1. INTRODUCTION

Polyhydroxyalkanoates (PHA) are an environmentally friendly alternative to the abusive use of petrochemical plastics since they are biodegradable but can be used for similar purposes such as manufacturing of packaging or biomedical devices. PHA are polyesters synthesised by bacteria as carbon storage compounds when levels of oxygen, nitrogen or phosphorus are low. PHA can be classified according to the amount of carbon atoms in the hydroxyacyl-CoA monomers as short-chain-length PHA (3 to 5) and medium-chain-length PHA (mlc-PHA) (6 to 14). The composition of the monomers depends on the microorganism and the carbon source used, since the last-mentioned can be transformed into hidroxyacyl-CoA precursors by different metabolic routes [1], [2], [3].

Figure 1
Connection of central metabolic pathways to PHA synthesis

Fatty acid de novo synthesis (FAS) pathway is particularly interesting because this carbon sources are generally present in inexpensive organic wastes; for example, glycerol is a by-product of biodiesel production [5], [6], [7]. Additionally, the polymer produced through this pathway, mcl-PHA, can be used as biodegradable alternatives for elastomer and rubber in cosmetics, paint formulations and medical devices [8]. PhaG is the enzyme that allows this connection between FAS pathway and PHA synthesis by transforming the intermediate 3-hydroxyacyl-ACP into 3-hydroxyacyl-CoA [5].

PhaG was characterised for the first time by Rehm et al. (1998) [5] in Pseudomonas putida KT2448 as a hydroxyacyl-CoA-ACP-transferase, however, Wang et al. (2012) [7] suggested that the PhaG protein function is rather a thioesterase. Bacteria missing this PhaG protein accumulated 85 % polymer with octanoate as substrate but only 3 % when gluconate, which is metabolised through FAS pathway, was provided as carbon source [5], indicating its importance for PHA synthesis from gluconate. Furthermore, PHA production from simple carbon sources was reestablished in P. oleovorans ATCC 29347 and P. fragi [9] and augmented up to 40 % in P. aeruginosa PAO1 [5], only by the insertion of the genes phaC + phaG or only phaG, respectively. PhaG protein has only been experimentally characterised and reported in Pseudomonas species, Burkholderia caryophylli and Aeromonas hydrophila [10], out of the 75 bacterial genera that have been reported as PHA producers [2].

There has not been evidence that R. eutropha H16, the most well studied bacterium regarding PHA metabolism and model for large-scale production, possess a phaG homologue. However, R. eutropha H16 has also shown its ability to utilise alternative carbon sources such as gluconate, glycerol or acetate for PHA synthesis [11]. Peplinski et al. (2010) [12] reported a relationship between (FAS) and PHA synthesis, based on the upregulation of genes involved in FAS pathway, such as accC2 and fabG, when R. eutropha H16 grew on sodium gluconate as carbon source and produced PHA. Additionally, other studies have shown that deletion of up to 9 phaA homologues does not suppress PHA production from sodium gluconate in this bacterium [13], [14]. These evidences suggest R. eutropha H16 may possess proteins able to perform the same function as PhaG.

Proteins with < 50 % similarity can be compared on the basis of conserved motifs which can reveal protein function even when this is not globally similar to any known protein [15]. Protein function prediction is a major issue in biology since the protein databases grow exponentially with the crescent availability of fast and inexpensive sequencing techniques while experimental protein characterisation techniques are still time consuming and expensive. Due to in silico analysis, novel and useful proteins can be targeted that will allow the development of new PHA production strategies. The goals of this research were: to identify genes in R. eutropha H16 codifying proteins with high similarity to PhaG, to analyse and compare those proteins in silico to all the already experimentally characterised PhaG proteins and to preliminarily confirm the presence of the genes codifying these in silico predicted proteins.

