Butyrate Production in Phylogenetically Diverse Firmicutes

mbt_244

Microbial Biotechnology (2011) doi:10.1111/j.1751-7915.2010.00244.x

1 Butyrate production in phylogenetically diverse

2 Firmicutes isolated from the chicken caecummbt_244 1..11

4 Venessa Eeckhaut,1* Filip Van Immerseel,1 present study indicates that butyrate producers 50 5 Siska Croubels,2 Siegrid De Baere,2 related to cluster XVI may play a more important role 51 6 Freddy Haesebrouck,1 Richard Ducatelle,1 in the chicken gut than in the human gut. 52 3 4 7 Petra Louis and Peter Vandamme 53 8 1Department of Pathology, Bacteriology and Avian Introduction 54 9 Diseases, Research Group Veterinary Public Health and 10 Zoonoses, and 2Department of Pharmacology, In the chicken gastrointestinal tract, the main sites of 55 11 Toxicology, Biochemistry and Organ Physiology, Faculty bacterial activity are the crop and the caeca and, to a 56 12 of Veterinary Medicine, Ghent University, Salisburylaan lesser extent, the small intestine. Lactobacillus spp. domi- 57 131 1 133, B-9820 Merelbeke, Belgium. nate the crop (Barnes et al., 1972; Watkins and Kratzer, 58 14 3Microbial Ecology Group, Rowett Institute of Nutrition 1983) and small intestinal tract (Lu et al., 2003), while the 59 15 and Health, University of Aberdeen, Greenburn Road, caecal microbiota are dominated by species of the 60 16 Bucksburn, Aberdeen AB219SB, UK. Clostridiales order, followed by Lactobacillales and Bacte- 61 17 4Laboratory of Microbiology, Faculty of Sciences, Ghent roidales (Dumonceaux et al., 2006). A molecular, culture- 62 18 University, K. L. Ledeganckstraat 35, B-9000 Ghent, independent study from Zhu and colleagues (2002) on the 63 19 Belgium. caecal microbiota of chickens identified 243 different 64 20 sequences, representing 50 phylogenetic groups or sub- 65 21 Summary groups of bacteria, with the majority of the caecal 66 sequences being near or above 95% identical to their 67 22 Sixteen butyrate-producing bacteria were isolated closest relatives in the database. Only about 10% of these 68 23 from the caecal content of chickens and analysed sequences corresponded with validly named species, 69 24 phylogenetically. They did not represent a coherent indicating that the knowledge of the intestinal microbiota 70 25 phylogenetic group, but were allied to four different of poultry is incomplete (Apajalahti et al., 2004; Bjerrum 71 26 lineages in the Firmicutes phylum. Fourteen strains et al., 2006). 72 27 appeared to represent novel species, based on a level The activity and composition of the caecal microbiota is 73 Յ 28 of 98.5% 16S rRNA gene sequence similarity largely influenced by diet-derived substrates that resist 74 29 towards their nearest validly named neighbours. The small intestinal digestion. Fermentation of these sub- 75 30 highest butyrate concentrations were produced by strates leads to formation of metabolites such as short- 76 31 the strains belonging to clostridial clusters IV and chain fatty acids (SCFAs) of which the concentration and 77 32 XIVa, clusters which are predominant in the chicken relative proportion is affected by the type and quantity of 78 33 caecal microbiota. In only one of the 16 strains tested, the available substrates (Wolin et al., 1999). The quanti- 79 34 the butyrate kinase operon could be amplified, while tatively most important SCFAs are acetic, propionic and 80 35 the butyryl-CoA : acetate CoA-transferase gene was butyric acid. Butyrate in particular is known to serve as the 81 36 detected in eight strains belonging to clostridial clus- direct energy source for the colonic epithelium (Roediger, 82 37 ters IV, XIVa and XIVb. None of the clostridial cluster 1980) and possesses anti-inflammatory properties result- 83 38 XVI isolates carried this gene based on degenerate ing from inhibition of the transcription factor NFkB activity 84 39 PCR analyses. However, another CoA-transferase (Place et al., 2005). In addition, butyric acid is capable to 85 40 gene more similar to propionate CoA-transferase was reinforce the colonic defence barrier by increasing the 86 41 detected in the majority of the clostridial cluster XVI production of mucins and host antimicrobial peptides 87 42 isolates. Since this gene is located directly down- (Barcelo et al., 2000; Schauber et al., 2003). Butyric acid 88 43 stream of the remaining butyrate pathway genes in also promotes the body weight of broilers and has an 89 44 several human cluster XVI bacteria, it may be inhibitory activity against Salmonella and Clostridium per- 90 45 involved in butyrate formation in these bacteria. The fringens (Leeson et al., 2005; Van Immerseel et al., 2005; 91 46 Hu and Guo, 2007; Timbermont et al., 2010). Little is 92 47 Received 4 September, 2010; accepted 22 November, 2010. *For 48 correspondence. E-mail [email protected]; Tel. (+32) known about the endogenous butyrate-producing capac- 93 49 92647361; Fax (+32) 92647789. ity in the lower intestinal tract of chickens, most likely 94

2 V. Eeckhaut et al.

1 Table 1. Number of isolates within a butyrate-production range per sampled chicken and number of isolates consuming at least 2 mM acetate. 2 3 Acetate consumption 4 Butyrate production (mM) (Ն 2 mM) among all isolates

