Vitamin text description
VITAMIN BIOSYNTHESIS AND METABOLIC PATHWAYS
L. plantarum is deficient for biotin (vitamin B7), niacin (vitamin B3), pantothenate (vitamin B5) and pyridoxine (vitamin B6) (12)(13); In natural fermentation processes these vitamins could be provided by the product that is fermented or other micro-organisms present during fermentation such as yeasts (14). Other studies have shown that L. plantarum can use precursors of these vitamins for vitamin production, indicating that at least some part of the vitamin pathways are present. For example dethio-biotin can be used for the synthesis of biotin in L. plantarum (15).
Vitamin B12 can be made by some bacteria, such as L. reuteri (16), but not by L. plantarum.
In 2005, an in silico genome-scale reconstruction of the metabolic pathways was made for L. plantarum strain WCFS1 (5). It predicted that the genome of L. plantarum WCFS1 has all the enzymes encoded to biosynthesize folic acid (vitamin B9), thiamine (vitamin B1), and pyridoxine/pyridoxamine (vitamin B6), and the authors were able to fill gaps in the biosynthesis pathways that where caused by poor annotation of the sequenced genome.
Vitamin K is a fat-soluble vitamin necessary for blood coagulation and in metabolic pathways in bone and other tissue. This vitamin has two natural forms: phylloquinone (K1), which is only produced by plants, and menaquinone (K2), which is produced by micro-organisms and animals.
In L. plantarum menaquinone is used as part of an electron transport chain for respiration in aerobic conditions (17).
There is no evidence for biosynthesis of vitamins A, C, D or E in L. plantarum.
Below we describe the presence/absence analysis of genes involved in metabolism of individual vitamins. A summary is presented in Table 7, and details can be found in Table 1.
Vitamin B1 (Thiamine)
Thiamine is a water-soluble vitamin. Our analysis shows that there are several putative gaps in the thiamine biosynthesis pathways starting from either purine metabolism, pyruvate or from cysteine as precursor (Figure 17). Starting from purine metabolism, the KEGG pathway database shows that there are no known orthologs of thiC in microorganisms, but this gene may be present but not assigned yet. Another gap is the absence of thiS and thiF (EC: 2.7.7.73). The function of these coincides with the initial sulfur chemistry of the molybdopterin pathway (moaD and moeB) (18). The moaD and moeB genes (OGs 3024 and 3015 respectively) are found in 16 strains (WCFS1, JDM1, 19.1, ER, NIZO2264, NIZO2484, NIZO2485, NIZO2535, NIZO2726, NIZO2776, NIZO2802, NIZO2830, NIZO2831, NIZO2891, NIZO3892 and NIZO3893), while in the other strains they are absent.
No known candidates are found for the thiG and thiH genes. However lp_1776 (OG_1574) of strain WCFS1 is orthologous to the B. subtilis yurR gene, which is a paralog of a gene encoding thiO. This makes it a potential candidate for the missing function, but the precise mechanism of sulfur transfer is still unclear. Orthologs of lp_1776 are found to be present in all strains except NIZO2263 and NIZO2264.
Another enzyme not found is thiamin monophosphate phosphohydrolase (EC 3.1.3.-). The KEGG pathway database again does not provide any known genes for this enzyme. It is most likely that due to the unclear nature of this enzyme that a candidate gene has not been assigned to this function yet. However, growth experiments show that this gene (and the encoded enzyme) is likely to be present (19). Another putative gap in thiamin metabolism is a phosphorylation step of thiamine phosphate, by thiamine-monophosphate kinase ThiL (EC: 2.7.4.16). This step can also be done by thiamine pyrophosphate kinase (EC: 2.7.6.2 – OG_1907) (19).
Taking the above substitutions into consideration, most strains appear to have complete thiamine biosynthesis pathways. Only strains CNW10, NAB1, NIZO2263, NIZO2264, NIZO2753, NIZO2776, 16 and NC8 have a truncated or absent gene(s) in the metabolic pathways. For strains NIZO2753 and 16 the thiamine-phosphate pyrophosphorylase (EC2.5.1.3 – OG_2269) is absent. In strains NIZO2776, CNW10 and NC8 the 1-deoxy-D-xylulose-5-phosphate synthase gene (EC: 2.2.1.7 – OG_2214) is truncated and in strain NIZO2264 it is absent. In strain NAB1 one cysteine desulfurase is truncated (sufS, EC 2.8.1.7), but the other cysteine desulfurases (iscS and csd2) are completely present in these strains. In strain NIZO2263 the tpk gene (EC 2.7.6.2 – OG_1907) is absent, but this is not required for biosynthesis.
