Introduction

Polyamines are aliphatic amines that contain more than two amino groups and are present in most organisms, including bacteria, archaea, and eukarya (Michael 2015). The common, major polyamines are putrescine, spermidine, and spermine. These molecules play roles in several biological processes, including the regulation of transcription and translation (Miller-Fleming et al. 2015); therefore, polyamines are important for basic cell functions.

In the human intestine, it is likely that the gut microbiota, composed of at least 160 different species in each individual (Qin et al. 2010), provides polyamines for the host (Noack et al. 1998; Matsumoto et al. 2012). Although polyamines can be supplied orally through the diet and can extend the life span of worms (Eisenberg et al. 2009), flies (Eisenberg et al. 2009), and mice (Soda et al. 2009), most dietary polyamines are rapidly absorbed in the small intestine (Uda et al. 2003). Therefore, the intestinal microbiota may play a key role in regulating polyamine concentrations in the lower part of intestine such as the large intestine. Furthermore, recent studies showed that increased polyamine concentrations in the intestine enhanced the longevity of mice (Matsumoto et al. 2011; Kibe et al. 2014).

Understanding the bacterial pathways involved in polyamine biosynthesis is essential for controlling polyamine concentrations in the intestine. However, few studies have examined polyamine biosynthetic pathways in intestinal bacterial species, particularly the predominant intestinal bacterial species. Although bacterial polyamine biosynthetic pathways have been widely studied in the model bacterium Escherichia coli (Tabor and Tabor 1985), this bacterial species is a minor component in the human intestine. The polyamine biosynthetic pathway in E. coli differs from those predicted to be present in the most predominant intestinal microbiota species (Hanfrey et al. 2011). Specifically, E. coli utilizes S-adenosylmethionine decarboxylase and spermidine synthase to biosynthesize spermidine from putrescine, whereas most intestinal bacteria are predicted to utilize carboxyspermidine dehydrogenase and carboxyspermidine decarboxylase (CASDC) for spermidine biosynthesis. However, CASDC has not been functionally characterized in intestinal bacteria, with the exception of the food-borne pathogen Campylobacter jejuni (Deng et al. 2010; Hanfrey et al. 2011).

Predominant intestinal microbiota species are mainly classified into two phyla: Firmicutes and Bacteroidetes. Among Bacteroidetes, species in the genus Bacteroides, which extensively harbor CASDC homologs (Hanfrey et al. 2011), are highly abundant in the adult human intestine (Qin et al. 2010), and bacteria in this genus are often utilized as model intestinal bacterial species. In fact, the colonization mechanisms of Bacteroides in the intestine have been widely studied (Goodman et al. 2009; Lee et al. 2013; Degnan et al. 2014). Here, we investigated the polyamine biosynthetic ability of Bacteroides thetaiotaomicron. This study revealed that CASDC is essential for spermidine biosynthesis in B. thetaiotaomicron and contributes to normal growth.

Materials and methods

Bacterial strains and culture conditions

The bacterial strains, plasmids, and primers used in this study are shown in Table 1. B. thetaiotaomicron was anaerobically cultured at 37 °C in liquid Gifu anaerobic medium (GAM; Nissui Pharmaceutical Co., Ltd., Tokyo, Japan), brain heart infusion (BHI) agar medium (Sigma-Aldrich Corp., St. Louis, MO, USA) containing 10 % horse blood (Nippon Bio-Supp. Center, Tokyo, Japan; referred to as BHI-blood agar medium), or polyamine-free minimal medium (pH 7.2) (Koropatkin et al. 2008) with minor modifications. The composition of the minimal medium was as follows: 0.5 % (w/v) glucose, 100 mM KH2PO4, 15 mM NaCl, 8.5 mM (NH4)2SO4, 4 mM l-cysteine, 1.9 μM hematin, 200 μM l-histidine, 1 μg/mL vitamin K3, 5 ng/mL vitamin B12, 100 μM MgCl2, 1.4 μM FeSO4, and 50 μM CaCl2. Anaerobic culture was performed in an anaerobic jar using the AnaeroPack (Mitsubishi Gas Chemical Co., Inc., Tokyo, Japan) or in the anaerobic chamber InvivO2 400 (10 % CO2, 10 % H2, and 80 % N2; Ruskinn Technology Ltd., Bridgend, UK). E. coli CC118 λpir was used as the cloning host. E. coli was aerobically cultured at 37 °C in Luria–Bertani medium. When appropriate, ampicillin (100 μg/mL for E. coli), erythromycin (25 μg/mL for B. thetaiotaomicron), gentamycin (200 μg/mL for B. thetaiotaomicron), and 5-fluoro-2′-deoxyuridine (200 μg/mL for B. thetaiotaomicron) were added to the medium.

Table 1 Bacterial strains, plasmids, and primers used in this study
Full size table

Generation of casdc deletion mutant and complementation strain of B. thetaiotaomicron

Deletion of casdc, the gene encoding CASDC, was performed using a double-crossover recombination technique that has been previously described (Koropatkin et al. 2008). In this technique, the gene encoding thymidine kinase (tdk) was used as a counter-selectable marker because the host strain B. thetaiotaomicron Δtdk (hereafter referred to as MS39) is resistant to the toxic nucleotide analog 5-fluoro-2′-deoxyuridine.

The plasmid for casdc deletion in B. thetaiotaomicron was constructed as follows. The DNA fragments corresponding to the upstream and downstream regions of casdc were amplified using the primer pairs Pr-1/Pr-2 and Pr-3/Pr-4, respectively. The genome of B. thetaiotaomicron JCM 5827T was used as the template. The amplified DNA strands were then fused by overlap PCR using the Pr-1/Pr-4 primer pair and inserted into the SalI site of the pExchange-tdk vector (Koropatkin et al. 2008) using an In-Fusion HD cloning kit (Clontech Laboratories, Inc., Mountain View, CA, USA). The resulting plasmid was named pMSK3.

Conjugative transfer of a suicide plasmid was carried out as follows. Cultured cells of E. coli S17-1 λpir harboring pMSK3 and B. thetaiotaomicron MS39 (Δtdk) were mixed in 1 mL of liquid GAM and inoculated onto BHI-blood agar medium. Cells were aerobically cultured at 37 °C for 1 day and suspended in 4 mL of liquid GAM. Suspended cells were plated on BHI-blood agar medium containing erythromycin and gentamycin, and pMSK3-inserted B. thetaiotaomicron strains (single-crossover strains harboring tdk) were selected. These strains were subjected to anaerobic culture in BHI-blood agar medium containing 5-fluoro-2′-deoxyuridine. As a result, mutants carrying a deletion of casdc (hereafter referred to as MS56; double-crossover strains) were generated.

The plasmid pMSK4 used for complementation analysis was constructed by inserting the casdc gene into the PstI- and NotI-digested pNBU2-bla-ermGb plasmid (Koropatkin et al. 2008). The casdc gene was PCR-amplified with the primer pair Pr-29/Pr-30 using B. thetaiotaomicron JCM 5827T genome DNA as the template. The resulting plasmid, pMSK4, was introduced into MS56 (Δtdk Δcasdc) by conjugative transfer, and erythromycin-resistant strains were selected. The integration of pMSK4 occurred at the NBU2 att1 site (tRNASer gene BT_t70) in the genome of MS56 (Δtdk Δcasdc) to generate complemented Δcasdc strains (hereafter referred to as MS93).