5 Butyrate Non-butyrate 6 Chicken (suffix) 0–2.0 2.1–5.0 5.1–10 10.1–15 > 15 producers producers

7 14-week-old layer (a) 37 16 8 4 0 12/28 0/37 8 4-week-old layer (b) 50 2 0 1 5 7/8 9/50 9 4-week-old broiler (c) 28 3 4 3 0 1/10 0/28 10 4-week-old broiler (d) 23 5 5 3 1 4/14 0/23 11 4-week-old broiler (e) 53 2 2 1 4 4/9 0/53 12 13 14 because only limited information is available on the iden- analysis of both types of fingerprints yielded comparable 53 15 tity and diversity of butyrate-producing bacteria in the results (data not shown) and 16 isolates with very distinct 54 16 chicken gut microbiota. Therefore the objective of the RAPD and REP-PCR fingerprints were selected for 55 17 present study was to isolate butyrate-producing bacteria further identification through 16S rRNA gene sequence 56 18 colonizing the caecum of chickens and to determine their analysis. Approximately 1300 bp of the 16S rRNA genes 57 19 pathway for butyrate production. were determined, compared with all sequence data in 58 public databases using the BLAST algorithm (Altschul 59 20 et al., 1990) and used for the construction of a phyloge- 22 60 21 Results netic tree (Fig. 1). The 16 isolates were distributed among 61 22 Isolation of butyrate-producing bacteria four clostridial clusters, cluster IV, XIVa, XIVb and XVI 62 (Collins et al., 1994), within the Firmicutes phylum. When 63 23 Dilutions of the caecal content were plated on M2GSC considering only validly named bacteria, three cluster IV 64 24 agar, since this medium was shown successful for the isolates (53-4c, 24-4c and 30-4c) were most closely 65 25 isolation of different types of butyrate producers from the related to Flavonifractor plautii (formerly Clostridium orbis- 66 26 human gut (Barcenilla et al., 2000). Sixty-five, 58, 38, 37 cindens) (Carlier et al., 2010) (97.7% and 95.7% 16S 67 27 and 62 colonies from a 14- and a 4-week-old layer pullet rRNA sequence similarity) and Pseudoflavonifractor cap- 68 28 and from 34-week-old broilers, respectively, were ran- illosus (formerly Bacteroides capillosus) (Carlier et al., 69 29 domly picked, grown overnight in M2GSC broth and 2010) (97.5% 16S rRNA sequence similarity) respec- 70 30 screened for fatty acid production. In accordance with the tively; one cluster IV isolate (40-4c) was most closely 71 31 study of Barcenilla and colleagues (2000), the cut-off related (97.8% 16S rRNA sequence similarity) to the 72 32 value was set at 2 mM butyrate for consideration as a butyrate-producing organism Subdoligranulum variabile; 73 33 butyrate producer. Twenty-six per cent of all tested iso- and two isolates (7-4c and 25-3b) were most closely 74 34 lates produced more than 2 mM butyrate, with the propor- related (91.5% and 92.7% 16S rRNA sequence similarity) 75 35 tion of butyrate-producing isolates varying between 14% to Eubacterium desmolans. From this cluster, isolate 76 36 and 43% for the five sampled chickens. The highest 25-3b was recently classified into a novel species within a 77 37 number of high-concentration butyrate producers novel genus, Butyricicoccus pullicaecorum (Eeckhaut 78 > 38 ( 15 mM) was isolated from the chickens, in which the et al., 2008). Three isolates (35-7e, 33-7e and 77-5d) 79 39 total number of butyrate producers was the lowest (b and were most closely related to members of the clostridial 80 40 e) (Table 1). At least 2 mM of the acetate present in the cluster XIVa, i.e. Anaerostipes caccae (94.5% 16S rRNA 81 41 M2GSC medium was consumed by 41% of all butyrate similarity), Eubacterium hallii (95.2% 16S rRNA similarity) 33 82 42 producers. All isolates from the 14-week-old layer and the and Clostridium populeti (92.5% 16S rRNA similarity) 83 43 three 4-week-old broiler chickens that consumed acetate respectively. In this cluster, the highest butyrate concen- 84 44 proved to be butyrate producers, in contrast with the iso- tration was produced by isolate 35-7e which was recently 85 45 lates from the 4-week-old layer where only 44% of the further characterized and classified into the novel species 86 46 acetate consumers were butyrate producers (Table 1). Anaerostipes butyraticus (Eeckhaut et al., 2010). Isolate 87 47 21-4c was most closely related (99.1% 16S rRNA 88 sequence identity) to Clostridium lactatifermentans,a 89 48 Identification of the butyrate producers lactate-fermenting bacterium of the clostridial cluster XIVb 90 49 Randomly amplified polymorphic DNA (RAPD) and (van der Wielen et al., 2002). All isolates within cluster 91 50 repetitive-extragenic-palindromic (REP) PCR fingerprints XIVa and XIVb showed the ability to utilize the lactic acid 92 51 were determined for the 45 out of the 68 butyrate- present in the medium (Table 2). Three of the cluster XVI 93 52 producing isolates that could be recultured. Numerical isolates (20-2a, 41-2a and 37-2a) were most closely 94

Butyrate-producing bacteria from the chicken caecum 3

Clostridium viride T2-7 DSM 6836T (X81125) 100 isolate 53-4c 100 isolate 24-4c isolate 30-4c 94 Pseudoflavonifractor capillosus ATCC 29799T (AY136666) 100 Flavonifractor plautii DSM 6740T (Y18187) - T 100 Eubacterium desmolans ATCC 43058 (L34618) cluster IV 100 Clostridiales strain A2-207 (AJ270471) 99 100 isolate 7-4c isolate 25-3b Butyricicoccus pullicaecorum CCUG 55265T (EU410376)

100 isolate 40-4c 100 Subdoligranulum variabile DSM 15176T (AJ518869) 100 Faecalibacterium prausnitzii 1-84 (AY169429) 100 Faecalibacterium prausnitzii L2-6 (AJ270470) Faecalibacterium prausnitzii M21/2 (AY305307) Clostridiales strain SS2/1 (AY305319) 100 isolate 35-7e Anaerostipes butyraticus DSM 22094T (FJ947528) 100 Anaerostipes sp. IE4 (AY960568) 91 Anaerostipes caccae L1-92 DSM 14662T (AJ270487) isolate 33-7e 100 100 Eubacterium hallii L2-7 (AJ270490) Eubacterium hallii DSM 3353T (L34621) cluster XIVa 96 isolate 77-5d T 100 Clostridium populeti ATCC 35295 (X71853) T 100 Roseburia hominis A2-183 DSM 16839 (AJ270482) 98 Roseburia intestinalis M50/1 (AY305308) 16/4 (AJ250365) 97 Butyrivibrio fibrisolvens 100 Clostridiales strain K10 (EU305624) 99 Clostridiales strain M62/1 (AY305309) Clostridiales strain SS3/4 (AY305316) T 100 Clostridium neopropionicum DSM 3847 (X76746) 100 Clostridium propionicum DSM 1682T (X77841) cluster XIVb T 100 Clostridium lactatifermentans DSM 14212 (AY033434) isolate 21-4c Erysipelotrichaceae strain 5_2_54FAA (NZ_GG749091) 100 Eubacterium dolichum DSM 3991T (L34682) Eubacterium tortuosum ATCC 25548T (L34683) isolate 60-7e

100 isolate 10-3b Eubacterium biforme DSM 3989T (M59230) T 100 Eubacterium cylindroides ATCC 27803 (L34617) cluster XVI 100 100 Eubacterium cylindroides T2-87 (AY305306) isolate 65-2a

100 isolate 20-2a Streptococcus pleomorphus ATCC 29734T (M23730)

100 isolate 41-2a isolate 37-2a Clostridium perfringens ATCC 13124T (M59103)

0.02

1 Fig. 1. Phylogenetic tree showing the relationship between the different butyrate-producing chicken isolates based on 16S rRNA gene 2 sequences. The tree was constructed by use of the neighbour-joining method. The 16S rRNA gene sequence of Clostridium perfringens 3 (ATCC 13124) was used as an outgroup to root the tree. Accession numbers are given in brackets. The numbers shown at the nodes of the 4 tree indicate bootstrap values out of 100 bootstraps resamplings (values under 90% are not shown). The strains isolated in this study are 5 shown in boldface and are labelled with an arrow when possessing a sequence related to the butyryl-CoA:acetate CoA transferase gene 6 (black arrow: CoATDF1, CoATDR2 primers; striped arrow: CoATDF1, CoATDR2 and BCoATscrF, BCoATscrR primers), the propionate 7 CoA-transferase gene (grey arrow) or the butyrate kinase gene (white arrow). Scale bar: 0.02 substitutions per nucleotide position.

4 V. Eeckhaut et al.

1 Table 2. Acidic fermentation products and relationships of the butyrate-producing strains. 2 3 Fermentation acids (mM)a,b Highest 16S rRNA gene sequence similarity

4 Propionic Formic Type strain of validly named species Accession 5 Strain Cluster Butyric acid Acetic acid acid Lactic acid acid (% 16S rRNA sequence similarity) number

6 53–4c IV 10.8 Ϯ 1.4 -4.6 Ϯ 0.7 -0.3 Ϯ 0.6 5.9 Ϯ 0.8 0.1 Ϯ 0.1 Pseudoﬂavonifractor capillosus (97.7) AY136666 7 24–4c 6.7 Ϯ 1.2 2.9 Ϯ 0.1 -0.8 Ϯ 1.8 3.7 Ϯ 0.01 1.3 Ϯ 0.02 Pseudoﬂavonifractor capillosus (95.7) AY136666 8 30–4c 11.8 Ϯ 0.7 0.9 Ϯ 0.3 0.6 Ϯ 0.1 3.6 Ϯ 0.1 0.1 Ϯ 0.1 Flavonifractor plautii (97.5) Y18187 9 7–4c 11.6 Ϯ 0.9 -0.7 Ϯ 0.2 -0.6 Ϯ 1.1 -0.4 Ϯ 0.3 0.5 Ϯ 0.2 Butyricicoccus pullicaecorum (97.9) EU410376 10 25–3b 15.3 Ϯ 0.2 -4.9 Ϯ 0.5 -0.6 Ϯ 1.5 -0.6 Ϯ 0.3 1.7 Ϯ 0.3 Butyricicoccus pullicaecorum (100) EU410376 11 40–4c 6.8 Ϯ 0.7 2.7 Ϯ 0.3 0.4 Ϯ 0.2 1.8 Ϯ 0.02 16 Ϯ 0.2 Subdoligranulum variabile (98.2) AJ518869 12 35–7e XIVa 21.7 Ϯ 0.6 -5.2 Ϯ 0.2 -0.4 Ϯ 0.3 -4.3 Ϯ 0.6 1.7 Ϯ 0.2 Anaerostipes butyraticus (100) FJ947528 13 33–7e 15.1 Ϯ 0.7 -1.3 Ϯ 0.2 -4.1 Ϯ 0.4 -8.1 Ϯ 0.1 0 Eubacterium hallii (95.2) L34621 14 77–5d 9.5 Ϯ 0.8 4.8 Ϯ 1.2 -1.9 Ϯ 1.6 -2.1 Ϯ 0.9 0.3 Ϯ 0.2 Clostridium populeti (92.5) X71853 15 21–4c XIVb 5.7 Ϯ 0.8 8.1 Ϯ 1.3 10.5 Ϯ 2.1 -3.3 Ϯ 0.8 0.2 Ϯ 0.1 Clostridium lactatifermentans (99.1) AY033434 16 60–7e XVI 3.1 Ϯ 0.8 0.2 Ϯ 0.01 -1.4 Ϯ 0.04 6.7 Ϯ 0.5 4.7 Ϯ 0.4 Eubacterium tortuosum (90.4) L34683 17 10–3b 2.5 Ϯ 0.9 0 -1.1 Ϯ 0.3 4.9 Ϯ 0.01 5.6 Ϯ 0.2 Eubacterium tortuosum (90.6) L34683 18 65–2a 3.6 Ϯ 0.3 -0.2 Ϯ 0.06 -0.8 Ϯ 0.1 8.3 Ϯ 0.2 2.9 Ϯ 0.07 Eubacterium cylindroides (98.7) L34617 19 20–2a 2.9 Ϯ 0.8 -0.4 Ϯ 0.5 -1.0 Ϯ 0.3 8.5 Ϯ 0.4 3.6 Ϯ 0.2 Streptococcus pleomorphus (98.5) M23730 20 41–2a 3.9 Ϯ 0.6 -0.9 Ϯ 0.2 -1.8 Ϯ 1.7 7.1 Ϯ 0.1 2.9 Ϯ 0.1 Streptococcus pleomorphus (95.5) M23730 21 37–2a 2.9 Ϯ 0.9 -0.8 Ϯ 0.4 -0.8 Ϯ 0.3 8.1 Ϯ 0.2 2.2 Ϯ 0.2 Streptococcus pleomorphus (95.2) M23730 22

23 a. Concentrations are averages of results of three replicates Ϯ standard deviations and were calculated by subtracting the values at t0 from these 24 at t24. 25 b. Strains were grown overnight in M2GSC medium supplemented with 8 mM DL-lactate.

27 related (98.5%, 95.5% and 95.2% 16S rRNA sequence IV, XIVa and XIVb strains 53-4c, 30-4c, 7-4c, 25-3b, 57 28 similarity) to the generically misnamed (Täpp et al., 2003) 35-7e, 33-7e, 77-5d and 21-4c. In these isolates (Fig. 1, 58 29 Streptococcus pleomorphus; one cluster XVI isolate (65- striped arrows), except for 21-4c and 53-4c (Fig. 1, black 59 30 2a) was most closely related to Eubacterium cylindroides arrows), an amplicon (about 530 bp) was also obtained 60 31 (96.5% 16S rRNA sequence identity), and two further using the BCoATscrF and BCoATscrR primers. Cluster IV 61 32 isolates (60-7e and 10-3b) were remotely (90.4% and strain 24-4c, nor any of the cluster XVI isolates yielded a 44 62 33 90.6% 16S rRNA sequence identity) related to Eubacte- PCR product with the two primer pairs targeting the 63 34 rium tortuosum (Fig. 1, Table 2). Given their distinct RAPD butyryl-CoA : acetate CoA-transferase gene. All the iso- 64 35 and REP-PCR fingerprints (data not shown) and phyloge- lated cluster XVI strains, except 60-7e, but also isolate 65 36 netic positions (Fig. 1), these 16 isolates are considered 77-5d and 21-4c from cluster XIVa and XIVb (Fig. 1, grey 66 37 to represent 16 distinct strains in the remainder of the text. arrows), showed an amplicon of the expected size (about 67 38 702 bp) after PCR using degenerate primers designed 68 against propionate CoA-transferases (PCT primers, 69 39 Detection of genes encoding enzymes involved in Table S1). All CoA-transferase gene amplicons were 70 40 butyrate production sequenced and deduced amino acid sequences sub- 71 41 All 16 chicken-derived butyrate-producing strains were jected to database searches using the BLASTP algorithm. 72 42 screened for the presence of the butyrate kinase operon Two phylogenetic trees were constructed based on the 73 43 and/or the butyryl-CoA : acetate CoA-transferase gene gene sequences related to butyryl-CoA : acetate CoA- 74 44 which carry out the final step of butyrate synthesis. transferase (Fig. 2A) and propionate CoA-transferases 75 45 In strain 40-4c an amplicon of the expected size (about (Fig. 2B). For most strains, the phylogeny of the CoA- 76 46 771 bp) was obtained after PCR using PTBfor1 and transferase gene sequences agreed well with the 16S 77 47 BUKrev2 primers targeted against the butyrate kinase rRNA gene-based phylogeny (strains 33-7e, 35-7e, 78 48 operon. The sequence of the amplicon was 98.6% similar 77-5d, 30-4c, 7-4c, 25-3b, Fig. 2A, and isolates 37-2a, 79 49 to the butyrate kinase gene of the human faecal bacterium 41-2a, 20-2a, 65-2a, 10-3b, Fig. 2B). However, the CoA- 80 50 S. variabile (DSM 15176), its closest phylogenetic neigh- transferase gene and 16S rRNA gene-based phylogenies 81 51 bour (Fig. 1, white arrow). The butyrate kinase operon of strains 53-4c and 21-4c were discordant, two strains 82 52 could not be amplified in any of the other strains with the (77-5d and 21-4c) carried more than one CoA-transferase 83 53 primer set used. gene and neither a CoA-transferase gene nor the butyrate 84 54 Butyryl-CoA : acetate CoA-transferase gene amplicons kinase operon could be amplified in strains 24-4c and 85 55 of the expected size (about 582 bp) using primer pair 60-7e. Inspection of draft genome sequences from human 86 56 CoATDF1 and CoATDR2 were found in clostridial cluster isolates belonging to cluster XVI revealed that the CoA 87

Butyrate-producing bacteria from the chicken caecum 5 A 100 Eubacterium hallii DSM 3353 (ZP_03715304) 100 Eubacterium halllii L2-7 (AAZ23219) isolate 33-7e Butyrivibrio fibrisolvens 16/4 (CBK74075) Roseburia hominis A2-183 (AAX19660) 100 Roseburia intestinalis M50/1 (CBL07589) isolate 35-7e 100 Anaerostipes caccae L1-92 (ZP_02418566) Clostridiales strain SS2/1 (ZP_02439482) isolate 77-5d 100 Faecalibacterium prausnitzii M21/2 (ZP_02092519) Faecalibacterium prausnitzii L2-6 (CBK99454) 100 Clostridiales strain SS3/4 (CBL41524) 100 Clostridiales strain M62/1 (ZP_06347958) 100 Clostridiales strain K10 (CBK78808) isolate 30-4c isolate 7-4c 100 97 100 isolate 25-3b isolate 53 4c Pseudoflavonifractor capillosus ATCC 29799T (ZP_02034783) Eubacterium hallii DSM 3353 (ZP 03718142) 100 isolate 21-4c 100 Megasphaera strain 28L (ZP_06559578) Anaerostipes caccae L1-92 (ZP_02417601)

