Accepted Article Title: Conversion of a mono- and diacylglycerol lipase into a triacylglycerol lipase by protein engineering
Authors: Dongming Lan; Grzegorz Maria Popowicz; Ioannis V. Pavlidis; Pengfei Zhou; Uwe T. Bornscheuer; Yonghua Wang
This manuscript has been accepted after peer review and the authors have elected to post their Accepted Article online prior to editing, proofing, and formal publication of the final Version of Record (VoR). This work is currently citable by using the Digital Object Identifier (DOI) given below. The VoR will be published online in Early View as soon as possible and may be different to this Accepted Article as a result of editing. Readers should obtain the VoR from the journal website shown below when it is published to ensure accuracy of information. The authors are responsible for the content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201500163 Link to VoR: http://dx.doi.org/10.1002/cbic.201500163
A Journal of
www.chembiochem.org
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION Conversion of a mono- and diacylglycerol lipase into a triacylglycerol lipase by protein engineering Dongming Lan,[a] Grzegorz Maria Popowicz,[b] Ioannis V. Pavlidis,[c] Pengfei Zhou,[d] Uwe T. Bornscheuer[c] and Yonghua Wang*[a]
Abstract: Despite the fact that most lipases are expected to be active against triglycerides, there is a small group of lipases which are active only on mono- and diacylglycerides. The reason behind the different substrate scope is not clear so far. In the present work we tried to identify the reasons for this differentiation, using the lipase from Malassezia globosa as a template. By means of protein engineering and with implementation of only one mutation we managed to convert this enzyme to a typical triacylglycerol lipase, although the wild type does not accept triglycerides. The variant Q282L accepts a broad spectrum of triglycerides, while the catalytic behavior is altered to some extent. From the in silico analysis it seems that specific hydrophobic interactions are the key to the altered substrate specificity.
Lipases (EC 3.1.1.3) are one of the most studied enzyme groups, with numerous applications developed so far.[1] Their substrate scope varies, but it is generally accepted that most lipases accept triacylglycerides (TAG). However, there is a small group of lipases which are not able to hydrolyze triglycerides, but only mono- and diacylglycerides (MAG and DAG). These lipases have attracted significant attention in medical research and industrial applications due to their altered specificity. For example, MAG and DAG specific lipases are involved in controlling the neuronal communication by biosynthesis and inactivation of 2-arachidonoylglycerol. [2] MAG-lipases are found to promote cancer pathogenesis by regulating a fatty acid network enriched in oncogenic signaling lipids. [3] Their important physiological functions make them being valuable targets for drug development.[4] In industrial application, MAG and DAG specific lipases are excellent biocatalysts that have been applied to the synthesis of high-pure DAG, [5] biodiesel [6] and food emulsifier [7]. However, the reason behind this peculiar substrate
[a]
[b]
[c]
[b]
specificity of these enzymes is yet unclear. The understanding of the molecular basis of the substrate selectivity mechanism can provide an insight to rationally design novel biocatalysts with desired substrate specificity via protein engineering. There are several examples of successful implementation of protein engineering techniques for the change of the substrate scope of lipases; some of them from our group. [8] However, there are no works on the modification of acylglycerol specificity, to help on the elucidation of the differences that lead to the difference of the specificity we mentioned above. Recently, the X-ray structure of a lipase from Malassezia globosa (SMG1) which is active on DAGs but not on TAGs, has been resolved by our group (PDB entries: 3UUE and 3UUF).[9] The crystal structure of this lipase was used in the present study to guide the rational design, in order to alter the acylglycerol specificity. As seen in Figure 1, several residues in the catalytic pocket were targeted for mutagenesis. More specifically, the loop of residues 103 to 107 is considered to be the lid of the lipase,[4] thus several positions were targeted on this loop. More than that, the neighboring residues of the catalytic serine, namely H170 and L172, were targeted. They are both bulky residues and we considered to change the motif of the catalytic serine. According to our previous work,[10] SMG1 fits to superfamily II of lipases (GHSLG), but in order to shift to smaller residues, we considered to change the motif to GESAG, which is characteristic for the superfamily III. Last but not least, the positions Y56, F278 and Q282 were targeted, as they point towards to the substrate pocket, and thus potentially pose steric hindrances. A full list of the variants prepared for this work is available in the supporting information section (Table SI-1).
Dr. D.M. Lan, Prof. Dr. Y. H. Wang College of Light Industry and Food Sciences South China University of Technology Guangzhou 510640, PR China E-mail: [email protected] Dr. G. M. Popowicz Institute of Structural Biology, Helmholtz Zentrum München Deutsches Forschungszentrum für Gesundheit und Umwelt (GmbH) Ingolstädter Landstraße 1, Germany Dr. Ioannis V. Pavlidis, Prof. Dr. Uwe T. Bornscheuer Dept. of Biotechnology and Enzyme Catalysis, Institute of Biochemistry University of Greifswald Felix-Hausdorff-Str. 4, D-17487, Greifswald, Germany P.F. Zhou School of Bioscience and Bioengineering South China University of Technology Guangzhou 510640, PR China Supporting information for this article is given via a link at the end of the document.