Vitamin B2 (Riboflavin)
Riboflavin is the central precursor of the cofactors FAD and FMN.
When the biosynthesis pathway of riboflavin is analyzed, some small differences between strains are apparent (Figure 18). Almost all strains are able to synthesize riboflavin from ribulose 5-phosphate or GTP, with the exception of strains NIZO3892, WCFS1, IPLA88 and possibly 16. NIZO3892 strain lacks all the riboflavin genes, and therefore it cannot synthesize riboflavin. In strains WCFS1 and IPLA88 the ribB gene (EC 2.5.1.9 – OG_2060) is a pseudogene. Strain WCFS1 lacks the ribD gene (EC 3.5.4.26/1.1.1.193 – OG_2061) completely, while strain 16 has a truncated ribD gene.
All strains are able to use riboflavin for the synthesis of FAD and FMN, since they all have a ribC1 (EC: 2.7.1.26 – OG_2190) and ribC2 gene (EC: 2.7.1.26/2.7.7.2 – OG_1438), including strain NIZO3892. In strains Nab1 and ST-III the ribC1 gene is truncated.
A putative riboflavin transporter (OG_193) is encoded by the ribU gene (20), which is present in all strains, but it is a pseudogene in 4 strains (NIZO2726, NIZO2776, NIZO2806, NIZO2830).
Vitamin B6 (Pyridoxal-5-phosphate)
Pyridoxal-5-phosphate is a water-soluble vitamin which is used as a cofactor in many reactions of amino acid metabolism. Most strains have all genes necessary for the conversion of pyriodoxamine, pyridoxal and pyridoxine into pyridoxal-5-phosphate (Figure 19). Only the pdxH gene (EC: 1.4.3.5 – OG_108) appears to be a pseudogene in strains NIZO2726, NIZO2776 and NIZO3893.
All the genes that lead to other pathways are absent in all strains. This leads to the assumption that L. plantarum requirespyridoxal-5-phosphate or one of these precursors to be added to the medium in order to make it grow.
Vitamin B3 (Niacin/Nicotinate/Nicotinamide)
Nicotinamide or nicotinate/niacin is a water-soluble vitamin that is incorporated into NAD and NADP. Most of the strains cannot synthesize vitamin B3 from quinolinate, since the essential nicotinate-nucleotide pyrophosphorylase (carboxylating) NadC (EC:2.4.2.19 – OG_3646) is absent in the chromosome of all strains. However, all strains have a putative transporter to import ribosyl nicotinamide (OG_1091), the precursor of NAD and NADP. In addition, 7 strains have plasmid-encoded proteins of vitamin B3 metabolism. Of these, five have a nicotinamidase (EC: 3.5.1.19 – OG_3749) to convert nicotinamide to nicotinate. Four of these five strains (NIZO1838, NIZO2802, NIZO3400 and NIZO3893) also have a plasmid-located cluster of 5 genes encoding a niacin transporter NiaP, and 4 enzymes of the biosynthesis pathway from L-aspartate to nicotinate D-ribonucleotide, i.e. L-asparate oxidase NadB (EC:1.4.3.16 – OG_3866), quinolate synthetase NadA (EC:2.5.1.72 – OG_3877), nicotinate-nucleotide pyrophosphorylase (carboxylating) NadC (EC:2.4.2.19 – OG_3646) and cysteine desulfurase (EC:2.8.1.7 – OG_3888). The latter enzyme presumably provides sulfur to the active site of NadA. Finally, strains NIZO2257 and NIZO2258 also have a plasmid-encoded NadC and NiaP proteins. Therefore, six strains are capable of biosynthesis of nicotinate D-ribonucleotide from aspartate and/or tryptophan.
All genes of nicotinate metabolism are found in most of the L. plantarum strains (Figure 20), and all strains can make NAD and NADP. However, the NADH pyrophosphatase gene(EC. 3.6.1.22 – OG_2474)is absent in 10 strains (NIZO1838, NIZO1840, NIZO2256, NIZO2257, NIZO2258, NIZO2484, NIZO2485, NIZO2757, NIZO2766 and NIZO2776). The bifunctional protein: 5'-nucleotidase; 2',3'-cyclic phosphodiesterase (EC 3.1.3.5 – OG_1343) is absent only in strain NIZO2484, where it appears to be part of a larger 8-gene deletion. In strain NIZO2264 one of the two purine nucleosidases (EC: 3.2.2.1 – OG_1969) is absent.
Vitamin B5 (Pantothenate)
Pantothenate is a water-soluble vitamin that is used by animals to synthesize coenzyme-A (CoA) and to metabolize proteins, carbohydrates and fats. Several gaps are seen in the pathway leading to the synthesis of pantothenate in L. plantarum. This leads to the conclusion that L. plantarum cannot biosynthesize pantothenate (Figure 21). All strains are able to use pantothenate to produce CoA. Strains WJL, NIZO1838, NIZO1839, NIZO2029, NIZO2256, NIZO2257, NIZO2741, NIZO2757, NIZO2766, NIZO2806, NIZO2814, NIZO2877 and NIZO2889 appear to have a truncated coaA gene (EC: 2.7.1.33 – OG_1251), but it is uncertain if this gene is truly truncated because these truncated genes are all found on the end of a contig and therefore may not fully assembled after sequencing. The phosphopantetheinyl transferase (EC: 2.7.8.- – OG_1935) gene is truncated only in strains 19.1, ER and NAB1.
Vitamin B7 (Biotin)
Biotin is a water-soluble vitamin that is a coenzyme for carboxylase enzymes, involved in the synthesis of fatty acids, isoleucine and valine.
L. plantarum misses the genes to biosynthesize biotin (Figure 22). Although all strains have genes involved in the transition of malonyl-[acyl-carrier protein] methyl ester to pimeloyl-[acyl-carrier protein] methyl ester, the rest of genes of the pathway leading to biotin are absent.
However, all strains have an intact gene for the biotin ECF transporter, substrate-specific component BioY (OG_805). When biotin is obtained from the environment all strains are able to use it for the production of holo-[carboxylases], since all strains have 2 paralogs encoding biotin-[acetyl-CoA-carboxylase] ligase (EC6.3.4.15).
Vitamine B9 (Folate)
Folate is a water-soluble vitamin and in humans in it is necessary for the production and maintenance of new cells, for DNA and RNA synthesis.
Almost all strains have the genes to biosynthesize folate from GTP (Figure 23). The folQ gene is part of the folate gene cluster folPQCEKB, but has not been entered yet into the KEGG pathway database. The folQ gene has experimentally been characterized to encode DHNTP pyrophosphohydrolase (EC 3.6.1.- – OG_1978) which removes pyrophosphate from dihydroneopterin triphosphate (DHNTP) to generate dihydroneopterin (21). Only strains 19.1 and NAB1 have pseudogenes in this pathway. Both have a truncated folC1 gene (EC: 6.3.2.12/6.3.2.17 – OG_347), while the folC2 gene is complete in both strains. Strain 19.1 also has a truncated folE gene (EC: 3.5.4.16 – OG_346).
During the biosynthesis of folate a gene called queD (EC: 4.2.3.12/4.1.2.50 – OG_2180) can interfere. It uses 7,8-dihydroneopterin 3’-triphospate, but instead of converting it to dihydroneopterin it converts it to 6-pyruvovyl-5,6,7,8-tetrahydropterin. This gene is found in almost all strains. In strains IPLA88, NIZO1840, NIZO2757 and NIZO2766, NIZO2776 this gene is truncated.
One carbon pool by folate
All strains have a similar set of genes concerning the one carbon pool by folate pathway (Figure 24). Only minor differences are seen. In strain 19.1 the glyA gene (EC: 2.1.2.1 – OG_1496) is truncated. In strains NIZO1840 and NIZO2263 the fhs gene (EC: 6.3.4.3 – OG_2172) is truncated and in strain NIZO2264 it is absent. In strain NIZO2484 the drfA gene (EC: 1.5.1.3) is truncated and in strain NIZO2259 the folD gene (EC: 1.5.1.5/3.5.4.9 – OG_238) is truncated.
Vitamin K2 (Menaquinone)
Most strains probably have a complete biosynthesis pathway from acetyl-CoA to menaquinone (vitamin K2) (Figure 25). Only one possible gap is seen in the pathway, which concerns the ispB gene (EC: 2.5.1.90). This enzyme converts farnesyl diphosphate into octaprenyl diphosphate. Several other enzymes that use farnesyl diphosphate to convert it into other diphosphates are found in the genome of L. plantarum (uppS, hepA, hepB and a polyprenyl synthetase). It is possible that these enzymes are wrongly annotated or multi-functional and that they me also be able to produce octaprenyl diphosphate.