0.1

B 100 isolate 37-2a 100 isolate 41-2a 97 isolate 20-2a 98 isolate 65-2a 100 Eubacterium cylindroides T2-87 (CBK88671) 100 Eubacterium biforme DSM 3989 (ZP_03488633-4) isolate 10-3b 95 Erysipelotrichaceae strain 5_2_54FAA (ZP 06645694) Eubacterium dolichum DSM 3991 (ZP_02078194) 90 Clostridium botulinum C str. Eklund (ZP_02620725) Clostridium tetani E88 (NP_781374) Alkaliphilus metalliredigens QYMF (YP_001322263) 100 isolate 21-4c Clostridium propionicum (CAB77207) 98 Anaerostipes caccae L1-92 (ZP_02420848) 100 Clostridiales strain SS2/1 (ZP_02440717) Listeria welshimeri SLCC5334 (YP_850387) isolate 77-5d 94 100 Anaerostipes caccae L1-92 (ZP_02420640)

0.05

1 Fig. 2. Phylogenetic tree of deduced protein sequences related to butyryl-CoA : acetate CoA-transferase (A) and propionate CoA-transferase 2 (B) respectively. The strains isolated in this study are shown in boldface, while the remaining strains are reference sequences. Accession 3 numbers are given in brackets. The numbers shown at the nodes of the tree indicate bootstrap values out of 100 bootstraps resamplings 4 (values under 90% are not shown). Scale bar: 0.1 substitutions per nucleotide position.

6 V. Eeckhaut et al.

THL BCD ETFb ETFa BHBD CRO CoAT It remains to be established whether they are also an 54 important component of the chicken microbiota. The 16S 55 1 Fig. 3. Arrangement of butyrate pathway genes in human cluster rRNA gene sequence similarity of the 16 butyrate- 56 2 XVI strains. THL, thiolase; BCD, butyryl-CoA dehydrogenase; producing strains towards the type strains of their nearest 57 3 ETFb, electron transfer protein b; ETFa, electron transfer protein a; validly named neighbours varied from 90.4% to 99.1%. As 58 4 BHBD, b-hydroxybutyryl-CoA dehydrogenase; CRO, crotonase; 5 CoAT, CoA-transferase. Corresponding genes from draft genome organisms sharing less than 98.5% of their 16S rRNA 59 6 sequences: E. cylindroides T2-87 (CBK88667–88671; part of BHBD sequence typically correspond with distinct species 60 7 sequence and CRO sequence are missing due to a stretch on N (Stackebrandt and Ebers, 2006) these 16 strains most 61 8 connecting contigs); E. dolichum DSM 3991 (EUBDOL_02008–14); 9 E. biforme DSM 3989 [EUBIFOR_01208–16; two THL genes are probably represent 14 novel bacterial species. 62 10 present in tandem (EUBIFOR_01208–9), the putative In the caeca of chickens, lactate is detected during the 63 11 CoA-transferase (EUBIFOR_01215–16) contains a frameshift]. first 15 days of live, thereafter only low lactate concentra- 66 64 12 tions are found, presumably as a result of metabolic 65 13 transferase gene (Fig. 2B) was located directly down- cross-feeding from lactate-producing bacteria such as 66 14 stream of the other butyrate pathway genes (Fig. 3), indi- lactobacilli and bifidobacteria to lactate-utilizing bacteria 67 15 cating that it may be linked to butyrate metabolism in (van der Wielen et al., 2000). So far only E. hallii, A. 68 16 these bacteria. caccae and Clostridiales strain SS2/1 within clostridial 77 69 17 cluster XIVa and Megasphaera elsdenii within clostridial 70 cluster IX are known to use lactate as precursor for 71 18 Discussion butyrate synthesis (Soto-Cruz et al., 2001; Duncan et al., 72 19 In several molecular-based studies on broiler gut micro- 2004). In the present study, all three clostridial cluster 73 20 biota using complete caecal as well as mucosa- XIVa strains consumed some of the DL-lactate which was 74 21 associated caecal 16S rRNA clone libraries, the largest added to the M2GSC medium. 75 22 set of retrieved sequences represented Clostridium For the production of intracellular butyrate, two distinct 76 23 cluster XIVa and Clostridium cluster IV bacteria (Gong pathways are described in clostridia (Gottschalk, 1986). In 77 24 et al., 2002; Lan et al., 2002; Zhu et al., 2002; Dumon- the human gut, only a few bacterial species, including 78 25 ceaux et al., 2006). Although in the human gut these two isolates related to Clostridium nexile and Coprococcus 79 26 phylogenetic lineages contain many butyrate-producing spp., use the butyrate kinase pathway (Louis et al., 2004). 80 27 bacteria (Louis and Flint, 2009), so far from the chicken Those strains are members of clostridial cluster XIVa, 81 28 caecum, only a few Faecalibacterium prausnitzii-like while the strain (40-4c) carrying the butyrate kinase 82 29 strains have been cultured and found to produce butyrate operon isolated in this study belongs to cluster IV. The 83 30 (Bjerrum et al., 2006). These butyrate producers were majority of the cultured human butyrate-producing strains 84 31 isolated during a study in which the overall microbial com- are found to carry the butyryl-CoA : acetate CoA- 85 32 munity composition in the different intestinal segments transferase gene (Louis et al., 2004; 2010; Charrier et al., 86 33 was analysed. In the present study we only looked for 2006) encoding the enzyme which consumes acetate 87 34 butyrate producers and found that 69 out of 260 isolates during the process of butyrate formation. The butyryl- 88 355 5 from two layer type and three broiler chickens produced at CoA : acetate CoA-transferase gene was detected in all 89 36 least 2 mM butyrate. So 26.5% were butyrate producers cluster XIVa and in the majority of the cluster IV strains 90 37 which is comparable with the 30.9% Barcenilla and col- examined here. Its phylogeny matched the 16S rRNA 91 38 leagues (2000) obtained in the human study. The clonal gene-based phylogeny for most isolates, except for 92 39 relatedness of 45 of these isolates (13 isolates could not isolate 53-4c. The CoA-transferase gene of strain 21-4c 93 40 be recultured after storage) was estimated using RAPD from cluster XIVb also differed markedly from the other 94 41 and REP-PCR fingerprinting; 16 strains with unique CoA-transferase sequences and clustered more closely 95 42 RAPD and REP-PCR fingerprints were selected for with a second class of CoA-transferase genes. For both 96 43 further identification using 16S rRNA gene sequence isolates, only one of the degenerate primer sets 97 44 analysis which revealed that they belonged to the (CoATDF1 and CoATDR2) leads to an amplicon, while the 98 45 clostridial clusters IV, XIVa, XIVb and XVI. Although 16S primers amplifying a narrower range of CoA-transferases, 99 46 rRNA gene sequences of F. prausnitzii- and S. variabile- BcoATscrF and BcoATscrR, did not result in an amplicon. 100 47 related bacteria within cluster IV are abundant in the Thus these genes may encode CoA-transferases with a 101 48 lumen as well as in the mucus of the chicken caecum different substrate specificity and it remains to be estab- 102 49 (Gong et al., 2002; Bjerrum et al., 2006), only one S. lished whether they are involved in butyrate formation. 103 50 variabile-related strain (97% 16S rRNA sequence similar- The butyryl-CoA : acetate CoA-transferase gene present 104 51 ity) was isolated in the present study. Although strains in cluster IV, XIVa and XIVb bacteria could not be found in 105 52 belonging to Roseburia are important butyrate producers cluster XVI strains by degenerate PCR using primer pairs 106 53 in the human colon, none was found in the present study. CoATDF1, CoATDR2 and BCoATscrF, BCoATscrR. 107

