Calibration of nitric oxide flux generation from diazeniumdiolate NO donors

Nitric Oxide 21 (2009) 69–75

Contents lists available at ScienceDirect

Nitric Oxide journal homepage: www.elsevier.com/locate/yniox

Analytical methods

Calibration of nitric oxide flux generation from diazeniumdiolate NO donors Qian Li a,d, Jack R. Lancaster Jr. a,b,c,d,* a

Department of Anesthesiology, University of Alabama at Birmingham, Birmingham, AL 35294-2172, USA Department of Physiology & Biophysics, University of Alabama at Birmingham, Birmingham, AL 35294-2172, USA Department of Environmental Health Sciences, University of Alabama at Birmingham, Birmingham, AL 35294-2172, USA d Center for Free Radical Biology, University of Alabama at Birmingham, Birmingham, AL 35294-2172, USA b c

a r t i c l e

i n f o

Article history: Received 14 January 2009 Revised 2 April 2009 Available online 12 April 2009 Keywords: Nitric oxide donor Diazeniumdiolate NONOate Oxymyoglobin assay Oxyhemoglobin assay Calibration

a b s t r a c t The 1-(secondary amino) diazen-1-ium-1,2-diolates (NONOates) are the most commonly utilized nitric oxide (NO, nitrogen monoxide) donor because of the ability of different NONOates to spontaneously break down liberating NO at different rates, which can be utilized to control NO fluxes. However, the parameters that determine these fluxes of NO generation, half-lives and stoichiometry of NO per donor, can vary significantly with specific experimental conditions in addition to the donor chosen. Here we report straightforward methods that can be used to determine these parameters. For donors of intermediate half-life (10–80 min) a real-time oxymyoglobin (oxyMb) assay can be analyzed to simultaneously determine both the half-life and the total amount of NO liberated, from which the NO flux can be obtained for any given donor concentration. The half-lives obtained by oxyMb assay are very similar to those obtained by following NONOate decomposition kinetics spectrophotometrically, and a survey of several NONOates from different commercial sources show consistent results. These data provide validation for the methodologies employed. In addition, procedures are described for calibration of donors with shorter (<10 min) and longer (>80 min) half-lives. These procedures can be used to reproducibly and routinely calibrate NO fluxes for a variety of donors under any specific condition. Ó 2009 Published by Elsevier Inc.

Introduction Nitric oxide (NO, nitrogen monoxide), a small diatomic molecule, is involved in essentially every aspect of physiology and pathophysiology [1]. In order to study its novel biological functions, several classes of NO donors have been developed and used as exogenous sources of NO [2,3]. Under biological conditions, NO is produced as a flux of varying rates, making it essential to have donors that also generate NO at varying (and controllable) rates. The 1-(secondary amino) diazen-1-ium-1,2-diolates (diazeniumdiolates, NONOates)1 are without doubt the NO donor class of choice because these compounds spontaneously liberate NO with first-order kinetics [4]. In addition, NONOates have half-lives of NO generation ranging from seconds to hours depending on the secondary amino group, and therefore can provide a wide range of NO exposure time to simulate various biological environments. The two characteristics for a given NONOate that determine the  NO flux are the rate constant (or half-life, t1/2) of NO liberation and the total amount of NO liberated after complete decomposition

* Corresponding author. Fax: +1 205 934 7437. E-mail address: [email protected] (J.R. Lancaster). 1 The 1-(secondary amine) NONOates are primarily donors of NO, while primary amine NONOates release nitroxyl as well [25]. 1089-8603/$ - see front matter Ó 2009 Published by Elsevier Inc. doi:10.1016/j.niox.2009.04.001

(which we will refer to as ‘‘NO content”). The NO content is given by the NONOate concentration ([D]0, concentration at zero time) multiplied by the stoichiometry of NO per donor (which we refer to as ‘‘a”), thus equal to a[D]0. Several physicochemical properties for a variety of NONOates have been summarized previously, including t1/2, a and spectrophotometric properties [5]. Theoretically, by utilizing published values for t1/2, a, and spectrophotometric measurement of [D]0, the NO flux generated from a given concentration of NONOate can be calculated. However, as has been indicated previously [5,6], specific experimental conditions significantly affect the rate and extent of NO formation (e.g., temperature, pH, ionic strength, and also presence of membranes [7]). In this context, it is important to standardize and calibrate NONOates from batch to batch under individual laboratory conditions, ideally utilizing a convenient method [6,8,9]. The oxyhemoglobin/oxymyoglobin (oxyHb/oxyMb) assay has been extensively used for measuring NO release in vitro and in vivo [10,11]. We describe here an application of this method and kinetic analysis which is a straightforward procedure that can be applied routinely under individual experimental conditions and allows simultaneous determination of both the rate constant and the NO content of a NONOate. To validate this method, we examine NONOates from different commercial sources, and also compare the results to those obtained by another method (NONOate decomposition kinetics).