2. METHODOLOGY

In silico analysis and PhaG homologues characterisation

Using PhaG protein sequence from P. putida KT2440 (Model PhaG protein [10]), accession number AAC34749.1, a standard and specialised protein BLAST (BLASTP) was carried on BacMap protein database, on both R. eutropha H16 chromosomes (Matrix: BLOSUM62, Mask: low complexity, Program: blastp, Database: Protein). Only those proteins with similarity > 40 % and e-value ≤ 1e-04 were selected. After a literature search and selection for already characterised PhaG proteins [10], [16], 9 PhaG homologues were analysed using the tool MOTIF finder from the GenomeNet Database Resources of the Kyoto University Bioinformatics Center. PhaG proteins for PROSITE patterns, NCBI Conserved Domains (CDD) and Protein families (Pfam), were aligned using the UniProt alignment tool (www.uniprot.org/align) and conserved aminoacids related to known catalytic aminoacid residues in thioesterases were located manually. Using Membrane protein IdeNtificatioN withOUt explicit use of hydropathy profiles and alignments (MINNOU server) an image from the secondary structure of the proteins was generated.

Culture conditions

R. eutropha H16 was maintained on tryptic soy agar (TSA) plates. Individual colonies were inoculated in tryptic soy broth (TSB) and incubated at 30 °C for 12 h. 2 mL of cultures with OD600: 0,5, were centrifuged, supernatant was discarded and cell pellet was washed twice in NaCl 0,9 % solution before DNA extraction.

DNA extraction and R. eutropha H16 phaG-like gene amplification

R. eutropha H16 DNA was used as template for the development and optimisation of phaG-like genes amplification. DNA was obtained by using DNeasy Blood and Tissue Kit (Qiagen) according to the manufacturer's protocol for Gram negative bacteria. Primers targeting phaG-like gene H16_A0147 were designed and desired sequence was sent to company Eurofins scientific to be synthesised. A PCR with different annealing temperatures was carried out to determine the annealing temperature. PCR amplifications were performed in a 20 μL reaction mixture containing 1X Taq buffer; 1.5 mM MgCl2, 0.2 mM of dNTPs, 1 μM of each primer, 0.03 U/ μL of Taq DNA polymerase and 1 μL of genomic DNA (25–30 ng). PCR conditions were: initial denaturation 95 °C for 30 s; denaturation 95 °C for 20 s; annealing 53-60 °C for 45 s; extension 72 °C for 60 s (30 cycles) and final extension 72 °C for 120 s. Four negative controls with no DNA were including at annealing temperatures of 54, 56, 58 and 60. The resulting PCR products were visualised in 1 % agarose gel and stained with EZ-Vision® in gel solution.

3. RESULTS

In silico analysis and PhaG homologues characterisation.

Two proteins, one codified by a gene from chromosome 1 and one codified by a gene from chromosome 2, were found with % sequence identity >20, % positive substitutions >40 and e-value <=1e-04 when compared to PhaG sequence and are presented in Table 1, as phaG-like proteins. According to Russell et al. (1997) [17] proteins that are able to perform the same function, either a remote homologue or analogue can show less than 50 % of sequence similarity due to shared active sites or conserved domains; the formal definition of remote homology is protein sequences that share an identity percentage of less than 25 % [18]. mhpC gene is already annotated in R. eutropha H16 genome as aminoacrylate hydrolase and located in chromosome 2 while all PHA-related genes are located in chromosome 1 in R. eutropha H16 [11]; for these reasons, it was not included in further analysis as potential PhaG analogue or remote homologue.

Already characterised PhaG proteins as well as H16_A0147 were individually tested for conserved domains and protein families. All PhaG homologues and H16_A0147 were found to have an α/β hydrolase1 domain as the principal and only domain in CDD. No matches were found for any of the proteins using PROSITE database. All PhaG homologues, as well as H16_A0147, exhibit significant similarity to Pfam Abhydrolase_6 and Hydrolase_4 (Table 2). No PhaG homologue showed significant similarity to a transferase Pfam or CDD. Interestingly, PhaG from Pseudomonas sp. USM 4-55 as well as H16_A0147 showed significant similarity to a PHA depolymerase domain.