Butyrate-producing bacteria from the chicken caecum 7

1 Instead, a CoA-transferase gene more closely related to pullet and a 4-week-old White Leghorn layer type pullet. For 54 2 propionate CoA-transferases was found in all but one of the third, fourth and ﬁfth sampling, the caeca of 4-week-old 55 3 the cluster XVI isolates. Analysis of draft genome broiler (Ross) chickens, purchased from different commercial 56 farms, were used. 57 4 sequences from cluster XVI butyrate producers revealed 58 5 that this gene is present directly downstream of the Caecal sampling 59 6 central pathway genes for butyrate formation, leading 7 from acetyl-CoA to butyryl-CoA. Based on degenerate Immediately after euthanasia, the caeca were transferred into 60 8 PCR experiments against several central pathway genes, an anaerobic (84% N2,8%CO2 and 8% H2) workstation 61 9 the arrangement in E. cylindroides strains had previously (Ruskinn Technology, Bridgend, UK). One gram of the caecal 62 content was homogenized in 9 ml of anaerobic M2GSC 63 10 been assumed to be similar to cluster I strains (Louis and medium pH 6 containing per 100 ml, 30 ml of rumen ﬂuid, 1 g 64 11 Flint, 2009). However, the evidence from draft genomes of casitone, 0.25 g of yeast extract, 0.2 g of glucose, 0.2 g 65 12 shows that cluster XVI strains isolated from human faeces of cellobiose, 0.2 g of soluble starch, 0.045 g of K2HPO4, 66

13 show a different arrangement (Fig. 3) which is also in line 0.045 g of KH2PO4, 0.09 g of (NH4)2SO4, 0.09 g of NaCl, 67

14 with the previous degenerate PCR results. Thus, E. cylin- 0.009 g of MgSO4·7H2O, 0.009 g of CaCl2, 0.1 mg of resa- 68 15 droides in fact carries the same gene arrangement as zurin, 0.4 g of NaHCO3 and 0.1 g of cysteine hydrochloride 69 16 Eubacterium dolichum rather than the one present in (Barcenilla et al., 2000). From this suspension, a 10-fold 70 serial dilution was made and from each dilution, 0.3 ml was 71 17 cluster I strains. Many CoA-transferases that have been spread onto agar plates containing M2GSC medium with 72 18 examined biochemically have a broad substrate specific- 1.5% agar. Plates were incubated at 41–42°C for 48 h. Single 73 19 ity in vitro. The butyryl-CoA : acetate CoA-transferase colonies were randomly picked from dilutions 10-7 or 10-8 and 74 20 found in many human gut butyrate producers exhibited a grown overnight in 10 ml of M2GSC broth. 75 21 similar affinity for propionyl-CoA as for butyryl-CoA (Char- 76 22 rier et al., 2006), while the propionate CoA-transferase SCFA and lactic acid quantification 77 23 from Clostridium propionicum also converts butyrate Quantitative determination of the acidic fermentation prod- 78 24 (Schweiger and Buckel, 1984). It is therefore hypoth- ucts (butyric acid, acetic acid, propionic acid, lactic acid and 79 25 esized that a CoA-transferase related to the gene from C. formic acid) was performed using a validated HPLC-UV 80 26 propionicum is responsible for butyrate formation in method after 24 h of growth in M2GSC broth or in M2GSC 81 27 cluster XVI bacteria, especially as its gene lies directly broth supplemented with 8 mM DL-lactic acid. After centrifu- 82 28 downstream of the butyrate central pathway genes in the gation of the overnight cultures, the supernatant was acidified 83 29 cluster XVI butyrate-producing bacteria that genome using concentrated hydrochloric acid and extracted with 84 diethylether for 20 min. After centrifugation, the organic layer 85 30 sequence is available for (Fig. 3). Clostridial cluster XIVb was transferred to another extraction tube and extracted 86 31 isolate 21-4c is the only isolate found here that produces again for 20 min with sodium hydroxide. After centrifugation 87 32 high amounts of propionate (Table 2) and it carries a the aqueous phase was transferred to an autosampler vial 88 33 CoA-transferase closely related to the one from C. propi- and concentrated hydrochloric acid was added. An aliquot 89 34 onicum, which may be responsible for butyrate production was injected on the HPLC-UV instrument. The HPLC instru- 90 35 in this strain. ment consisted of a P1000XR type quaternary gradient 91 36 To our knowledge, this is the first study investigating the pump, an AS3000 type autosampler, a UV1000 type ultravio- 92 let detector and a SN4000 type system controller, all from 93 37 diversity and phylogenetic relationship of culturable ThermoFisher Scientific (Breda, the Netherlands). Chromato- 94 38 butyrate-producing bacteria from chicken caeca. It would graphic separation was achieved using a PLRP-S column 95 39 be useful to confirm the importance of the strains found (250 ¥ 4.6 mm i.d., Varian, Middelburg, the Netherlands). A 96 40 here on a bigger number of animals using PCR gradient elution was performed using sodium dihydrogen 97 41 approaches for the detection of the functional genes on phosphate in HPLC grade water and HPLC grade acetonitrile 98 42 the caecal content and clone libraries to see if the same as mobile phase A and B respectively. The detector was set 99 43 sequences can be retrieved. Although more extensive at a wavelength of 210 nm. For data processing, the Chrom- 100 quest software (ThermoFisher Scientific) was used. Quanti- 101 44 surveys are likely to reveal additional phylogenetic groups fication was performed using linear calibration curves 102 45 of butyrate-producing bacteria, the present study indi- (calibration range: 0.5–50.0 mM for each compound). Limits 103 46 cates that butyrate producers related to cluster XVI may of quantification (LOQ) were set at 0.5 mM for formic acid, 104 47 play a more important role in the chicken gut than in the lactic acid, acetic acid and propionic acid and at 1.0 mM for 105 48 human colon. butyric acid. Limits of detection (LOD) were: 0.19 mM for 106 49 formic acid, 0.18 mM for lactic acid, 0.19 mM for acetic acid, 107 0.14 mM for propionic acid and 0.33 mM for butyric acid. 108 50 Experimental procedures 109 51 Chickens Identification of the butyrate-producing isolates 110

52 Caecal samples from ﬁve chickens were taken. The ﬁrst two RAPD and REP-PCR were used to investigate the genetic 111 53 samples were taken from a 14-week-old Isa Brown layer type diversity and clonality of the isolates. Genomic DNA was 112

8 V. Eeckhaut et al.

1 extracted using alkaline lysis as previously described et al., 2007) using the neighbour-joining method, 100 times 61 2 (Scheirlinck et al., 2007). The DNA was used for REP-PCR bootstrap, pairwise gap deletion and Poisson correction. 62