High-performance liquid chromatography (HPLC)

An HPLC system (Chromaster, Hitachi Ltd, Tokyo, Japan) equipped with a cation-exchange column (#2619PH, 4.6 × 50 mm; Hitachi) maintained at 67 °C was used to measure polyamine concentrations. Polyamines were eluted by mobile phase A (45.2 mM tri-sodium citrate, 63.3 mM sodium chloride, and 60.9 mM citric acid) and mobile phase B (200 mM tri-sodium citrate, 2 M sodium chloride, 5 % ethanol, and 5 % 1-propanol). The concentration of mobile phase B was linearly increased from 50 to 85 % during minutes 0–6, maintained at 85 % during minutes 6–12, increased to 100 % during minutes 12–18, maintained at 100 % during minutes 18–45, and then returned to 50 % for minutes 45–60. Eluted polyamines were derivatized with o-phthalaldehyde using the post-column method and were detected using a fluorescence detector (λ ex 340 nm, λ em 435 nm). For the derivatization of polyamines, reaction solution 1 (0.4 N NaOH) and reaction solution 2 (234 mM boric acid, 0.05 % Brij-35, 5.96 mM o-phthalaldehyde, 0.2 % 2-mercaptoethanol) were mixed with the eluate at 67 °C. The concentration of each polyamine was calculated based on a standard curve created using standards of known concentrations. The standards used and their retention times were as follows: agmatine, 34.8 min; cadaverine, 20.6 min; carboxyspermidine, 8.4 min; putrescine, 15.2 min; spermidine, 26.0 min; and spermine, 39.1 min.

Analysis of polyamines in cells and culture supernatant by HPLC

B. thetaiotaomicron strains were cultured overnight in 4 mL of liquid GAM. Cultured cells were harvested by centrifugation at 3400×g for 3 min and washed once with polyamine-free minimal medium. Washed cells were diluted to an initial optical density at 600 nm (OD600) of 0.03 in 30 mL of the same minimal medium and cultured at 37 °C. Growth was monitored by measuring OD600 with a spectrophotometer. Cultures were withdrawn at the indicated times and centrifuged at 18,700×g for 3 min at 4 °C in order to collect the cells and supernatant.

To analyze intracellular polyamines, the collected cells were washed once with phosphate-buffered saline (18,700×g, 4 °C, 5 min). The cells were then resuspended in 300 μL of 5 % (v/v) trichloroacetic acid and incubated in boiling water for 15 min. After centrifugation (18,700×g, 4 °C, 5 min), the supernatants were filtered through a Cosmonice filter W (Nacalai Tesque Inc., Kyoto, Japan) and subjected to HPLC analysis, and cell debris was dissolved in 300 μL of 0.1 N NaOH. Protein concentration in the NaOH solution was measured by the Bradford method using a Bio-Rad protein assay kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA) with bovine serum albumin as a standard. The resulting concentration of intracellular polyamines was determined as nmol/mg cellular protein or nmol/mL culture.

To analyze polyamines in the culture supernatant, the supernatants collected from cultures in minimal medium were mixed with trichloroacetic acid to give a final concentration of 10 % (v/v) and centrifuged twice at 18,700×g at 4 °C (first centrifugation, 5 min; second centrifugation, 15 min). Next, the solutions were filtered through a Cosmonice filter W (Nacalai Tesque) and subjected to HPLC analysis. The resulting concentration of polyamines in the culture supernatant was determined in nmol/mL culture, which is equal to the μM.

