Previous Article | Next Article ![]()
Applied and Environmental Microbiology, May 2002, p. 2404-2410, Vol. 68, No. 5
0099-2240/02/$04.00+0 DOI: 10.1128/AEM.68.5.2404-2410.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Department of Biochemistry and Cell Biology, Rice University, Houston, Texas 77251
Received 11 December 2001/ Accepted 26 February 2002
|
|
|---|
|
|
|---|
![]() View larger version (20K): [in a new window] |
FIG. 1. Structures of streptamine analogs examined as possible methyl acceptors. I, streptamine; II, 2-deoxystreptamine; III, 2-epi-streptamine; IV, N3-methyl-2-deoxy-D-streptamine; V, N1-methyl-2-deoxy-D-streptamine; VI, N1-ethyl-2-deoxy-D-streptamine; VII, N1-methyl-2-epi-D-streptamine; VIII, N3-methyl-2-epi-D-streptamine; IX, N1,N3-dimethyl-2-epi-streptamine. Vertical lines represent hydroxyl groups.
|
![]() View larger version (32K): [in a new window] |
FIG. 2. Enzymatic reactions previously discovered by my laboratory to be involved in biosynthesis of the streptidine core of streptomycin (13, 16, 21) and the bluensidine core of bluensomycin (14, 17). Enzymes U and V are absent in streptomycin producers, and enzymes H, I, and J are absent in bluensomycin producers (17). Note that in streptomycin producers two analogous series of reactions, in the same order, convert an -OH group to a guanidino group, with intriguing evolutionary implications (14, 17). Vertical lines represent hydroxyl groups. Enzyme D, L-glutamine:aminocyclitol aminotransferase, is the only enzyme common to the biosyntheses of all major aminoglycoside antibiotics (1, 9, 16, 18, 20, 24). KGN is 2-ketoglutaramate. N1-Amidino-D-streptamine (IGN) is the aminocyclitol core of one of the destomycin antibiotics (3), also with interesting evolutionary implications. Enzyme F, amidinotransferase, is an easily assayed (19, 24) generic biosynthetic marker for streptomycin producers (2). Enzyme W also phosphorylates streptomycin (16).
|
|
|
|---|
Growth of cultures and preparation of extracts.
Stock cultures were grown at room temperature on slants of yeast-malt extract agar (Difco catalog no. 0770-01). After 10 to 13 days aliquots were transferred to 2-liter Erlenmeyer flasks containing 500 ml of 0.05% glucose-1% myo-inositol-1% Soytone-0.2% yeast extract and grown at 26°C on a New Brunswick rotary shaker (240 rpm) to early stationary phase (19), approximately 42 to 51 h. Mycelia were harvested by suction filtration or centrifugation. Washed mycelial pads were stored at -20°C and then sonicated with 2 volumes of cold water and centrifuged at 4°C for 20 min at 20,000 x g, and the supernatants were stored at -20°C. Each extract was tested for biosynthetic competence by activity of glutamine:aminocyclitol aminotransferase (18).
Assay for N-methyltransferase activity (Tables 1
to 4).
The assay mixtures employed were as follows: Ado[14C]Met, 126,000 cpm/5 µl, 10 µl; 0.1 M K phosphate, pH 7.4, or 0.2 M NaHCO3, 5 µl; streptamine analog, 1 mg/ml, 5 µl; H2O, 5 µl; and sonicate supernatant, 10 µl. The mixture was incubated at 34°C for 2 h. Tris buffers, pH 7.4 and 8.5, were also employed in preliminary experiments. The pH 7.4 phosphate buffer was used for streptamine and 2-epi-streptamine, while the NaHCO3 buffer was used for 2-deoxystreptamine and methylated streptamine analogs. After incubation, the N-[14C]methylstreptamine derivatives, with a 2+ charge at pH 3.6, were separated from reactants and other labeled products by high-voltage paper electrophoresis at pH 3.6 (ammonium formate, 0.2 ionic strength) with a Savant horizontal plate apparatus (30 V/cm) and Whatman no. 1 filter paper matrix, with a 46-cm path marked in 1-cm intervals. Picric acid was the visible reference compound during electrophoresis (15). After drying, 1-cm segments were cut, placed in scintillation vials containing 5 ml of Liquifluor (Fisher), and counted with a Beckman LS 3801 liquid scintillation counter. Compounds with 2+ charges were located by comparison with N1-[14C]amidino-D-streptamine (IGN, Fig. 2). Up to six assays can be run in parallel.