Table 1

PhaG-like proteins in R. eutropha H16 genome.

Table 2

E-values from PhaG homologues and R. eutropha H16_RS00705 compared to Pfam database.

α/β hydrolase superfamily is a large family of proteins where Thioesterases are included. Predicted secondary structures for P. putida KT 2447 PhaG and H16_A0147 are presented in Figure 2. Visually, predicted secondary structures are very similar among these two proteins in contrast with their low similarity percentages, further supporting that low global protein sequence similarity does not necessarily mean two proteins differ in their spatial arrangement, which is a key protein function determinant.

Figure 2
Predicted secondary structure image generated with MINNOU Server [19], β-sheets are represented as green arrows and α-helices as red waves. The third line indicates the confidence level of the predicted structure for that particular position. H16_A0147 gene product (A) and PhaG protein from P. putida KT2447 (B).

Confirmation of H16_A0147 gene presence

A PCR product with around 800 bp was obtained using the designed primers on DNA extracted from R. eutropha H16, in accordance with the in silico obtained information about H16_A0147 sequence (Table 3).

Table 3

Amplification of a phaG-like gene from R. eutropha H16 genome. Gradient temperature PCR was done.

4. DISCUSSION

In silico analysis and PhaG homologues characterisation.

The comparison of the Pfam analysis among the PhaG homologues and H16_A0147 indicates they belong to the same protein family, α/β hydrolase fold family, in spite of the low sequence similarity, further suggesting they might share catalytic active sites. Regarding the H16_A0147 gene, a thioesterase domain [20] was found in its sequence but no domain indicating a transferase function.

PhaG had been originally classified as an (R)-3-hydroxyacyl- ACP-CoA transferase based on the ability of a partially purified extract to convert 3-hydroxydecanoyl-CoA into 3-hydroxydecanoyl-ACP when ACP was present [5]. However, overexpression of phaG in E. coli resulted in the extracellular accumulation of 3-hydroxydecanoic acid [7] and low PHA production (0,9 mg/L) compared to E. coli harboring both phaG and PP0763 (25 mg/L) a predicted medium-chain-fatty-acid CoA ligase [21] suggesting that PhaG has rather thioesterase activity than transferase activity and separates the ACP moiety from the 3-hydroxyacyl before another enzyme, probably a ligase, catalyses the joining of the CoA moiety to form 3-hydroxyacyl-CoA. Our results are according to those from Wang et al. (2012) [21] and support the possibility that PhaG-like proteins in R. eutropha H16 perform a thioesterase function instead of a transferase function, since no transferase conserved domain was found. Further studies are needed in order to characterise the enzyme activity of this protein.

α/β hydrolase fold family is one of the largest groups of structurally related proteins. Thioesterases are included together with other hydrolytic enzymes such as acetylcholinesterases, carboxylesterases, dienelactones hydrolases, lipases, cutinases, serine carboxypeptidases, proline iminopeptidases, proline oligopeptidases and epoxide hydrolases, and also enzymes that require HCN, H2O2 or O2 instead of H2O such as haloalkane dehalogenases, haloperoxidases and hydroxynitrile lyases [22] but the family also includes non-catalytic enzymes [23]. Thioesterases can hydrolyse the ester bond between a carbonyl group and a sulfur atom, the substrates include both CoA and ACP moieties, and are classified in 23 families with low sequence homology but similar tertiary structure [20].

Genome context was also analysed for H16_A0147 gene and compared with the location of phaG from P. putida KT 2447. H16_A0147 gene is surrounded upstream by other genes codifying hypothetical proteins, and downstream by gene pspE1 codifying a rhodanese-related sulfurtransferase and by bhmT codifying a methyltransferase (Figure 3). phaG gene is located in the lagging strand unlike H16_A0147 and is surrounded upstream by a gene codifying a hypothetical protein and by gene yddV codifying a diguanylate cyclase. Downstream is surrounded by gene ychM codifyng a sulfate transporter and by a gene codifying a hypothetical protein (Figure 5). No similarities were found regarding genome context of the two genes.