Evaluation of CASDC activity in B. thetaiotaomicron

The CASDC activity in B. thetaiotaomicron strains was evaluated using a cell-free extract. B. thetaiotaomicron strains were cultured at 37 °C in 30 mL of minimal medium until the OD600 reached 0.5–0.6. The cells were harvested by centrifugation (10,000×g, 4 °C, 3 min) and washed once with 85 mM HEPES buffer (pH 8.2) containing 4 mM dithiothreitol, 50 μM pyridoxal 5′-phosphate (PLP), and 0.1 mM EDTA. Washed cells were resuspended in 3 mL of the same buffer, and the cell membranes were disrupted by sonication using the QSonica Q500 sonicator (QSonica LLC, Newton, CT, USA). Cell debris was removed by centrifugation at 12,000×g at 4 °C for 30 min. Supernatants were replaced with 85 mM HEPES buffer (pH 8.2) containing 4 mM dithiothreitol and 50 μM PLP using an Amicon Ultra-15 centrifugal filter unit (10 kDa cut off; Millipore Corp., Billerica, MA, USA) and were concentrated to a protein concentration of 7.2 mg/mL using an Amicon Ultra-15 centrifugal filter unit (10 kDa cut off; Millipore). The resulting solutions were used as cell-free extracts of B. thetaiotaomicron. A reaction mixture containing 1 mM carboxyspermidine, 6.48 mg/mL cell-free extract, 76.5 mM HEPES buffer (pH8.2), 3.6 mM dithiothreitol, and 45 μM PLP was incubated at 37 °C for 9 h. The enzymatic reaction was stopped by adding trichloroacetic acid to give a final concentration of 10 % (v/v). This solution was appropriately diluted with 85 mM HEPES buffer (pH 8.2) containing 10 % (v/v) trichloroacetic acid and applied to HPLC analysis after centrifugation (18,700×g, 4 °C, 15 min) and filtration with a Cosmonice filter W (Nacalai Tesque). CASDC activity was evaluated by analyzing the decrease in carboxyspermidine and increase in spermidine in the solution. The specific activity was expressed as μmol of spermidine formed per min per mg of protein in the cell-free extract.

Prediction of polyamine biosynthetic pathways and polyamine transport systems in B. thetaiotaomicron

The polyamine biosynthetic pathways and polyamine transport systems in B. thetaiotaomicron VPI-5482T (same strain as ATCC 29148T and JCM 5827T) were predicted using in silico analysis, for which a BLAST search (http://blast.ncbi.nlm.nih.gov) was carried out using protein sequences involved in the biosynthesis or transport of polyamines as query sequences (Supplementary Table S1). To investigate the presence of the ATP-binding cassette transporters PotABCD and PotFGHI, only solute-binding proteins PotD and PotF, respectively, were used as queries. B. thetaiotaomicron VPI-5482T proteins with a BLAST score of over 100 bits were identified to be homologs of the query proteins. When two or more query proteins were homologous to same protein of B. thetaiotaomicron VPI-5482T, functional assignment of a more similar query protein to the B. thetaiotaomicron protein was utilized.

Results

Polyamine biosynthetic pathway and transport systems in B. thetaiotaomicron

Protein BLAST searches suggested that B. thetaiotaomicron harbors a polyamine biosynthetic pathway and transport systems as illustrated in Fig. 1. This polyamine biosynthetic pathway is homologous to that of the food-borne pathogen C. jejuni (Hanfrey et al. 2011) and includes the spermidine biosynthetic enzymes carboxyspermidine dehydrogenase and CASDC. In addition, genes encoding the putative agmatine/putrescine antiporter AguD (Driessen et al. 1988) and putative spermidine ATP-binding cassette transporter PotABCD (Furuchi et al. 1991) are present in the genome of B. thetaiotaomicron, suggesting that this organism can not only synthesize polyamines but also transport them.

Fig. 1
figure 1

Predicted polyamine biosynthetic pathway and polyamine transport systems in B. thetaiotaomicron. Enzyme and transporter names are indicated in white letters, while chemical compound names are shown in black letters. Reactions and transports predicted by in silico analysis are presented by dotted arrows. Reaction experimentally verified in this study is indicated with a solid arrow. AguD agmatine/putrescine antiporter, AIH agmatine deiminase/iminohydrolase, CASDC carboxyspermidine decarboxylase, CASDH carboxyspermidine dehydrogenase, NCPAH N-carbamoylputrescine amidohydrolase, PotABCD spermidine ATP-binding cassette transporter, SpeA arginine decarboxylase

