cell components. Enteric pathogens such as Salmonella spp. gain a competitive advantage in the intestine by utilizing these substrates, an advantage enhanced by the host inflammatory response (15,(34)(35)(36). It is conceivable that the competitive fitness of L. monocytogenes can be enhanced by activation of BMC-dependent eut and pdu, with corresponding substrates provided by enzymatic activities of gut microbiota, such as the release of ethanolamine following membrane phospholipid degradation and the release of rhamnose following mucus glycan hydrolysis activity, and propanediol as a fermentation product (15). Notably, despite the presence of a complete vitamin B 12 synthesis cluster, we found that eut (9), pdu (8), and pdu-stimulated rhamnose utilization in L. monocytogenes in the present study requires supplementation of B 12 to the medium. This points to an important role of B 12 in activation of L. monocytogenes BMCmediated metabolic pathways containing B 12 -dependent signature aldehyde reductases. Vitamin B 12 can be found in foods, including meat and dairy products (23,37), and is also found in human intestine, where part of the B 12 is derived from gut microbiota that have the capacity to produce B 12 (12,23). The fact that in the present study we observed induction of the B 12 synthesis pathway in cells grown in MWB plus rhamnose but no activation of B 12 -dependent pdu, whereas activation was found with B 12 added to the medium, points to an intricate regulation of the B 12 synthesis pathway and its connection to BMCs activation. In addition to earlier studies on transcriptional and translational control of BMC eut and pdu in L. monocytogenes (1,12,15,23,38), studies are required to assess for example impact of extracellular and intracellular B 12 concentrations on activation of BMC pathways and their role in L. monocytogenes ecophysiology and virulence.

MATERIALS AND METHODS
Strains, culture conditions, and growth measurements. All experiments in this study were carried out with L. monocytogenes EGDe anaerobically grown at 30°C in defined medium MWB (Modified Welshimer's broth) (39). Overnight-grown cells in Luria broth (LB) were washed three times in phosphate-buffered saline before inoculation into MWB. MWB was supplemented with 20 mM L-rhamnose as the sole carbon source with or without the addition of 20 nM vitamin B 12 . Anaerobic conditions were achieved by using an Anoxomat anaerobic culture system with a gas mixture composed of 10% CO 2 , 5% H 2 , and 85% N 2 . MWB with 20 mM rhamnose and 20 nM vitamin B 12 was defined as a rhamnose pduinduced condition, while MWB with 20 mM rhamnose was defined as a rhamnose pdu-noninduced condition. OD 600 measurements in MWB were performed every 12 h for 3 days. Plate counting in MWB to quantity the CFU was performed every 24 h for 3 days. All growth measurements were performed with three independent experiments with three technical repeats.
Analysis of metabolites for rhamnose metabolism using HPLC. Samples were taken from the cultures at 0, 24, 48, and 72 h. After centrifugation, the supernatant was collected for the HPLC measurements of rhamnose, acetate, lactate, 1,2-propanediol, 1-propanol, and propionate. The experiment was performed with three biological replicates. In addition, the standard curves of all the metabolites were measured in the concentrations 0.1, 1, 5, 10, and 50 mM. HPLC was performed using an Ultimate 3000 HPLC (Dionex) equipped with an RI-101 refractive index detector (Shodex, Kawasaki, Japan), an autosampler, and an ion-exclusion Aminex HPX-87H column (7.8 mm by 300 mm) with a guard column (Bio-Rad, Hercules, CA). As the mobile phase, 5 mM H 2 SO 4 was used at a flow rate of 0.6 ml/min, and the column was kept at 40°C. The total run time was 30 min, and the injection volume was 10 ml. All HPLC measurements were performed with three independent experiments with three technical repeats.
TEM. L. monocytogenes EGDe cultures were grown anaerobically at 30°C under rhamnose pdu-induced or rhamnose pdu-noninduced conditions. Samples were collected at 48 h of incubation. About 10mg of dry cells was fixed for 2 h in 2.5% (vol/vol) glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.2). After a rinse in the same buffer, postfixation was done in 1% (wt/vol) OsO 4 for 1 h at room temperature. The samples were dehydrated by ethanol and were then embedded in resin (Spurr HM20) for 8 h at 70°C. Thin sections (,100 nm) of polymerized resin samples were obtained with microtomes. After being stained with 2% (wt/vol) aqueous uranyl acetate, the samples were analyzed with a JEOL 1400 plus TEM at a 120-kV setting (8,9). The observation of BMCs structures was performed within three biological replicates, and determination of the fraction of BMC-positive cells was based on the analysis of 300 cells in respective TEM pictures for both pdu-induced and pdu-noninduced conditions, as previously described (8).
Proteomics. L. monocytogenes cultures were anaerobically grown at 30°C under rhamnose pduinduced and rhamnose pdu-noninduced conditions. Samples were collected at 48 h of incubation and then washed twice with 100 mM Tris (pH 8). About 10 mg (wet weight) of cells in 100ml of 100 mM Tris was sonicated for 30 s twice to lyse the cells. Samples were prepared according to the filter-assisted sample preparation protocol (FASP) with the following steps: reduction with 15 mM dithiothreitol, alkylation with 20 mM acrylamide, and digestion with sequencing-grade trypsin overnight (40). Each prepared peptide sample was analyzed by injecting (18ml) into a nano-LC-MS/MS (Thermo nLC1000 connected to a LTQ-Orbitrap XL) as described previously (8,9). Liquid chromatography-mass spectrometry (LC-MS) data with all MS/MS spectra were analyzed with the MaxQuant quantitative proteomics software package as described before (8,9,41). A protein database with the protein sequences of L. monocytogenes EGDe (ID UP000000817) was downloaded from UniProt. Filtering and further bioinformatics and statistical analysis of the MaxQuant ProteinGroups file were performed with Perseus (42). Reverse hits and contaminants were filtered out. Protein groups were filtered to contain minimally two peptides for protein identification, of which at least one is unique and at least one is unmodified. A volcano plot was prepared based on the Student t test difference of a Pdu-induced versus a Pdu-noninduced control. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (43) partner repository with the data set identifier PXD025734 (https://www.ebi.ac.uk/pride/archive/projects/PXD025734).
Bioinformatics and statistical analysis. Pathview R package (44) to visualize the proteomics data: the UniProt protein IDs from Table S1 in the supplemental material were collected and retrieved to Entre IDs. A list of Entrez IDs, protein expression indicated by LFQ intensity (see Table S2), was mapped to the L. monocytogenes EGDe KEGG pathway database using the tool Pathview (R version 3.2.1). The box represents genes, and the different colors indicate the level of expression with default setting.
Statistical analyses were performed in Prism 8.0.1 for Windows (GraphPad Software). As indicated in the figure legends, statistical significances were determined using a Holm-Sidak t test and are indicated in the figures (***, P , 0.001; *, P , 0.05; ns, P . 0.05 ).

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. TEXT S1, TXT file, 0.002 MB. FIG S1, TIF file, 1.1 MB.