Figure 3 A
Localisation and surrounding genes in their respective strains for (A) H16_A0147 in R. eutropha H16 genome and (B) phaG (PPU3428_kt) in P. putida KT2447 genome (Source: Bacmap [24]).

Figure 3 B
Localisation and surrounding genes in their respective strains for (A) H16_A0147 in R. eutropha H16 genome and (B) phaG (PPU3428_kt) in P. putida KT2447 genome (Source: Bacmap [24]).

Confirmation of PhaG-like gene presence

Using the primers designed the in silico sequence to target specifically the gene H16_A0147, a PCR product was obtained with the expected length, at annealing temperatures of 58, 59 and 60, indicating lower temperatures are not optimal for primer alignment to the DNA targeted region. It is necessary to perform DNA sequencing on this PCR product in future studies to confirm that the sequence belongs to this particular gene. This confirmation could imply that this microorganism has an important element for the metabolic pathway implicated in the connection of FAS with PHA synthesis.

FAS pathway is necessary to provide fatty acids to the cell, required for phospholipid synthesis and, thus, membrane formation for cell growth and division, functions that are highly active during exponential phase and decline during stationary phase when PHA production begins [12], [25], [26]. However, among these studies, Peplinski et al. (2010) [12] and Shimizu et al. (2013) [26] reported genes accC2 and fabG involved in FAS pathway were upregulated during PHA production in P. putida KT2447. These genes are expressed during the initial steps of FAS, and could be allowing accumulation of 3-hydroxyacyl-ACP, which is the substrate for PhaG. According to that, Wang et al. (2012) [21] demonstrated that phaG, phaC1 and phaC2 genes expression showed n-fold induction of 220, 2.6 and 4.3, respectively, during PHA production phase in the same bacteria. We suggest this could be happening in R. eutropha H16 metabolism, and could be further studied by gene expression analysis targeting H16_A0147, accC2 and fabG.

This is the first study that shows the presence and the characteristics of a putative phaG-like gene in R. eutropha H16 and represents the first step for the identification of a possible connection between FAS pathway and PHA synthesis in this novel bacterium. Further studies are necessary to confirm the potential relationship of the gene product of H16_A0147 gene with the PHA metabolism. Characterising a PhaG-like protein in R. eutropha H16 could allow the use of residues from biodiesel industry [6], [7], or phenylacetic acid, which is a contaminating compound [27], for PHA synthesis in the future.

5. CONCLUSIONS

H16_A0147 was identified as a potential gene codifying for a PhaG-like protein with Thioesterase activity. Further studies are neccessary to confirm this metabolic link. If this link is confirmed, expression of H16_A0147 may be manipulated in R. eutropha H16 for the production of PHA using waste products such as glycerol or contaminants like phenylacetic acid as carbon sources.

Acknowledgments

Carolina Ramírez, Luisa Múnera, Dr. Nancy Pino, the School of Microbiology and laboratory GDCON, all from University of Antioquia, contributed to this study and are gratefully acknowledged.

7. REFERENCIAS BIBLIOGRÁFICAS

[1] Aldor I.S. y Keasling, J.D. (2003). Process design for microbial plastic factories: metabolic engineering of polyhydroxyalkanoates. Curr. Opin. Biotechnol. 14 475–483.

[2] Reddy, C.S.K. Ghai, R., Rashmi. y Kalia, V.C. (2003). Polyhydroxyalkanoates: an overview. Biores Tech 87, 137–146.

[3] Urtivia, V., Villegas, P., González, M., Seeger, M. (2014). Bacterial production of the biodegradable plastics polyhydroxyalkanoates. Intern J Biol Macromol. 70: 208–213.