|
View this table: [in a new window] |
TABLE 1. Methylation of 2-deoxystreptamine derivatives by extracts of the hygromycin B producer S. hygroscopicus with Ado[14C]Met as methyl donord
|
|
View this table: [in a new window] |
TABLE 2. Methylation of other streptamine analogs by extracts of the hygromycin B producer, with Ado[14C]Met as methyl donora
|
|
View this table: [in a new window] |
TABLE 3. Methylation of 2-epi-streptamine (III) by extracts of the spectinomycin producer S. flavopersicus with Ado[14C]Met as methyl donora
|
|
View this table: [in a new window] |
TABLE 4. Methylation of other streptamine analogs by extracts of the spectinomycin producer with Ado[14C]Met as methyl donora
|
Assay of N-[14C]methylstreptamine analogs as amino donors for glutamine:aminocyclitol aminotransferase by a pseudo-half-reaction with scyllo-inosose as amino acceptor (Table 5).
Assay mixtures were as follows: N-[14C]methyl-2-deoxy-D-streptamine product of S. hygroscopicus N-methyltransferase (88,000 cpm/5 µl) or other N-[14C]methylstreptamine analog, 10 µl; 0.1 M K phosphate buffer, pH 7.4, plus neutralized pyridoxal-P, 2.5 mg/ml, 5 µl; scyllo-inosose, 10 mg/ml, 10 µl; and S. hygroscopicus sonicate supernatant, 10 µl. The mixture was incubated at 34°C for 2 h. Labeled reactants and products were separated by high-voltage paper electrophoresis at pH 3.6 as before. The labeled amino donor with a 2+ charge was converted to a labeled keto product with a 1+ charge after its transamination. The fraction of label transferred was 50%.
|
View this table: [in a new window] |
TABLE 5. Abilities of the N-[14C]methyl streptamine analogs formed by N-methyltransferases from the hygromycin B and spectinomycin producers to serve as amino donors to scyllo-inosose in pseudo-half-reactions catalyzed by glutamine:aminocyclitol aminotransferase from the hygromycin B producera
|
|
View this table: [in a new window] |
TABLE 6. Abilities of N-[14C]methylstreptamine analogs formed by N-methyltransferases from producers of hygromycin B and spectinomycin to serve as phosphate acceptors, catalyzed by a streptamine kinase activity in extracts of the streptomycin producer S. bikiniensisa
|
|
View this table: [in a new window] |
TABLE 7. Use of a coupled streptamine kinase-arginine:streptamine-P amidinotransferase system to distinguish enantiomers of N-monoalkylated streptamine analogsa
|
|
|
|---|
The question next arose as to whether the N1- or N3-amino group was preferentially methylated. The results are shown in experiment 2 of Table 1, employing monoalkylated 2-deoxystreptamine derivatives of known structures as potential methyl acceptors. Whereas N3-methyl-2-deoxy-D-streptamine (IV) is a weak acceptor, N1-ethyl-2-deoxy-D-streptamine (VI) is as active an acceptor as is 2-deoxystreptamine. It can therefore be concluded that, since reaction 1 occurs, reaction 2 also occurs. Reactions 1 and 2 are as follows:
![]() | (1) |
![]() | (2) |
Reaction 2 represents the final step in biosynthesis of the aminocyclitol core (IV) of hygromycin B.
Abilities of other aminocyclitols to undergo methylation by extracts of a hygromycin B producer.
