The FASEB Journal • Research Communication
S-nitrosated ␣-1-acid glycoprotein kills drug-resistant bacteria and aids survival in sepsis Kaori Watanabe,*,1 Yu Ishima,*,‡,1 Takaaki Akaike,† Tomohiro Sawa,†,§ Teruo Kuroda,储 Wakano Ogawa,储 Hiroshi Watanabe,*,‡ Ayaka Suenaga,* Toshiya Kai,*,¶ Masaki Otagiri,*,#,** and Toru Maruyama*,‡,2 *Department of Biopharmaceutics and †Department of Microbiology, Graduate School of Pharmaceutical Sciences, and ‡Center for Clinical Pharmaceutical Sciences, School of Pharmacy, Kumamoto University, Kumamoto, Japan; §Precursory Research for Embryonic Science and Technology (PRESTO), Japan Science and Technology Agency (JST), Saitama, Japan; 储Department of Molecular Microbiology, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, Okayama, Japan; ¶Tohoku Nipro Pharmaceutical Corporation, Fukushima, Japan; and #Faculty of Pharmaceutical Sciences and **Drug Delivery System (DDS) Research Institute, Sojo University, Kumamoto, Japan Treating infections with exogenous NO, which shows broad-spectrum antimicrobial activity, appears to be effective. Similar to NO biosynthesis, biosynthesis of ␣-1-acid glycoprotein variant A (AGPa), with a reduced cysteine (Cys149), increases markedly during inflammation and infection. We hypothesized that AGPa is an S-nitrosation target in acute-phase proteins. This study aimed to determine whether Snitrosated AGPa (SNO-AGPa) may be the first compound of this novel antibacterial class against multidrug-resistant bacteria. AGPa was incubated with RAW264.7 cells activated by lipopolysaccharide and interferon-␥. The antimicrobial effects of SNO-AGPa were determined by measuring the turbidity of the bacterial suspensions in vitro and survival in a murine sepsis model in vivo, respectively. Results indicated that endogenous NO generated by activated RAW264.7 cells caused S-nitrosation of AGPa at Cys149. SNO-AGPa strongly inhibited growth of gram-positive, gram-negative, and multidrug-resistant bacteria and was an extremely potent bacteriostatic compound (IC50: 10ⴚ9 to 10ⴚ6 M). The antibacterial mechanism of SNO-AGPa involves S-transnitrosation from SNO-AGPa to bacterial cells. Treatment with SNO-AGPa, but not with AGPa, markedly reduced bacterial counts in blood and liver in a mouse sepsis model. The sialyl residues of AGPa seem to suppress the antibacterial activity, since SNO-asialo AGPa was more potent than SNO-AGPa.—Watanabe, K., Ishima, Y., Akaike, T., Sawa, T., Kuroda, T., Ogawa, W., Watanabe, H., Suenaga, A., Kai, T., Otagiri, M., Maruyama, T. Snitrosated ␣-1-acid glycoprotein kills drug-resistant bacteria ABSTRACT
Abbreviations: ␣1PI, ␣1-protease inhibitor; AGP, ␣-1-acid glycoprotein; AGPa, ␣-1-acid glycoprotein variant A; CLP, cecal ligation and puncture; DAF-FM DA, 4-amino-5-methylamino2=,7=-difluorofluorescein diacetate; DTPA, diethylenetriaminepentaacetic acid; DTT, dithiothreitol; GS-NO, S-nitrosoglutathione; IAN, isoamyl nitrite; NEM, N-ethylmaleimide; NOC7, 1-hydroxy-2-oxo-3-(N-methyl-3-aminopropyl)-3-methyl-1triazene; PPB, potassium phosphate buffer; SNO-AGPa, Snitrosated ␣-1-acid glycoprotein variant A; SNO-asialo-AGPa, S-nitrosated asialo-␣-1-acid glycoprotein 0892-6638/13/0027-0391 © FASEB
and aids survival in sepsis. FASEB J. 27, 391–398 (2013). www.fasebj.org Key Words: nitric oxide 䡠 acute-phase protein 䡠 post-translational modification According to statistics from the World Health Organization in 2008, ⬎10 million people die annually of infections caused by microorganisms (1). Unfortunately, the discovery of new antimicrobials progresses slowly, while microbial resistance to all clinical antimicrobial agents continues to emerge rapidly (2). NO plays a crucial role in specific and nonspecific immunity in humans and possesses broad-spectrum antimicrobial activity (3– 8). However, because of the short biological half-life of NO, a carrier of NO or NO-generating agent may be developed for clinical application of NO as an antimicrobial drug. Among various derivatives of NO, nitrosonium cation exists preferentially as a component of S-nitrosothiols, which have a 100-fold stronger bacteriostatic effect compared with NO alone (3, 4, 9). Thus, from a clinical perspective, one would expect S-nitrosothiols to be effective novel antimicrobial agents. The plasma level of ␣-1-acid glycoprotein (AGP), an acute-phase protein, increases markedly during bacterial infection, as does that of NO. In most individuals, AGP exits as a mixture of two main genetic variants, F1*S and A (AGPa). The molar ratio of the variants F1*S and AGPa in blood typically ranges from 3:1 to 2:1 (10). While AGP variant F1*S does not possess free 1
These authors contributed equally to this work. Correspondence: Department of Biopharmaceutics, Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1 Oe-honmachi, Kumamoto 862-0973, Japan. E-mail:
[email protected] doi: 10.1096/fj.12-217794 This article includes supplemental data. Please visit http:// www.fasebj.org to obtain this information. 2
391