70

Q. Li, J.R. Lancaster Jr. / Nitric Oxide 21 (2009) 69–75

Experimental procedures Materials Myoglobin (Mb) from equine skeletal muscle, diethylenetriamine pentaacetic acid (DTPA) and sodium hydrosulfite (Na2S2O4) were purchased from Sigma–Aldrich, Inc. (St. Louis, MO, USA). Sodium hydroxide (NaOH) and sodium phosphate were obtained from Fisher Scientific (Fair Lawn, NJ, USA). Disodium 1-[2-(carboxylato)pyrrolidin-1-yl]diazen-1-ium-1,2-diolate-methanol (PROLI/ NO) was from Alexis Biochemicals (San Diego, CA, USA). (Z)-1-[N, N-diethylamino]diazen-1-ium-1,2-diolate-diethylammonium salt (DEA/NO), (Z)-1-[N-(3-aminopropyl)-N-(n-propyl)amino]diazen-1ium-1,2-diolate (PAPA/NO) and (Z)-1-[N-(3-aminopropyl)-N-(i-propyl)amino]diazen-1-ium-1,2-diolate (NOC-5) were supplied by Alexis Biochemicals (San Diego, CA, USA), Dojindo Laboratories (Kumamoto, Japan), Cayman Chemical (Ann Arbor, MI, USA) or Sigma–Aldrich, Inc. (St. Louis, MO, USA) as indicated. Aliquots of NONOates in 10 mM NaOH were used as stock solutions and were stable for months at 80 °C. Preparation of oxyMb The preparation of oxyMb from Mb has been described in detail [10]. Briefly, excess sodium hydrosulfite was added to Mb solution in 100 mM sodium phosphate buffer containing 100 lM DTPA, pH 7.4 (working buffer). The color change of the solution from dark brown to wine-red indicates the reduction of metmyoglobin (metMb) and the formation of deoxymyoglobin (deoxyMb). After two sequential desalting steps by PD10 column containing SephadexTM G-25 Medium from GE Healthcare (Uppsala, Sweden), deoxyMb was oxygenated and the concentration of oxyMb was determined spectrophotometrically by averaging the values from three wavelengths (e418nm = 128 mM1cm1, e543nm = 13.6 mM1cm1 and e581nm = 14.6 mM1cm1) [10,12]. Determination of working wavelengths and conversion extinction coefficients The working wavelengths and the molar extinction coefficients for the conversion of oxyMb to metMb (De = eoxyMb  emetMb) were measured as described previously for hemoglobin [10]. Known concentrations of oxyMb were oxidized completely by excess NO generated from PROLI/NO (half-life as short as 1.8 s at pH 7.4, and 37 °C [13]). In the visible region, the absorbance changes at 582 and 545 nm were found highest, therefore, these two wavelengths were chosen as the working wavelengths. Other wavelengths with high differences such as the Soret (405 nm) can also be applied [10,11]. The absorbance changes at working wavelengths were plotted vs. oxyMb concentrations, and the conversion extinction coefficients were derived from the slope of the straight line obtained by linear regression fit. De = 11.50 ± 0.03 and 8.60 ± 0.03 mM1 cm1 (mean ± SD, n = 3) were determined for 582 and 545 nm, respectively in working buffer.

sition, the absorbance decrease of 12.5–25 lg/ml NONOate at 250 nm was followed. Data were analyzed by Microcal (TM) Origin (Version 6.0 from Microcal Software, Inc., Northampton, MA, USA) using non-linear regression function BoxLucas1 as described in the Results Section [14]. At least five half-lives of data were used for analysis. Statistics Data in Table 1 are presented as mean ± SD from at least three independent measurements. Values from this work in Table 2 were obtained from all measurements listed in Table 1 without distinguishing different vendors. SD for a was calculated by SD(a)2/ Mean(a)2 = SD(a[D]0)2/Mean(a[D]0)2 + SD([D]0)2/Mean([D]0)2, where SD(a), SD(a[D]0) and SD([D]0) are the SD for a, a[D]0 and [D]0, respectively; Mean(a), Mean(a[D]0) and Mean([D]0) are the mean values for a, a[D]0 and [D]0, respectively. Results Standardization and calibration of NO content and half-life by oxyMb assay The oxyHb assay utilizes the facile NO reaction with oxyHb yielding methemoglobin (metHb) and nitrate (rate constant k2 3.4  107 M1 s1) [15] and is commonly used to determine NO by measuring oxyHb conversion to metHb spectrophotometrically [10]. With proper data analysis this method can be used to simultaneously determine both t1/2 and a[D]0 for a NONOate. The method involves following oxyMb2 conversion to metMb after addition of NONOate. Under these conditions, the reactions taking place are given by Eqs. (1) and (2): k1