The results of experiment 1 of Table 2 show that streptamine (I) can be readily methylated in vitro, despite the fact that N-methylated streptamine cores have not yet been detected among the isolated destomycins (10). scyllo-Inosamine (IN, Fig. 2), even at 50 times the concentration of streptamine, did not inhibit methylation of streptamine. Surprisingly, 2-epi-streptamine (III) is an excellent methyl acceptor, as good as 2-deoxystreptamine is. However, a mixture of N1- and N3-methyl-2-epi-D-streptamine enantiomers (VII and VIII) was not significantly methylated. It is concluded that, since reaction 1 occurs, both reaction 3 and reaction 4 can also be catalyzed in good yields by this hygromycin B producer. Reactions 3 and 4 are as follows:
![]() | (3) |
![]() | (4) |
Search for enzymes that can methylate 2-epi-streptamine (III) in extracts of a spectinomycin producer.
Biosynthetically competent extracts of S. flavopersicus, which could convert myo-[14C]inositol (mI, Fig. 2) plus NAD+, oxalacetate, and L-glutamine or N3-methyl-2-deoxy-D-streptamine (IV) to scyllo-[14C]inosamine (IN, Fig. 2) (18), were tested for their ability to methylate 2-epi-streptamine (III). The results are shown in experiment 1 of Table 3. A remarkable 51% of the donor methyl label was transferred to 2-epi-streptamine. When a mixture of N1- and N3-methyl-2-epi-D-streptamine enantiomers (VII and VIII) was the methyl acceptor, 27% of the methyl label was transferred. When the N1,N3-dimethyl-2-epi-streptamine (IX) core of spectinomycin was tested, no labeled methyl group was transferred, thereby ruling out any O-methylation.
Abilities of other aminocyclitol derivatives to be methylated by extracts of the spectinomycin producer.
Experiment 1 of Table 4 shows that streptamine is a very good methyl acceptor. Most importantly for our third goal and contrary to our expectations (see L. Slechta and J. H. Coats, Abstr. 14th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 294, 1974), 2 deoxystreptamine (II), N1-ethyl-2-deoxy-D-streptamine (VI), and N3-methyl-2-deoxy-D-streptamine (IV) were all excellent methyl acceptors as well.
From the results of Tables 3 and 4 it can be inferred that the following reactions can be catalyzed by extracts of this spectinomycin producer.
![]() | (5) |
![]() | (6) |
![]() | (7) |
![]() | (8) |
![]() | (3) |
![]() | (9) |
![]() | (10) |
![]() | (11) |
![]() | (2) |
![]() | (12) |
![]() | (1) |
Reactions 5 to 8 represent the final steps in the biosynthesis of the aminocyclitol core (IX) of spectinomycin. Although it is beyond the scope of this study to identify the gene or genes that code for these N-methyltransferase activities, it is suggested that a comparison of the results of purifications based on the assays represented by reactions 1 and 12 should resolve the question of whether one or two different enzymes, or isozymes, are involved. The scyllo configurations and symmetry of streptamine and 2-deoxystreptamine are such that it is possible that flip-flopping the rings could permit one enzyme to add both methyl groups.
Abilities of N-[14C]methylstreptamine analogs to serve as surrogate substrates for L-glutamine:aminocyclitol aminotransferase.
The N-[14C]methyl derivatives of 2-deoxystreptamine, streptamine, and 2-epi-streptamine produced in 100-fold scaled-up incubations with each of the two N-methyltransferase preparations were separated from unreacted methyl donor and other labeled products by cation-exchange column chromatography. The isolated labeled streptamine analogs were then tested for their ability to serve as amino donors in pseudo-half-reactions catalyzed by glutamine:aminocyclitol aminotransferases in the same mycelial extracts. The results are shown in Table 5, with extracts of the hygromycin B producer as the source of aminotransferase. Similar results were obtained when extracts of the spectinomycin producer were the source of aminotransferase, with yields of 19 to 24%.
All six labeled aminocyclitol isolates contained high levels of active amino donors in this assay, which also provided the first evidence for this aminocyclitol aminotransferase in the destomycin family of antibiotics. Since we have previously shown that nonlabeled N3-methyl-2-deoxy-D-streptamine (IV) is an excellent amino donor for aminocyclitol aminotransferases in producers of streptomycin (18), spectinomycin (18), and bluensomycin (18), as well as producers of the 2-deoxystreptamine antibiotics gentamicin and neomycin (6, 7, 16), the results of Table 5 can be represented by reactions 13 to 15.
![]() | (13) |
![]() | (14) |
![]() | (15) |
Abilities of N-[14C]methylstreptamine analogs to be phosphorylated by a D-streptamine 6-kinase in extracts of a streptomycin producer.
Streptomycin and bluensomycin producers have been found to possess a total of four different aminocyclitol kinase activities (enzymes E, J, V, and W, Fig. 2). In addition there is a kinase that phosphorylates streptamine and 2-deoxystreptamine at position 6 (13, 23, 25) that is different from enzyme W (Fig. 2) (20) but might be the same as enzyme E (Fig. 2). Kinases E, J, and V are each different enzymes (17). It was of interest to determine whether any of the six N-[14C]methylstreptamine analogs used for the above transaminations might provide an easily obtained substrate for assays of D-streptamine 6-kinase activity. The results are shown in Table 6. At these low surrogate substrate concentrations the only reaction with a high-enough kcat/Km value was that shown as reaction 16.
![]() | (16) |
This relative specificity raised the question of whether at higher substrate concentrations other N1-alkylated D-streptamine analogs with lower kcat/Km values could be phosphorylated by this kinase (21). These higher concentrations would require use of nonlabeled streptamine derivatives and a coupled D-streptamine 6-kinase-[guanidino-14C]arginine:D-streptamine-6-P amidinotransferase assay to detect the products (13, 21). The results of such experiments are shown in Table 7. Whereas N3-methyl-2-deoxy-D-streptamine (IV) was inactive in this assay, N1-ethyl-2-deoxy-D-streptamine (VI) and the N1-methyl-2-epi-D-streptamine (VII) enantiomer of the DL-N-methyl-2-epi-streptamine racemic mixture (VII + VIII) were indeed sequentially phosphorylated and transamidinated (enzyme F, Fig. 2), as depicted in reactions 17 to 22.
![]() | (17) |
![]() | (18) |
![]() | (19) |
![]() | (20) |
![]() | (21) |
![]() | (22) |
These results confirm our earlier conclusions that this kinase phosphorylates the hydroxyl group para to a free amino group, which can then be amidinated by arginine:aminocyclitol-P amidinotransferase (enzyme F, Fig. 2) (21, 23-25). This ability to distinguish between N3- and N1-methyl-2-deoxy-D-streptamine enantiomers (IV and V) and separate them would have also been useful during the controversies concerning the respective structures of the aminocyclitol cores of hygromycin B and destomycin A (5). (Indeed, the 12th edition of the Merck Index [1996] still had the wrong structure for destomycin A.)
Some other potential uses for streptamine N-methyltransferases.
Streptamine and its analogs have properties compatible with possible roles as proprietary nutrient reserves (17), osmolytes, and/or extended-range pH buffers (pKas of 7.0 and 9.0; see SigmaUltra buffer no. 9410; Sigma Chemical Co.) during some stages in the feast-or-famine life cycles of streptomycetes in soils. Streptamine N-methyltransferases can provide the first specific assays for evaluation of such postulated roles.
In the biosynthesis of aminoglycoside antibiotics the next step following synthesis of the aminocyclitol cores usually involves an NDP-sugar:streptamine glycosyltransferase. The structure of kanamycin A (10) suggests that glucose itself is the first substituent on its 2-deoxystreptamine core, and mutasynthesis experiments have shown that N-methyl-2-deoxystreptamines can be incorporated into a number of antibiotics (4, 9). Reaction of a biosynthetically competent (aminotransferase-positive) extract of a kanamycin producer with the N-[14C]methyl-2-deoxy-D-streptamine products of spectinomycin N-methyltransferase(s) plus an NDP-glucose should produce the predicted labeled conjugate. The labeled conjugate in turn could readily serve as a substrate for the detection of later biosynthetic reactions. It remains to be seen whether insertion of a 2-deoxystreptamine N1-methyltransferase gene into producers of other 2-deoxystreptamine antibiotics would result in antibiotics useful against certain resistant organisms.
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»