Full size image

B. thetaiotaomicron accumulates spermidine

To evaluate polyamine-producing ability, B. thetaiotaomicron JCM 5827T and MS39 (Δtdk) were grown in polyamine-free minimal medium and the cells and supernatants were subjected to HPLC analysis. The organisms secreted only minimal amounts of polyamines (<0.25 nmol/mL culture), but rather accumulated spermidine intracellularly as the sole polyamine (JCM 5827T, 138.5 nmol/mg cellular protein or 5.5 nmol/mL culture; MS39 (Δtdk), 135.5 nmol/mg cellular protein or 6.9 nmol/mL culture) (Fig. 2). The concentration of intracellular spermidine of MS39 (Δtdk) was high during the mid-log phase and then slightly decreased after entry to the stationary phase (Fig. 3). These results revealed that B. thetaiotaomicron mainly synthesizes spermidine as a final polyamine product but does not export it, at least under the conditions used in this study.

Fig. 2
figure 2

Polyamine profiles in culture supernatant and cells of B. thetaiotaomicron grown in polyamine-free minimal medium. JCM 5827T and MS39 (Δtdk) were cultured in polyamine-free minimal medium until the OD600 reached 0.5–0.6. The cells and supernatants were collected by centrifugation and subjected to HPLC analysis to investigate the polyamine profiles in the a culture supernatant and b cells. Spd indicates the peak corresponding to spermidine

Full size image
Fig. 3
figure 3

Intracellular spermidine content of B. thetaiotaomicron strains grown in polyamine-free minimal medium. Cultures of MS39 (Δtdk), MS56 (Δtdk Δcasdc), and MS93 (Δtdk Δcasdc att1::casdc +) grown in polyamine-free minimal medium were centrifuged, and the cells were collected. These cells were subjected to HPLC analysis to investigate the concentration of intracellular spermidine. The open triangle represents MS39 (Δtdk). The closed and open circles indicate MS56 (Δtdk Δcasdc) and MS93 (Δtdk Δcasdc att1::casdc +), respectively. Data are shown as the mean ± standard deviation (n = 3)

Full size image

Accumulation of intracellular spermidine in B. thetaiotaomicron depends on the CASDC-encoding gene

Deletion of the casdc gene encoding CASDC was performed in B. thetaiotaomicron to validate the function of the polyamine (spermidine) biosynthetic pathway shown in Fig. 1. Gene disruption was carried out in a Δtdk background for easier strain construction. Δtdk appeared to have no effect on the polyamine biosynthetic ability of B. thetaiotaomicron (Fig. 2). In the resulting mutant MS56 (Δtdk Δcasdc), the concentration of intracellular spermidine was reduced to a nearly undetectable level (Fig. 3). Putrescine and carboxyspermidine were also not detectable in MS56 (Δtdk Δcasdc) (data not shown), but agmatine accumulation was observed in the cell (25.4–42.6 nmol/mg cellular protein or 0.85–4.35 nmol/mL culture) and culture supernatant (maximum 108.8 nmol/mL culture at 30 h of culture time). In contrast, the complemented strain MS93 (Δtdk Δcasdc att1::casdc +) carrying a single copy of casdc in the genome accumulated spermidine in the cell at approximately the same concentration as the parental strain MS39 (Δtdk) (Fig. 3). Agmatine, which was detected in MS56 (Δtdk Δcasdc), was not detected in the cells or supernatant of MS93 (Δtdk Δcasdc att1::casdc +) (data not shown). These results indicate that CASDC contributes to the accumulation of spermidine in B. thetaiotaomicron cells.