3 with the (GTG)5 primer and RAPD typing using the Butyrate pathway gene arrangements of clostridial cluster 63 4 10-nucleotide primer 272 (Table S1) (Versalovic et al., 1994; XVI strains E. dolichum, E. biforme and E. cylindroides were 64 5 Coenye et al., 2002). The resulting RAPD and REP-PCR examined in draft genome sequences available in the 65 6 fingerprints were analysed using the BioNumerics V4.61 soft- GenBank database by performing BLASTP analyses with the 66 7 ware package (Applied Maths, Sint-Martens-Latem, Belgium) corresponding genes from human isolate L2-50 (Louis et al., 67 8 as described before (Scheirlinck et al., 2007). The similarities 2007; accession number of corresponding DNA sequence 68 9 among the banding patterns were calculated using Pearson’s DQ987697) and the propionate CoA-transferase gene from 69 10 correlation coefficient (expressed as a percentage-of- C. propionicum (Selmer et al., 2002; Accession No. 70 11 similarity value) and a dendrogram was constructed based on CAB77207). All hits were between 45% and 79% identical. 71 12 the UPGMA algorithm (Pearson and Lipman, 1988). 72 13 The extracted DNA was further used as target for amplifi- Nucleotide sequence accession numbers 73 14 cation of 16S rRNA genes using the ‘universal’ eubacterial 15 primers fD1 and rD1 (corresponding to positions 8 and 1541 The sequences of the 16S rRNA genes and the CoA- 74 16 in the Escherichia coli numbering system) (Table S1) (Weis- transferase genes used in the construction of the phyloge- 75 17 burg et al., 1991). The purified amplicons were sequenced netic trees are together with the sequence of the butyrate 76 18 using the BigDye Terminator sequencing kit with primers pD, kinase gene deposited under GenBank Accession 77 19 g*, 3 and O* on an ABI PRISM Genetic Analyser (Table S1) No. HQ452851–HQ452864, HQ452835–HQ452849 and 78 20 (Coenye et al., 1999). The sequences obtained were com- HQ452850 respectively. 79 21 pared with entries in the EMBL and GenBank database using 80 22 the BLAST program (Altschul et al., 1990). The most similar 23 database sequences were aligned with the isolate sequences Acknowledgements 81 24 using the CLUSTALW program (Chenna et al., 2003). To outline DVM Davy Persoons is acknowledged for supplying the 82 25 the phylogenetic relationship, a neighbour-joining tree chickens. This work was funded by the Belgian Federal 83 26 (Saitou and Nei, 1987) was constructed via the PHYLIP Public Service for Health, Food Chain Safety and Environ- 84 27 package (Felsenstein, 1989), using DNADIST for distance ment. The Rowett Institute of Nutrition and Health receives 85 28 analysis (Kimura, 1980) and bootstrap values of 100 financial support from the Scottish Government Rural and 86 29 replicates. Environmental Research and Analysis Directorate. 87 30 88 31 Amplification and phylogenetic analysis of the genes References 89 32 involved in butyrate formation Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, 90 33 The route for butyrate synthesis in the phylogenetically D.J. (1990) Basic local alignment search tool. J Mol Biol 91 34 diverse butyrate-producing strains was determined using 215: 403–410. 92 35 degenerate primers PTBfor2 and BUKrev1 for amplification Apajalahti, J., Kettunen, A., and Graham, H. (2004) Charac- 93 36 of part of the operon encoding phosphotransbutyrylase and teristics of the gastrointestinal microbial communities, with 94 37 butyrate kinase, via a ramped annealing approach as special reference to the chicken. Worlds Poult Sci J 60: 95 38 described previously (Louis et al., 2004). Two degenerate 223–232. 96 39 primer pairs were used to amplify part of the butyryl- Barcelo, A., Claustre, J., Moro, F., Chayvialle, J.A., Cuber, 97 40 CoA : acetyl CoA-transferase gene. Primer set CoATDF1, J.C., and Plaisancié, P. (2000) Mucin secretion is modu- 98 41 CoATDR2 (Table S1) was designed using a broad range of lated by luminal factors in the isolated vascularly perfused 99 42 CoA-transferase-related sequences and had a non- rat colon. Gut 46: 218–224. 100 43 degenerate clamp region at the 5′ end, based on the Barcenilla, A., Pryde, S.E., Martin, J.C., Duncan, S.H., 101 44 sequence from Roseburia sp. A2-183 (Charrier et al., 2006). Stewart, C.S., and Flint, H.J. (2000) Phylogenetic relation- 102 45 Conserved regions in butyryl-CoA : acetyl CoA-transferases ships of dominant butyrate-producing bacteria from the 103 46 but not in 4-hydroxybutyrate CoA transferase and acetyl-CoA human gut. Appl Environ Microbiol 66: 1654–1661. 104 47 hydrolase were used for the design of the primers BCoATscrF Barnes, E.M., Mead, G.C., Barnum, D.A., and Harry, E.G. 105 48 and BCoATscrR (Table S1) (Louis and Flint, 2007). Degener- (1972) The intestinal flora of the chicken in the period of 2 106 49 ate primers PCTfor1 and PCTrev2 (Table S1) were designed to 6 weeks of age, with particular reference to the anaero- 107 50 against conserved regions of CoA-transferase genes related bic bacteria. Br Poult Sci 13: 311–326. 108 51 to a propionate CoA-transferase from C. propionicum Belenguer, A., Duncan, S.H., Calder, A.G., Holtrop, G., Louis, 109 52 (AJ276553) (Charrier et al., 2006). PCR products of the P., Lobley, G.E., and Flint, H.J. (2006) Two routes of meta- 88 110 53 expected size were purified and sequenced using the BigDye bolic cross-feeding between Bifidobacterium adolescentis 111 54 Terminator sequencing kit. Contigs were generated with and butyrate-producing anaerobes from the human gut. 112 55 Lasergene 6 (DNASTAR) and deduced protein sequences Appl Environ Microbiol 5: 3593–3599. 113 56 were compared with entries in the GenBank database using Bjerrum, L., Engberg, R.M., Leser, T.D., Jensen, B.B., 114 57 BLASTP (Altschul et al., 1990). The deduced protein Finster, K., and Pedersen, K. (2006) Microbial community 115 58 sequences and reference sequences were aligned with composition of the ileum and cecum of broiler chickens as 116 59 CLUSTALW (Chenna et al., 2003) and manually inspected. revealed by molecular and culture-based techniques. Poult 117 60 Phylogenetic trees were constructed with MEGA4 (Tamura Sci 85: 1151–1164. 118