Figure 1. Catalytic pocket of SMG1. In the centre the residues of the catalytic triad are depicted in grey, while the targeted residues for mutagenesis are depicted in cyan and the lid region is colored orange.
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION All SMG1 variants were prepared and expressed in Pichia pastoris and purified by immobilized metal affinity chromatography, taking advantage of the C-terminal His-Tag. The alteration of the selectivity scope was monitored in the esterification reaction of glycerol with fatty acids (FAs) varying in carbon length and degree of saturation (C8:0–C22:6) in a solvent free system at 30°C. Strikingly, the residues on the lid did not alter the selectivity; all of them produced DAGs, but no triglycerides were formed (Table SI-1). However, we were able to find two positions that seem to alter the selectivity. The variant Y56A did only produce monoglycerides, but this seems to be an effect of the low activity observed, as it is much less active than the wild type or all other variants. The most interesting variant was Q282L, which enabled SMG1 to produce triglycerides. Wildtype SMG1 and the variant Q282L have the same fatty acid (FA) selectivity during this competitive acylation reaction; both enzymes exhibit a preference for decanoic acid for the saturated FAs, but they are more efficient with unsaturated FAs (oleic acid C18:1 and linoleic acid C18:2; see Figure SI-1). As the single mutant Q282L is the only one active towards the synthesis of triglycerides, its hydrolytic activity was also evaluated, using several triglycerides. As can be seen in Table 1, Q282L was able to hydrolyze a broad scope of triglycerides, while the wild type enzyme did not exhibit any catalytic activity. The catalytic efficiency of this variant may seem low, leading to about 20 % conversion after three days, but this is to be expected for novel activities that the wild-type enzyme does not have. The variant seems to have a preference for longer acyl chains, such as in triolein (C18:1) or olive oil. It needs to be stated that olive oil is composed mostly of triglycerides of oleic acid and linoleic acid, which were shown to be preferred in the synthetic reaction (Figure SI-1). Based on these results, the variant Q282L can be considered a true TAG-lipase.
Table 2. Free fatty acid content after the hydrolysis of 1,2- and 1,3-DAGs by SMG1 wild-type and mutant Q282L. Standard deviation was less than 2.7 %. Reaction time (min)
1,2-dilaurin
1,3-dilaurin
Wild type
Q282L
Wild type
Q282L
0
0%
0%
0%
0%
15
83.1 %
5.9 %
55.8 %
1.8 %
30
94.6 %
6.5 %
94.27 %
5.6 %
60
99.9 %
9.8 %
99.9 %
6.6 %
The biochemical properties of Q282L mutant were investigated and compared to these of the wild-type enzyme. The Q282L mutant has the same optimal temperature (25°C) and pH (6.0) than the wild-type. To study the substrate selectivity of variant Q282L, the hydrolytic activity towards p-nitrophenyl (pNP) esters with variable acyl chain length (C4 to C16) was evaluated. As seen in Figure 2, the mutation Q282L led to a change of the substrate preference from medium (C6-C8) to shorter ones (C4). Interestingly, a significant improvement on the hydrolytic activity was observed. The specific activity of the Q282L variant was increased 14-fold and 4-fold compared to that of the wild-type enzyme for the C-4 and C-8 substrates, respectively (Table SI-2).
Table 1. Free fatty acid content of several substrates, after hydrolysis with SMG1 Q282L. Standard deviation is less than 2.5 %. The wild type enzyme did not exhibit any activity under the same conditions. Reaction time (h)
Tricaprylin
Tridecanoin
trimyristin
Triolein
Olive oil
0
0%
0%
0%
0%
0%
12
0.6%
5.1 %
4.2%
13.6%
18.6%
24
3.8%
6.5 %
7.2 %
18.0%
20.4%
36
6.8%
8.3%
9.7 %
21.3%
22.6 %
48
7.2%
8.8 %
11.1%
20.1%
22.7%
60
8.1%
9.5 %
12.6%
19.4
22.7%
72
8.2%
9.6 %
12.2%
19.5
22.8%
For a more detailed characterization of the interesting variant, the hydrolysis of 1,2-DAG and 1,3-DAG by Q282L variant was performed. As shown in Table 2, both 1,2-DAG and 1,3-DAG were completely hydrolyzed after 1 h by the wild-type enzyme, while only 9.8 % and 6.6 % free FAs were released by the Q282L variant, respectively. This underlines the shift of the substrate specificity and the much lower activity toward diacylglycerols by the Q282L variant.
Figure 2. The substrate specificities of SMG1 wild-type and mutant Q282L toward p-nitrophenyl esters of different acyl chain length (saturated). The activity of each enzyme for the preferred substrate was set as 100%, and all other values had been standardized to this reference value.