In vitro assay of CASDC in B. thetaiotaomicron

An in vitro assay of CASDC was performed using carboxyspermidine (substrate) and cell-free extracts prepared from MS39 (Δtdk), MS56 (Δtdk Δcasdc), and MS93 (Δtdk Δcasdc att1::casdc +). Incubation of the reaction mixture without cell-free extract (negative control) did not decrease carboxyspermidine and increase spermidine (Fig. 4a). When the cell-free extract from MS39 (Δtdk) was added, carboxyspermidine levels decreased and spermidine levels increased (Fig. 4b), indicating the presence of CASDC activity (76.9 μmol/min/mg protein). In contrast, the cell-free extract from MS56 (Δtdk Δcasdc) did not exhibit a decrease in carboxyspermidine (Fig. 4c; CASDC activity 0.1 μmol/min/mg protein). Incubation with the cell-free extract from the complemented strain MS93 (Δtdk Δcasdc att1::casdc +) showed similar characteristics to MS39 (Δtdk) (Fig. 4d), indicating the restoration of CASDC activity (104.1 μmol/min/mg protein). These results strongly suggest that CASDC is the sole enzyme that coverts carboxyspermidine to spermidine in B. thetaiotaomicron.

Fig. 4
figure 4

CASDC activity using cell-free extracts from each B. thetaiotaomicron strain. The following cell-free extracts were used to evaluate CASDC activity: a no bacterial strain, b MS39 (Δtdk), c MS56 (Δtdk Δcasdc), and d MS93 (Δtdk Δcasdc att1::casdc +). Each strain was cultured in polyamine-free minimal medium until the OD600 reached 0.5–0.6. The cell-free extract was then prepared as described in the “Materials and methods.” The enzymatic reaction was performed in the presence of 1 mM carboxyspermidine and 6.48 mg/mL cell-free extract at 37 °C for 9 h. After stopping the reaction by adding trichloroacetic acid to a final concentration of 10 %, the reaction mixture was analyzed by HPLC. C-Spd carboxyspermidine, Spd spermidine

Full size image

casdc contributes to the normal growth of B. thetaiotaomicron

The contribution of casdc to the growth of B. thetaiotaomicron was investigated using MS39 (Δtdk), MS56 (Δtdk Δcasdc), and MS93 (Δtdk Δcasdc att1::casdc +) cultured in polyamine-free minimal medium (Fig. 5). MS56 (Δtdk Δcasdc) showed a decreased growth rate compared to the parental strain MS39 (Δtdk). Growth of the complemented strain MS93 (Δtdk Δcasdc att1::casdc +) was faster than growth of MS39 (Δtdk) and MS56 (Δtdk Δcasdc). The generation times of MS39 (Δtdk), MS56 (Δtdk Δcasdc), and MS93 (Δtdk Δcasdc att1::casdc +) were 116.3, 163.1, and 105.2 min, respectively. The culture times required to reach the maximum OD600 were as follows: MS39 (Δtdk), 21 h; MS56 (Δtdk Δcasdc), 24 h; MS93 (Δtdk Δcasdc att1::casdc +), 18 h. These results reveal that casdc contributes to the normal growth of B. thetaiotaomicron.

Fig. 5
figure 5

Growth curve of B. thetaiotaomicron strains in polyamine-free minimal medium. Cells from MS39 (Δtdk), MS56 (Δtdk Δcasdc), and MS93 (Δtdk Δcasdc att1::casdc +) were precultured in liquid GAM, washed once with minimal medium, and diluted at an initial OD600 of 0.03 in the same minimal medium. The cells were then anaerobically cultured at 37 °C, and growth was monitored by measuring the OD600. The symbols are the same as those described in the legend of Fig. 3. Data are shown as the mean ± standard deviation (n = 3)

Full size image

Discussion

In this study, functional analysis of a spermidine biosynthetic gene casdc was performed in B. thetaiotaomicron. The genus Bacteroides is predominant in the intestine of humans; 20 of the 56 most abundant species in the human intestinal microbiota are members of this genus (Qin et al. 2010). However, polyamine biosynthesis in Bacteroides has not been thoroughly investigated. Although E. coli, C. jejuni, and Vibrio cholerae have well-characterized polyamine biosynthetic pathways (Tabor and Tabor 1985; Lee et al. 2009; Deng et al. 2010; Hanfrey et al. 2011), these bacterial species are minor components of the microbiota or are pathogens in the human intestine, suggesting that their polyamine biosynthesis does not significantly affect intestinal polyamine concentrations. Considering the important physiological role of polyamines in the gut, there is a clear need to analyze the polyamine biosynthetic pathways in predominant intestinal bacterial species such as those in the Bacteroides genus.