Butyrate-producing bacteria from the chicken caecum 9

1 Bond, P.L., Hugenholtz, P., Keller, J., and Blackall, L.L. (1995) and comparison with bacteria in the cecal lumen. FEMS 62 2 Bacterial community structures of phosphate-removing Microbiol Lett 208: 1–7. 63 3 and nonphosphate-removing activated sludges from Gottschalk, G. (1986) Bacterial Metabolism. Berlin, Heidel- 64 4 sequencing batch reactors. Appl Environ Microbiol 61: berg, Germany; New York, USA: Springer, pp. 228–237. 99 65 5 1910–1916. Hu, Z., and Guo, Y. (2007) Effects of dietary sodium butyrate 66 6 Carlier, J.-P., Bedora-Faure, M., K’ouas, G., Alauzet, C., and supplementation on the intestinal morphological structure, 67 7 Mory, F. (2010) Proposal to unify Clostridium orbiscindens absorptive function and gut flora in chickens. Anim Feed 68 8 Winter et al. 1991 and Eubacterium plautii (Séguin 1928) Sci Technol 132: 240–249. 69 9 Hofstad and Aasjord 1982, with description of Flavonifrac- Kimura, M. (1980) A simple method for estimating evolution- 70 10 tor plautii gen. nov., comb. nov., and reassignment of ary rates of base substitutions through comparative studies 71 11 Bacteroides capillosus to Pseudoflavonifractor capillosus of nucleotide sequences. J Mol Evol 16: 111–120. 72 12 gen. nov., comb. nov. Int J Syst Evol Microbiol 60: 585– Lan, P.T., Hayashi, H., Sakamoto, M., and Benno, Y. (2002) 73 13 590. Phylogenetic analysis of cecal microbiota in chicken by the 74 14 Charrier, C., Duncan, G.J., Reid, M.D., Rucklidge, G.J., use of 16S rDNA clone libraries. Microbiol Immunol 46: 75 15 Henderson, D., Young, P., et al. (2006) A novel class of 371–382. 76 16 CoA-transferase involved in short-chain fatty acid metabo- Leeson, S., Namkung, H., Antongiovanni, M., and Lee, E.H. 77 17 lism in butyrate-producing human colonic bacteria. Micro- (2005) Effect of butyric acid on the performance and 78 18 biology 152: 179–185. carcass yield of broiler chickens. Poult Sci 84: 1418–1422. 79 19 Chenna, R., Sugawara, H., Koike, T., Lopez, R., Gibson, T.J., Louis, P., and Flint, H.J. (2007) Development of a semi- 80 20 Higgins, D.G., and Thompson, J.D. (2003) Multiple quantitative degenerate real-time PCR-based assay for the 81 21 sequence alignment with the Clustal series of programs. estimation of butyryl-CoA CoA-transferase genes in 82 22 Nucleic Acids Res 31: 3497–3500. complex bacterial samples. Appl Environ Microbiol 73: 83 23 Coenye, T., Falsen, E., Vancanneyt, M., Hoste, B., Govan, 2009–2012. 84 24 J.R.W., Kersters, K., and Vandamme, P. (1999) Classifica- Louis, P., and Flint, H.J. (2009) Diversity, metabolism and 85 25 tion of Alcaligenes faecalis-like isolates from the environ- microbial ecology of butyrate-producing bacteria from the 86 26 ment and human clinical samples as Ralstonia gilardii sp. human large intestine. FEMS Microbiol Lett 294: 1–8. 87 27 nov. Int J Syst Bacteriol 49: 405–413. Louis, P., Duncan, S.H., McCrae, S.I., Millar, J., Jackson, 88 28 Coenye, T., Spiker, T., Martin, A., and LiPuma, J.J. (2002) M.S., and Flint, H.J. (2004) Restricted distribution of the 89 29 Comparative assessment of genotyping methods for epi- butyrate kinase pathway among butyrate-producing bacte- 90 30 demiologic study of Burkholderia cepacia Genomovar III. ria from the human colon. J Bacteriol 186: 2099–2106. 91 31 J Clin Microbiol 40: 3300–3307. Louis, P., McCrae, S.I., Charrier, D., and Flint, H.J. (2007) 92 32 Collins, M.D., Lawson, P.A., Willems, A., Cordoba, J.J., Organization of butyrate synthetic genes in human colonic 93 33 Fernandez-Garayzabel, J., Garcia, P., et al. (1994) The bacteria: phylogenetic conservation and horizontal gene 94 34 phylogeny of the genus Clostridium: proposal of five new transfer. FEMS Microbiol Lett 269: 240–247. 95 35 genera and eleven new species combinations. Int J Syst Louis, P., Young, P., Holtrop, G., and Flint, H.J. (2010) Diver- 96 36 Bacteriol 44: 812–826. sity of human colonic butyrate-producing bacteria revealed 97 37 Dumonceaux, J.T., Hill, J.E., Hemmingsen, S.M., and Van by analysis of the butyryl-CoA : acetate CoA-transferase 98 38 Kessel, A.G. (2006) Characterisation of intestinal micro- gene. Environ Microbiol 12: 304–314. 99 39 biota and response to dietary virginiamycin supplementa- Lu, J., Idris, U., Harmon, B., Hofacre, C., Maurer, J.J., and 100 40 tion in the broiler chicken. Appl Environ Microbiol 72: Lee, M.D. (2003) Diversity and succession of the intestinal 101 41 2815–2823. bacterial community of the maturing broiler chicken. Appl 102 42 Duncan, S.H., Louis, P., and Flint, H.J. (2004) Lactate- Environ Microbiol 69: 6816–6824. 103 43 utilizing bacteria, isolated from human feces, that produce Pearson, W.R., and Lipman, D.J. (1988) Improved tools for 104 44 butyrate as a major fermentation product. Appl Environ biological sequence comparison. Proc Natl Acad Sci USA 105 45 Microbiol 70: 5810–5817. 85: 2444–2448. 106 46 Eeckhaut, V., Van Immerseel, F., Teirlynck, E., Pasmans, F., Place, R.F., Noonan, E.J., and Giardina, C. (2005) HDAC 107 47 Fievez, V., Snauwaert, C., et al. (2008) Butyricicoccus pul- inhibition prevents NF-kB activation by suppressing pro- 108 48 licaecorum gen. nov., sp. nov., an anaerobic, butyrate- teasome activity: down-regulation of proteasome subunit 109 49 producing bacterium isolated from the caecal content of a expression stabilizes IkBa. Biochem Pharmacol 70: 394– 110 50 broiler chicken. Int J Syst Evol Microbiol 58: 2799–2802. 406. 111 51 Eeckhaut, V., Van Immerseel, F., Pasmans, F., De Brandt, E., Pryde, S.E., Duncan, S.H., Hold, G.L., Stewart, C.S., and 112 52 Haesebrouck, F., Ducatelle, R., and Vandamme, P. (2010) Flint, H.J. (2002) The microbiology of butyrate formation in 10 113 53 Anaerostipes butyraticus sp. nov., an anaerobic, butyrate the human colon. FEMS Microbiol Lett 217: 133–139. 114 54 producing bacterium from Clostridium cluster XIVa, iso- Roediger, W.E.W. (1980) The colonic epithelium in ulcerative 115 55 lated from broiler chicken caecal content. Int J Syst Evol colitis: an energy deficient disease? Lancet 2: 712– 116 56 Microbiol 60: 1108–1112. 715. 117 57 Felsenstein, J. (1989) PHYLIP – phylogeny inference package Saitou, N., and Nei, M. (1987) The neighbor-joining method: 118 58 (version 3.2). Cladistics 5: 164–166. a new method for reconstructing phylogenetic trees. Mol 119 59 Gong, J., Forster, R.J., Yu, H., Chambers, J.R., Sabour, P.M., Biol Evol 4: 406–425. 120 60 Wheatcroft, R., and Chen, S. (2002) Diversity and phylo- Schauber, J., Svanholm, C., Termén, S., Iffland, K., Menzel, 121 61 genetic analysis of bacteria in the mucosa of chicken ceca T., Scheppach, W., et al. (2003) Expression of the catheli- 122