D ! a NO þ P

ð1Þ k2

NO þ oxyMb ! metMb þ NO3

ð2Þ



where D is the NO donor (NONOate here), P is the product of NONOate decomposition, a is the stoichiometry of NO per donor, and k1 and k2 are the rate constants of D decomposition and NO/ oxyMb reaction, respectively. The concentration of donor at any time is given by Eq. (3):

½D ¼ ½D0 

½oxyMb0  ½oxyMb

a

ð3Þ

where [D]0 and [oxyMb]0 are the initial concentrations of D and oxyMb, respectively. The rate of change of oxyMb is given by Eq. (4):

d½oxyMb ¼ k2 ½NO½oxyMb dt

ð4Þ

For the derivation we make the steady-state assumption3 [16,17]:

d½NO ¼ ak1 ½D  k2 ½NO½oxyMb ¼ 0 dt

ð5Þ

and combining with Eq. (4), Kinetic measurements All kinetic measurements were performed at 25 °C in working buffer using a Shimadzu UV-2501PC Spectrophotometer with temperature control (Kyoto, Japan). For the oxyMb assay, 995 ll of 50– 60 lM oxyMb solution was placed in both reference and sample cuvettes. After the baseline scan, 5 ll of 0.5–1 mg/ml NONOate in 10 mM NaOH was added into sample cuvette, while 5 ll of 10 mM NaOH was added into reference cuvette. The absorbance changes at both 582 and 545 nm were followed immediately after quick addition and mixing. For the kinetics of NONOate decompo-

2 We utilize Mb because, unlike hemoglobin, it does not possess cysteine, and thus avoids the potential complication of additional chemistry such as nitrosothiol formation [26]. 3 The steady-state assumption is based on the phenomenon that very soon after initiation of the reaction the rate of consumption will increase due to increasing [NO]. Within a short period of time this consumption rate will reach a point where it will equal the rate of production. After this point the [NO] will not change but will be very low until the completion of either reaction (Eqs. (1) and (2)), because oxyMb scavenging is much faster than the NO generation, that is, NO is trapped by oxyMb once it is produced from donor.

71

Q. Li, J.R. Lancaster Jr. / Nitric Oxide 21 (2009) 69–75 Table 1  NO release parameters of NONOates from different commercial sources. NONOate (5 mg/ml)

Vendor

Calibration methods Spectrophotometry [D]0 (mM)

d

a

Donor decompositionb

OxyMb assayb

t1/2 (min)

a[D]0 (mM)

t1/2 (min)

ac

DEA/NO (24.24 mM)

Alexis Cayman Calbiochem Sigma

23.80 ± 0.01 22.33 ± 0.96 23.16 ± 1.63 24.44 ± 0.68

9.92 ± 0.05

48.30 ± 1.46 46.90 ± 2.26 47.92 ± 0.54 47.84 ± 1.80

10.06 ± 0.21 10.13 ± 0.39 10.06 ± 0.09 10.18 ± 0.24

2.03 ± 0.06 2.10 ± 0.14 2.07 ± 0.15 1.96 ± 0.09

PAPA/NO (28.38 mM)d

Alexis Cayman

28.89 ± 0.17 30.94 ± 1.20

61.71 ± 0.39

44.85 ± 0.73 48.01 ± 1.19

61.98 ± 0.27 61.81 ± 2.24

1.55 ± 0.03 1.55 ± 0.07

NOC-5 (28.38 mM)d

Alexis Calbiochem Dojindo Sigma

28.01 ± 0.30 27.68 ± 0.13 27.08 ± 0.15 29.20 ± 0.52

80.18 ± 0.12

48.53 ± 0.20 48.74 ± 0.95 46.39 ± 1.18 50.32 ± 1.88

79.02 ± 1.05 79.99 ± 1.06 79.04 ± 1.83 79.98 ± 2.04

1.73 ± 0.02 1.76 ± 0.04 1.71 ± 0.04 1.72 ± 0.07

Data are presented as mean ± SD from at least three independent measurements. a In 10 mM NaOH; calculated using eD for PAPA/NO reported by Hrabie et al. [18], DEA/NO and NOC-5 by Sigma (listed in Table 2). b In 100 mM sodium phosphate buffer containing 100 lM DTPA at pH 7.4 (working buffer) and 25 °C. c Calculated by a[D]0/[D]0 (Column 5/Column 3). d Concentration calculated by mass/molecular weight.