The thermal, medium and pH stability of the variant was also evaluated. Despite the fact that the melting temperature of the wild-type and the Q282L variant was similar (48°C and 49°C respectively), the residual activity of the variant was significantly lower after incubation at 45°C. As can be seen in Figure 3, the wild-type is quite stable for about 90 min, while the Q282L variant lost about 85 % of its activity in the first 30 min. It is possible that a minor conformation change derived from the single mutation was responsible for the rapid deactivation of Q282L. This finding was similar to that of a lipase from
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION Pseudomonas fragi that had similar apparent melting temperature with its mutant, but showed lower thermostability.[11]
measured by using pNP-butyrate as substrate at 25°C and pH 6.0 with a reaction time of 5 min.
The pH stability of Q282L and SMG1 were determined by preincubating enzymes in different pH buffers, ranging from pH 5.0 to pH 9.0, for 12 h at 4°C. The residual activities were measured at pH 6.0. Both enzymes are quite stable at pH between 7 and 9, retaining about 70 % of their initial activity after 12 h incubation (Figure 5). However, moving to lower pH, the stability decreases. The variant Q282L seems to be a little more susceptible to pH, as it has a lower stability compared to the wild-type at pH 6, but at lower pH values the decrease in activity is comparable. The pH stability profile of Q282L is similar to that of the lipase from Bacillus thermocatenulatus (BTL2). BTL2 is stable over a pH range from 7.0 to 11.0, but unstable at lower pH.[14]
Figure 3. Thermal stability of SMG1 wild-type (squares) and Q282L variant (triangles). Enzymes were incubated at 45°C, and samples were withdrawed in specific time intervals. Residual activity was measured using pNP-butyrate as substrate at 25°C and pH 6.0 with a reaction time of 5 min
As lipases are often used in organic solvents, the susceptibility of the Q282L variant to methanol, ethanol, acetonitrile and 2propanol was investigated. The activity of variant Q282L and SMG1 wild-type were measured after treatment with 30 % (v/v) of various water-miscible organic solvents, as these ones have a more prominent effect on the activity of the enzymes.[12] It needs to be clarified that the addition of the organic solvent did not change the pH of the mixture. As can be seen in Figure 4, methanol and ethanol have a more prominent effect compared to isopropanol, an observation that was also reported in literature.[13] The variant Q282L showed less tolerance against 2propanol than SMG1 wild-type, while it seems more robust towards acetonitrile, which has the most dramatic effect on the activity of both enzymes.
Figure 4. Effect of organic solvents on the stability of SMG1 wild type and Q282L mutant. Enzymes were incubated for 30 min at 37°C in 0.1 M phosphate buffer, pH 6.0, containing 30% (v/v) of various organic solvents. The solvent concentration in the assay was 0.2% (v/v) and the reaction mixture without organic solvent was used as control. The residual activity was
Figure 5. Effect of pH on the stability of SMG1-wild type and Q282L mutant. Lipases were incubated in buffers of various pH for 12h at 4°C. The relative activity (activity after incubation to the one prior to it) was measured by using pNP-butyrate as substrate at 25°C and pH 6.0 with a reaction time of 5 min.
From all aforementioned results, it seems that the Q282L mutation is critical as it not only converts the lipase to a triacylglyceride lipase, but also changes to some extent the catalytic behavior of the enzyme. In order to suggest a hypothesis on the effect of this mutation, we performed some bioinformatics analysis. The shift from glutamine to leucine does not make significant alterations in the steric hindrances, as the two amino acids do not differ significantly in bulkiness. However, their hydrophobicity is completely different; Glutamine is one of the most hydrophilic amino acids, while leucine is highly hydrophobic. As can be seen in Figure 6, one of the acyl chains of the triglyceride is in contact with the residue in position 282. We suggest that leucine in this position facilitates the hydrophobic interactions and allows the conversion of triglycerides. In the case of the wild-type enzyme, the hydrophilicity of glutamine deters the acyl group from this position, and thus the triglycerides cannot be accepted as substrates. It needs to be underlined that the position 282 is directly after the catalytic histidine. This could be the reason for the altered catalytic activity of the variant Q282L; the interactions of the residue with the substrate could induce minor alterations of the position of the histidine and thus result in significant changes in the catalytic efficiency. More than that, as can be seen in Figure 6, the glutamine can have a hydrogen bond with
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION the histidine, resulting in a more stable conformation. This hydrogen bond is formed from the side chain, and thus it is not present in the Q282L variant. This observation could explain also the lower stability of the variant at higher temperatures, as indicated in Figure 3.
[2]
[3] [4] [5] [6] [7] [8]
[9] [10] Figure 6. Tetrahedral intermediate of tridecanoate (cyan) and SMG1 variants (grey). They catalytic residues are shown in grey. Glutamine in position 282 (orange) is pointing towards one of the acyl chains, producing undesired repulsive interactions. The leucine in the same position (green) has the same bulkiness, but is able to establish hydrophobic interactions with the acyl chain.