B. thetaiotaomicron JCM 5827T and MS39 (Δtdk) cultured in polyamine-free minimal medium accumulated spermidine as the only intracellular polyamine (Figs. 2 and 3). This observation is consistent with that of a previous study of polyamines in B. thetaiotaomicron (Noack et al. 1998). In MS56 (Δtdk Δcasdc), agmatine was the only polyamine that had accumulated in the cell; putrescine, carboxyspermidine, and spermidine were not detected. Loss of function of CASDC is not directly involved in agmatine accumulation because purified recombinant CASDC was not active on agmatine (data not shown). These results suggest that this bacterium possesses a regulatory system that can strictly control the level of each intracellular polyamine.

Despite the accumulation of intracellular spermidine in B. thetaiotaomicron JCM 5827T and MS39 (Δtdk), these strains did not export polyamines into the culture supernatant. However, B. thetaiotaomicron contains potential novel agmatine export systems, as deletion of casdc in this species resulted in agmatine export. There may also be other unidentified polyamine export systems in B. thetaiotaomicron. In fact, export of spermidine in B. thetaiotaomicron DSM 2079T has been suggested in previous studies, where the monoassociation of germ-free rats with B. thetaiotaomicron increased spermidine concentrations in the cecal content (Noack et al. 2000) compared to in germ-free rats that had not been exposed to this microbe (Noack et al. 1998). Spermidine export in B. thetaiotaomicron may be activated in the intestine, where the environment is largely different from that of the polyamine-free minimal medium, or spermidine may be released from lysed B. thetaiotaomicron cells in the intestine.

CASDC of B. thetaiotaomicron converted carboxyspermidine to spermidine in the cell (Figs. 3 and 4). This result is consistent with those of previous studies in which deletion of casdc from C. jejuni abolished the accumulation of spermidine in cells (Hanfrey et al. 2011), and this enzyme was shown to utilize carboxyspermidine as a substrate for spermidine synthesis (Deng et al. 2010). The genus Vibrio, which is occasionally found in the intestine, also exhibits CASDC activity (Lee et al. 2009). Although CASDC activity in other intestinal bacteria remains to be experimentally confirmed, the putative CASDC-encoding genes are conserved in 38 of the 56 most abundant bacterial species in the human intestine (Hanfrey et al. 2011). Among these 38 species, some species, belonging to the genera Alistipes, Bacteroides, and Parabacteroides, have been reported to accumulate cellular spermidine (Hosoya and Hamana 2004; Hamana et al. 2008). Therefore, spermidine biosynthesis using CASDC may be a common characteristic of many intestinal bacterial species.

In a previous study, casdc from B. thetaiotaomicron (BT_0674) was identified to be the gene required for growth in polyamine-free minimal medium, in which genome-wide screening using a transposon mutant library was performed (Goodman et al. 2009). Moreover, a similar phenotype was observed in a casdc deletion mutant of C. jejuni, which is auxotrophic for spermidine (Hanfrey et al. 2011). In our study, deletion of casdc caused growth defects of B. thetaiotaomicron (Fig. 5), indicating that biosynthesis and accumulation of spermidine in the cell is important, although not essential, for normal growth.

The function of CASDC determined in this study is important for increasing the understanding of the polyamine biosynthetic pathways in the intestinal microbiota. Although the genetic and biochemical analyses of polyamine biosynthesis in B. thetaiotaomicron are incomplete and the intestinal microbiota may possess novel polyamine biosynthetic pathways that have not been identified, our study revealed that casdc of B. thetaiotaomicron is essential for spermidine biosynthesis and contributes to normal growth.