cysteine residue, AGPa has one reduced cysteine residue at position 149 (Cys149). In addition, the carbohydrate part of AGPa represents 45% of the attached molecular mass in the form of 5 or 6 highly sialylated complex-type N-linked glycans (10 –12). All these findings led us to the hypothesis that AGPa would acquire antibacterial activity through S-nitrosation and may be a suitable NO carrier in infectious disease therapeutics. MATERIALS AND METHODS Materials Lyophilized AGP was donated by the Chemo-Sero-Therapeutic Research Institute (Kumamoto, Japan). Dithiothreitol (DTT) and glutathione were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sulfanilamide, naphthyl ethylenediamine dihydrochloride, HgCl2, NaNO2, and NaNO3 were obtained from Nacalai Tesque (Kyoto, Japan). Isoamyl nitrite (IAN) was purchased from Wako Chemicals (Osaka, Japan). S-nitrosoglutathione (GS-NO), 1-hydroxy-2-oxo-3-(N-methyl3-aminopropyl)-3-methyl-1-triazene (NOC7), and diethylenetriaminepentaacetic acid (DTPA) were obtained from Dojindo Laboratories (Kumamoto, Japan). 4-Amino-5-methylamino2=,7=-difluorofluorescein diacetate (DAF-FM DA) was purchased from Sekisui Medical Co. Ltd. (Tokyo, Japan). Other chemicals were of highest grade commercially available. Bacterial strains and growth Escherichia coli, Salmonella typhimurium LT2, Bacillus subtilis, and Streptococcus pyogenes were provided by T.A. Pseudomonas aeruginosa PAO1, Klebsiella pneumoniae MGH78578, Staphylococcus aureus OM481, S. aureus OM584, S. aureus OM505, and S. aureus OM623 were provided by T. Kuroda. Bacteria were grown in L medium (1% polypeptone, 0.5% yeast extract, and 0.5% NaCl, pH 7.0), except for measurement of IC50, at 37°C under aerobic conditions. Growth of cells was monitored turbidimetrically at 630 nm. Preparation of S-nitrosated AGPa (SNO-AGPa) First, to reduce Cys149 of AGPa, AGP (300 M) was treated with 900 M DTT in 0.1 M potassium phosphate buffer (PPB) containing 5 mM ethylenediaminetetraacetic acid, pH 7.4, for 5 min at 37°C. SNO-AGPa was purified by gel filtration (PD-10 Desalting Columns; GE Healthcare, Tokyo, Japan), was eluted with 0.1 M PPB (pH 8.0) containing 1 mM DTPA. Contents of free cysteine residue in AGPa was quantified colorimetrically by the method using 5,5=-dithiobis-2-nitrobenzoic acid and was found to be 1.00 ⫾ 0.05 mol residues/mol AGPa. To synthesize SNO-AGPa, DTT-treated AGPa (300 M) was reacted with 3 mM GS-NO in 0.1 M PPB containing 1 mM DTPA (pH 8.0) for 30 min at 37°C. SNO-AGPa was purified by gel filtration, was eluted with 0.1 M PPB (pH 8.0) containing 1 mM DTPA, and was concentrated by ultrafiltration (Microcon YM-10; Amicon; Millipore, Billerica, MA, USA); it was stored at ⫺80°C until use. The S-nitroso moiety of SNO-AGPa was quantified by the method described below and was found to be 0.40 ⫾ 0.05 mol SNO/mol AGPa. Quantification of SNO-AGPa SNO-AGPa was quantified by using HPLC coupled with a flow reactor system, as reported previously (13). Briefly, the eluate 392
Vol. 27
January 2013
from the HPLC column was mixed with HgCl2 solution to decompose SNO-AGPa to yield NO2⫺ in a reaction coil. The NO2⫺ generated was then detected after reaction with Griess reagent in the flow reactor system. The columns used for analyses were gel filtration columns ( 8⫻30 cm; Diol-60; YMC Co., Ltd., Kyoto, Japan). Generation of SNO-AGPa in cultured macrophages RAW264.7 cells were cultured at 37°C in humidified 5% CO2 95% air in Dulbecco’s modified Eagle’s medium containing 10% FBS, 100 U/ml penicillin, 100 g/ml streptomycin, and 0.5 mM l-arginine. The cells were plated in 96-well culture plates at a density of 2.5 ⫻ 105 cells/100 l/well and were allowed to adhere for 9 h. Then, the medium was replaced with fresh medium, and cells were stimulated with LPS (10 g/ml) and IFN-␥ (10 ng/ml) for 7 h. After the cells were washed with PBS, they were incubated with 100 M AGPa for 10 min at 37°C in the dark. Supernatants were collected, and SNO-AGPa was detected by using HPLC coupled with a flow reactor system. Physicochemical characterization of SNO-AGPa Gross conformational changes induced by S-nitrosation of AGPa were assessed by SDS-PAGE in reducing and nonreducing conditions at 4°C in the dark, with use of a standard curve prepared with a set of marker proteins (Full Range Rainbow Marker; Amersham, Little Chalfont, UK). CD spectra of AGPa and SNO-AGPa were measured at 25°C by using a J-820-type spectropolarimeter (Jasco, Tokyo, Japan). For calculation of the mean residue ellipticity (), the molecular weight of AGPa was taken as 44,000. Far-UV spectra were recorded at protein concentrations of 5 M, in 0.1 M PPB (pH 8.0). S-nitroso moieties during storage in pH 5 or pH 7 solution was examined as follows. They were dissolved in PBS (pH 5 or pH 7) containing 1 mM DTPA and were left in the dark at 37°C for a maximum of 23 d. At appropriate times after the start of incubation, aliquots of the SNO-AGPa solutions were taken and injected into the HPLC flow reactor system to detect SNO-AGPa. The effect of lyophilization was studied by redissolving lyophilized samples in PBS containing 1 mM DTPA (pH 8.0). In addition, S-nitroso moieties of SNO-AGPa were quantified before and after the dissolution. Separation of the variants of AGP Variants were separated via the method of Hervé et al. (14). An iminodiacetate Sepharose gel loaded with copper(II) ions and equilibrated in buffer 1 (20 mM sodium phosphate buffer, pH 7.0, containing 0.5 M sodium chloride) was packed into a column ( 2⫻30 cm). AGP (20 mg dissolved in 1.0 ml of buffer) was applied to the column at a flow rate of 1.0 ml/min. Fractions (10 ml/tube) were collected, and their respective absorbance values were determined spectrometrically at 280 nm. As soon as the AGPa was eluted, elution buffer 2 (buffer 1 plus 20 mM imidazole) was applied to the column to remove the bound F1*S variant. The peak fractions of each eluate were collected, dialyzed against deionized water, and lyophilized. Identification of the site of S-nitrosation of AGPa AGP variants (AGPa and F1*S) and mutant (C149R-A) (100 M) were reacted with 1 mM GS-NO in 0.1 M PPB (pH 8.0) containing 1 mM DTPA for 30 min at 37°C. The S-nitroso peaks of proteins were detected by means of the HPLC flow reactor system.
The FASEB Journal 䡠 www.fasebj.org
WATANABE ET AL.
Antibacterial activity of SNO-AGPa In vitro antibacterial activity of SNO-AGPa was examined according to previously reported methods, with slight modifications (3, 4, 9). We used M9 ⫹ 2 mM MgSO4, 0.2% glucose, 0.2 mM CaCl2, 0.02% vitamin B1 (E. coli and S. typhimurium LT2), or added 0.1% yeast extract medium (S. aureus OM481, S. aureus OM584, S. aureus OM505, S. aureus OM623, P. aeruginosa PAO1, and K. pneumoniae MGH78578), or added 0.1% peptone medium (B. subtilis and S. pyogenes) during incubation of bacteria with SNO-AGPa. In brief, bacteria were cultured overnight in M9 medium and were then washed 3 times with M9 medium. Bacterial suspensions (OD630: 0.04 – 0.05; 0.05 ml in M9 medium) were mixed with 0.05 ml of sterile PBS containing various concentrations of SNO-AGPa or AGPa in 96-well plates. At 9 h after incubation with SNO-AGPa, the numbers of bacteria exposed to SNO-AGPa were determined by measuring the turbidity of the culture suspensions. OD630 values were measured by using a microplate reader (Model 450; Bio-Rad Laboratories, Hercules, CA, USA).
Preparation of the cecal ligation and puncture (CLP) model and measurement of viable bacteria in blood Bacterial infection was induced by CLP, according to a previously reported method, with minor modifications (17). Sepsis was induced in ICR mice, which were then given SNO-AGP (50 nmol AGPa/kg). Two punctures and cecal ligation were then performed by using a 21-gauge needle. Survival was determined daily for 4 d after the puncture. Blood samples (0.3 ml) from surviving mice were collected from the tail vein at 9 h after the puncture. After blood samples were collected, aliquots of blood (0.1 ml) were immediately placed on Luria-Bertani agar plates (9.9 ml), which were incubated at 37°C for 24 h. Mice were killed at 9 h after CLP. Liver tissues were collected, weighed, carefully washed 4 times with sterile PBS, and homogenized in sterile PBS (1 ml/g). Serial dilutions of liver homogenates in PBS were placed on Luria-Bertani agar plates, which were incubated at 37°C for 24 h. The numbers of bacterial colonies were then counted and expressed as colony-forming units per milliliter or colony-forming units per gram of tissue, respectively.
Preparation of S-nitrosated asialo-AGPa (SNO-asialo-AGPa) Asialo-AGPa was prepared according to the method of Primozic and McNamara (15). AGPa solution (5 ml, 40 mg/ml in 0.1 M acetate buffer, pH 5.0) was incubated with 2 U neuraminidase that had been attached to beads (Clostridium perfringens) at 37°C for 24 h. After the reaction, AGPa and neuraminidase were separated by filtration through 0.2-m filters. The filtrate was dialyzed against deionized water and then lyophilized. S-nitrosation of asialo-AGPa was achieved in the same manner as with AGPa. The resulting asialo-AGPa was confirmed by thiobarbituric acid methods, which indicated that ⬎95% of sialic acid had been removed (16).
Statistical analysis Data are shown as means ⫾ se for the indicated number of animals. Significant differences among each group were determined by means of the 2-tailed unpaired Student’s t test. The Spearman test was used for correlation analyses. Survival was compared by using Kaplan-Meier survival curves and the log-rank test. A probability value of P ⬍ 0.05 was considered to indicate statistical significance.
RESULTS Analysis of NO transfer from SNO-AGPa into bacterial cells NO incorporated into bacterial cells was identified directly by using DAF-FM DA, which is a fluorescent indicator of NO. The bacterial suspensions (OD630: 0.04 – 0.05; 0.05 ml in M9 medium) were mixed with 0.05 ml of PBS containing various concentrations of SNO-AGPa or SNO-asialo-AGPa and were incubated in a 96-well plate for 6 h at 37°C. Bacteria were then treated with 10 M DAF-FM DA for 1 h at 37°C. After cells were washed with PBS, fluorescence intensity was measured by using a fluorescence plate reader (Tecan SpectraFluor Xfluor4). The numbers of bacteria were determined as described above, by measuring the turbidity of the suspension. S-transnitrosation from SNO-AGPa to cysteines of bacterial cells Decay of SNO-AGPa at 25 M in M9 was monitored during incubation with E. coli ATCC (OD630: 0.04 – 0.05) at 37°C, and the amount of SNO-AGPa remaining in the supernatant was quantified as described above. Decay of SNO-AGPa was also analyzed with the use of bacteria pretreated with 1 mM N-ethylmaleimide (NEM). NEM-treated cells were washed 3 times with PBS and subjected to the reaction with SNO-AGPa. These cells originally contained 5.0 nmol sulfhydryl/108 cfu, as determined by the dithiobisnitrobenzoic acid method using bacterial lysate obtained by sonication for 1 min on ice with an ultrasonic processor (Model GE50; Tokyo Rikakikai Co., Ltd., Tokyo, Japan); 93% of the cysteine residues of bacterial surface proteins were blocked by NEM treatment. S-NITROSATED ␣-1-ACID GLYCOPROTEIN
S-nitrosation of AGPa To determine whether AGPa could be S-nitosated, AGPa was incubated with NOC7, IAN, and GS-NO, after which S-nitrosation efficiency was evaluated by means of HPLC coupled with a flow reactor system using HgCl2 and Griess reagent (13). Results showed that the S-nitroso content of AGPa increased with rising pH values (Fig. 1A) and that GS-NO effectively caused SNO-AGPa in a concentrationand time-dependent manner (Fig. 1B, C). Under optimal in vitro S-nitrosation conditions, the S-nitroso moiety of AGPa was ⬃0.40 ⫾ 0.05 mol SNO/mol AGPa. To examine whether AGPa was also S-nitrosated by endogenous NO, we added AGPa to RAW264.7 cells activated by LPS and IFN-␥. As Fig. 1D shows, the HPLC flow reactor system detected SNO-AGPa as a single peak with a retention time of 14 min, which indicates that endogenous NO efficiently S-nitrosated AGPa. We next assessed the physicochemical characteristics of SNO-AGPa. SDS-PAGE clearly showed that the SNOAGPa synthesized in this study was homogeneous and had a mobility identical to that of parental AGPa under reducing and nonreducing conditions (Fig. 1E). To confirm the effect of S-nitrosation on AGPa structure in more detail, CD spectroscopy showed that S-nitrosation of AGPa had no detectable effect on the secondary structure of native AGPa (Fig. 1F). The S-nitroso moieties of SNO-AGPa were quantified after lyophilization or after one freeze-thaw cycle; results showed that 393
Figure 1. S-nitrosation of AGPa. A) The S-nitroso moiety of SNO-AGPa was incubated with NO donors in solutions of different pH values for 3 h. AGPa, 100 M; NO donors, 1 mM. B, C) AGPa was incubated with GS-NO to show concentration (B) and time dependence (C). D) SNO-AGPa formation by IFN-␥- and LPS-stimulated RAW264.7 cells. RAW264.7 cells were preincubated with LPS and IFN-␥ for 7 h. Retention times for SNO-AGPa and NO2⫺ were ⬃14 and 23 min, respectively. E) SDS-PAGE of AGPa before and after S-nitrosation. F) Far-UV CD spectra of AGPa and SNO-AGPa. G) Stability of the SNO moiety of SNO-AGPa after lyophilization or one freeze-thaw cycle. H) Identification of the S-nitrosated site on AGPa. I) Stability of SNO-AGPa and GS-NO in solutions of different pH values. Data are expressed as means ⫾ se (n⫽3).
almost 100% of SNO-AGPa was stable under both conditions (Fig. 1G). To obtain information about the S-nitrosation site in AGPa, we used GS-NO treatment of 3 AGPs: the AGP variant F1*S, which did not possess the reduced cysteine residue; AGPa; and the AGPa mutant in which the cysteine was replaced with arginine, the variant named C149R-A (18). We analyzed these variants by using the HPLC flow reactor system. As seen in Fig. 1H, a nitrosated peak was detected in AGPa but not in variants F1*S and C149R-A, which indicated that the site of S-nitrosation of AGPa was an SH group of Cys149 in AGPa. We also investigated the stability of SNO-AGPa and GS-NO at pH 5 and pH 7. As Fig. 1I shows, SNO-AGPa was more stable than GS-NO at both pH values. SNO-AGPa was more stable in acidic conditions than at neutral pH (Fig. 1I). Antibacterial activity of SNO-AGPa and AGPa Several reports have detailed the antiproliferative effects of S-nitrosothiols on bacteria, parasites, and mammalian 394
Vol. 27
January 2013
cells, with the effective antibacterial concentrations of low-molecular-weight S-nitrosothiols ranging from 100 M to mM values (IC50: 10⫺4 to 10⫺3 M; refs. 19 –23). Under the present experimental conditions, NO itself had little antibacterial effect (IC50: 10⫺2 M; refs. 3–5). SNO-AGPa strongly inhibited the growth of not only various drug-sensitive gram-positive and gram-negative bacteria, including B. subtilis, S. pyogenes, E. coli ATCC, S. typhimurium ATCC, and P. aeruginosa ATCC, but also drug-resistant bacteria, such as S. aureus OM481, S. aureus OM505, S. aureus OM584, S. aureus OM623 [methicillin-resistant S. aureus (MRSA)], and K. pneumonia MGH78578, for which the IC50 values were 10⫺9 to 10⫺6 M (Fig. 2, Supplemental Fig. S1, and Table 1). In contrast, AGPa before S-nitrosation or SNO-AGPa after UV treatment had little bacteriostatic effect (Fig. 2 and Supplemental Figs. S1 and S2), which indicates that the antiproliferative effects of SNO-AGPa can be attributed to the SNO moiety. Therefore, SNO-AGPa appears to be the most potent bacteriostatic compound among S-nitrosated molecules reported so far (3, 4, 9).