[11] [12] [13] [14]
In conclusion, we could show that the transformation of the SMG1 lipase to a triglyceride lipase is possible with semirational design by only exchanging one residue. From the bioinformatics analysis it seems that the catalytic pocket can accommodate triglycerides, but the catalysis was hindered from disfavored interactions. This underlines the fact that the obvious (bulkiness) is not always the right answer, and that the network of residue interactions is critical for the understanding of the substrate specificity.
Experimental Section The full experimental section is available in the Supporting Information file.
Acknowledgements This work was supported by the National Science Funds for the Excellent Youth Scholars (31222043), National Natural Science Foundation of China (21406076), research fund for the Doctoral Program of Higher Education of China (20130172120014). Keywords: Protein engineering • mono- and diacylglycerol lipase • substrate specificity [1]
a) R.D. Schmid, R. Verger, Angew Chem. 1998, 110, 1694-1720; Angew. Chem. Int. Ed. 1998, 37, 1609-1633; b) K.E. Jaeger, T. Eggert, Curr. Opin. Biotechnol. 2002 13, 390-397; c) A. K. Singh, M. Mukhopadhyay, Appl. Biochem. Biotechnol. 2012, 166, 486–520.
a) T.P. Dihn, D. Carpenter, F.M. Leslie, T.F. Freund, I. Katona, S.L. Sensi, S. Kathuria, D. Piomelli, Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 10819–10824; b) T. Bisogno, F. Howell, G. Williams, A. Minassi, M. G. Cascio, A. Ligresti, I. Matias, A. Schiano-Moriello, P. Paul, E. J. Williams, U. Gangadharan, C. Hobbs, V. Di Marzo, P. Doherty, J. Cell Biol. 2003, 163, 463-468. D. K. Nomura, J. Z. Long, S. Niessen, H. S. Hoover, S. W. Ng, B. F. Cravatt, Cell. 2010, 140, 49-61. R. A. Kohnz, D. K. Nomura, Chem. Soc. Rev. 2014, 43, 6859-6869. W. Wang, T. Li, X. Qin, Z. Ning, B. Yang, Y. Wang, J. Mol. Catal. B Enzym. 2012, 77, 87–91. S. Hama, T. Numata, S. Tamalampudi, A. Yoshida, H. Noda, A. Kondo, H. Fukuda, J Mol Catal B-Enzym. 2009, 58, 93-97. J. J. Huang, Z. Yang, F. F. Guan, S. S. Zhang, D. Cui, G. H. Guan, Y. Li, Process Biochem. 2013, 48, 1899-1904. a) H.B. Brundiek, A.S. Evitt, R. Kourist, U.T. Bornscheuer, Angew. Chem. 2012, 124, 425-428; Angew. Chem. Int. Ed. 2012, 51, 412–414; b) D. Reyes-Duarte, J. Polaina, N. López-Cortés, M. Alcalde, F.J. Plou, K. Elborough, A. Ballesteros, K.N. Timmis, P.N. Golyshin, M. Ferrer, Angew. Chem. 2005, 117, 7725-7729; Angew. Chem. Int. Ed. 2005, 44, 7553–7557; c) J. Schmitt, S. Brocca, R.D. Schmid, J. Pleiss, Protein Eng. Des. Sel. 2002, 15, 595–601. T. Xu, L. Liu, S. Hou, J. Xu, B. Yang, Y. Wang, J. Liu, J. Struct. Biol. 2012, 178, 363–369. R. Kourist, H. Jochens, S. Bartsch, R. Kuipers, S. K. Padhi, M. Gall, D. Böttcher, H.-J. Joosten, U.T. Bornscheuer, ChemBioChem 2010, 11, 1635-1643. P. Gatti-Lafranconi, S. M. Caldarazzo, A. Villa, L. Alberghina, M. Lotti, FEBS Lett. 2008, 582, 2313-2318. P. P. Wangikar, P. C. Michels, D. S. Clark, J. S. Dordick. J. Am. Chem. Soc. 1997,119, 70–76. D. S. Dheeman, S. Antony-Babu, J. M. Frias, G. T. M. Henehan, J. Mol. Catal. B-Enzym. 2011, 72, 256-262. D. T. Quyen, C. Schmidt-Dannert, R. D. Schmid, Prot. Expr. Purif. 2003, 28, 102-110.
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION Entry for the Table of Contents
COMMUNICATION A mono- and diglyceride lipase was engineered to become a true triacylglyceride lipase by introducing only one single point mutation (Q282L). The variant has broad substrate specificity on triglycerides. The mutation indicates that the main reason that the wild-type enzyme does not accept triglycerides is not their bulkiness, but specific hydrophobic interactions.
Dongming Lan, Grzegorz Maria Popowicz, Ioannis V. Pavlidis, Pengfei Zhou, Uwe T. Bornscheuer, Yonghua Wang* Page No. – Page No. Conversion of a mono- and diacylglycerol lipase into a triacylglycerol lipase by protein engineering
Authors: Dongming Lan; Grzegorz Maria Popowicz; Ioannis V. Pavlidis; Pengfei Zhou; Uwe T. Bornscheuer; Yonghua Wang
This manuscript has been accepted after peer review and the authors have elected to post their Accepted Article online prior to editing, proofing, and formal publication of the final Version of Record (VoR). This work is currently citable by using the Digital Object Identifier (DOI) given below. The VoR will be published online in Early View as soon as possible and may be different to this Accepted Article as a result of editing. Readers should obtain the VoR from the journal website shown below when it is published to ensure accuracy of information. The authors are responsible for the content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201500163 Link to VoR: http://dx.doi.org/10.1002/cbic.201500163
A Journal of
www.chembiochem.org
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION Conversion of a mono- and diacylglycerol lipase into a triacylglycerol lipase by protein engineering Dongming Lan,[a] Grzegorz Maria Popowicz,[b] Ioannis V. Pavlidis,[c] Pengfei Zhou,[d] Uwe T. Bornscheuer[c] and Yonghua Wang*[a]
Abstract: Despite the fact that most lipases are expected to be active against triglycerides, there is a small group of lipases which are active only on mono- and diacylglycerides. The reason behind the different substrate scope is not clear so far. In the present work we tried to identify the reasons for this differentiation, using the lipase from Malassezia globosa as a template. By means of protein engineering and with implementation of only one mutation we managed to convert this enzyme to a typical triacylglycerol lipase, although the wild type does not accept triglycerides. The variant Q282L accepts a broad spectrum of triglycerides, while the catalytic behavior is altered to some extent. From the in silico analysis it seems that specific hydrophobic interactions are the key to the altered substrate specificity.
Lipases (EC 3.1.1.3) are one of the most studied enzyme groups, with numerous applications developed so far.[1] Their substrate scope varies, but it is generally accepted that most lipases accept triacylglycerides (TAG). However, there is a small group of lipases which are not able to hydrolyze triglycerides, but only mono- and diacylglycerides (MAG and DAG). These lipases have attracted significant attention in medical research and industrial applications due to their altered specificity. For example, MAG and DAG specific lipases are involved in controlling the neuronal communication by biosynthesis and inactivation of 2-arachidonoylglycerol. [2] MAG-lipases are found to promote cancer pathogenesis by regulating a fatty acid network enriched in oncogenic signaling lipids. [3] Their important physiological functions make them being valuable targets for drug development.[4] In industrial application, MAG and DAG specific lipases are excellent biocatalysts that have been applied to the synthesis of high-pure DAG, [5] biodiesel [6] and food emulsifier [7]. However, the reason behind this peculiar substrate
[a]
[b]
[c]
[b]
specificity of these enzymes is yet unclear. The understanding of the molecular basis of the substrate selectivity mechanism can provide an insight to rationally design novel biocatalysts with desired substrate specificity via protein engineering. There are several examples of successful implementation of protein engineering techniques for the change of the substrate scope of lipases; some of them from our group. [8] However, there are no works on the modification of acylglycerol specificity, to help on the elucidation of the differences that lead to the difference of the specificity we mentioned above. Recently, the X-ray structure of a lipase from Malassezia globosa (SMG1) which is active on DAGs but not on TAGs, has been resolved by our group (PDB entries: 3UUE and 3UUF).[9] The crystal structure of this lipase was used in the present study to guide the rational design, in order to alter the acylglycerol specificity. As seen in Figure 1, several residues in the catalytic pocket were targeted for mutagenesis. More specifically, the loop of residues 103 to 107 is considered to be the lid of the lipase,[4] thus several positions were targeted on this loop. More than that, the neighboring residues of the catalytic serine, namely H170 and L172, were targeted. They are both bulky residues and we considered to change the motif of the catalytic serine. According to our previous work,[10] SMG1 fits to superfamily II of lipases (GHSLG), but in order to shift to smaller residues, we considered to change the motif to GESAG, which is characteristic for the superfamily III. Last but not least, the positions Y56, F278 and Q282 were targeted, as they point towards to the substrate pocket, and thus potentially pose steric hindrances. A full list of the variants prepared for this work is available in the supporting information section (Table SI-1).
Dr. D.M. Lan, Prof. Dr. Y. H. Wang College of Light Industry and Food Sciences South China University of Technology Guangzhou 510640, PR China E-mail: [email protected] Dr. G. M. Popowicz Institute of Structural Biology, Helmholtz Zentrum München Deutsches Forschungszentrum für Gesundheit und Umwelt (GmbH) Ingolstädter Landstraße 1, Germany Dr. Ioannis V. Pavlidis, Prof. Dr. Uwe T. Bornscheuer Dept. of Biotechnology and Enzyme Catalysis, Institute of Biochemistry University of Greifswald Felix-Hausdorff-Str. 4, D-17487, Greifswald, Germany P.F. Zhou School of Bioscience and Bioengineering South China University of Technology Guangzhou 510640, PR China Supporting information for this article is given via a link at the end of the document.