A study of electron transport chains in L. plantarum strain WCFS1 showed that adding menaquinone (vitamin K2) to the medium increases the biomass and pH-values (17). It states that strain WCFS1 has an incomplete menquinone pathway, but this is not supported by the genes found in KEGG and by our analysis, as only the ispB gene (EC:2.5.1.90) is not found. But a trans-hexaprenyltranstransferase has a 49% similarity with the ispB gene of L. helveticus. Furthermore this gene is named octaprenyl-diphosphate synthase, but alternative names are dimethylallyltransferase, geranyltranstransferase, polyprenyl synthetase and farnesyldiphosphate synthase. A polyprenyl synthase is found (OG_242, Table 1) and it has a 82% sequence similarity to a octaprenyl-diphosphate synthase gene of L. pentosus.
Only nine strains have pseudogenes or genes that are absent is this pathway. This varies from a single absent gene uppS (EC: 2.5.1.31 – OG_1537) in strain NIZO2263, which does not affect the biosynthesis directly, to an absent genomic region resulting in the loss of four genes, i.e idi1 (EC: 5.3.3.2 – OG_229), mvaD (EC: 4.1.1.33 – OG_1948), mvaK1 (EC: 2.7.4.2 – OG_230) and mvaK2 (EC: 2.7.1.36 – OG_1947) in strains NIZO2264 and NIZO2806.
Strains NIZO2855, NIZO2877 and ATCC14917 have a truncated hepB2 (EC: 2.5.1.30 – OG_1994), but the hepB1 (EC: 2.5.1.30 – OG_307) is complete. Both genes share the same function indicating that the loss of one gene does not result in loss of function.
Other strains that are missing genes are NIZO2484 (missing idi1 (EC: 5.3.3.2 – OG_229)), while strains NIZO2535 and NIZO2855 both miss mvaA (EC: 1.1.1.88 – OG_1729).
Almost all strains have three menA genes (EC: 2.5.1.74). They differ in length (305, 306 and 326 amino acids long) and amino acid sequence identity (e-value between 4e-50 and 6e-34). One paralog (OG_2088) is absent only in strains NIZO2806 and NIZO3894. Some other strains have yet another copy or a fragment of the menA gene in their genome.
Cofactors: Molybdopterin
The genome of strain L. plantarum WCFS1 was shown to contain a cluster of 29 genes encoding a respiratory nitrate reductase and enzymes for biosynthesis of its required cofactor molybdopterin (22). The complete set of genes, including the molybdopterin biosynthesis pathway, is only found in 16 strains. In strains WCFS1, JDM1, NIZO2264 NIZO2484, NIZO2485, NIZO2535, NIZO2535, NIZO2726, NIZO2776, NIZO2830, NIZO2831, NIZO2891 and NIZO3893 the complete set of genes is intact, while strains 19.1, ER, NIZO2802 and NIZO3892 appear to have one truncated gene in this gene cluster. However, the presence of the entire gene cluster suggests that the pathway is also functional in the latter 4 strains.
12. Kandler O, Weiss N. Genus lactobacillus. Bergey's manual of systematic …; 1986.
13. Ledesma OV, HOLGADO ADR. A synthetic medium for comparative nutritional studies of lactobacilli. Journal of Applied …. 1977.
14. Ruiz-Barba JL, Jimenez-Diaz R. Availability of essential B-group vitamins to Lactobacillus plantarum in green olive fermentation brines. Applied and environmental …. 1995.
15. Bowman WC, DeMoll E. Biosynthesis of biotin from dethiobiotin by the biotin auxotroph Lactobacillus plantarum. Journal of Bacteriology. 1993.
16. Santos F, Vera JL, van der Heijden R, Valdez G. The complete coenzyme B12 biosynthesis gene cluster of Lactobacillus reuteri CRL1098. …. 2008.
17. Brooijmans R, de Vos WM. Lactobacillus plantarum WCFS1 electron transport chains. Applied and …. 2009.
18. Begley TP, Downs DM, Ealick SE, McLafferty FW. Thiamin biosynthesis in prokaryotes. … of microbiology. 1999.
19. Morett E, Korbel JO, Rajan E, Saab-Rincon G. Systematic discovery of analogous enzymes in thiamin biosynthesis. Nature. 2003.
20. Wels M, Francke C, Kerkhoven R. Predicting cis-acting elements of Lactobacillus plantarum by comparative genomics with different taxonomic subgroups. Nucleic acids …. 2006.
21. Klaus S, Wegkamp A, Sybesma W. A nudix enzyme removes pyrophosphate from dihydroneopterin triphosphate in the folate synthesis pathway of bacteria and plants. Journal of Biological …. 2005.
22. Kleerebezem M, Boekhorst J, van Kranenburg R, Molenaar D, Kuipers OP, Leer R, et al. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci USA. 2003 Feb 18;100(4):1990–5.