The application of L‐glutaminase for the synthesis of the immunomodulatory γ‐D‐glutamyl‐L‐tryptophan and the kokumi‐imparting γ‐D‐glutamyl peptides

Abstract Glutaminase of Bacillus amyloliquefaciens has been used to synthesize the immunomodulatory γ‐D‐glutamyl‐L‐tryptophan (γ‐D‐Glu‐L‐Trp) and the kokumi‐active γ‐D‐glutamyl peptides. The optimum yield of γ‐D‐Glu‐L‐Trp was 55.76 mM in corresponding to a minimum yield of by‐product (γ‐D‐Glu‐γ‐D‐Glu‐L‐Trp) in the presence of 75 mM D‐Gln and 100 mM L‐Trp. The glutaminase has a low Km values for the donors (D‐Gln and L‐Gln:5.53 and 0.98 mM), but high ones for the acceptors (L‐Trp, L‐Phe, L‐Met, L‐Val and γ‐[D‐Glu]( n =1,2,3)‐L‐Val/L‐Phe/L‐Met, ranging from 32.51 to 193.05 mM). The highest Km value appearing when n = 2 (γ‐[D‐Glu]( n =0,1,2)‐L‐Val/L‐Phe/L‐Met) suggested the rising difficulty for synthesis when the number of donor increases in the reaction mixtures. The γ‐[D‐Glu]( n =1,2,3)‐L‐Val/L‐Phe/L‐Met at 5 mM can impart the blank chicken broth an enhancing monthfulness, thickness, and umaminess taste.

Although the production of by-products would be reduced with the D-Gln as γ-glutamyl donor, the practical application of D-Gln for the synthesis of γ-D-glutamyl peptides is rare except for γ-D-Glu-L-Trp (SCV-07) which was mainly synthesized using GGT as the catalyst in the presence of D-Gln and L-Trp (Saini, Bindal, & Gupta, 2017;Suzuki, Kato, & Kumagai, 2004). SCV-07, as an immunomodulator peptide, is often used as a therapeutic to attenuate the severity and duration of both acute and fractionated radiation-induced oral mucositis in a clinical trial of head and neck cancer patients (Adkins et al., 2010;Alterovitz, Tuthill, Rios, Modelska, & Sonis, 2011;Watkins, Pouliot, Fey, Tuthill, & Sonis, 2010); inhibit growth of a variety of tumor cell line types, including in leukemia, lymphoma, herpes simplex virus type 2, head and neck cancer xenograft models (Papkoff et al., 2011;Tuthill, Paploff, Watkins, & Sonis, 2011, Ii et al. 2008, as well as stimulate the differentiation of T-lymphocyte and the expression of Thy1-antigen on bone marrow cells, and enhance the production of interleukin 2 and interferon in mice (Simavirtsev et al., 2003).
The biggest characteristic of the transpeptidation by this glutaminase-catalysis is the production of multiple by-products like

γ-[L-Glu]2-Val/Phe/Met/Tyr, γ-[L-Glu]3-Val/Phe/Met/Tyr and γ-[L-
Glu]4-Val/Phe/Met/Tyr. Although these by-products were coined as kokumi-imparting peptides, the yield of the "so called" product, γ-glutamyl dipeptides, was seriously restricted. Whether D-Gln can be as a γ-glutamyl donor to reduce the production of by-products, as well as for the synthesis of γ-D-Glu-L-Trp is worth trying. Therefore, the glutaminase form Bacillus amyloliquefaciens was used as a catalysis to synthesize γ-D-Glu-L-Trp and various γ-D-glutamyl peptides.

| Enzymatic synthesis of γ-D-glutamyl peptides
: Synthesis of γ-D-Glu-L-Trp: a reaction mixture that contained D-Gln (100 mM, pH 10.0), L-Trp (100 mM, pH 10.0), and glutaminase (0.05 U/ml) was placed in a water bath shaking table for incubating for 8 hr at 37 ºC, and then the mixture was inactivate enzyme in a boiling water bath at 90 ºC for 15 min. The optimum ratio of substrate was also analyzed including the various D-Gln (75, 100, 150 mM) and a fixed L-Trp (100 mM). The isolation of γ-D-Glu-Dipeptides was carried in accordance with the method of Suzuki with some modification Suzuki et al., 2004). A 50 ml reaction mixture was applied to a column column, referring to Yang's methods Yang, Sun-Waterhous, Xie, et al., 2018;Yang et al., 2017). The substrate specificity of the enzyme with different acceptors were measured as described previously with some modification (Shuai, Zhang, Mu, & Jiang, 2011). Briefly, an enzymatic reaction condition was as follows: all the γ-GpNA, γ-glutamyl acceptors and enzyme in the borate-NaOH (50 mM, pH 10.0) buffer solution, then adding 0.1 M HCl to terminate the reaction after incubation at 37 ºC for 30 min. The content of p-nitroaniline was determined by a UV spectrophotometer (UV765, Shanghai, China) at 410 nm. The release of 1 μmol of p-nitroaniline per min from γ-GpNA was defined as the amount of one unit (U) of enzyme through the transpeptidation reaction.