Figure 1. Catalytic pocket of SMG1. In the centre the residues of the catalytic triad are depicted in grey, while the targeted residues for mutagenesis are depicted in cyan and the lid region is colored orange.
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION All SMG1 variants were prepared and expressed in Pichia pastoris and purified by immobilized metal affinity chromatography, taking advantage of the C-terminal His-Tag. The alteration of the selectivity scope was monitored in the esterification reaction of glycerol with fatty acids (FAs) varying in carbon length and degree of saturation (C8:0–C22:6) in a solvent free system at 30°C. Strikingly, the residues on the lid did not alter the selectivity; all of them produced DAGs, but no triglycerides were formed (Table SI-1). However, we were able to find two positions that seem to alter the selectivity. The variant Y56A did only produce monoglycerides, but this seems to be an effect of the low activity observed, as it is much less active than the wild type or all other variants. The most interesting variant was Q282L, which enabled SMG1 to produce triglycerides. Wildtype SMG1 and the variant Q282L have the same fatty acid (FA) selectivity during this competitive acylation reaction; both enzymes exhibit a preference for decanoic acid for the saturated FAs, but they are more efficient with unsaturated FAs (oleic acid C18:1 and linoleic acid C18:2; see Figure SI-1). As the single mutant Q282L is the only one active towards the synthesis of triglycerides, its hydrolytic activity was also evaluated, using several triglycerides. As can be seen in Table 1, Q282L was able to hydrolyze a broad scope of triglycerides, while the wild type enzyme did not exhibit any catalytic activity. The catalytic efficiency of this variant may seem low, leading to about 20 % conversion after three days, but this is to be expected for novel activities that the wild-type enzyme does not have. The variant seems to have a preference for longer acyl chains, such as in triolein (C18:1) or olive oil. It needs to be stated that olive oil is composed mostly of triglycerides of oleic acid and linoleic acid, which were shown to be preferred in the synthetic reaction (Figure SI-1). Based on these results, the variant Q282L can be considered a true TAG-lipase.
Table 2. Free fatty acid content after the hydrolysis of 1,2- and 1,3-DAGs by SMG1 wild-type and mutant Q282L. Standard deviation was less than 2.7 %. Reaction time (min)
1,2-dilaurin
1,3-dilaurin
Wild type
Q282L
Wild type
Q282L
0
0%
0%
0%
0%
15
83.1 %
5.9 %
55.8 %
1.8 %
30
94.6 %
6.5 %
94.27 %
5.6 %
60
99.9 %
9.8 %
99.9 %
6.6 %
The biochemical properties of Q282L mutant were investigated and compared to these of the wild-type enzyme. The Q282L mutant has the same optimal temperature (25°C) and pH (6.0) than the wild-type. To study the substrate selectivity of variant Q282L, the hydrolytic activity towards p-nitrophenyl (pNP) esters with variable acyl chain length (C4 to C16) was evaluated. As seen in Figure 2, the mutation Q282L led to a change of the substrate preference from medium (C6-C8) to shorter ones (C4). Interestingly, a significant improvement on the hydrolytic activity was observed. The specific activity of the Q282L variant was increased 14-fold and 4-fold compared to that of the wild-type enzyme for the C-4 and C-8 substrates, respectively (Table SI-2).
Table 1. Free fatty acid content of several substrates, after hydrolysis with SMG1 Q282L. Standard deviation is less than 2.5 %. The wild type enzyme did not exhibit any activity under the same conditions. Reaction time (h)
Tricaprylin
Tridecanoin
trimyristin
Triolein
Olive oil
0
0%
0%
0%
0%
0%
12
0.6%
5.1 %
4.2%
13.6%
18.6%
24
3.8%
6.5 %
7.2 %
18.0%
20.4%
36
6.8%
8.3%
9.7 %
21.3%
22.6 %
48
7.2%
8.8 %
11.1%
20.1%
22.7%
60
8.1%
9.5 %
12.6%
19.4
22.7%
72
8.2%
9.6 %
12.2%
19.5
22.8%
For a more detailed characterization of the interesting variant, the hydrolysis of 1,2-DAG and 1,3-DAG by Q282L variant was performed. As shown in Table 2, both 1,2-DAG and 1,3-DAG were completely hydrolyzed after 1 h by the wild-type enzyme, while only 9.8 % and 6.6 % free FAs were released by the Q282L variant, respectively. This underlines the shift of the substrate specificity and the much lower activity toward diacylglycerols by the Q282L variant.
Figure 2. The substrate specificities of SMG1 wild-type and mutant Q282L toward p-nitrophenyl esters of different acyl chain length (saturated). The activity of each enzyme for the preferred substrate was set as 100%, and all other values had been standardized to this reference value.