Table 2 Physicochemical data for NONOates reported by others and this work. NONOate

Report

kmax (nm) (eD (mM1 cm1))

22–25 °C DEA/NO

PAPA/NO

NOC-5

Alexis Cayman Calbiochem Sigma [8] [5] [9] [6] This work Alexis Cayman [18] [5] This work Alexis Calbiochem Dojindo Sigma [18] This work

250 (6.5)

a

t1/2 (min) 22 °C

16a 16a

37 °C

25 °C

2a 2a

1.5

2.1e 2[8] (2–4)e 1.4g 2.22 ± 0.02a,i

1.5 ± 0.11e,f 1.5

b

248 (9.2)c 250 (6.5)e 250 (6.5)e

16 16d 16e,[18] 9.6g,h

248 (8.89 ± 0.43)k

250 250 252 248

(8.05) (8.05)k (8.1)e (8.48 ± 0.39)k

76.6a 76.6a 76.6a

8.25 ± 0.29a,i 9.92 ± 0.05i,l 10.11 ± 0.22l,m

1a,j 1.97 ± 0.06l,n

15a 15[21] (8–30)e 61.71 ± 0.39i,l 61.90 ± 1.43l,m

2 1.64 ± 0.07l,n

25o 93b

2p 2

25o 248 (8.68)k 250 (7.44)k 248 (8.56 ± 0.26)k

93 93.0a 80.18 ± 0.12i,l 79.51 ± 1.42l,m

1.71 ± 0.06l,n

All values are from the original reports unless otherwise indicated. Reports from vendors are obtained from their online information or product inserts. Most of them are likely obtained from literature reports instead of measurement by vendors themselves which, however, were not clearly indicated. Values referred to as ‘‘This work” are obtained from all measurements listed in Table 1 without distinguishing different vendors. a In 0.1 M phosphate buffer, pH 7.4. b In PBS, pH 7.4. c In water. d In phosphate buffer. e In pH 7.4 buffer. f a[D]0 was measured by chemiluminescence, how [D]0 was obtained was not indicated. g Determined potentiometrically in100 mM phosphate buffer at pH 7.4. h At 21 °C. i By donor decomposition kinetics. j At both 25 and 37 °C, a[D]0 was measured as nitrite concentration after complete decomposition by Griess assay, how [D]0 was obtained was not indicated. k In 10 mM NaOH. l In 100 mM sodium phosphate, 100 lM DTPA, pH 7.4 (working buffer), 25 °C. m By oxyMb assay. n Calculated by a[D]0/[D]0 (Column 5/Column 1 in Table 1). o In 0.1 M PBS, pH 7.4. p Inder physiological conditions.

72

Q. Li, J.R. Lancaster Jr. / Nitric Oxide 21 (2009) 69–75

d½oxyMb ¼ ak1 ½D dt

ð6Þ

Combining Eqs. (3) and (6):

  d½oxyMb ½oxyMb0  ½oxyMb ¼ ak1 ½D0  dt a
¼ k1 a½D0  ½oxyMb0 þ ½oxyMb

ln½D ¼ ln½D0  k1 t

ð14Þ

ð7Þ

½D=½D0 ¼ ek1 t

ð15Þ

ð8Þ

Assuming P is the only product from D decomposition except NO and the D:P stoichiometry is 1:1 (Eq. (1)), the absorbance at any time t is given by

Taking the definite integral from t = 0 to t = t:

lnða½D0  ½oxyMb0 þ ½oxyMbÞ  lnða½D0 Þ ¼ k1 t

Considering the presence of trace metMb contamination even in freshly purified oxyMb solution, the initial absorbance of the solution is given by

A0 ¼ eoxyMb ½oxyMb0 þ emetMb ½metMb0

ð9Þ

where eoxyMb and emetMb are the extinction coefficients of oxyMb and metMb, respectively. [metMb]0 is the initial concentration of metMb. At time t, the absorbance of the reaction solution will be

A ¼ eoxyMb ½oxyMb þ emetMb ð½oxyMb0  ½oxyMb þ ½metMb0 Þ ð10Þ The difference in absorbance is given by

A0  A ¼ ðeoxyMb  emetMb Þð½oxyMb0  ½oxyMbÞ

ð11Þ

ð12Þ

bx

Being of the general form y = a(1  e ), non-linear regression analysis can be applied with the following substitutions:

y ¼ A x¼t ð13Þ

ð16Þ

where Ac is the absorbance from any contaminant(s) present and eD and eP are the extinction coefficients of donor and product, respectively. It is assumed that Ac does not change with time. At the end of the reaction (t = 1), the absorbance is given by

A1 ¼ eP ½D0 þ AC

ð17Þ

Combining with Eq. (16)

A  A1 ¼ ðeD  eP Þ½D

ð18Þ

Combining Eqs. (14) and (18)

ð19Þ

The rate constant of donor decomposition k1 is the slope of the plot of ln(A  A1) vs. t. While an advantage to this method is that it allows k1 determination with a linear transformation, a disadvantage is that the measurement of A1 can be time-consuming for long half-life NONOates. In this case, the data are analyzed in terms of A0  A. A0 is the absorbance of the NONOate solution before decomposition (at t = 0), and can be obtained by immediate measurement following addition of stock solution (or a solution at high pH).

