Indirect determination of nitric oxide production by

ANALYTICAL BIOCHEMISTRY Analytical Biochemistry 328 (2004) 14–21 www.elsevier.com/locate/yabio

Indirect determination of nitric oxide production by reduction of nitrate with a freeze–thawing-resistant nitrate reductase from Escherichia coli MC1061 Sergio Arias-Negrete,a,¤ Luis A. Jiménez-Romero,a Martha O. Solís-Martínez,a Joel Ramírez-Emiliano,a Eva E. Avila,a and Patricia Cuéllar-Matab a

Instituto de Investigación en Biología Experimental, Universidad de Guanajuato, 36000 Guanajuato, Gto., Mexico b Facultad de Química, Universidad de Guanajuato. Guanajuato, Gto., Mexico Received 28 January 2003

Abstract Preparation of a nitrate reductase lysate of Escherichia coli MC1061 to measure nitrate and nitrite in biologic Xuids is described. To obtain the crude bacterial lysate containing nitrate reductase activity, E. coli MC1061 was subjected to 16–20 freeze–thawing cycles, from ¡70 to 60 °C, until nitrite reductase activity was 625%. Nitrate reductase activity was detected mainly in the crude preparation. To validate the nitrate reduction procedure, standard nitrate solutions (1.6–100 M) were incubated with the nitrate reductase preparation for 3 h at 37 °C, and nitrite was estimated by the Griess reaction in a microassay. Nitrate solutions were reduced to nitrite in a range of 60–70%. Importantly, no cofactors were necessary to perform nitrate reduction. The biological samples were Wrst reduced with the nitrate reductase preparation. After centrifugation, samples were deproteinized with either methanol/ether or zinc sulfate and nitrite was quantiWed. The utility of the nitrate reductase preparation was assessed by nitrate + nitrite determination in serum of animals infected with the protozoan Entamoeba histolytica or the bacteria E. coli and in the supernatant of cultured lipopolysaccharide-stimulated RAW 264.7 mouse macrophages. Our results indicate that the nitrate reductase-containing lysate provides a convenient tool for the reduction of nitrate to determine nitrate + nitrite in biological Xuids by spectrophotometric methods.  2004 Elsevier Inc. All rights reserved. Keywords: Escherichia coli nitrate reductase; Escherichia coli nitrite reductase; Griess reaction; Nitric oxide; Nitrate; Nitrite

Nitric oxide (NO)1 is a reactive gas that participates in several vital physiological functions in the body such as maintenance of vascular tone, neurotransmission, and immunoregulatory eVects relevant to the control of infections [1,2]. NO originates from oxygen and L-arginine in a reaction enzymatically catalyzed by three NO synthases (NOS, EC 1.14.13.39). The neuronal (nNOS or type I) and endothelial (eNOS or type III) NOS isoforms, which are constitutively present [3], produce small amounts of NO [4]. The third NOS isoform is inducible (iNOS or type II), and it produces large amounts of NO after cell stimulation with proinXammatory substances ¤

Corresponding author. E-mail address: [email protected] (S. Arias-Negrete). 1 Abbreviations used: NO, nitric acid; NOS, NO synthase; NaR; nitrate reductase; NiR, nitrite reductase; LPS, lipopolysaccharide; PBS, phosphate-buVered saline. 0003-2697/$ - see front matter  2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2004.01.026

and/or cytokines [3]. The quantitative determination of NO released from tissues or cultured cells is necessary to understand the complex role of this biological messenger. There are several methods to determine NO production, such as electron paramagnetic resonance spectroscopy [5], chemiluminescence [6], the oxyhemoglobin assay [7], and electrochemistry [8]. Because NO has a short half-life (2–30 s), it is preferable to determine nitrite (NO2¡), the stable end product of NO metabolism, which may be further oxidized to nitrate (NO3¡) in the presence of heme iron-containing proteins [1]; estimation of these anions is a common indirect method to monitor NO production in biological Xuids. The relative proportions of these two end products are variable and dependent on the origin of the sample, so that the best indication of the extent of NO metabolism can be obtained only when both anions are measured by

S. Arias-Negrete et al. / Analytical Biochemistry 328 (2004) 14–21

methods such as capillary electrophoresis [9], HPLC [10], or the Griess reaction [4]. QuantiWcation of nitrite by the Griess reaction is a simple and popular spectrophotometric method [4]. However, some of these methodologies require reduction of nitrate to nitrite. To reduce nitrate, the nicotinamide adenine dinucleotide phosphate (NADPH)-dependent nitrate reductase from Aspergillus niger and a formate-nitrate reductase prepared from Escherichia coli have been utilized. However, the high cost of the Aspergillus nitrate reductase is disadvantageous, and formate-nitrate reductase from E. coli requires several cofactors to reduce nitrate [4]. Moreover, cadmium sponge [11], vanadium (III) chloride [12], or copper–cadmium alloy [13] have also been utilized to reduce nitrate to nitrite. However, because of the toxicity of cadmium [14], the enzymatic reduction is preferred over the chemical reduction. The purpose of this work was to devise a procedure to prepare a lysate of E. coli containing nitrate reductase (NaR-lysate) to easily, inexpensively, and routinely reduce nitrate to nitrite for measuring by the Griess reaction total stable metabolites of NO accumulated in biological samples. E. coli MC1061 contains both nitrate and nitrite reductase (NaR and NiR). NaR of E. coli was found to be highly resistant to abrupt thermal changes, whereas NiR was found to be heat-labile, which resulted in advantages for our purposes. The NaR-lysate reduced 60–70% of the 1.6– 100 M nitrate to nitrite in 3 h at 37 °C. Residual NiR activity of the crude enzymatic preparation ranged from 15 to 25%. Nitrate reduction by the NaR-lysate is a simpler and more aVordable procedure to determine nitrate plus nitrite concentration in biological Xuids. Here we show that this preparation can be coupled to the Griess reaction to eVectively determine total nitrite in serum obtained from experimentally infected animals and in LPS-stimulated RAW 264.7 mouse macrophages.

Materials and methods All reagents utilized in this study were of analytical grade. Amebas Entamoeba histolytica trophozoites were cultured in screw-capped borosilicate tubes with TYI-S-33 medium at 37 °C for 72 h [15]. To harvest the amebas, culture tubes were chilled at 4 °C in an ice water bath for 5 min, centrifuged at 220g for 3 min, and then adjusted at 5 £ 106/ml to be used to inoculate hamsters. Bacteria E. coli MC1061 (ATCC 5338) was cultured in Luria Bertani broth at 37 °C at 200 rpm; bacteria were harvested

15

by centrifugation at 3200g for 20 min at 4 °C, washed twice with sterile phosphate-buVered saline (PBS; 14 mM phosphate buVer, 154 mM NaCl, pH 7.2), and resuspended in this buVer. This suspension was utilized to inoculate mice to induce in vivo nitrite production. Measurement of nitrite concentration by the Griess reaction For every determination of nitrite in the biological samples, or during standardization procedures, a nitrite standard curve was prepared in duplicate in a 96-well Xat-bottomed microtiter plate (Nunc); 100 l of Griess reagent [50 l of 1% sulfanilamide (prepared in 5% phosphoric acid) and 50 l 0.1% N-naphthylethylenediamine (prepared in milliQ water)] was added to 100 l nitrite solution (1.6–100 M Wnal concentration). The reaction mixture was incubated at room temperature for 15 min protected from light [4]. The optical density at 540 nm (OD540) of standard solutions and samples was measured in a microplate reader (Labsystems Multiskan MS, Helsinki, Finland). To determine nitrite plus nitrate in biological samples, 100 l of the sample was treated with the NaR-lysate of E. coli and then deproteinized. The Griess reagent was added and the detection was carried out as described above. Deproteinization Two procedures to deproteinizate the biological samples were performed. In the Wrst procedure organic solvents were used: 900 l methanol:diethylether (3:1) was added to 100 l of the biological sample and incubated at ¡20 °C. Serum samples were incubated overnight and cell culture supernatants for 1 h [16]. In the second procedure supernatants were deproteinized with 30% w/v ZnSO4 by incubating 100-l samples with 5 l ZnSO4 [17]. In both cases, to sediment the precipitated proteins, the samples were centrifuged at 12,000g for 10 min at 4 °C and nitrite concentration was determined in the supernatant. Preparation and standardization of the nitrate reductase lysate of E. coli MC1061 E. coli MC1061 was cultured overnight in Luria Bertani broth at 37 °C at 200 rpm. Bacteria were harvested by centrifugation, washed three times with sterile MilliQ water and once with PBS, and then frozen at ¡70 °C. To inactivate the NiR activity of the E. coli preparation, the bacteria suspension was subjected to approximately 16– 20 freeze–thawing cycles: the cell suspension was frozen at ¡70 °C for at least 60 min and afterward was immersed in a water bath at 60 °C, until the bacterial pellet had completely thawed. These cycles were repeated until NiR activity was 15–25%. The NaR remained

16


active. The NaR-lysate was kept in aliquots at ¡70 °C until use. The E. coli suspension contained 1.2 £ 1010 cells per millimeter, equivalent to 23.8 g/l of protein. To determine the amount of bacteria to be utilized in the reduction of nitrate, the following procedure was used: 100 l of 200 M sodium nitrate dissolved in PBS was incubated with increasing amounts of the NaR-lysate (200–1000 g of protein). The volume was adjusted to 200 l with PBS; the Wnal concentration was 100 M nitrate. Samples were incubated for 3 h at 37 °C and centrifuged at 14,000g; the nitrite concentration was determined in the supernatant by the Griess reaction. The remaining NiR activity in the NaR-lysate was quantiWed by using a 200 M sodium nitrite solution (in PBS) as described above. In a typical preparation of the NaRlysate the nitrate reduction was 60–70%, and the residual NiR activity was 025%. These values were taken into account to calculate the actual concentration of nitrite in the samples tested. Nitrate reduction in biological samples The crude preparation of NaR-lysate of E. coli MC1061 was utilized to enzymatically reduce nitrate in biological samples. To 100 l of serum or cell culture supernatants, 700 g of nitrate reductase lysate (35 l volume of 1.2 £ 1010 bacteria suspension) was added and incubated at 37 °C for 3 h, samples were deproteinized to perform the Griess reaction. For comparative purposes, reduction of nitrate by cadmium was performed according to OYcial Methods of Analysis of the Association of OYcial Analytical Chemists (AOAC) indications [11] with some minor modiWcations: To 100 l of standard solutions of nitrate (1.6–100 M) was added 30 l of 2.6% w/v ammonium chloride and 10 l of 2.1% w/v sodium tetraborate solution. Then cadmium sponge (0.4 g) was added and incubated for 90 min under agitation; afterward, an aliquot of 100 l was taken for nitrite determination. Before cadmium was added to the samples, pH was neutralized by repeated washes with distilled water, until pH was above 6.5 [18]. When reduction of nitrate by the use of cadmium was performed, the solutions for the standard curve were similarly prepared. Induction of in vivo and in vitro nitrite and nitrate production To determine nitrate and nitrite in biological samples in which NO production was expected, three distinct protocols using diVerent proinXammatory stimuli were performed: (1) Adult mice (Balb/c strain) were injected intravenously with a volume of 150 l containing 108 E. coli MC1061 strain cells. Twenty hours after inoculation, venous blood was obtained, and nitrate plus nitrite was determined in serum. (2) Adult hamsters (Mesocricetus auratus; 70 g weight) were inoculated directly into the

liver with 5 £ 105 E. histolytica HM1 strain trophozoites [19], and 6 days after inoculation, the animals were anesthetized to obtain blood by cardiac puncture. (3) Lipopolysaccharide-stimulated RAW 264.7 murine macrophages were cultured in Dulbecco’s modiWed Eagle’s medium supplemented with 10% fetal bovine serum (Gibco BRL 16000-044), 50 units/ml penicillin, and 50 g/ml streptomycin in a 5% CO2–95% air incubator at 37 °C. Cells were plated at 105 per well in sterile 96-well plates, the plates were incubated for 24 h, and before stimulation cultured medium was replaced by fresh medium. Murine macrophages were stimulated with 10 ng/ml LPS (from E. coli 0111:B4; Sigma L-3012) and LPS plus 50 units of mouse recombinant IFN- (r-IFN, Sigma I 4777) for 24 h. A cell-free aliquot of the supernatant was utilized to quantify nitrate and nitrite. Calculations The coeYcient of variation (CV) was calculated as (SD/mean)100 for the standard curve of nitrite [20]. Sigma Plot software version 2.0 was used to calculate mean + SE. Paired Student t test was employed to determine signiWcance (p) of increase in serum nitrite of infected animals or in cultured mouse macrophages.

Results and discussion The measurement of nitrate and nitrite in biological Xuids has been used as a correlation of NO synthesis [16,21,22]. The spectrophotometric determination of nitrate and nitrite in biological samples, by the Griess reagent, requires reduction of nitrate to nitrite [4]. The objective of this study was to evaluate the ability of the NaR-lysate to reduce nitrate to nitrite to determine total nitrite in biologic Xuids. The samples were Wrst enzymatically treated with the NaR-lysate, so that both the nitrite produced from the nitrate reduction and the nitrite contained in the sample were determined by the Griess reaction. According to the results, the standard determination of nitrite was linear when measured between 1.6 and 100 M (y D 0.01296x C 0.01482; coeYcient of correlation R2 was 0.9995; from seven independent determinations in duplicate; Fig. 1). All of the values from these determinations showed a slope of 0.01296 § 0.0006, with a CV of 4.6%. Considering that the molar absorptivity (), also known as the extinction coeYcient, is characteristic of the chemical absorbing species and dependent upon the wavelength of absorption but not upon concentration or path length, this coeYcient can be calculated from the slope of the standard curve of the nitrite [23] determined by the Griess reaction. Nitrite concentrations can be calculated by using the expression c D A/, where c is the concentration (M), A is the absorbance at 540 nm, and D 12900 M¡1cm¡1. The CV for the


17

Fig. 1. Linear regression of standard curves from nitrite determined by the Griess reaction, y D 0.0129x C 0.015; coeYcient of correlation R2 D 0.9995. Each point represent the mean § SD from seven independent determinations in duplicate. Solutions of sodium nitrite (0– 100 M) were used as standard.

determination of nitrite by the Griess reaction was 1.8– 3.9% for 3.12–100 mM nitrite, and for the lowest concentration of nitrite (1.6 M), the CV increased by up to 5.8%; these values are similar to another previously reported to quantify nitrite spectrophotometrically [16]. Most of the E. coli wild-type strains contain both nitrate and nitrite reductase [24] as E. coli MC 1061. This represented a problem for these test purposes, since the latter enzyme could interfere with the determination of nitrate and nitrite by lowering the actual nitrite concentration in the biological samples tested. Therefore, it was important to diminish the NiR of the lysate of E. coli. To diminish the NiR activity, the lysate was subjected to several (14–20) cycles of freeze–thawing until the NiR activity was reduced by 75–85% from its original level, and the NaR activity was 70–90% (Fig. 2A). Theses percentages varied depending upon bacterial batch preparation. Therefore, the NaR activity resisted thermal changes, whereas the NiR activity was unstable. Other bacterial preparations that conserved NaR required more than 16 freeze–thawing cycles (data not shown). It is important to note that if the freeze–thawing procedure were extended to enrich the NaR, both NaR and NiR could be inactivated (data not shown). Consequently, it was necessary to monitor NaR and NiR enzymatic activities after every 4 or 5 freeze–thawing cycles. These determinations needed to be done for each batch to do the appropriate corrections to quantify the nitrate and nitrite in the biological samples by the Griess reaction. The amount of NaR-lysate necessary to give the maximum reduction of nitrate and the least loss of nitrite was determined. Accordingly, NaR-lysate (300–1000 g