[4] Lau, N., Foong, C.P., Kurihara, Y., Sudesh, K. y Matsui, M. (2014). RNA-Seq Analysis Provides Insights for Understanding Photoautotrophic Polyhydroxyalkanoate Production in Recombinant Synechocystis Sp. PLOS one. Vol 9(1).

[5] Rehm, B.H.A., Kruger, N. y Steinbuchel, A. (1998). A New Metabolic Link between Fatty Acid de novo Synthesis and Polyhydroxyalkanoic Acid Synthesis: the phaG gene from Pseudomonas putida KT2440 encodes a 3-hydroxyacyl- acyl carrier protein-coenzyme a transferase. J. Biol. Chem. 273:24044-24051.

[6] Nomura, C.T., Tanaka, T., Eguen, T.E., Appah, A.S., Matsumoto, K., Taguchi, S., Ortiz, L. y Doi, Y. (2008). FabG Mediates Polyhydroxyalkanoate Production from Both Related and Nonrelated Carbon Sources in Recombinant Escherichia coli LS5218. Biotechnol. 24: 42-351.

[7] Wang, Q., Zhuang, Q., Liang, Q. y QI, Q. (2013). Polyhydroxyalkanoic acids from structurally-unrelated carbon sources in Escherichia coli. Appl Microbiol Biotechnol. Vol 97:3301–3307.

[8] Leong, Y.K., Show P.L., Ooi, C.W., Ling, T.C. y Lan, J.C. (2014). Current trends in polyhydroxyalkanoates (PHAs) biosynthesis: Insights from the recombinant Escherichia coli. J Biotech 180, 52-65.

[9] Fiedler, S., Steinbuchel, A. y Rehm, B.H. (2000). PhaG-Mediated Synthesis of Poly(3-Hydroxyalkanoates) Consisting of Medium-Chain- Length Constituents from Nonrelated Carbon Sources in Recombinant Pseudomonas fragi. Appl. Environ. Microbiol. 66(5):2117-2124.

[10] Röttig, A., y Steinbüchel, A. (2013). Acyltransferases in Bacteria. Microbiol Mol Bio Reviews : MMBR, 77(2), 277–321.

[11] Pohlmann, A., Fricke, W. F., Reinecke, F., Kusian, B., Liesegang, H., Cramm, R., Eitinger, T., Ewering, C., Potter, M., Schwartz, E., Strittmatter, A., Voss, I., Gottschalk, G., Steinbuchel, A., Friedrich, B. y Bowien, B. (2006). Hydrogen-based biotechnology: genome sequence of the bioplastic-producing ‘‘Knallgas’’ bacterium Ralstonia eutropha H16. Nat Biotechnol 24, 1257–1262.

[12] Peplinski, K., Ehrenreich, A., Doring, C., Bomeke, M., Reinecke, F., Hutmacher, C. y Steinbuchel, A. (2010). Genome-wide transcriptome analyses of the ‘Knallgas’ bacterium Ralstonia eutropha H16 with regard to polyhydroxyalkanoate metabolism. Microbiology, 156: 2136–2152.

[13] Lindenkamp, N., Peplinski, K., Volodina, E., Ehrenreich, A. y Steinbuchel, A. (2010). Impact of Multiple -Ketothiolase Deletion Mutations in Ralstonia eutropha H16 on the Composition of 3-Mercaptopropionic Acid-Containing Copolymers. Appl. Environ. Microbiol. Vol. 76(16): 5373–5382.

[14] Lindenkamp, N., Volodina, E. y Steinbuchel, A. (2012). Genetically Modified Strains of Ralstonia eutropha H16 with -Ketothiolase Gene Deletions for Production of Copolyesters with Defined 3-Hydroxyvaleric Acid Contents. App. Environ. Microbio. 78(15):5375-83.

[15] Nevill-Manning, C., Wu, T. y Brutlag, D. (1998) Highly specific protein sequence motifs for genome analysis. Proc. Natl Acad. Sci. 95:5865–5871.