The FASEB Journal 䡠 www.fasebj.org
WATANABE ET AL.
Figure 2. Antibacterial activity of SNO-AGPa (solid circles) and AGPa (open circles) against bacteria. A) B. subtilis. B) S. pyogenes. C) S. aureus OM481. D) S. aureus OM505. E) E. coli ATCC. F) S. typhimurium LT2. G) P. aeruginosa PAO1. H) K. pneumoniae MGH78578. Data are expressed as means ⫾ se (n⫽15). Bacterial suspensions (OD630: 0.04 – 0.05; 50 l in medium) were mixed with 50 l of sterile PBS containing various concentrations of SNO-AGPa or AGPa in 96-well plates. Numbers of bacteria exposed to SNO-AGPa were determined by measuring turbidity of the suspensions. OD630 values were measured by using a microplate reader. Bacterial growth, expressed as percentage of control, was determined at 9 h after incubation with SNO-AGPa or AGPa.
Mechanism and effects of SNO-asialo-AGPa, NO, and SNO-AGPa on bacteria At the local sites of infection, the sialic acid-degrading enzyme (sialidase) concentration is higher than that in normal tissue (24 –29). Therefore, to clarify the effect of the sialic acid units of SNO-AGP on the antibacterial activity of SNO-AGP, we enzymatically prepared asialoAGPa (30) by removing sialic acid units from the TABLE 1. IC50 values for gram-negative and gram-positive bacteria after SNO-AGPa treatment Gram stain
Negative E. coli S. typhimurium LT2 P. aeruginosa K. pneumoniae MGH78578 Positive B. subtilis S. pyogenes S. aureus OM481 S. aureus OM505 S. aureus OM584 S. aureus OM623
S-NITROSATED ␣-1-ACID GLYCOPROTEIN
IC50 (nM)
65 250 30 60 4 800 4000 6000 5000 3000
terminal ends of the oligosaccharide chains, followed by S-nitrosation, yielding a molecule that we called SNO-asialo-AGPa. As Fig. 3A illustrates, SNO-asialoAGPa had a bacteriostatic effect against E. coli, with an IC50 value of 8 ⫻ 10⫺10 M, which was ⬃80 times greater than that of SNO-AGPa. We also examined the transfer of NO from SNOAGPa to bacteria by using DAF-FM DA, which is a fluorescent indicator of intracellular NO accumulation (31). Results demonstrated that SNO-asialo-AGPa transferred NO to bacteria more effectively than did SNOAGPa (Fig. 3B). Furthermore, we found that the fluorescence intensity correlated significantly with bacterial growth (Fig. 3C), which suggests that the antibacterial activity of SNO-AGPa depended on the ability of SNOAGPa to transport NO to bacteria. To clarify the mechanism of NO transfer from SNOAGPa to bacteria, we examined S-transnitrosation from SNO-AGPa by measuring the content of SNO-AGPa in the medium. The decrease in SNO-AGPa was more rapid than the decrease in SNO-human serum albumin, and NEM treatment of bacteria abrogated this phenomenon (Fig. 3D). These data suggest that S-transnitrosation from SNO-AGPa to cysteines of bacterial surface proteins mediates the antibacterial activity of SNO-AGPa. 395
Figure 3. Mechanism and effects of SNO-asialo-AGPa, NO, and SNO-AGPa on bacteria. A) Antibacterial activity of SNO-asialo-AGPa against E. coli ATCC. Data are expressed as means ⫾ se (n⫽15). B) Transfer of NO from SNO-AGPa to bacteria, as assessed by DAF-FM DA indication of NO incorporation into cells. Data are expressed as means ⫾ se (n⫽3). C) Relationship between fluorescence intensity and bacterial growth. Linear regression of logarithmic values was calculated by using the leastsquares method (r⫽0.93). D) S-transnitrosation from SNO-AGPa to bacteria as shown by the amounts of SNO-AGPa remaining in the supernatant. Bacteria were reacted with NEM to block cysteines of the bacterial surface proteins. SNO-AGPa was incubated with the NEM-treated or nontreated bacteria for the indicated periods. Data are expressed as means ⫾ se (n⫽3). *P ⬍ 0.05 vs. NEM-treated bacteria.
In vivo antibacterial activity of SNO-AGPa Finally, we evaluated the antibacterial activity of SNOAGPa in vivo by using CLP mice as a model of sepsis; this model is believed to closely mimic the nature and course of sepsis in patients after trauma (17). Mice received saline, AGPa, or SNO-AGPa at 3 h after surgery. Figure 4A shows that 90% of saline-treated mice died within 24 h after induction of CLP. In contrast, survival of mice with sepsis that had had SNO-AGPa treatment was significantly longer than that of saline- or AGPa-treated mice, with 50% of the group receiving SNO-AGPa surviving (Fig. 4A). We also measured the number of bacteria in both blood and liver of mice at 9 h after induction of CLP. As seen in Fig. 4B, C, administration of SNO-AGP significantly reduced the number of bacteria in blood and liver, respectively. Such an enhanced antiproliferative effect by SNOAGPa against bacteria may have contributed to improved survival observed in these model mice.