Fig. 2. Nitrate and nitrite reductase activity of E. coli MC 1061 lysate (NaR-lysate) after freeze–thawing cycles. (A) The E. coli suspension was frozen at ¡70 °C for at least 60 min and then immersed in a water bath at 60 °C until the pellet was completely thawed. Nitrate (NaR, 䊉) and nitrite (NiR, 䊏) reductase activity were tested before and after 5, 9, and 14 freeze–thawing cycles. NaR-lysate (500 g) was incubated with 100 M nitrate and nitrite solutions for 3 h at 37 °C. NaR activity was calculated as the synthesis of nitrite. NiR activity was calculated as a loss of nitrite from the standard solution. Nitrite was determined by the Griess reaction. (B) NaR-lysate (250–1000 g) was incubated with 100 M nitrate and nitrite solutions for 3 h at 37 °C. NaR-lysate received 14 freeze–thawing cycles. Reducing activity of NaR (䊉) and NiR (䊏) was calculated using 100 M nitrite as a reference solution. Representative results from four independent NaR lysate preparations.

protein) was added to 100 M sodium nitrate solution. It was found that 500 g of NaR-lysate reduced the nitrate by 90%. The residual NiR activity was 15–20% (Fig. 2B). These values must be taken into account to calculate nitrite + nitrate in the samples tested. In addition, if the concentration of NaR-lysate is increased, the NiR activity could mask the reduction of nitrate by giving a deceptively low percentage of reduction.

18


Nitrate and nitrite solutions (from 1.6 to 100 M) were incubated with 500 g of the NaR-lysate to determine NaR and NiR activities. It was found that the NaR activity reduced nitrate to nitrite up to 60%, and the NiR reduced the nitrite by 20%. These controls must be done for each determination to make the proper calculation considering both the NO3 conversion and the loss of NO2 (Fig. 3). The incubation time for the NaR-lysate to reduce nitrate was 3 h at 37 °C; the incubation time was reduced to 1 h and the results were comparable (data not shown). Importantly, no cofactors to be added for the nitrate reductase lysate to reduce nitrate were necessary. This indicates that any cofactors needed were possibly conserved in this enzymatic preparation. Since NaR-lysate is a bacterial suspension, it was important to determine whether the NaR activity was contained in the insoluble or in the soluble fraction. Consequently, the bacterial suspension was subjected to centrifugation (12,000g, for 10 min). The cell pellet and the supernatant were separated, and the NaR activity was evaluated; 95% of the NaR activity was found in the pellet, whereas in the supernatant only 5% of its activity was detected. This suggests that the NaR activity is contained in the insoluble fraction of the bacterial lysate. To compare the reducing ability of the bacterial lysate with that of cadmium, 100 M Nitrate was treated with Cd; a reduction of 65 § 9% was attained (Table 1).

Fig. 3. Nitrate and nitrite reductase activity of NaR-lysate. NaRlysates were subjected to 14 and 20 freeze–thawing cycles. Five hundred micrograms of NaR-lysate was added to 100 l of 0–100 M nitrate and nitrite solutions and incubated at 37 °C for 3 h. After centrifugation at 14,000g, the supernatants were separated. Determination of nitrite was done by the Griess reaction (䉱) y D 0.01141x C 0.0195, R2 D 0.9972. NaR activity was determined as synthesis of nitrite (䊉) y D 0.00881x C 0.00623, R2 D 0.9983. NiR activity was calculated as loss of nitrite of the standard solution (䊏) y D 0.005324x C 0.01706, R2 D 0.928. The results represent the average § SD of three independent batches of NaR-lysate.

Table 1 Conversion eYciency of nitrate reductase lysate from Escherichia coli and cadmium sponge Experimental condition Biological reduction Nitrite (100 M) + None Nitrate (100 M) + NaR Lysate Nitrite (100 M) + NaR Lysate Chemical reduction Nitrate (100 M) + None Nitrate (100 M) + cadmium

Absorbance at 540 nm

Percentage of conversion

1.22 § 0.13 0.7 § 0.15 0.86 § 0.09

¡ 56.6 § 12.5 (NO¡ 3 to NO2 )

29 § 7.6 (Loss of NO¡ 2)

1.02 § 0.01 0.673 § 9.8

65 § 9

These values represent the mean § SE of four independent experiments performed in duplicate. Nitrite was determined by a standard curve as described under Materials and methods.

