Amino acid metabolism
Lactic acid bacteria (LAB) have multiple amino acid and vitamin auxotrophies, and it is believed that evolution on rich and complex media resulted in selection of specific auxotrophies which are closely correlated to their biotope (3).Several studies have be performed to find the amino acid auxotrophies for L. plantarum (4)(5)(6).All three studies used different strains and they agree that L. plantarum is auxotrophic for valine, isoleucine and glutamate. Furthermore, there were strain-specific auxotrophies for phenylalanine, methionine, arginine, threonine and cysteine.
Below we describe the analysis of genes involved in amino acid metabolism in the 54 L. plantarum genomes. A summary can be found in Table 6.
Branched-chain amino acids (Valine, Leucine and Isoleucine)
The biosynthesis pathways for valine, leucine and isoleucine are found to be absent in all L. plantarum strains, as reported in previous studies (4)(5).
In the genomes of L. plantarum all genes necessary for histidine biosynthesis are found in a single large cluster on the chromosome. This gene cluster is present in all strains (Figure 4). However, several strains contain putative pseudogenes in this cluster. In strains NIZO2484, NIZO2485 and NIZO2776 the hisCgene (EC: 126.96.36.199 – OG_883) is interrupted. Strain NIZO2776 has a truncated hisGgene (EC: 188.8.131.52 – OG_1018). Strain 19.1 has a truncated hisDgene (EC: 184.108.40.206 – OG_1011), which is necessary for the last steps in the histidinebioysynthesis pathway. Besides the biosynthesis of histidine, all strains have genes to methylate histidine. No genes are found for the decarboxylation of histidine into histamine.
Alanine and aspartate
All strains are able to biosynthesize aspartate (Figure 5). Besides biosynthesis, most strains are able to convert L-aspartate into L-asparagine and several other products, such as fumarate and oxaloacetate. There are three pathways from aspartate to fumarate. One pathway uses adenylosuccinate as an intermediate and another pathway uses argininosuccinate as an intermediate. These pathways are found in all strains. The last pathway is the direct conversion into fumarate by aspA (EC: 220.127.116.11 – OG_2139). In strains UMCA_3037 and NIZO2259 these genes are pseudogenes. Most strains use an aspartate aminotransferase (EC:18.104.22.168 – OG_1954) to convert aspartate to oxaloacetate, but in strains NIZO2264 and NIZO2806 this gene is absent. Furthermore strains NIZO1838, NIZO2802 and NIZO3893 contain an L-aspartate oxidase (EC 22.214.171.124 – OG_3886) for the same reaction.
All strains are able to produce L-asparagine through three pathways. Few strains have two asparagine synthase genes (ansB1 and ansB2 (EC: 126.96.36.199 – OGs 4210, 5710)) (Table 1). Most strains can convert L-aspartate into D-aspartate through the racD gene (EC: 188.8.131.52), but in strains JDM1, NIZO2257, NIZO2258, NIZO2726, NIZO2776, NIZO2830 and NIZO2831 this gene is split. In strains NIZO2535 and NIZO2891 a paralog of this gene is present. All strains can convert L-alanine into D-alanine through the alanine racemasealr (EC: 184.108.40.206 – OG_1889). Only in strains 16 and P8 a L-aspartate-beta-decarboxylase (EC: 220.127.116.11 – OG_4451) is found, which can convert aspartate into alanine.
All strains should be able to biosynthesize alanine (Figure 5), but it is not clear from which precursors and which enzymes.
Glutamate is synthesized from 2-oxoglutarate and NH3 by the gdhgene (EC: 18.104.22.168 – OG_526), which is found in all strains (Figure 6). However, since L. plantarum has an incomplete citrate cycle, no 2-oxoglutarate is formed. So L. plantarum cannot biosynthesize glutamate, which is in agreement with growth experiments (4)(5). The gdhgene is most likely involved in amino acid degradation instead of the biosynthesis of glutamate (7).
After glutamate is acquired from the environment, all strains are able to convert it into glutamine. This glutamine can be converted into precursors for the purine, pyrimidine, amino sugars, proline and arginine metabolic pathways.
Most strains are able to convert glutamate into succinate. Only strains NIZO2456 and NIZO2494 have a gadBpseudogene (EC: 22.214.171.124 – OG_6316). In strains NIZO1838, NIZO2236, NIZO2484, NIZO2485 and NIZO2776 the gapDgene (EC: 126.96.36.199 – OG_2392) appears to be absent.
The biosynthetic pathways of lysine with aspartate or homoserine as precursors is highly conserved in the 54 L. plantarum strains (Figure 7). There are three pathways in the biosynthesis pathway of lysine to get from 2,3,4,5-tetrahydrodipicolinate to LL-2,6-diaminoheptanedioate: the direct pathway, the succinylase pathway and the acetylase pathway. The genes of all the pathways are present in all the strains. Which pathway is taken by L. plantarum is unknown.
Besides the biosynthesis of lysine, the direct precursor of lysine is also used as a precursor for peptidoglycan biosynthesis. Most strains have both genes towards peptidoglycan biosynthesis, but the murF gene (EC: 188.8.131.52 – OG_1893) is a pseudogene in strains 19.1 and NAB2.
Arginine and proline
Two growth studies showed differences in arginine requirements in L. plantarum: strain NIZO2726 was prototrophic for arginine, while strain WCFS1 was auxotrophic (4)(5). In the present study, the full pathway of biosynthesis of arginine from glutamate was found to be present in all strains (Figure 8). Regulatory effects or mutation in the protein sequence could be the cause of arginine deficiency, but the exact reason is unknown.
All strains have all the genes necessary for the biosynthesis of proline from glutamate. Only strain NAB1 has a split proA gene (EC: 184.108.40.206 – OG_1234).
All strains appear to have a complete biosynthesis pathway for methionine from aspartate (Figure 9).
All strains are able to synthesize cysteine from serine (Figures 10). The only difference is in the number of paralogs (and pseudogenes) of the cystathionine beta-lyase genes (EC: 220.127.116.11). All strains have cblB, cblA1 and cblA2 genes (OGs 1836, 1882, 184). The cblA3 gene (OG_2957) is only found in strains CNW10, NIZO2263, NIZO2726, NIZO2741, NIZO2802, NIZO2806, NIZO2830, NIZO2831, NIZO2855, NIZO2877, NIZO3892, ATCC14917, NC8, JDM1 and WCFS1.
Serine, Glycine and Threonine
Serine and threonine can be biosynthesized by all strains (Figure 11 and 12). For several enzymes there are 2 or more paralogs in the genomes. Only strain 19.1 may have a split glyA gene (EC: 18.104.22.168 – OG_1496), preventing the biosynthesis of glycine. In strain NIZO3892 the gene encoding D-serine dehydratase (dsdA, EC: 22.214.171.124 – OG_580) is a pseudogene.
Aromatic amino acids (Tryptophan, tyrosine and phenylalanine)
Most strains have a complete shikimate pathway converting erythrose 4-phosphate and phosphoenolpyruvate into chorismate (Figure 13). Some strains however lack genes or have putative pseudogenes in this pathway (Table 6). All stains are able to convert chorismate into tyrosine, although the hisC gene is split in stains NIZO2484, NIZO2485, and NIZO2776. According to the sequence data, only strains NIZO1837, NIZO2260 and NIZO3894 have prephenatedehydratase (EC: 126.96.36.199 – OG_4186) making it able to biosynthesize phenylalanine. But according to a growth study strain WCFS1 was able to grow when phenylalanine was omitted from the medium (5). A protein Blast against the NCBI database did not show any proteins similar to prephenatedehydratase. The gap in this pathway could not be filled.
The genes necessary for the biosynthesis of tryptophan are adjacent to each other in a small region on the genome. Part of this region is missing only in strain NIZO3400, i.e. genes trpC, trpD, trpG and part of trpF. The other strains have all genes for the biosynthesis of tryptophan, although strains 19.1 and NIZO2814 have a split trpDgene (188.8.131.52 – OG_1049).
Amino acid decarboxylases/antiporters
In several LAB adjacent genes are found for an amino acid decarboxylase and an antiporter(8)(9)(10). The decarboxylase removes a CO2 from the amino acid. During this process a proton from the cytoplasm is added to the decarboxylated form of the amino acid. This makes the cytoplasm less acidic. The decarboxylated amino acid is then transported out of the bacterium while another amino acid molecule is transported into the bacterium. This system allows the bacteria to survive in more acidic conditions than other LAB by maintaining a higher internal pH.
No amino acid decarboxylase/antiporter combinations are found on the chromosomes of L. plantarum strains, but two combinations are found on plasmids. In strain NIZO2801 a paralog of glutamate decarboxylase (EC 184.108.40.206 – OG_6316) is found on a plasmid, accompanied by a glutamate/gamma-aminobutyrate (GABA) antiporter. In strains P8 and Lp16, an aspartate-beta-decarboxylase (OG_4451) accompanied by an aspartate-alanine antiporter (OG_4452) are found on plasmids.
3. Bringel F. Carbamoylphosphate and naturalauxotrophies in lacticacidbacteria. Lait. EDP Sciences; 1998;78(1):31–7.
4. Morishita T, Deguchi Y, Yajima M, Sakurai T, Yura T. Multiple nutritionalrequirements of lactobacilli: geneticlesionsaffectingaminoacidbiosyntheticpathways. 1981.
5. Teusink B, van Enckevort F, Francke C. In silico reconstruction of the metabolicpathways of Lactobacillus plantarum: comparingpredictions of nutrientrequirementswiththosefromgrowthexperiments. Appliedand …. 2005.
6. Saguir FM, de Nadra M. Improvement of a chemicallydefined medium for the sustainedgrowth of Lactobacillus plantarum: nutritionalrequirements. Currentmicrobiology. 2007.
7. Helinck S, Le Bars D, Moreau D. Ability of thermophiliclacticacidbacteria to producearoma compounds fromaminoacids. Appliedand …. 2004.
8. Cotter PD, Gahan C, Hill C. A glutamate decarboxylase system protects Listeria monocytogenes in gastricfluid. Mol Microbiol. 2001.
9. Biase D, Pennacchietti E. Glutamate decarboxylase‐dependentacidresistance in orallyacquiredbacteria: function, distribution and biomedical implications of the gadBCoperon. Mol Microbiol. 2012.
10. Molenaar D, Bosscher JS, Brink Ten B. Generation of a proton motive force by histidine decarboxylation and electrogenic histidine/histamine antiport in Lactobacillus buchneri. Journal of …. 1993.