The thermal, medium and pH stability of the variant was also evaluated. Despite the fact that the melting temperature of the wild-type and the Q282L variant was similar (48°C and 49°C respectively), the residual activity of the variant was significantly lower after incubation at 45°C. As can be seen in Figure 3, the wild-type is quite stable for about 90 min, while the Q282L variant lost about 85 % of its activity in the first 30 min. It is possible that a minor conformation change derived from the single mutation was responsible for the rapid deactivation of Q282L. This finding was similar to that of a lipase from
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION Pseudomonas fragi that had similar apparent melting temperature with its mutant, but showed lower thermostability.[11]
measured by using pNP-butyrate as substrate at 25°C and pH 6.0 with a reaction time of 5 min.
The pH stability of Q282L and SMG1 were determined by preincubating enzymes in different pH buffers, ranging from pH 5.0 to pH 9.0, for 12 h at 4°C. The residual activities were measured at pH 6.0. Both enzymes are quite stable at pH between 7 and 9, retaining about 70 % of their initial activity after 12 h incubation (Figure 5). However, moving to lower pH, the stability decreases. The variant Q282L seems to be a little more susceptible to pH, as it has a lower stability compared to the wild-type at pH 6, but at lower pH values the decrease in activity is comparable. The pH stability profile of Q282L is similar to that of the lipase from Bacillus thermocatenulatus (BTL2). BTL2 is stable over a pH range from 7.0 to 11.0, but unstable at lower pH.[14]
Figure 3. Thermal stability of SMG1 wild-type (squares) and Q282L variant (triangles). Enzymes were incubated at 45°C, and samples were withdrawed in specific time intervals. Residual activity was measured using pNP-butyrate as substrate at 25°C and pH 6.0 with a reaction time of 5 min
As lipases are often used in organic solvents, the susceptibility of the Q282L variant to methanol, ethanol, acetonitrile and 2propanol was investigated. The activity of variant Q282L and SMG1 wild-type were measured after treatment with 30 % (v/v) of various water-miscible organic solvents, as these ones have a more prominent effect on the activity of the enzymes.[12] It needs to be clarified that the addition of the organic solvent did not change the pH of the mixture. As can be seen in Figure 4, methanol and ethanol have a more prominent effect compared to isopropanol, an observation that was also reported in literature.[13] The variant Q282L showed less tolerance against 2propanol than SMG1 wild-type, while it seems more robust towards acetonitrile, which has the most dramatic effect on the activity of both enzymes.
Figure 4. Effect of organic solvents on the stability of SMG1 wild type and Q282L mutant. Enzymes were incubated for 30 min at 37°C in 0.1 M phosphate buffer, pH 6.0, containing 30% (v/v) of various organic solvents. The solvent concentration in the assay was 0.2% (v/v) and the reaction mixture without organic solvent was used as control. The residual activity was
Figure 5. Effect of pH on the stability of SMG1-wild type and Q282L mutant. Lipases were incubated in buffers of various pH for 12h at 4°C. The relative activity (activity after incubation to the one prior to it) was measured by using pNP-butyrate as substrate at 25°C and pH 6.0 with a reaction time of 5 min.
From all aforementioned results, it seems that the Q282L mutation is critical as it not only converts the lipase to a triacylglyceride lipase, but also changes to some extent the catalytic behavior of the enzyme. In order to suggest a hypothesis on the effect of this mutation, we performed some bioinformatics analysis. The shift from glutamine to leucine does not make significant alterations in the steric hindrances, as the two amino acids do not differ significantly in bulkiness. However, their hydrophobicity is completely different; Glutamine is one of the most hydrophilic amino acids, while leucine is highly hydrophobic. As can be seen in Figure 6, one of the acyl chains of the triglyceride is in contact with the residue in position 282. We suggest that leucine in this position facilitates the hydrophobic interactions and allows the conversion of triglycerides. In the case of the wild-type enzyme, the hydrophilicity of glutamine deters the acyl group from this position, and thus the triglycerides cannot be accepted as substrates. It needs to be underlined that the position 282 is directly after the catalytic histidine. This could be the reason for the altered catalytic activity of the variant Q282L; the interactions of the residue with the substrate could induce minor alterations of the position of the histidine and thus result in significant changes in the catalytic efficiency. More than that, as can be seen in Figure 6, the glutamine can have a hydrogen bond with
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION the histidine, resulting in a more stable conformation. This hydrogen bond is formed from the side chain, and thus it is not present in the Q282L variant. This observation could explain also the lower stability of the variant at higher temperatures, as indicated in Figure 3.
[2]
[3] [4] [5] [6] [7] [8]
[9] [10] Figure 6. Tetrahedral intermediate of tridecanoate (cyan) and SMG1 variants (grey). They catalytic residues are shown in grey. Glutamine in position 282 (orange) is pointing towards one of the acyl chains, producing undesired repulsive interactions. The leucine in the same position (green) has the same bulkiness, but is able to establish hydrophobic interactions with the acyl chain.
[11] [12] [13] [14]
In conclusion, we could show that the transformation of the SMG1 lipase to a triglyceride lipase is possible with semirational design by only exchanging one residue. From the bioinformatics analysis it seems that the catalytic pocket can accommodate triglycerides, but the catalysis was hindered from disfavored interactions. This underlines the fact that the obvious (bulkiness) is not always the right answer, and that the network of residue interactions is critical for the understanding of the substrate specificity.
Experimental Section The full experimental section is available in the Supporting Information file.
Acknowledgements This work was supported by the National Science Funds for the Excellent Youth Scholars (31222043), National Natural Science Foundation of China (21406076), research fund for the Doctoral Program of Higher Education of China (20130172120014). Keywords: Protein engineering • mono- and diacylglycerol lipase • substrate specificity [1]
a) R.D. Schmid, R. Verger, Angew Chem. 1998, 110, 1694-1720; Angew. Chem. Int. Ed. 1998, 37, 1609-1633; b) K.E. Jaeger, T. Eggert, Curr. Opin. Biotechnol. 2002 13, 390-397; c) A. K. Singh, M. Mukhopadhyay, Appl. Biochem. Biotechnol. 2012, 166, 486–520.
a) T.P. Dihn, D. Carpenter, F.M. Leslie, T.F. Freund, I. Katona, S.L. Sensi, S. Kathuria, D. Piomelli, Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 10819–10824; b) T. Bisogno, F. Howell, G. Williams, A. Minassi, M. G. Cascio, A. Ligresti, I. Matias, A. Schiano-Moriello, P. Paul, E. J. Williams, U. Gangadharan, C. Hobbs, V. Di Marzo, P. Doherty, J. Cell Biol. 2003, 163, 463-468. D. K. Nomura, J. Z. Long, S. Niessen, H. S. Hoover, S. W. Ng, B. F. Cravatt, Cell. 2010, 140, 49-61. R. A. Kohnz, D. K. Nomura, Chem. Soc. Rev. 2014, 43, 6859-6869. W. Wang, T. Li, X. Qin, Z. Ning, B. Yang, Y. Wang, J. Mol. Catal. B Enzym. 2012, 77, 87–91. S. Hama, T. Numata, S. Tamalampudi, A. Yoshida, H. Noda, A. Kondo, H. Fukuda, J Mol Catal B-Enzym. 2009, 58, 93-97. J. J. Huang, Z. Yang, F. F. Guan, S. S. Zhang, D. Cui, G. H. Guan, Y. Li, Process Biochem. 2013, 48, 1899-1904. a) H.B. Brundiek, A.S. Evitt, R. Kourist, U.T. Bornscheuer, Angew. Chem. 2012, 124, 425-428; Angew. Chem. Int. Ed. 2012, 51, 412–414; b) D. Reyes-Duarte, J. Polaina, N. López-Cortés, M. Alcalde, F.J. Plou, K. Elborough, A. Ballesteros, K.N. Timmis, P.N. Golyshin, M. Ferrer, Angew. Chem. 2005, 117, 7725-7729; Angew. Chem. Int. Ed. 2005, 44, 7553–7557; c) J. Schmitt, S. Brocca, R.D. Schmid, J. Pleiss, Protein Eng. Des. Sel. 2002, 15, 595–601. T. Xu, L. Liu, S. Hou, J. Xu, B. Yang, Y. Wang, J. Liu, J. Struct. Biol. 2012, 178, 363–369. R. Kourist, H. Jochens, S. Bartsch, R. Kuipers, S. K. Padhi, M. Gall, D. Böttcher, H.-J. Joosten, U.T. Bornscheuer, ChemBioChem 2010, 11, 1635-1643. P. Gatti-Lafranconi, S. M. Caldarazzo, A. Villa, L. Alberghina, M. Lotti, FEBS Lett. 2008, 582, 2313-2318. P. P. Wangikar, P. C. Michels, D. S. Clark, J. S. Dordick. J. Am. Chem. Soc. 1997,119, 70–76. D. S. Dheeman, S. Antony-Babu, J. M. Frias, G. T. M. Henehan, J. Mol. Catal. B-Enzym. 2011, 72, 256-262. D. T. Quyen, C. Schmidt-Dannert, R. D. Schmid, Prot. Expr. Purif. 2003, 28, 102-110.
ChemBioChem
10.1002/cbic.201500163
COMMUNICATION Entry for the Table of Contents
COMMUNICATION A mono- and diglyceride lipase was engineered to become a true triacylglyceride lipase by introducing only one single point mutation (Q282L). The variant has broad substrate specificity on triglycerides. The mutation indicates that the main reason that the wild-type enzyme does not accept triglycerides is not their bulkiness, but specific hydrophobic interactions.
Dongming Lan, Grzegorz Maria Popowicz, Ioannis V. Pavlidis, Pengfei Zhou, Uwe T. Bornscheuer, Yonghua Wang* Page No. – Page No. Conversion of a mono- and diacylglycerol lipase into a triacylglycerol lipase by protein engineering