mM of these synthetic γ-D-glutamyl peptides (γ-[L-Glu]n-L-
Phe/L-Met/L-Val) were added into a blank chicken broth to evaluate the flavor characteristic. Taste profiles mainly included mouthfulness, thickness and umaminess. Thereinto, mouthfulness describes an overall perception associated with food texture, structure and morphological complexity, and thickness which was measured as the increased taste intensity evaluated 10 s after food tasting describes rich complexity. The panels were consisted of 10 males and 7 females (age 30-40) from a professional company. 10 ml of solutions was given to these panelist individually and then put into the mouth for 15 s before swallowing, finally recorded the taste experienced in 25 s. Panelists were asked to rate the intensity of a given taste on a scale from 0 (not detectable) to 5 (strongly detectable).

| Statistical analysis
Statistical analysis was performed in triplicate and analyzed by Microsoft Excel 2000. Duncan test was used to determine significant differences at p < .05 between samples using SPSS 16.0 statistical software.

| The enzymatic synthesis of γ-D-glutamyl peptides γ-D-Glu-L-Trp.
Three main intense chromatographic peaks were observed, the most intense peak corresponded to γ-D-Glu-L-Trp ( GGT-catalyzing (Suzuki et al., 2004). Meanwhile, the by-product, γ-D-Glu-γ-D-Glu-L-Trp, was about 1 mM yield during the first 2 hours of incubation, and then dramatically increased to 7.72 mM during the next 3 hr (from the 2nd to 5th h), finally maintained stability (p > .05).

The observation illustrates there was a time difference between obtaining the maximum of γ-D-Glu-L-Trp (3 hr of incubation) and γ-D-
Glu-γ-D-Glu-L-Trp (5 hr (Table 1). Their fragments mainly contained [M-AA(Acceptor)-COOH] + (m/z = 84, b-CO 2 ion), [AA(Acceptor)-COOH] + (y-CO 2 type ion), [AA(Acceptor)-NH 3 ] + (y-CO 2 type ion), [AA(Acceptor)+H] + (y type ion) in each of the mass spectrum. Their yield was 30.23 ± 4.18, 23.12 ± 3.97, 17.19 ± 2.04 mM, respectively (Table 2), accounting for 117.90%, 115.43%, and 120.63% of their L-configurations in the same reaction condition (L-configurations' synthesis have been reported by Yang). The result indicated the yield of γ-glutamyl dipeptides slightly increased with D-Gln as a γ-glutamyl donor instead of L-Gln in γ-glutamyl peptides synthesis, which was in line with the reports (Suzuki et al., 2002(Suzuki et al., , 2003. 24%~74.96% of that of γ-L-glutamyl oligopeptides. These results indicated the by-products would be still synthesized with a decreasing number of by-products and the production of by-products. In other words, unlike GGT, glutaminase can catalyze γ-D-glutamyl di-/tri-peptide as the acceptor to synthesize the short-chain γ-D-glutamyl oligopeptides (Suzuki et al., 2002(Suzuki et al., , 2003. In addition, the γ-D-Glu-D-Gln and other poly-D-glutamylated species were not detected, indicating D-Gln can not be an acceptor to participate in a γ-glutamyl transfer reaction by glutaminase-catalyzing.
According to the multiple products and the catalytic properties of glutaminase, the synthetic route of the products is the synthetic γ-D-[Glu] (n-1) -L-AA as an acceptor to participate in the γ-glutamyl transfer reaction to form the by-products (γ-D-[Glu] n -L-AA) Yang, Sun-Waterhous, Xie, et al., 2018;Yang et al., 2017). However, the number of the products and their content with the D-Gln as the acceptor was less than that of L-Gln in the same reaction condition. It seems to indicate that γ-D-[Glu] n -L-AA has a low affinity for this enzyme. Therefore, the affinity of these γ-D-[Glu] n -L-AA to this enzyme and the differences of the affinity of the L-and D-configurations to this enzyme has been determined.    (Saini et al., 2017), but be in close proximity to that of GGT of Bacillus subtilis SK11.004 (Shuai et al., 2011). The K m values for the acceptors were range from 32.51 to 193.05, whose magnitude is in accordance with that of GGT of Bacillus subtilis SK11.004 (Shuai et al., 2011). In addition, the substrate specificity of the glutaminase for the transpeptidase activity was also determined and the result revealed that various amino acids could serve as γ-glutamyl acceptors with the D-Gln as the γ-glutamyl donor (Table 4). Besides Gly-Gly con-  Note: Transferase activity was measured by a spectrophotometric method as described in Materials and Methods. Activity is expressed relative to that found with 20 mM Gly-Gly (100%).  (Suzuki et al., 2003).

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing.