Effective complementarily-addressed photomodification of nucleic acids by oligonucleotide derivatives, containing aromatic azido groups]

Highly effective site-specific photomodification of a DNA-target was carried out with oligonucleotide reagents carrying aromatic azido groups. Oligonucleotide derivatives with a photoactive function R on the 5'-terminal phosphate and at C-5 atom of deoxyuridine were synthesized: R1NH(CH2)3NHpd(TCCACTT) and d(ULNHRCCACTT), where R1 is p-azidotetrafluorobenzoyl, R2 is 2-nitro, 5-azidobenzoyl, R3 is p-azidobenzoyl; LNH = -CH2NH-, -CH2OCH2CH2NH- or -CH2NHCOCH2CH2NH-. The prepared compounds form stable complementary complexes and effect site-specific photomodification of the target DNA. The modification of pentadecanucleotide d(TAAGTGGAGTTTGGC) with the reagents was investigated. Maximum extent of modification strongly depended on the reagent's type, the photoreagent with R1 being the most effective. Whatever the binding site was, this agent provided a 65-70% modification in all cases except LNH = -CH2NH-, when the yield was twice lower. For the reagents bearing R1 the modification sites were identified. Selective modification at the G9 residue was detected in the case of LNH = -CH2OCH2CH2NH- and when a photoactive group was linked to the terminal phosphate.     

Sensitized photomodification of DNA with binary systems of oligonucleotide conjugates. IV. Photoinduced electron transfer

The photomodification of single-stranded DNA sensitized to visible light (450-580 nm) by a binary system of oligonucleotide conjugates complementary to adjacent DNA sequences was studied. One oligonucleotide carries a residue of the photoreagent p-azidotetrafluorobenzaldehyde hydrazone at its 3'-terminal phosphate, and the other has a residue of the sensitizer, perylene or 1,2-benzanthracene, at the 5'-terminal phosphate. The rate of photomodification sensitized by the perylene derivative is 300,000-fold higher than the rate of photomodification in the absence of the sensitizer. Since the excitation energy of perylene is lower than the energy necessary for the initiation of azide photodecomposition, it is likely that the sensitization in the complementary complex occurs by electron transfer from the azido group of the photoreagent to the excited sensitizer. The sensitization by the 1,2-benzanthracene oligonucleotide derivative occurs by means of singlet-singlet energy transfer, which enables this sensitizer to act as a unconsumable catalyst each molecule of which is able to initiate the photomodification of more than 20 DNA molecules. By both mechanisms, the photomodification occurs with high specificity on the G11 residue of the target DNA. The degree of sensitized photomodification reaches 72%.     

Sensitized photomodification of DNA with binary systems of oligonucleotide conjugates. V. Effect of the structure of the target DNA. Quantitative photomodification]

A sensitized photomodification of several single-stranded target DNAs by binary systems of oligonucleotide conjugates complementary to the adjacent regions of DNA was performed. One of the conjugates contained a sensitizer (pyrene, anthracene, or 1,2-benzanthracene), and another conjugate contained a photoreagent 4-azidotetrafluorobenzalhydrazone. The sensitized photomodification is initiated by irradiation at 365-580 nm due to effective energy transfer from the excited sensitizer to the photoreagent in a complementary complex of the binary system with the target DNA where the sensitizer and photoreagent are brought sterically together. Conditions for the quantitative photomodification of a single-stranded DNA by the binary system of oligonucleotide conjugates were found. The maximum degree of photomodification depends on the number of guanosine residues in the (pG)n sequence of the target DNA at the modification site: at n = 1 the yield of covalent adducts was 62-68%, at n = 2, 75-82%, and at n = 4, 98-99%.     

Structure of photoreactive binary system of oligonucleotide conjugates assembled on the target nucleotide sequence

Recently we have developed an approach to superspecific photomodification of nucleic acids by binary systems of oligonucleotides conjugated to precursor groups capable of assembling into photoactivatable structure upon simultaneous binding of the conjugates to the target. We have investigated the solution structure of a model binary system 1:2:3, where 1 is the target 12-mer 5'-pdGTATCAGTTTCT, 2 is the photoreactive conjugate 5'-dAGAAACp-NH(CH2)2NH-Az and 3 is the sensitizing conjugate 5'-Pyr-pdTGATAC (Az is p-azidotetrafluorobenzoyl group and Pyr is the pyrenyl-1-methylamino group). The photoreaction within this complex results in crosslinking of reagent 2 with N7-position of the G7 residue of the target thus indicating that the photoreactive Az residue is located in the major groove near the G7 residue. The center-to-center distances between the Pyr and Az moieties in complex 1:2:3 independently determined by the Pyr-group fluorescence quenching and the Az-group sensitized photodecomposition were 11.2 and 12.6 A, respectively.     

Sensitized photomodification of DNA with binary systems of oligonucleotide conjugates. III. Double-quantum sensitization

Site-specific modification of single-stranded DNA by oligonucleotide derivatives of p-azido-O-(4-aminobutyl)tetrafluorobenzaldoxime sensitized by an oligonucleotide derivative of pyrenylethylamine was studied. Upon irradiation with the long-wave UV light (365-390 nm) of a DNA target-oligonucleotide reagent complementary complex, a considerable increase in the rate of sensitized photomodification at the G11 residue of the target relative to the direct photomodification was observed owing to the singlet-single energy transfer from the sensitizer onto the photoreagent. Upon simultaneous irradiation of the complex with UV and visible light in the region of the triplet-triplet absorption of pyrene (360-580 nm), an additional increase in the modification rate and a change in its site-direction (from the G11 to T13 residue) occurred through the two-photon triplet-triplet sensitization. The total extent of the structure photomodification amounted to 80%.     

Effect of linker length on the photomodification process of tyrosine residues in N-(tyrosyl)-N'-(5-azido-2-nitrobenzoyl)diaminoalkanes]

N-(Tyrosyl)-N'-(5-azido-2-nitrobenzoyl)-1,4-diaminobutane, containing a Tyr residue connected with the photoreactive aryl azide group through the diaminobutylene linker, was synthesized as a model for studying the photomodification of Tyr residues in proteins. This compound and the compound with a shorter, 1,2-diaminoethylene linker, obtained previously, were subjected to computer modeling to find their minimal-energy conformations. The aromatic rings of Tyr and 5-azido-2-nitrobenzoic acid residues in the latter compound were localized in parallel planes at a distance of approximately 0.3 nm between them and were shown to be implicated in stacking interactions. On the contrary, the planes of aromatic rings of the former compound with a longer, diaminobutylene linker were found to be situated perpendicularly to each other, with the distance between the centers of the rings being approximately 0.6 nm. The computer analysis was confirmed by experimental results: when studying the photomodification of the compound with the diaminobutylene linker, neither stable products of the Tyr photomodification nor unstable products capable of transformation into stable products in the dark were found. On the contrary, such products were previously identified in the case of the compound with diaminoethylene linker. The formation of amino, nitro, azoxy, and azo derivatives was common for the photomodification of both compounds.     

The Effect of Linker Length on the Photomodification of Tyr Residue in N-(Tyrosyl)-N"-(5-azido-2-nitrobenzoyl)-diaminoalkanes

N-(Tyrosyl)-N"-(5-azido-2-nitrobenzoyl)-1,4-diaminobutane containing a Tyr residue connected with the photoreactive aryl azide group through a diaminobutylene linker was synthesized as a model for studying the photomodification of Tyr residues in proteins. This compound and the compound with a shorter, 1,2-diaminoethylene linker, obtained previously, were subjected to a computer modeling to find their minimal energy conformations. The aromatic rings of Tyr and 5-azido-2-nitrobenzoic acid residues in the latter compound were localized in parallel planes at a distance of approximately 0.3 nm between them and were shown to be implicated in stacking interaction. On the contrary, the planes of aromatic rings of the former compound with a longer diaminobutylene linker were found to be situated perpendicularly to each other, with the distance between the centers of the rings being approximately 0.6 nm. The computer analysis was confirmed by experimental results: when studying the photomodification of the compound with the diaminobutylene linker, neither stable products of the Tyr photomodification nor unstable products capable of transformation into stable products in the dark were found. On the contrary, such products were previously identified in the case of the compound with diaminoethylene linker. The formation of amino, nitro, azoxy and azo derivatives was common for the photomodification of both compounds.     

Site-specific photomodification of nucleic acids with arylazide and perfluoroarylazide oligonucleotide derivatives. II. Specificity in relation to nucleosides]

Oligonucleotide reagents bearing aromatic azido groups of different structures were shown to be suitable for nucleoside specific photomodification of nucleic acids. Modification of the pentadecanucleotide targets d(TAAGTGGAGTTTGGC), d(TAAGTGGAAAAAAAA), d(TAAGTGGACCCCCCC) and d(TAAGTGGATTTTTTT) was investigated with reagents d(UCH2OCH2CH2NHCORCCACTT) carrying a photoactive group R(R1-n-azidotetrafluorophenyl-reagent (I), R2-2-nitro-5-azidophenyl-reagent (II) and R3-n-azidophenyl-reagent (III)) at C-5-modified deoxyuridine. Photomodification did not exceed 5% for the targets in case of reagent (III); the modification extent was 25-50% depending on the target sequence for reagent (II); reagent (I) with perfluoro azido group was the most effective, that provided 60-70% of modification. Reagents (I) and (II) were found to be sensitive to the nucleoside sequence of the target: the most vulnerable sites for reagent (I) and (II) were guanine and cytosine residues, respectively. These bases were modified predominantly when being adjacent to the addressed site of the target.     

Carboxymethylation of a minor ribonuclease from Aspergillus saitoi.

(1) RNase Ms was inactivated by iodoacetate. The inactivation was most rapid at pH 6.0, and was inhibited in the presence of a denaturant such as 8 m urea or 6 m guanidine-HCL. (2) Competitive inhibitors protected RNase Ms from inactivation by iodoacetate; the effect was in the order 2',(3')-GTP greater than 2',(3')-AMP, 2',(3')-UMP greater than or equal to 2',(3')-CMP. The order is not consistent with that of the binding constants of the 4 nucleotides towards RNase Ms (A is greater than C greater than G greater than U). (3) RNase Ms was inactivated with the concomitant incorporation of one molar equivalent of carboxymethly group. The following evidence indicated that the carboxymethyl group was incorporated into the carboxyl group of an aspartic acid or glutamic acid residue. (i) The carboxymethyl group incorporated into RNase Ms was liberated by treatment with 0.1 n NaOH or 1 m hydroxylamine. (ii) The amino acid composition of carboxymethylated RNase Ms (CM RNase Ms) after acid hydrolysis is similar to that of RNase Ms. (4) 14C-Labeled CM RNase Ms was digested successively with alkaline protease and amino-peptidase M. The radioactive amino acid released was eluted just before aspartate on an amino acid analyzer. After hydrolysis with 6 n HCL, glutamic acid was produced exclusively from the radioactive amino acid. The specific radioactivity of this amino acid calculated from the radioactivity and glutamic acid formed was practctically the same as that of CM RNase Ms. Thus, it was concluded that a carboxymethyl group was incorporated at the carboxyl group of a glutamic acid residue of RNnase Ms. (5) CM RNase Ms bound with 2'-AMP to the same extent as native RNase Ms, but bound to a lesser extent with 2',(3')-GMP. (6) Although the conformation of CM RNase Ms as judged from the CD spectrum was practically the same as that of native RNase Ms, the reactivity of CM RNase Ms towards dinitrofluorobenzene was different from that of native RNase Ms, indicating some difference in the conformation. (7) These results indicate that one glutamic acid residue is involved in the active of RNase Ms.     

Sensitized photomodification of DNA with binary sysytems of oligonucleotide conjugates. VI. Effect of substituents in the anthracene residue of the sensitizer

Photomodification of ssDNA by binary systems of oligonucleotide conjugates complementary to the adjacent sequences of the target DNA was studied. One of the conjugates comprised a substituted anthracene as a sensitizer; the other, p-azidotetrafluorobenzaldehyde 3-aminopropionylhydrazone as a photoreagent. The sensitized photomodification is initiated by the 365-580-nm light through an efficient energy transfer from the photoexcitated sensitizer onto the photoreagent in a complementary complex of the binary system with the DNA target where the sensitizer and the photoreagent are sterically converged. Influence of substituents in the anthracene residue on the efficiency of the DNA sensitized photomodification was considered. The oligonucleotide conjugate of anthracene-9-al 3-aminopropionylhydrazone allows highly specific initiation of the sensitized photomodification upon irradiation with visible light at > 460 nm in conditions generating no photoreaction in the sensitizer's absence.     

Site-specific laser modification (cleavage) of oligodeoxynucleotides

Sequence-specific photomodification of oligodeoxynucleotide pAGAGTATTGACTTA ("a target") has been carried out with the aid of complementary fluorescent probes. Such a probe consisted of oligodeoxynucleotide pAATACTCT and a chromophore group attached to its 5' end. Three different derivatives of ethidium bromide were used as a chromophore. The photomodification was induced by nitrogen laser radiation (337 nm, 15 MW/cm2). The irradiation induces the following photodamages: target cleavage at the specific binding site with a cutting off of the 8-mer from its 5' end (yield up to 12%), formation of specific covalent adduct target-probe with a yield of 20-70%, and piperidine-sensitive target modifications with a 7-27% yield (for different chromophores). The total yield of specific photodamages of all kinds is 50-80%. The target cleavage and generation of piperidine-sensitive modifications are optically nonlinear processes. Piperidine treatment of the irradiated samples led to specific cleavage of the target with the yield up to 40%. All kinds of observed modifications are not influenced by high concentrations of free radical scavengers: 1.3M tBuOH and 10 mM cystamine. The pattern of cleavage indicates that the most probable position of the chromophore is between T8 and G9 of the target, i.e., the chromophore stacks on top of the last A.T base pair of the duplex. The aggregate of evidence is in agreement with the mechanism of nonlinear photomodification (the cleavage and generation of piperidine-sensitive modifications) based on the transfer of two-photon excitation energy from the chromophore to the target.     

Highly effective complementary addressed laser modification (cleavage) of oligodeoxynucleotides

Complementary addressed nonlinear photomodification of oligodeoxynucleotide dAGAGTATTGACTTA ("target") has been carried out by means of fluorescent derivatives of oligonucleotide dpAATACTCT ("addressed chromophore"). Three different ethidium derivatives were used as a chromophore. The photomodification was induced by nitrogen laser radiation (337 nm, 15 MW/cm2), which led to the target cleavage in the addressation region with the yield of the main fragment (8 bases long) about 10%, formation of specific covalent adduct target-addressed chromophore with the yield 20-70%, "hidden" (not revealed by gel electrophoresis) target damages with 7-27% yield (for different chromophores). The total yield of specific (i. e. localized in the vicinity of the addressation site) modification was 50-80%. The target cleavage and hidden damage generation are optically nonlinear processes. Piperidine treatment of the irradiated samples caused addressed cleavage of the target with up to 40% yield. All kinds of observed modification are not effected by high concentrations of free radical scavengers, 1,3 M t-BuOH or 10 mM cystamine. The bulk of the data is in agreement with the mechanism of nonlinear photomodification (the cleavage and hidden damage generation) based on the transfer of two-photon excitation energy from the chromophore to the target.     

Sensitized site-specific photomodification of ssDNA by binary systems of oligonucleotide conjugates: VI. Effect of substituents in the sensitizer anthracene residue

Photomodification of ssDNA by binary systems of oligonucleotide conjugates complementary to the adjacent sequences of the target DNA was studied. One of the conjugates comprised a substituted anthracene as a sensitizer; the other,p-azidotetrafluorobenzaldehyde 3-aminopropionylhydrazone as a photoreagent. The sensitized photomodification is initiated by the 365–580-nm light through an efficient energy transfer from the photoexcitated sensitizer onto the photoreagent in a complementary complex of the binary system with the DNA target where the sensitizer and the photoreagent are sterically converged. Influence of substituents in the anthracene residue on the efficiency of the DNA sensitized photomodification was considered. The oligonucleotide conjugate of anthracene-9-al 3-aminopropionylhydrazone allows highly specific initiation of the sensitized photomodification upon irradiation with visible light at >460 nm in conditions generating no photoreaction in the sensitizer’s absence. For Part V, see [1]; prefix “d” in designations of oligonucleotides is omitted.     

Asymmetric epoxidation of alpha-olefins having neighboring sugar chiral templates and alternating copolymerization with dicarboxylic anhydrides

Sugar-substituted epoxides 5-8 were synthesized by asymmetric epoxidation (in CH(2)Cl(2)/water) of alpha-olefins having neighboring sugars (1-4) by use of an achiral oxidant (MCPBA), in which the sugar moiety acted as a chiral template. The diastereoselectivities depend on the methylene spacer between vinyl group and carbohydrate derivatives. The methylene spacer between sugar and vinyl groups influenced the diastereoselectvity. In the case of epoxidation of 4 at 27 degrees C for 24 h, the diastereoselectivity was the highest (99/1). Copolymerizations of 5-8 with succinic anhydride were attained at 100 degrees C for 72 h to give poly(ethylene succinate) having pendant carbohydrate [poly(SAn-alt-5), M(n) = 1.4 x 10(3); poly(SAn-alt-6), M(n) = 2.2 x10(3); poly(SAn-alt-7), M(n) = 2.9 x 10(3); poly(SAn-alt-8), M(n) = 1.8 x 10(3)]. The methylene spacer between sugar and epoxide has an effect on the reactivity of epoxide in copolymerization as well as the diastereoselectivity. Alternating copolymerization of 7 and glutaric anhydride gave a polyester of M(n) 4.2 x10(3).     

Synthesis of oligodeoxyribonucleotide with aliphatic amino or phosphate group at the 5'' end by the phosphotriester method on a polystyrene support.

Lipophilic protecting groups mTrNH(CH2)n X (mTr:monomethoxytrityl, X = NH,O,S, n = 2,3,4,6) were attached to the 5'-phosphoryl group of 3'-O-protected thymidine. When the diamine derivatives (X = NH2) were used, the time course of the stability of mTr groups on the amino group and the phosphoramidate linkage with 80% aq. AcOH was measured. It was found that the mTr group was removed from the amino group rapidly and that the phosphoramidate linkage was more stable. It's stability depended upon the length of the CH2 linker. Oligonucleotides with an aliphatic amino group at their 5'-ends were synthesized by the phosphotriester method on a polystyrene support using protected nucleotides with P-O or P-S linkages. In the case of product with a P-S linkage, 5'-O-phosphorylated nonadecanucleotide was also prepared by I2-H2O treatment.     

Optimising the signal peptide for glycosyl phosphatidylinositol modification of human acetylcholinesterase using mutational analysis and peptide-quantitative structure-activity relationships.

Glycosyl phosphatidylinositol (GPI)-modified proteins have a C-terminal signal peptide (GPIsp) that mediates the addition of a GPI-anchor to an amino acid residue at the cleavage and modification site (omega-site). Within the GPIsp, a stretch of hydrophilic amino acid residues are found which constitutes the spacer region that separates the omega-site residue from a hydrophobic C-terminus. Deletions and insertions into the spacer region of human acetylcholinesterase (AChE) show that the length of this spacer region is very important for efficient GPI-modification. Surprisingly, the natural length of the spacer region in human AChE was not optimal for the highest degree of GPI modification. The importance of the two adjacent residues downstream of the omega-site, the omega+1 and omega+2 residues, was investigated by peptide-quantitative structure-activity relationships (Peptide-QSAR). A model was made that predicts the efficiency of the GPI modification when these residues are substituted with others, and suggests important features for these residues. The most preferred omega+1 and omega+2 residues, predicted by the model, in combination with an ideal spacer length resulted in an optimised GPIsp. This mutant protein is more efficiently GPI-modified than any mutant AChE tested thus far.     

Amino-terminal nucleotide-binding sequences of a Lactobacillus deoxynucleoside kinase complex isolated by novel affinity chromatography

A highly efficient new affinity medium for deoxycytidine kinase, deoxycytidine 5'-tetraphosphate-Sepharose (dCp4-Sepharose), has been constructed. A dCp4-Sepharose column effects a one-step, 19,000-fold, purification to homogeneity of dCyd kinase from the ammonium sulfate fraction of Lactobacillus acidophilus R-26 extract, with 60% recovery. dCTP, a potent end-product inhibitor, is used as an eluent, and it also stabilizes the extremely labile purified enzyme. A noncompeting deoxyadenosine kinase activity accompanies the deoxycytidine kinase activity eluted. Native polyacrylamide gel electrophoresis shows a single protein band, which coincides with both deoxycytidine kinase and deoxyadenosine kinase activities at several gel concentrations. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis reveals a single polypeptide band of 26,000 daltons. Since the native enzyme is known to have an Mr of 50,000, it appears that the enzyme is composed of two subunits of similar size. Sequence analysis of the intact protein from the N-terminus reveals but a single amino acid species per residue up to the 17th residue; at the 18th, 21st, 26th, and 27th residue positions of the sequence, however, there appear to be two different amino acids in almost equal amounts. This may indicate that the enzyme is composed of two nonidentical subunits having the same amino acid sequence near the N-terminus. Residues 6-13 contain the highly conserved Gly-X-X-Gly-X-Gly-Lys sequence found at the active sites of kinases and other nucleotide-binding proteins.     

Intracomplex alkylation of octanucleotide pd(TGTTTGGC) by a 5'-(4-N-methyl-N-(2-chloroethyl)-amino)benzylphosphamide derivative of heptanucleotide pd(CCAAACA). Preparation of a covalent adduct and study of its spatial structure in an aqueous solution by two-dimensional (1)H-NMR spectroscopy]

Covalent adduct--the product of intracomplex alkylation at N-3-position of dC-8-nucleoside residue of target octanucleotide pd[TGTTTGGC] was completely synthesized by means of 4-[N-methyl-N-(2-chloroethyl)amino]benzyl-5'-phosphamido derivative of heptanucleotide pd[CCAAACA]. Its melting temperature was shown to be 70 degrees C. Tm did not depend on covalent adduct concentration and was by 40 degrees C higher than that for unmodified duplex pd[TGTTTGGC].pd[CCAAACA] at concentration of 0.5 x 10(-4) M. The spatial structure of the covalent adduct in aqueous solution was investigated by two-dimensional 3H-NMR spectroscopy. The assignment of oligonucleotide protons as well as protons of a modifying group was carried out using COSY, COSY-DQF and NOESY experiments. Conformational analysis of proton-proton coupling constants for H1', H2'a, H2'b and H3' protons showed the sugar residues to be in 2'-endo conformation. Analysis of NOE connectivities observed between the protons of the alkylating group and oligonucleotide protons yielded conclusion, regarding the 4-[N-methyl-N-(2-chloroethyl)amino]benzylamido 5'-residue being localized in the region of the lacked nucleoside residue of the heptanucleotide chain about 5 A apart from the dC-1 residue and from cytosine base of the alkylated dC-8 residue.     

ATP-dependent Saccharomyces cerevisiae phospho enol pyruvate carboxykinase: isolation and sequence of a peptide containing a highly reactive cysteine.

Saccharomyces cerevisiae phospho enol pyruvate carboxykinase (EC 4.1.1.49), inactivated by N-(iodoacetyl)-N'-(5-sulfo-1-naphthyl)ethylenediamine, incorporated 0.95 mol of the fluorescent moiety per mol of enzyme subunit. Reagent incorporation was completely protected by the presence of ADP plus MnCl2. The labeled protein was digested with trypsin after carboxymethylation. Two labeled peptides were isolated by reverse-phase high-performance liquid chromatography and were sequenced by gas-phase automatic Edman degradation. Both peptides contained overlapping amino acid sequences from Asn-358 to Lys-375, thus identifying Cys-364 as the reactive amino acid residue. The position of the target amino acid residue is immediately preceding a putative phosphoryl-binding sequence proposed for some nucleotide-binding proteins.     

Fatty acyl-gramicidin S derivatives with both high antibiotic activity and low hemolytic activity

In the present study, novel eight GS derivatives having the octanoyl-(Lys)(n)- moieties, cyclo{-Val-Orn-Leu-d-Phe-Pro(4β-NH-X)-Val-Orn-Leu-d-Phe-Pro-} {X=-H (1), and -(Lys)(n)-CO(CH(2))(6)CH(3)n=0 (2), 1 (3), 2 (4), and 3 (5)} and cyclo{-Val-Orn-Leu-d-Phe-Pro(4α-NH-X)-Val-Orn-Leu-d-Phe-Pro-} {X=-H (6), and -(Lys)(n)-CO(CH(2))(6)CH(3)n=1 (7), and 2 (8)} were synthesized. Among them, 4, 5 and 8 result the high antibiotic activity against both Gram-positive and Gram-negative microorganisms tested. In addition, 4 and 5 showed very low hemolytic activity compared with that of GS. Thus, the introduction of the excess amino groups and the fatty acyl moiety to the γ-NH(2) group of Pro(5) residue in GS molecule lowered the unwanted hemolytic activity and enhanced the desired antibiotic activity.