[16] Arsad H. (2009) Cloning and Characterisation of (R)-3-hydroxyacyl-acyl Carrier Proteincoenzyme A Transferase Gene (phaG) from Pseudomonas sp. USM 4-55. Trop. Life. Sci. Res. 20(2):1-14.

[17] Russell, B., Saqi, M.A., Sayle, R.A., Bates, P.A. y Sternberg, M.J.E. (1997). Recognition of analogous and homologous protein folds: Analysis of sequence and structure conservation. J. Mol. Biol. 269:423-439.

[18] Bedoya, O. y Tischer, I. (2015) Reducing dimensionality in remote homology detection using predicted contact maps. Comput. Biol. Med. 59:64-72.

[19] Cao, B., Porollo, A., Adamczak, R., Jarrell, M. y Meller, J. (2006). Enhanced Recognition of Protein Transmembrane Domains with Prediction-based Structural Profiles. Bioinformatics. 22:303-9

[20] Cantu, D.C., Chen, Y. y Reilly, P.J. (2010). Thioesterases: a new perspective based on their primary and tertiary structures. Protein Sci. 19:1281–1295.

[21] Wang, Q., Tappel, RC., Zhu, C. y Nomura, CT. (2012). Development of a new strategy for production of medium-chain-length polyhydroxyalkanoates by recombinant Escherichia coli via inexpensive non-fatty acid feedstocks. Appl Environ Microbiol. 78(2):519-27.

[22] Bugg, TD. (2004). Diverse catalytic activities in the alphabeta-hydrolase family of enzymes: activation of H2O, HCN, H2O2, and O2. Bioorg Chem. 32(5):367-375.

[23] Nardini, M., Dijkstra, BW. (1999). α/β Hydrolase fold enzymes: the family keeps growing. Curr Opin Struct Biol. 9(6):732-737

[24] Stothard, P., Van Domselaar, G., Shrivastava, S., Guo, A., O'Neill, B., Cruz, J., Ellison, M. y Wishart, DS. (2005) BacMap: an interactive picture atlas of annotated bacterial genomes. Nucleic Acids Res 33:D317-D320.

[25] Brigham, C.J., Speth, D.R., RHA, C. y Sinskeya, A.J. (2012). Whole-Genome Microarray and Gene Deletion Studies Reveal Regulation of the Polyhydroxyalkanoate Production Cycle by the Stringent Response in Ralstonia eutropha H16. App Environ Microbiol. 78;8033–8044

[26] Shimizu, R., Chou, K., Orita, I., Suzuki, Y., Nakamura, S. y Fukui, T. (2013). Detection of phase-dependent transcriptomic changes and Rubisco-mediated CO2 fixation into poly (3-hydroxybutyrate) under heterotrophic condition in Ralstonia eutropha H16 based on RNA-seq and gene deletion analyses. BioMed Cen Microbiol. 13:169.

[27] Tobin, KM., O'Leary, ND., Dobson, AD. y O'Connor, KE. (2007). Effect of heterologous expression of phaG [(R)-3-hydroxyacyl-ACP-CoA transferase] on polyhydroxyalkanoate accumulation from the aromatic hydrocarbon phenylacetic acid in Pseudomonas species. FEMS Microbiol Lett. 268(1):9-15

Author notes

melissa.uribe1@udea.edu.co

Additional information

Cómo citar este artículo: M. Uribe, A. Villa.“In silico analysis of phag-like protein in ralstonia eutropha H16, potentially involved in polyhydroxyalkanoates synthesis”, Revista Politécnica, vol. 15, no.29 pp.55-64, 2019. DOI: 10.33571/rpolitec.v15n29a5

Alternative link

https://revistas.elpoli.edu.co/index.php/pol/article/view/1371/1224 (html)

https://revistas.elpoli.edu.co/index.php/pol/article/view/1371/1214 (pdf)