DISCUSSION Our study revealed 4 major findings. First, Cys149 in AGPa was S-nitrosated by exogenous NO donors and endogenous NO generated by activated RAW264.7 cells. Second, SNO-AGPa was quite stable after both lyophilization and one freeze-thaw cycle. Of interest is that SNO-AGPa was more stable in acidic conditions than at neutral pH. These data suggest that SNO-AGPa is more stable at the site of inflammation than in normal tissues, because inflammation generally in396
Vol. 27
January 2013
duces local acidic conditions (32). Third, SNO-AGPa and SNO-asialo-AGPa strongly inhibited the growth of not only various drug-sensitive gram-positive and gramnegative bacteria but also drug-resistant bacteria in vitro. Fourth, SNO-AGPa significantly improved survival in a murine model of severe sepsis. These data suggest that AGPa is a suitable NO carrier and would allow NO to exhibit continuous biological activity. As a result of increased production of both an acute-phase protein and NO in the acute-phase of number of disorders including inflammation and bacterial infection, the cysteine in acute-phase proteins has been proposed as one of the putative target sites for S-nitrosation. So far, among acute-phase proteins, ␣1protease inhibitor (␣1PI) was found to be S-nitrosated, and SNO-␣1PI acquired new diverse biological actions, such as inhibition of platelet aggregation and bacteriostasis. The present study reports, for the first time to the best of our knowledge, that AGPa, with a single cysteine at position 149, was S-nitrosated. Our recent crystallographic analysis of AGPa clearly showed that Cys149 was located in the flexible loop (Cys147-Pro151) and that it was partially exposed to solvent (Supplemental Fig. S3). Such microenvironments surrounding Cys149 are consistent with the structural features of the previously reported S-nitrosation site in proteins (33, 34). Thus, like SNO-␣1PI, SNO-AGPa may also be produced as an NO sink in inflammatory conditions or at sites of infection. SNO-AGPa demonstrated a potent bacteriostatic effect against a wide range of bacteria (IC50: 10⫺9 to 10⫺6 M), and this effect is more powerful than the previously reported effects of NO alone and of S-nitrosothiols; e.g.,
The FASEB Journal 䡠 www.fasebj.org
WATANABE ET AL.
Figure 4. In vivo antibacterial activity of SNO-AGPa. A) Survival of mice in the sepsis model (n⫽10). Two punctures and cecal ligation were performed with a 21-gauge needle. Septic mice were then given SNO-AGPa 3 h after the puncture. Survival was determined daily for 4 d after the puncture. B) Effect of SNO-AGPa on the number of bacteria in the blood and at 9 h after CLP. C) Effect of SNO-AGPa on the number of bacteria in the liver at 9 h after CLP. Data are expressed as means ⫾ se (n⫽8 –10). **P ⬍ 0.01, ***P ⬍ 0.001 vs. saline-treated mice.
the IC50 values of S-nitrosated human serum albumin against several bacteria were in the range of 10⫺6 to 10⫺5 M. Under similar experimental conditions, SNOAGPa exhibited superior antibacterial activity against S. typhimurium: ⬎10 times more effective compared with SNO-␣1PI, which showed bacteriostatic activity at 10⫺5 M (3, 4, 9). A more interesting result was that desialylation of SNO-AGPa enhanced the bacteriostatic effect against E. coli, which was ⬃80 times more potent than the effect of SNO-AGPa (Fig. 3A). Because sialidase activity is enhanced during infection, e.g., in patients with sepsis (27–29), sialidase would be expected to induce the formation of SNO-asialo-AGPa and hence manifests more potent antibacterial activity under pathophysiological conditions. Our previous data showed that S-transnitrosation from ␣1PI to bacteria is a crucial step for the antibacterial action of SNO-␣1PI. A similar mechanism would be also involved for the antibacterial activity of SNOAGPa, because the inhibition of bacterial growth by SNO-AGPa correlated well with the transport of NO to S-NITROSATED ␣-1-ACID GLYCOPROTEIN
bacteria (Fig. 3C). Interestingly, such NO transport ability of SNO-asialo-AGPa was stronger than that of SNO-AGPa. Moreover, such S-transnitrosation from SNO-AGPa to bacteria was markedly inhibited by the modification of the cysteines of bacterial cells. These findings led us to the suggestion that the sialic acid units of AGPa may have a negative effect on NO transport ability, and thus on bacteriostatic activity. In addition, the antibacterial activity of SNO-AGPa may be mediated by the SH group in cysteines of bacterial cells. Recently, Sakarya et al. (35) found that S. aureus adherence to pharynx cells was enhanced by the removal of cell surface sialic acids. Thus, increase of S-transnitrosation to cysteines of bacterial cells, and hence enhanced bacteriostatic effect by SNO-asialo-AGPa, may be due to the enhancement of SNO-AGPa adherence to bacteria. We also found that SNO-AGPa, not AGPa administered 3 h after surgery, exhibited superior antibacterial activity in vivo in CLP model mice and significantly improved their survival (Fig. 4). Previously, Muchitsch et al. (36) showed that the survival rate (48 h) in rats with septic peritonitis significantly increased when exogenous AGP (5 mol/kg i.v.) was given 15 min prior to and 24 h after cecal puncture. Similarly, Hochepied et al. (37) reported that the administration of AGP (12.5 mol/kg i.v.) 2 h before a challenge of K. pneumoniae significantly increased survival. At a glance, these findings suggest that biological functions of AGP, such as anti-inflammatory and immunomodulatory activities, rather than the antibacterial activity of SNOAGPa, are largely responsible for the improved survival of CLP model mice observed in this study. However, as compared to these studies, we administered a very small amount of SNO-AGPa (0.05 mol AGPa/kg), which was approximately one-tenth of the endogenous AGP level at the acute phase of infection. Therefore, the SNO moiety, not AGPa, could be largely contributing to the altered survival of the bacterial sepsis model produced by SNO-AGPa. This finding is consistent with the results obtained from in vitro experiments that the antiproliferative effects of SNO-AGPa against bacteria can be attributed to the SNO moiety. In summary, the present findings suggest that SNOAGPa and its asialo derivatives are not only useful NO carriers but also members of a new class of S-nitrosated proteins possessing potent antibacterial activity. Thus, SNO-AGPa and its asialo derivatives may participate in a host defense system as endogenous antibacterial agents and may have potential as novel agents in antimicrobial therapy. Moreover, because the mechanism of the bactericidal activity of NO is likely to be different from that of existing antibacterial drugs, combined therapy with these agents may be possible. The authors thank Ms. Judith B. Gandy for editing the manuscript. Thanks are also due to members of the Gene Technology Center in Kumamoto University for their important contributions to the experiments. The authors also thank the members of the Chemo-Sero-Therapeutic Research Institute (Kumamoto, Japan) for donation of lyophilized AGP. This research was supported in part by grants-in-aid for scientific research from the Japan Society for the Promotion of Science 397
(JSPS; KAKENHI 18390051 and 22790162). This work was also supported in part by grants from the Uehara Memorial Fund and the Kumayaku Alumni Research Fund.
21.
REFERENCES 1. 2. 3.
4. 5.
6. 7. 8. 9.
10. 11.
12.
13.
14.
15. 16. 17. 18.
19.
398
World Health Organization (2008) World Health Report 2008, World Health Organization, Geneva, Switzerland Levy, S. B. (2002) The Antibiotic Paradox, 2nd Ed., Perseus Books Group, Cambridge, MA, USA Miyamoto, Y., Akaike, T., Alam, M. S., Inoue, K., Hamamoto, T., Ikebe, N., Yoshitake, J., Okamoto, T., and Maeda, H. (2000) Novel functions of human alpha(1)-protease inhibitor after S-nitrosylation: inhibition of cysteine protease and antibacterial activity. Biochem. Biophys. Res. Commun. 267, 918 –923 Miyamoto, Y., Akaike, T., and Maeda, H. (2000) S-nitrosylated human alpha(1)-protease inhibitor. Biochim. Biophys. Acta. 1477, 90 –97 Kuo, C-F., Wu, J. J., Lin, K. Y., Tsai, P. J., Lee, S. C., Jin, Y. T., Lei, H. Y., and Lin, Y. S. (1998) Role of streptococcal pyrogenic exotoxin B in the mouse model of group A streptococcal infection. Infect. Immun. 66, 3931–3935 Fang, F. C. (1997) Perspectives series: host/pathogen interactions. Mechanisms of nitric oxide-related antimicrobial activity. J. Clin. Invest. 99, 2818 –2825 Bogdan, C. (2001) Nitric oxide and the immune response. Nat. Immunol. 2, 907–916 Foley, E., and O’Farrell, P. H. (2003) Nitric oxide contributes to induction of innate immune responses to gram-negative bacteria in Drosophila. Genes Dev. 17, 115–125 Ishima, Y., Sawa, T., Kragh-Hansen, U., Miyamoto, Y., Matsushita, S., Akaike, T., and Otagiri, M. (2007) S-Nitrosylation of human variant albumin Liprizzi (R410C) confers potent antibacterial and cytoprotective properties. J. Pharmacol. Exp. Ther. 320, 969 –977 Fournier, T., Medjoubi, N. N., and Porquet, D. (2000) Alpha-1acid glycoprotein. Biochim. Biophys. Acta. 1482, 157–171 Martìnez Cordero, E., Gonza`lez, M. M., Aguilar, L. D., Orozco, E. H., and Herna`ndez Pando, R. (2008) Alpha-1-acid glycoprotein, its local production and immunopathological participation in experimental pulmonary tuberculosis. Tuberculosis 88, 203– 211 Silva Miranda, M., Rodriguez, K. W., Martinez Cordero, E., and Rojas-Espinosa, O. (2006) Expression of cyclooxygenase-2, alpha 1-acid-glycoprotein and inducible nitric oxide synthase in the developing lesions of murine leprosy. Int. J. Exp. Pathol. 87, 485–494 Akaike, T., Inoue, K., Okamoto, T., Nishino, H., Otagiri, M., Fujii, S., and Maeda, H. (1997) Nanomolar quantification and identification of various nitrosothiols by high performance liquid chromatography coupled with flow reactors of metals and Griess reagent. J. Biochem. 122, 459 –466 Hervé, F., Duché, J. C., Barré, J., Millot, M. C., and Tillement, J. P. (1992) pH titration curves of the desialylated human alpha 1-acid glycoprotein variants by combined isoelectrofocusingelectrophoresis: utilization in the development of a fractionation method for the protein variants by chromatography on immobilized metal affinity adsorbent. J. Chromatogr. 577, 43–59 Primozic, S., and McNamara, P. J. (1985) Effect of the sialylation state of alpha 1-acid glycoprotein on propranolol binding. J. Pharm. Sci. 74, 473–475 Warren, L. (1959) The thiobarbituric acid assay of sialic acids. J. Biol. Chem. 234, 1971–1975 Hubbard, W. J., Choudhry, M., Schwacha, M. G., Kerby, J. D., Rue, L. W. 3rd., Bland, K. I., and Chaudry, I. H. (2005) Cecal ligation and puncture. Shock 1, 52–57 Nishi, K., Fukunaga, N., Ono, T., Akuta, T., Yumita, N., Watanabe, H., Kadowaki, D., Suenaga, A., Maruyama, T., and Otagiri, M. (2010) Construction of an expression system for human alpha(1)-acid glycoprotein in E. coli: the roles of oligosaccharide moieties in structural and functional properties. Drug Metab. Pharmacokinet. 25, 200 –207 Mnaimneh, S., Geffard, M., Veyret, B., Vincendeau, P. (1997) Albumin nitrosylated by activated macrophages possesses antip-
Vol. 27
January 2013
20.
22. 23. 24. 25. 26.
27.
28.
29.
30.
31. 32. 33.
34.
35.
36. 37.
arasitic effects neutralized by anti-NO-acetylated-cysteine antibodies. J. Immunol. 158, 308 –314 Ewing, J. F., Young, D. V., Janero, D. R., Garvey, D. S., and Grinnell, T. A. (1997) Nitrosylated bovine serum albumin derivatives as pharmacologically active nitric oxide congeners. J. Pharmacol. Exp. Ther. 283, 947–954 De Groote, M. A., Granger, D., Xu, Y., Campbell, G., Prince, R., and Fang, F. C. (1995) Genetic and redox determinants of nitric oxide cytotoxicity in a Salmonella typhimurium model. Proc. Natl. Acad. Sci. U. S. A. 92, 6399 –6403 De Groote, M. A., Testerman, T., Xu, Y., Stauffer, G., and Fang, F. C. (1996) Homocysteine antagonism of nitric oxide-related cytostasis in Salmonella typhimurium. Science 272, 414 –417 Incze, K., Farkas, J., Mihályi, V., and Zukál, E. (1974) Antibacterial effect of cysteine-nitrosothiol and possible percursors thereof. Appl. Microbiol. 27, 202–205 Hennet, T., Chui, D., Paulson, J. C., and Marth, J. D. (1998) Immune regulation by the ST6Gal sialyltransferase. Proc. Natl. Acad. Sci. U. S. A. 95, 4504 –4509 Nan, X., Carubelli, I., and Stamatos, N. M. (2007) Sialidase expression in activated human T lymphocytes influences production of IFN-gamma. J. Leukoc. Biol. 81, 284 –296 Cross, A. S., Sakarya, S., Rifat, S., Held, T. K., Drysdale, B. E., Grange, P. A., Cassels, F. J., Wang, L. X., Stamatos, N., Farese, A., Casey, D., Powell, J., Bhattacharjee, A. K., Kleinberg, M., and Goldblum, S. E. (2003) Recruitment of murine neutrophils in vivo through endogenous sialidase activity. J. Biol. Chem. 278, 4112–4120 Lambré, C. R., Pilatte, Y., Le Maho, S., Greffard, A., De Crémoux, H., and Bignon, J. (1988) Sialidase activity and antibodies to sialidase-treated autologous erythrocytes in bronchoalveolar lavages from patients with idiopathic pulmonary fibrosis or sarcoidosis. Clin. Exp. Immunol. 73, 230 –235 Piagnerelli, M., Boudjeltia, K. Z., Brohee, D., Piro, P., Carlier, E., Vincent, J. L., Lejeune, P., and Vanhaeverbeek, M. (2003) Alterations of red blood cell shape and sialic acid membrane content in septic patients. Crit. Care Med. 31, 2156 –2162 Krivan, H. C., Roberts, D. D., and Ginsburg, V. (1988) Many pulmonary pathogenic bacteria bind specifically to the carbohydrate sequence GalNAc beta 1-4Gal found in some glycolipids. Proc. Natl. Acad. Sci. U. S. A. 85, 6157–6161 Matsumoto, K., Nishi, K., Kikuchi, M., Watanabe, H., Nakajou, K., Komori, H., Kadowaki, D., Suenaga, A., Maruyama, T., and Otagiri, M. (2010) Receptor-mediated uptake of human alpha1acid glycoprotein into liver parenchymal cells in mice. Drug Metab. Pharmacokinet. 25, 101–107 Kojima, H., Urano, Y., Kikuchi, K., Higuchi, T., Hirata, Y., and Nagano, T. (1999) Fluorescent indicators for imaging nitric oxide production. Angew. Chem. Int. Ed. Engl. 38, 3209 –3212 Yatvin, M. B., Kreutz, W., Horwitz, B. A., and Shinitzky, M. (1980) pH-sensitive liposomes: possible clinical implications. Science 210, 1253–1255 Nishi, K., Ono, T., Nakamura, T., Fukunaga, N., Izumi, M., Watanabe, H., Suenaga, A., Maruyama, T., Yamagata, Y., Curry, S., and Otagiri, M. (2011) Structural insights into differences in drug-binding selectivity between two forms of human alpha1acid glycoprotein genetic variants, the A and F1*S forms. J. Biol. Chem. 286, 14427–14434 Greco, T. M., Hodara, R., Parastatidis, I., Heijnen, H. F., Dennehy, M. K., Liebler, D. C., and Ischiropoulos, H. (2006) Identification of S-nitrosylation motifs by site-specific mapping of the S-nitrosocysteine proteome in human vascular smooth muscle cells. Proc. Natl. Acad. Sci. U. S. A. 103, 7420 –7425 Sakarya, S., Ertugrul, M. B., Oztürk, T., and Gökbulut, C. (2010) Effect of pharynx epithelial cells surface desialylation on receptor-mediated adherence of Staphylococcus aureus. J. Appl. Microbiol. 108, 1313–1322 Muchitsch, E. M., Auer, W., and Pichler, L. (1998) Effects of alpha 1-acid glycoprotein in different rodent models of shock. Fundam. Clin. Pharmacol. 12, 173–181 Hochepied, T., Van, Molle, W., Berger, F. G., Baumann, H., and Libert, C. (2000) Involvement of the acute phase protein alpha 1-acid glycoprotein in nonspecific resistance to a lethal gramnegative infection. J. Biol. Chem. 275, 14903–14909
The FASEB Journal 䡠 www.fasebj.org
Received for publication August 29, 2012. Accepted for publication October 1, 2012.
WATANABE ET AL.