A0 ¼ eD ½D0 þ AC

a ¼ a½D0 ðeoxyMb  emetMb Þ b ¼ k1

A ¼ eD ½D þ eP ð½D0  ½DÞ þ AC

  lnðA  A1 Þ ¼ ln ðeD  eP Þ½D0  k1 t

If difference spectroscopy is utilized, A0 = 0 and combining Eqs. (8) and (11) yields

A ¼ a½D0 ðeoxyMb  emetMb Þð1  ek1 t Þ

this method, using the difference between A and either the absorbance at infinite time (A  A1), or the absorbance at zero time (A0  A). Based on first-order kinetics (Eq. (1)):

ð20Þ

Combining with Eq. (16),

A0  A ¼ ðeD  eP Þð½D0  ½DÞ

ð21Þ

Therefore, when A(y) is plotted against t(x) and the curve is fitted to this non-linear regression function, a and b can be obtained and, consequently, the rate constant of donor decomposition k1 (from which the donor half-live can be calculated t1/2 = ln 2/k1) and also a[D]0 (from which a can be determined knowing [D]0). OxyMb solution is used in both reference and sample cuvettes, therefore the procedure is referred to the split-beam mode according to Feelisch et al. [10]. However, the principle of the equation derived above will apply to the other two spectrophotometric methods [10]. For higher accuracy, we followed two wavelengths (582 and 545 nm) simultaneously [12], allowing two determinations per experiment which was averaged. Fig. 1A shows a representative tracing of the absorbance at 582 nm of a solution of 55 lM oxyMb upon addition of 2.5 lg/ml DEA/NO from Calbiochem in working buffer at 25 °C. Fig. 1B shows a non-linear regression fit (R2 = 0.99998) of the data plotted according to Eq. (12). From the parameter b the rate constant of DEA/NO decomposition k1 equals 6.96  102 min1, that is, the half-life is 9.96 min; from the parameter a, the NO content in 2.5 lg/ml DEA/NO, a[D]0, is calculated as 23.70 lM (Eq. (13)).

Fig. 2 shows a 250 nm decay trace of 12.5 lg/ml DEA/NO from Alexis in working buffer at 25 °C and the non-linear regression analysis (R2 = 0.99999). The half-life of 9.91 min is obtained from parameter b. In addition to the rate constant or the half-life, the initial concentration of donor [D]0 can also be obtained if eD and eP are known (using parameter a from the A0  A method or intercept from the A  A1 method).

Half-life measurement by NONOate decomposition kinetics

Consistency and validation of methodology

In order to validate the oxyMb assay, the NONOate decomposition kinetics was also determined by following the absorbance of the NONOates at 250 nm [8]. However, in this case only the rate constant, but not the NO content, can be determined. There are two ways to plot absorbance data (A) vs. time to determine k1 by

We have applied these methods to a survey of several NONOates from different commercial sources (Table 1). The results of determination of the half-life for decomposition using the donor decomposition kinetics is shown in Column 4, and the values for t1/2 and a[D]0 using the oxyMb method are shown in Columns 6

Combining with Eq. (15),

A0  A ¼ ðeD  eP Þ½D0 ð1  ek1 t Þ

ð22Þ

Similar to the oxyMb assay derivation, this can also be analyzed as described above by a non-linear regression, where

y ¼ A0  A x¼t a ¼ ½D0 ðeD  eP Þ b ¼ k1

ð23Þ

73

Q. Li, J.R. Lancaster Jr. / Nitric Oxide 21 (2009) 69–75

B

0.00

0.30

-0.05

0.25

-0.10

0.20

- A582nm

A582nm

A

-0.15 -0.20

0.10 0.05

-0.25 -0.30

0.15

0.00 0

10

20

30

40

50

60

0

Time (min)

10

20

30

40

50

Time (min)

Fig. 1. Representative oxyMb assay for the NO content and the NO release rate constant of NONOate. (A) oxyMb (55 lM) decay at 582 nm in the presence of 2.5 lg/ml DEA/ NO from Calbiochem in 100 mM phosphate buffer containing 0.1 mM DTPA at pH 7.4 and 25 °C. (B) A vs. t plot of the data in Fig. 1A (circles) and the non-linear fit (line) according to Eqs. (12) and (13) (R2 = 0.99998), from which the parameter a (0.27251 ± 0.00011) and b (0.06958 ± 0.00008 min1) were determined. Consequently, the NO content (23.70 lM) and the half-life (9.96 min) were obtained.