Our results agree with those reported [18]. The results of our study indicate that both enzymatic and chemical reduction are comparable, but enzymatic reduction is preferred, thereby avoiding the concern for cadmium toxicity [14] and its disposal. Alternative methods have been described for reduction of nitrate through copper– cadmium alloy followed by color development by the Griess reagent [13]; similarly vanadium (III) was utilized to reduce nitrate [12]. The nitrate reductase from Aspergillus species and that of Pseudomonas oleovorans, in addition to the formate-nitrate reductase of E. coli, have been utilized to eYciently reduce nitrate in biological samples [4,24]. The preparation of NaR from diVerent denitrifying microorganisms is time consuming and involves many biochemical procedures [4,25]. The method used in this work represents an interesting alternative when a large number of samples are being processed and a great amount of enzyme is needed, or, in some cases, where the pure enzymes are not aVordable, or the bacteria to prepare the nitrate reductase are not easily available. Some wildtype strains of E. coli contain formate-nitrite reductase that interferes with nitrite quantiWcation, giving an artiWcially low concentration. Moreover, a nitrate-reductasepositive, nitrite-reductase-deWcient strain of E. coli is also recommended, but not easily available [4]. NaR from E. coli MC1061 used in this study was very resistant to thermal inactivation since an abrupt temperature shift from ¡70 to 60 °C repeated from 16 to 20 times did not signiWcantly aVect its activity, but it did reduce NiR activity. If the temperature required to melt the cell suspensions is increased, the result will be the loss of both NaR and NiR activities. The NaR present in the bacterial lysate was stable for more than 6 months when frozen in aliquots at ¡70 °C. Interestingly, Alcaligenes sp. nitrite reductase activity was enhanced by freeze–thawing cycles [26], whereas nitrate reductase from barley


Hordeum vulgare leaves showed low thermostability [27], and the archaeon Pyrobaculum aerophilum has a thermostable nitrate reductase [28]. Moreover, several physicochemical factors account for denaturation of enzymes through freeze–thawing cycles, such as aggregation of proteins, denaturing by highly concentrated salts during freezing, pH change during freezing, or variation in the state of hydration of the protein through the change phase of solvent water by freezing (reviewed by [26]). From the present study, it is suggested that NiR is possibly inactivated by some of these mechanisms, while NaR is more resistant. LPS from gram-negative bacteria, cytokines, some parasites, ovarian steroids, and catecholamines induce in vivo NO production for host defense and physiological processes [18,29–31]. In general, nitrite levels in serum are low, and the proportion of nitrite and nitrate synthesized varies depending on the origin of the sample [18]. For these reasons, and in many studies, nitrate plus nitrite concentrations are determined as indicators of endogenous NO production [32]. In humans and some other vertebrates serum nitrite concentration ranges from 3 to 8 M as spectrophotometrically determined [13]. It was also found that nitrite accounts for 10–40% of nitrite + nitrate synthesized, depending on the animal species tested [13]. To evaluate NaR-lysate reducing activity, the nitrate + nitrite concentration was determined in some biological samples. Accordingly, mice were experimentally infected with E. coli, and hamsters were injected in the liver with E. histolytica trophozoites. Serum from these animals was collected and treated with the NaR-lysate, and nitrite + nitrate levels were determined. Normal mice had 17 § 1 M serum nitrite + nitrate. Mice injected intravenously with E. coli showed, after 24 h postinoculation, a serum nitrite + nitrate level of 58.6 § 15.7 M, indicating a 150% increase (p 0 0.0254) (Table 2). Hamsters infected with E. histolytica trophozoites directly in the liver had 30.7 § 2.4 M serum nitrite + nitrate, whereas in normal hamsters nitrate + nitrite was 20 § 1.6 M, indicating an increase of 50% (p 0 0.025) (Table 2), which agrees with previous reports about NO production in experimental amoebiasis [17].

Table 2 Serum nitrate + nitrite (M) in experimentally infected animals Species/infection

Normal

Infecteda

Mice + E. coli

17 § 1

Hamster + E. histolytica

20 § 1.6

58.6 § 16 (150%) (p 0 0.025) 30.7 § 2.4 (50%) (p 0 0.025)

a Each value represents the mean § SE of serum nitrite + nitrate levels of Wve infected animals and six noninfected controls. Serum nitrite for mice and hamsters is undetectable. The numbers in parentheses indicate the percentage increase in the infected animals as compared with the uninfected controls; p was determined by Student’s t test.

19

No serum nitrite in normal hamsters or mice was detected in this study by the Griess reagent. This Wnding indicates the importance of determining nitrite + nitrate levels in blood and plasma. The role of NO has been determined in pulmonary infection caused by Cryptococcus neoformans. Serum nitrite + nitrate levels ranging from 100 to 200 M, depending on the course of infection, were reported [33]. In humans, serum nitrite concentration ranges from 0.55 to 4.2 M and nitrate from 10.3 to 47.8 M; these concentrations were determined by using diVerent methodologies, and the samples were subjected to diVerent pretreatments, which could explain that variability [reviewed in 10]. Moreover, it was observed that serum nitrate levels were higher in old patients with cardiovascular diseases (44 M) and lower (20 M) in healthy adults. Both groups had the same serum nitrite level (1.4 M). This result suggests that oxidation of nitrite to nitrate could depend on health status and age [10]. Patients infected with virus dengue fever showed increased nitrite and nitrate serum levels (50 M) as compared with healthy controls (25 M), whereas patients with dengue hemorrhagic fever had nitrite + nitrate serum levels similar to healthy persons [34]. Studies on the participation of NO in human diseases, experimental infections, and immune response in vitro indicate the importance of determining nitrite + nitrate in serum or culture supernatants. The proportion of nitrite and nitrate derived from in vitro NO synthesis varies depending on the biological system tested. Mouse macrophages from the C3H/He strain which are LPS-sensitive synthesized nitrate and nitrite, whereas the LPS-resistant C3H/HeJ macrophages produced low levels of nitrate but no nitrite [35]. To determine whether the NaR-lysate could be useful to indirectly detect in vitro NO synthesis by immune cells stimulated with proinXammatory substances, the RAW 264.7 mouse macrophages were stimulated with LPS or LPS plus rIFN-. In the supernatants of cultured mouse macrophages, the direct measurement of nitrite was performed, and the following concentrations were obtained: in nonstimulated macrophages (control) the nitrite concentration was 2 M, whereas in LPS-stimulated macrophages the nitrite concentration was 28 M, indicating a 13-fold increase. The LPS + rIFN--stimulated cells had 50 M nitrite, indicating a 24-fold stimulation with respect to nonstimulated macrophages (Table 3). When these supernatants were treated with the NaR-lysate, and the nitrite + nitrate concentration was determined, it was found that nonstimulated macrophages had 2 M total nitrite. In the LPS-stimulated cells the nitrite + nitrate levels were 28 M, a 13.7-fold increase with respect to nonstimulated macrophages, and that for LPS plus rIFN- (70 M nitrate + nitrite) was a 27-fold augmentation (Table 3). In the nonstimulated macrophages and in those stimulated with LPS, only nitrite was produced. Treatment with the NaR-lysate did not increase

20


Table 3 Determination of nitrite and nitrite + nitrate production by cultured RAW 264.7 mouse macrophages stimulated with LPS and rIFN- Experimental condition

Nitritea (M/L)

No stimulation LPS LPS + rIFN-

2 § 0.2 28.7 § 2 49.5 § 1.3

Stimulation (Fold)

Nitrate + Nitriteb (M/L)

Stimulation (Fold)

13.8 23.9

2.16 § 2.1 28.2 § 3.9 59.3 § 27.4

13.7 27.5

Cells cultured in Dulbecco’s modiWed Eagle’s medium (DMEM) plus 10% fetal calf serum were stimulated for 24 h with 10 ng/ml LPS or 10 ng/ml LPS plus 50 U rIFN-. Nitrite concentration was determined with the Griess reagent before (a) and after (b) reduction of supernatant with NaRlysate (nitrate + nitrite). Nitrite was not detected in complete DMEM but contained nitrate (10.3 M), this concentration was subtracted from the supernatants treated with NaR-lysate. Each value represents the mean § SE of a representative experiment performed in triplicate. DiVerences among control and stimulation were statistically signiWcant; p 0 0.001 was calculated by Student’s t test.

these values. These results indicate that NO produced by mouse macrophages is mostly converted to nitrite when stimulated with LPS alone, whereas stimulation with LPS + IFN- induced production of nitrite and a low amount of nitrate (10 M approximately). There are reports on in vitro stimulation of macrophages where only nitrite was determined in the supernatant [31,36]. The nitrite values reported are in the same range determined in this study. QuantiWcation of NO end products in humans or other animal serum has shown that nitrite accounts for 10–20% and that 80–90% is metabolized to nitrate [10]. Nitrite is oxidized to nitrate in the liver [24]. Therefore, it is highly recommended to reduce nitrate to nitrite to determine the concentration of nitrite (nitrite + nitrate) in biological samples to actually measure NO production. In some studies the production of NO is determined as the accumulation of nitrite (NO2), the end product of NO metabolism [31], whereas in others a nitrate reduction is performed and the results are reported as nitrate + nitrite or NOx products [33]. Moreover, our data obtained in this work demonstrate that the NaR-lysate is a good option to enzymatically reduce nitrate to nitrite; otherwise, basal nitrite levels can be underestimated because some percentage of the synthesized NO could be converted to nitrate. Proteins can interfere with spectrophotometic determination of nitrite by the Griess reagent. Deproteinization of biological samples can be accomplished by either methanol/ether or zinc sulfate [16,17]. The use of organic solvents dilutes 10 times the biological samples, whereas zinc sulfate dilutes the sample by less than 5%. The advantage of precipitation with zinc sulfate over methanol/ ether is the immediate precipitation at room temperature. There are more sensitive methods to directly detect NO or its stable end products; however, their utilization requires complex equipment [37,38]. The spectrophotometric determination of nitrite by the Griess reaction was satisfactory for our purposes. It is advantageous that NaR of E. coli was found to be thermoresistant, and the NiR was thermolabile. In such a way the use of E. coli MC1061 as a source of NaR proved to be a good alternative for performing nitrate reduction in standard solutions and in biological samples and for determining nitrate plus nitrite anions as a

result of NO synthesis. The E. coli MC1061 crude preparation represents a convenient, aVordable, and reliable method to obtain the NaR and indirectly measure NO production in vivo and in in vitro models where experimental infections or proinXammatory substances are employed.

Acknowledgments We are grateful to Dr. J. García-Soto for his valuable comments. We also thank O.H. García-Negrete and J.A. Anguiano-Torres for excellent technical support. J. Ramírez-Emiliano is a recipient of a scholarship from the Mexican Council of Science and Technology (CONACYT), and M.O. Solís-Martínez is a recipient of a scholarship from the Council of Science and Technology of Guanajuato (CONCYTEG). We also thank James H. Bieshaar for his editorial help. This work was supported by Grant 27843N from CONACYT, México and the “Programa de Investigación Institucional 2000” de la Universidad de Guanajuato, México.

References [1] S.S. Gross, M.S. Wolin, Nitric oxide: pathophysiological mechanisms, Annu. Rev. Physiol. 57 (1995) 737–769. [2] O.W. GriYth, D.J. Stuehr, Nitric oxide synthases: properties and catalytic mechanism, Annu. Rev. Physiol. 57 (1995) 707–736. [3] H. Luss, N.C. Nussler, H.G. Beger, A.K. Nussler, Expression and detection of inducible nitric oxide synthase in experimental models of inXammation, Methods 10 (1996) 51–60. [4] M.A. Titheradge, The enzymatic measurement of nitrate and nitrite, Methods Mol. Biol. 100 (1998) 83–91. [5] S.T. Ohnishi, Measurement of NO using electron paramagnetic resonance, Methods Mol. Biol. 100 (1998) 129–153. [6] E.D. Michelakis, L.A. Archer, The measurement of NO in biological systems using chemiluminescence, Methods Mol. Biol. 100 (1998) 111–127. [7] M. Salter, R.G. Knowles, Assay of NOS activity by the measurement of conversion of oxyhemoglobin to methemoglobin by NO, Methods Mol. Biol. 100 (1998) 61–65. [8] R. Kikura-Hanajiri, R.S. Martin, S.M. Lunte, Indirect measurement of nitric oxide production by monitoring nitrate and nitrite using microchip electrophoresis with electrochemical detection, Anal. Chem. 74 (2002) 6370–6377.

S. Arias-Negrete et al. / Analytical Biochemistry 328 (2004) 14–21 [9] E. Morcos, N.P. Wiklund, Nitrite and nitrate measurement in human urine by capillary electrophoresis, Electrophoresis 22 (2001) 2763–2768. [10] V. Jedlickova, Z. Paluch, S. Alusik, Determination of nitrate and nitrite by high-performance liquid chromatography in human plasma, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 780 (2002) 193–197. [11] G.H. Richardson, in: K. Helrich (Ed.), OYcial Methods of Analysis of the Association of OYcial Analytical Chemists, Vol. 2, Association of OYcial Analytical Chemists, Inc., Arlington, VA, 1990, pp. 843–844. [12] K.M. Miranda, M.G. Espey, D.A. Wink, A rapid, simple spectrophotometric method for simultaneous detection of nitrate and nitrite, Nitric Oxide 5 (2001) 62–71. [13] K.V.H. Sastry, R.P. Moudgal, J. Mohan, J.S. Tyagi, G.S. Rao, Spectrophotometric determination of serum nitrite and nitrate by copper-cadmium alloy, Anal. Biochem. 306 (2002) 79–82 (doi:10.1006/abio.2002.5676.). [14] R.A. Goyer, in: C.D. Klaassen (Ed.), Casarett and Doull’s Toxicology: The basic science of poisons, McGraw-Hill, New York, 1996, pp. 699–702. [15] L.S. Diamond, D.R. Harlow, C.C. Cunnick, A new medium for the axenic cultivation of Entamoeba histolytica and other Entamoeba, Trans. R. Soc. Trop. Med. Hyg. 72 (1978) 431–432. [16] I. Guevara, J. Iwanejko, A. Dembinska-Kiec, J. Pankiewickz, A. Wanat, P. Anna, I. Golabek, S. Bartus, M. Malczewska-Malec, A. Szcudlik, Determination of nitrite/nitrate in human biological material by the simple Griess reaction, Clin. Chim. Acta 274 (1998) 177–188. [17] J. Pacheco-Yepez, R. Campos-Rodriguez, M. Shibayama, J. Ventura-Juarez, J. Serrano-Luna, V. Tsutsumi, Entamoeba histolytica : production of nitric oxide and in situ activity of NADPH diaphorase in amebic liver abscess of hamsters, Parasitol. Res. 87 (2001) 49–56. [18] R.L. van Bezooijen, I. Que, A.G.H. Ederveen, H.J. Kloosterboer, S.E. Papapoulos, C.W.G.M. Löwik, Plasma nitrate + nitrite levels are regulated by ovarian steroids but do not correlate with trabecular bone mineral density in rats, J. Endocrinol. 159 (1998) 27–34. [19] V. Tsutsumi, R. Mena-Lopez, F. Anaya-Velazquez, A. MartinezPalomo, Cellular bases of experimental amebic liver abscess formation, Am. J. Pathol. 117 (1984) 81–91. [20] D. Kealey, P.J. Haines, Instant notes: Analytical Chemistry, BIOS ScientiWc Publishers, Farmham, UK, 2002. [21] H. Baba, I. Kawamura, C. Kohda, T. Nomura, Y. Ito, T. Kimoto, I. Watanabe, S. Ichiyama, M. Mitsuyama, Induction of gamma interferon and nitric oxide by truncated pneumolysin that lacks pore-forming activity, Infect. Immun. 70 (2002) 107–113. [22] J.B.J. Hibbs, C. Westenfelder, R. Taintor, Z. Vavrin, C. Kablitz, R.L. Baranowski, J.H. Ward, R.L. Menlove, M.P. McMurry, J.P. Kushner, et al., Evidence for cytokine-inducible nitric oxide synthesis from L-arginine in patients receiving interleukin-2 therapy, J. Clin. Invest. 89 (1992) 866–877. [23] D. Sutton, Electronic spectra of transition metal complexes, McGraw-Hill, London, 1968. [24] D.L. Granger, J.B.J. Hibbs, L.M. Broadnax, Urinary nitrate excretion in relation to murine macrophage activation. InXuence of die-

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37] [38]

21

tary L-arginine and oral NG-monomethyl-L-arginine, J. Immunol. 146 (1991) 1294–1302. H. Uchimura, H. Enjoji, T. Seki, A. Taguchi, N. Takaya, H. Shoun, Nitrate reductase-formate dehydrogenase couple involved in the fungal denitriWcation by Fusarium oxysporum, J. Biochem. (Tokyo) 131 (2002) 579–586. M. Masuko, H. Iwasaki, T. Sakurai, S. Suzuki, A. Nakahara, EVects of freezing on puriWed nitrite reductase from a denitriWer, Alcaligenes sp. NCIB 11015, J. Biochem. Tokyo 98 (1985) 1285– 1291. J.L. Serra, J.M. Ibarlucea, J.M. Arizmendi, M.J. Llama, PuriWcation and properties of the assimilatory nitrite reductase from barley Hordeum vulgare leaves, Biochem. J. 201 (1982) 167–170. S. Afshar, E. Johnson, S. de Vries, I. Schroder, Properties of a thermostable nitrate reductase from the hyperthermophilic archaeon Pyrobaculum aerophilum, J. Bacteriol. 183 (2001) 5491–5495. S. Majumdar, R. Gupta, N. Dogra, Interferon-gamma- and lipopolysaccharide-induced tumor necrosis factor-alpha is required for nitric oxide production: tumor necrosis factor-alpha and nitric oxide are independently involved in the killing of Mycobacterium microti in interferon-gamma- and lipopolysaccharide-treated J774A.1 cells, Folia Microbiol. (Praha) 45 (2000) 457–463. M. Qadoumi, I. Becker, N. Donhauser, M. RollinghoV, C. Bogdan, Expression of inducible nitric oxide synthase in skin lesions of patients with american cutaneous leishmaniasis, Infect. Immun. 70 (2002) 4638–4642. D. Chi, M. Qui, G. Krishnaswamy, C. Li, W. Stone, Regulation of nitric oxide production from macrophages by lipopolysaccharide and catecholamines, Nitric Oxide 8 (2003) 127–132. C.P. Simmons, N.S. Goncalves, M. Ghaem-Maghami, M. BajajElliott, S. Clare, B. Neves, G. Frenkel, G. Dougan, T.T. MacDonald, Impaired resistance and enhanced pathology during infection with a noninvasive, attaching-eVacing enteric bacterial pathogen, Citrobacter rodentium, in mice lacking IL-12 or IFNgamma, J. Immunol. 168 (2002) 1804–1812. J. Rivera, J. Mukherjee, L.M. Weiss, A. Casadevall, Antibody eYcacy in murine pulmonary Cryptococcus neoformans infection: a role for nitric oxide, J. Immunol. 168 (2002) 3419–3427. N. Valero, L.M. Espina, G. Añez, E. Torres, J.A. Mosquera, Short report: increased level of serum nitric oxide in patients with dengue, Am. J. Trop. Med. Hyg. 66 (2002) 762–764. D.J. Stuehr, M.A. Marletta, Mammalian nitrate biosynthesis: mouse macrophages produce nitrite and nitrate in response to Escherichia coli lipopolysaccharide, Proc. Natl. Acad. Sci. USA 82 (1985) 7738–7742. X. Zhou, D. Potoka, P. Boyle, E. Nadler, K. McGinnis, H. Ford, Aminoguanidine renders inducible nitric oxide synthase knockout mice more susceptible to Salmonella typhimurium infection, FEMS Microbiol. Lett. 2 (2002) 93–97. A.M. Komarov, In vivo detection of nitric oxide distribution in mice, Mol. Cell. Biochem. 234–235 (2002) 387–392. J. Woitzik, N. Abromeit, F. Schaefer, Measurement of nitric oxide metabolites in brain microdialysates by a sensitive Xuorometric high-performance liquid chromatography assay, Anal. Biochem. 289 (2001) 10–17.