10 V. Eeckhaut et al.

1 cidin LL-37 is modulated by short chain fatty acids in Versalovic, J., Scheider, M., De Bruijn, F.J., and Lupski, J.R. 40 2 colonocytes: relevance of signalling pathways. Gut 52: (1994) Genomic fingerprinting of bacteria using repetitive 41 3 735–741. sequence-based polymerase chain reaction. Methods Mol 42 4 Scheirlinck, I., Van der Meulen, R., Van Schoor, A., Vancan- Cell Biol 5: 25–40. 43 5 neyt, M., De Vuyst, L., Vandamme, P., and Huys, G. (2007) Watkins, B.A., and Kratzer, F.H. (1983) Effect of oral dosing 44 6 Influence of geographical origin and flour type on the diver- of Lactobacillus strains on gut colonization and liver biotin 45 7 sity of lactic acid bacteria in traditional Belgian sourdoughs. in broiler chicks. Poult Sci 62: 2088–2094. 46 8 Appl Environ Microbiol 73: 6262–6269. Weisburg, W.B., Barns, S.M., Pelletier, D.A., and Lane, D.J. 47 9 Schweiger, G., and Buckel, W. (1984) On the dehydration of (1991) 16S ribosomal DNA amplification for phylogenetic 48 10 (R)-lactate in the fermentation of alanine to propionate by study. J Bacteriol 173: 697–703. 49 11 Clostridium propionicum. FEBS Lett 171: 79–84. van der Wielen, P.W.J., Biesterveld, S., Notermans, S., 50 12 Selmer, T., Willanzheimer, A., and Hetzel, M. (2002) Propi- Hofstra, H., Urlings, B.A.P., and van Knapen, F. (2000) The 51 13 onate CoA-transferase from Clostridium propionicum. role of volatile fatty acids in development of the caecal 52 14 Cloning of gene and identification of glutamate 324 at the microflora in broiler chickens during growth. Appl Environ 53 15 active site. Eur J Biochem 269: 372–380. Microbiol 66: 2536–2540. 54 16 Soto-Cruz, O., Chávez-Rivera, R., and Saucedo-Castañeda, van der Wielen, P.W.J., Rovers, G.M.L.L., Scheepens, 55 17 G. (2001) Stimulation of the Megasphaera elsdenii’s J.M.A., and Biesterveld, S. (2002) Clostridium lactatifer- 56 18 butyrate production in continuous culture by a yeast addi- mentans sp. nov., a lactate-fermenting anaerobe isolated 57 19 tive. Braz Arch Biol Technol 44: 179–184. from the caeca of a chicken. Int J Syst Evol Microbiol 52: 58 20 Stackebrandt, E., and Ebers, J. (2006) Taxonomic param- 921–925. 59 21 eters revisited: tarnished golden standards. Microbiol Wolin, J.M., Miller, L.M., Yerry, S., Zhang, Y., Bank, S., and 60 22 Today 33: 152–155. Weaver, G.A. (1999) Changes of fermentation pathways of 61 23 Tamura, K., Dudley, J., Nei, M., and Kumar, S. (2007) fecal microbial communities associated with a drug treat- 62 24 MEGA4: molecular evolutionary genetics analysis (MEGA) ment that increases dietary starch in the human colon. Appl 63 25 software version 4.0. Mol Biol Evol 24: 1596–1599. Environ Microbiol 65: 2807–2812. 64 26 Täpp, J., Thollesson, M., and Herrmann, B. (2003) Phyloge- Zhu, X.Y., Zhong, Y., Pandya, Y., and Joerger, R.D. (2002) 65 27 netic relationships and genotyping of the genus Strepto- 16S rRNA-based analysis of microbiota from the cecum of 66 28 coccus by sequence determination of the RNase P RNA broiler chickens. Appl Environ Microbiol 68: 124–137. 67 29 gene, rnpB. Int J Syst Evol Microbiol 53: 1861–1871. 68 30 Timbermont, L., Lanckriet, A., Dewulf, J., Nollet, N., Supporting information 69 31 Schwarzer, K., Haesebrouck, F., et al. (2010) Control 32 of Clostridium perfringens induced necrotic enteritis in Additional Supporting Information may be found in the online 70 33 broilers by target released butyric acid, fatty acids and version of this article: 71 34 essential oils. Avian Pathol 39: 117–121. Table S1. Sequences of the used primers. 72 35 Van Immerseel, F., Boyen, F., Gantois, I., Timbermont, L., 36 Bohez, L., Pasmans, F., et al. (2005) Supplementation of Please note: Wiley-Blackwell are not responsible for the 73 37 coated butyric acid in the feed reduces colonization and content or functionality of any supporting materials supplied 74 38 shedding of Salmonella in poultry. Poult Sci 84: 1851– by the authors. Any queries (other than missing material) 75 39 1856. should be directed to the corresponding author for the article. 76

© 2010 The Authors Journal compilation © 2010 Society for Applied Microbiology and Blackwell Publishing Ltd, Microbial Biotechnology Toppan Best-set Premedia Limited Journal Code: MBT Proofreader: Jason Article No: 244 Delivery date: 21 December 2010 Page Extent: 10 Copyeditor: Harry AUTHOR QUERY FORM

Dear Author, During the preparation of your manuscript for publication, the questions listed below have arisen. Please attend to these matters and return this form with your proof. Many thanks for your assistance.

Query Query Remark References

q1 AUTHOR: Please conﬁrm the ﬁrst two affiliation addresses are correct

q2 AUTHOR: Altshul et al., 1990 has been changed to Altschul et al., 1990 so that this citation matches the Reference List. Please conﬁrm that this is correct.

q3 AUTHOR: ‘halii’ has been changed to ‘hallii’. Is this OK?

q4 AUTHOR: Should ‘nor’ be changed to ‘or’?

q5 AUTHOR: Should ‘two layer type’ be changed to ‘two layer types’? q6 AUTHOR: Should ‘15 days of live’ be changed to ‘15 days of being live’? q7 AUTHOR: ‘Clostridiales’ here has been italicized. Is this OK? q8 AUTHOR: Belenguer et al., 2006 has not been cited in the text, please indicate where it should be cited; or delete from the reference list. q9 AUTHOR: (Gottschalk, 1986) Please either provide the chapter title and Editor(s), or delete the page range q10 AUTHOR: Pryde et al., 2002 has not been cited in the text, please indicate where it should be cited; or delete from the reference list. q11 AUTHOR: Should ‘0.1 substitutions per nucleotide position’ be changed to ‘0.1 (A) or 0.05 (B) substitutions per nucleotide position’?

USING E-ANNOTATION TOOLS FOR ELECTRONIC PROOF CORRECTION Required Software Adobe Acrobat Professional or Acrobat Reader (version 7.0 or above) is required to e-annotate PDFs. Acrobat 8 Reader is a free download: http://www.adobe.com/products/acrobat/readstep2.html Once you have Acrobat Reader 8 on your PC and open the proof, you will see the Commenting Toolbar (if it does not appear automatically go to Tools>Commenting>Commenting Toolbar). The Commenting Toolbar looks like this:

Note Tool — For making notes at specific points in the text Marks a point on the paper where a note or question needs to be addressed.

How to use it: 1. Right click into area of either inserted text or relevance to note 2. Select Add Note and a yellow speech bubble symbol and text box will appear 3. Type comment into the text box 4. Click the X in the top right hand corner of the note box to close.

Replacement text tool — For deleting one word/section of text and replacing it Strikes red line through text and opens up a replacement text box.

How to use it: 1. Select cursor from toolbar 2. Highlight word or sentence 3. Right click 4. Select Replace Text (Comment) option 5. Type replacement text in blue box 6. Click outside of the blue box to close

Cross out text tool — For deleting text when there is nothing to replace selection Strikes through text in a red line.

How to use it: 1. Select cursor from toolbar 2. Highlight word or sentence 3. Right click 4. Select Cross Out Text

Approved tool — For approving a proof and that no corrections at all are required.

How to use it: 1. Click on the Stamp Tool in the toolbar 2. Select the Approved rubber stamp from the ‘standard business’ selection 3. Click on the text where you want to rubber stamp to appear (usually first page) Page 1 of 2

Highlight tool — For highlighting selection that should be changed to bold or italic. Highlights text in yellow and opens up a text box.

How to use it: 1. Select Highlighter Tool from the commenting toolbar 2. Highlight the desired text 3. Add a note detailing the required change

Attach File Tool — For inserting large amounts of text or replacement figures as a files. Inserts symbol and speech bubble where a file has been inserted.

How to use it: 1. Click on paperclip icon in the commenting toolbar 2. Click where you want to insert the attachment 3. Select the saved file from your PC/network 4. Select appearance of icon (paperclip, graph, attachment or tag) and close

Pencil tool — For circling parts of figures or making freeform marks Creates freeform shapes with a pencil tool. Particularly with graphics within the proof it may be useful to use the Drawing Markups toolbar. These tools allow you to draw circles, lines and comment on these marks.

How to use it: 1. Select Tools > Drawing Markups > Pencil Tool 2. Draw with the cursor 3. Multiple pieces of pencil annotation can be grouped together 4. Once finished, move the cursor over the shape until an arrowhead appears and right click

5. Select Open Pop-Up Note and type in a details of required change

6. Click the X in the top right hand corner of the note box to close.

Help For further information on how to annotate proofs click on the Help button to activate a list of instructions:

Page 2 of 2