B

A

0.5

0.5

0.4

(A0 - A)250 nm

0.6

A250nm

0.4 0.3 0.2 0.1

0.3 0.2 0.1 0.0

0.0 0

20

40

60

80

0

Time (min)

20

40

60

Time (min)

Fig. 2. Representative NONOate decomposition kinetics for the NO release rate constant. (A) 250-nm decay trace of 12.5 lg/ml DEA/NO from Alexis in 100 mM phosphate buffer containing 0.1 mM DTPA at pH 7.4 and 25 °C. (B) A0  A vs. t plot of the data in Fig. 2A (circles) and the non-linear fit (line) according to Eqs. (22) and (23) (R2 = 0.99999). The half-life of 9.91 min can be calculated from parameter b (0.06995 ± 0.00007 min1).

and 5, respectively. There is excellent agreement between the two methods for the values of t1/2 (compare Columns 4 and 6). In addition, the half-lives are very consistent for the same NONOate from different vendors. As described above, the oxyMb analysis yields only the value for a[D]0 so the value for a requires an accurate value for [D]0. For pure NONOate, [D]0 can be obtained either by mass/molecular weight or spectrophotometrically with extinction coefficient. An accurate value for [D]0 of impure NONOate can be obtained spectrophotometrically if the contamination has negligible interference at the maximum absorbance wavelength (kmax). However, as can be seen in Table 2, the extinction coefficients for NONOates are not consistent among reports. Since we know neither the purity of commercial NONOates nor the accurate extinction coefficients, we used both of the methods for comparison. The spectra of these NONOates in working buffer at early time (data not shown) are the same as in 10 mM NaOH. To minimize the decomposition of NONOate during spectral recording, we performed our measurement for [D]0 in 10 mM NaOH. For a 5 mg/ml solution of DEA/ NO, PAPA/NO and NOC-5, there is in general excellent agreement for the NONOate concentration determined by mass/molecular weight (Table 1, Column 1) with that determined from absorbance using the extinction coefficients for PAPA/NO by Hrabie et al. [18], DEA/NO and NOC-5 by Sigma (Column 3). This indicates that these values for eD are likely to be accurate and also that the NONOates

are likely to be pure (this point is considered further in the Discussion). The close agreement between the two measurements for [D]0 (Columns 1 and 3) also provides the possibility that these values will yield accurate values for a, which is shown for each NONOate in Column 7. Even though the value for a is close to 2 for DEA/NO, less than 2 for PAPA/NO and NOC-5, they are not different for a given donor from different vendors. Overall, we feel that those similarities described above validate our oxyMb assay. We emphasize that the purpose of these calculations is not to generate accurate parameters such as extinction coefficients and a, but to compare and validate our method for determining the rate and extent of NO generation of a given solution of donor, whether the concentration is determined spectrophotometrically or by mass/molecular weight. Discussion 

NO release from NONOate can be accurately followed by direct electrochemical measurement [6,9,19] or by chemiluminescence [5], and similar kinetic analysis can yield both the NO content and the rate constant k1. However, neither NO electrode nor the  NO gas-phase chemiluminescence are commonly available to most laboratories. We describe here methods employing only spectrophotometric measurements that can be used for routine calibration.

74

Q. Li, J.R. Lancaster Jr. / Nitric Oxide 21 (2009) 69–75

The accuracy of this method relies on the assumption that the oxyMb reaction is fast enough to scavenge all the NO that has been released. NO autoxidation is the other possible mechanism for NO disappearance. In this case, the rate of disappearance is given by 4k[O2][NO]2 where k = 3  106 M2 s1 [20]. To compete with oxyMb scavenging (rate k2[oxyMb][NO] where k2 = 3.4  107 M1 s1), it can be shown that [NO] must be P4000  [oxyMb] with 240 lM of O2 in buffer. Thus, NO autoxidation will not compete with the oxyMb reaction under normal conditions. The good agreement of the values for half-lives measured by donor decomposition and oxyMb assay (Table 1) verifies this conclusion that oxyMb trapped essentially all the NO. Since both t1/2 and a have been obtained and the eD at kmax can be determined spectrophotometrically assuming 100% purity and using the concentration calculated by mass/molecular weight, we compared our results to other reports (Table 2). As shown in Column 3, there is good agreement between all reported values for kmax, however, there are differences in the values for eD. For DEA/NO and NOC-5, our values for eD compare closely with those reported by Sigma, but are somewhat higher than those reported in the literature. Our value for eD for PAPA/NO is also somewhat higher than reported in the literature. Although the origin of these slight differences is not clear, the close agreement between the concentrations obtained by mass/molecular weight and absorbance for different sources (Table 1 Columns 1 and 3) suggests that the higher values for eD found here and reported by Sigma are more accurate (although see below). With regard to the rate constants for decomposition, our halflives at 25 °C for the three donors are intermediate between the values reported for room temperature (22–25 °C) and for 37 °C (Table 2 Columns 4–7). The close agreement between the half-lives determined by two independent methods (donor decomposition and the oxyMb assay) for all donors (column 7) we feel validate our values. The differences between our values (Column 7) and the previously reported values (Columns 4–6) probably lie in slightly differing conditions (including temperature, ionic strength and pH4). This variability underscores the importance of experimental calibration of NONOate kinetics under the precise conditions being utilized experimentally. In order to avoid the variability of reported values for eD as discussed above, we used [D]0 by mass/molecular weight (Column 1 in Table 1) instead of those by spectrophotometry for calculation of a in Table 2. The values for a for PAPA/NO (1.64 ± 0.07) and NOC-5 (1.71 ± 0.06) are both lower than the theoretical value of 2. However, our a value for DEA/NO is significantly higher than that reported by Maragos et al. [8] (2.04 ± 0.12 vs. 1.5 ± 0.11). The origin of this discrepancy may lie in the differences in the values for eD utilized to calculate [D]0. As described above, our value for eD for DEA/NO is 8.89 ± 0.43 mM1 cm1 (Table 2, column 3) while the value utilized by Maragos et al. [8] is 6.5 mM1 cm1. An underestimated eD will yield a lower value for a. As discussed in Results, without knowing the purity of the NONOate, the correct eD cannot be determined. However, one advantage of our oxyMb assay is that it can provide the NO content without knowing eD or a, which are not consistent among reports. In other words, our method can yield NO flux rate for any given solution independent of the purity or the accuracy of eD. For the oxyMb assay, there is an optimum ‘‘window” of time over which data can be collected. Specifically, the time course should not be exceedingly long (e.g., no longer than several hours) to avoid slower side reactions and yet long enough to obtain data

4 It is the protonated form of NONOate that decomposes. This also means that using Griess assay or other acidic-based method will not yield the rate of decomposition under neutral conditions.

over several half-lives of the donor. This restriction necessarily confines the applicability of the method to a subset of NONOate donors, specifically, those with half-life greater than 10 min or less than 80 min. However, a NONOate with a very short half-life, such as PROLI/NO (half-life as short as 1.8 s at pH 7.4 and 37 °C [13]), is usually used as the equivalent of a bolus addition of NO instead of generating an NO flux. In this case, the rate of decomposition is not relevant and only the NO content is needed. This quantity can be obtained by the end point measurement of the oxyMb assay since

a½D0 ¼ A1 =ðeoxyMb  emetMb Þ

ð24Þ

from Eq. (12) at t = 1. For donors of very long half-life (such as DETA/NO, t1/2 = 20 h at 37 °C [21]) the common usage is for a sustained, relatively constant rate of NO generation, thus involving an initial time frame substantially shorter than one half-life. Under these conditions, the linear rate of NO generation can be obtained by simple measurement of oxyMb oxidation. Finally, we emphasize that our procedure will yield only the flux rate of NO generation; the concentration of NO will be determined by the rates of both NO generation and NO disappearance. It is generally very difficult to quantify the rate of NO disappearance, since it is determined by a variety of processes including chemical reaction inside and outside cells and also physical loss due to volatilization [22]. This latter process will be very sensitive to the physical characteristics of the experimental system, most especially the ratio of the surface area in contact with headspace to the solution volume [23,24]. Although not presented here, it is mathematically possible to determine the rate of NO loss from knowing the NO concentration and the rate of NO production. Acknowledgments We thank Dr. Larry K. Keefer for helpful comments. This work was supported by NIH Grants HL71189 and HL074391. References [1] P. Pacher, J.S. Beckman, L. Liaudet, Nitric oxide and peroxynitrite in health and disease, Physiol. Rev. 87 (2007) 315–424. [2] P.G. Wang, M. Xian, X.P. Tang, X.J. Wu, Z. Wen, T.W. Cai, A.J. Janczuk, Nitric oxide donors: chemical activities and biological applications, Chem. Rev. 102 (2002) 1091–1134. [3] M.R. Miller, I.L. Megson, Recent developments in nitric oxide donor drugs, Br. J. Pharmacol. 151 (2007) 305–321. [4] .D. Morley, L.K. Keefer, Nitric oxide/nucleophile complexes: a unique class of nitric oxide-based vasodilators, J. Cardiovasc. Pharmacol. 22 (Suppl. 7) (1993) S3–S9. [5] L.K. Keefer, R.W. Nims, K.M. Davies, D.A. Wink, ‘‘NONOates” (1-substituted diazen-1-ium-1, 2-diolates) as nitric oxide donors: convenient nitric oxide dosage forms, Methods Enzymol. 268 (1996) 281–293. [6] K. Schmidt, W. Desch, P. Klatt, W.R. Kukovetz, B. Mayer, Release of nitric oxide from donors with known half-life: a mathematical model for calculating nitric oxide concentrations in aerobic solutions, Naunyn Schmiedebergs Arch. Pharmacol. 355 (1997) 457–462. [7] B.T. Dinh, S.E. Price, A. Majul, M. El Hajj, V. Morozov, J.A. Hrabie, K.M. Davies, Diazeniumdiolate reactivity in model membrane systems, Nitric Oxide 18 (2008) 113–121. [8] C.M. Maragos, D. Morley, D.A. Wink, T.M. Dunams, J.E. Saavedra, A. Hoffman, A.A. Bove, L. Isaac, J.A. Hrabie, L.K. Keefer, Complexes of NO with nucleophiles as agents for the controlled biological release of nitric oxide. Vasorelaxant effects, J. Med. Chem. 34 (1991) 3242–3247. [9] J.D. Artz, G.R. Thatcher, NO release from NO donors and nitrovasodilators: comparisons between oxyhemoglobin and potentiometric assays, Chem. Res. Toxicol. 11 (1998) 1393–1397. [10] M. Feelisch, D. Kubitzek, J. Werringloer, The oxyhemoglobin assay, in: M. Feelisch, J.S. Stamler (Eds.), Methods in Nitric Oxide Research, John Wiley & Sons Ltd., England, 1996, pp. 455–478. [11] S.S. Gross, E.A. Jaffe, R. Levi, R.G. Kilbourn, Cytokine-activated endothelial cells express an isotype of nitric oxide synthase which is tetrahydrobiopterindependent, calmodulin-independent and inhibited by arginine analogs with a rank-order of potency characteristic of activated macrophages, Biochem. Biophys. Res. Commun. 178 (1991) 823–829. [12] T.M. Rothgeb, F.R. Gurd, Physical methods for the study of myoglobin, Methods Enzymol. 52 (1978) 473–486.

Q. Li, J.R. Lancaster Jr. / Nitric Oxide 21 (2009) 69–75 [13] J.E. Saavedra, G.J. Southan, K.M. Davies, A. Lundell, C. Markou, S.R. Hanson, C. Adrie, W.E. Hurford, W.M. Zapol, L.K. Keefer, Localizing antithrombotic and vasodilatory activity with a novel, ultrafast nitric oxide donor, J. Med. Chem. 39 (1996) 4361–4365. [14] D.A. Ratkowsky, Handbook of Non-linear Regression Models, M. Dekker, New York, 1990. pp. 75–98. [15] R.F. Eich, T.S. Li, D.D. Lemon, D.H. Doherty, S.R. Curry, J.F. Aitken, A.J. Mathews, K.A. Johnson, R.D. Smith, G.N. Phillips, J.S. Olson, Mechanism of NO-induced oxidation of myoglobin and hemoglobin, Biochemistry 35 (1996) 6976–6983. [16] B. Li, Y. Shen, B. Li, Quasi-steady-state laws in enzyme kinetics, J. Phys. Chem. A 112 (2008) 2311–2321. [17] Experimental Kinetics and Gas Reactions, in: R.A. Alberty, R.J. Silbey (Eds.), Physical Chemistry, John Wiley & Sons, Inc., 2004, pp. 642-643. [18] J.A. Hrabie, J.R. Klose, D.A. Wink, L.K. Keefer, New nitric oxide-releasing zwitterions derived from polyamines, J. Org. Chem. 58 (1993) 1472–1476. [19] K. Schmidt, P. Klatt, B. Mayer, Reaction of peroxynitrite with oxyhaemoglobin: interference with photometrical determination of nitric oxide, Biochem. J. 301 (Pt. 3) (1994) 645–647. [20] M.N. Moller, Q. Li, D.A. Vitturi, J.M. Robinson, J.R. Lancaster Jr., A. Denicola, Membrane ‘‘lens” effect: focusing the formation of reactive nitrogen oxides from the *NO/O2 reaction, Chem. Res. Toxicol. 20 (2007) 709–714.

75

[21] D.L. Mooradian, T.C. Hutsell, L.K. Keefer, Nitric oxide (NO) donor molecules: effect of NO release rate on vascular smooth muscle cell proliferation in vitro, J. Cardiovasc. Pharmacol. 25 (1995) 674–678. [22] D.D. Thomas, X. Liu, S.P. Kantrow, J.R. Lancaster Jr., The biological lifetime of nitric oxide: implications for the perivascular dynamics of NO and O2, Proc. Natl. Acad. Sci. USA 98 (2001) 355–360. [23] M. Laurent, M. Lepoivre, J.P. Tenu, Kinetic modelling of the nitric oxide gradient generated in vitro by adherent cells expressing inducible nitric oxide synthase, Biochem. J. 314 (Pt. 1) (1996) 109–113. [24] B. Chen, W.M. Deen, Effect of liquid depth on the synthesis and oxidation of nitric oxide in macrophage cultures, Chem. Res. Toxicol. 15 (2002) 490–496. [25] K.M. Miranda, T. Katori, C.L. Torres de Holding, L. Thomas, L.A. Ridnour, W.J. McLendon, S.M. Cologna, A.S. Dutton, H.C. Champion, D. Mancardi, C.G. Tocchetti, J.E. Saavedra, L.K. Keefer, K.N. Houk, J.M. Fukuto, D.A. Kass, N. Paolocci, D.A. Wink, Comparison of the NO and HNO donating properties of diazeniumdiolates: primary amine adducts release HNO in vivo, J. Med. Chem. 48 (2005) 8220–8228. [26] L. Jia, C. Bonaventura, J. Bonaventura, J.S. Stamler, S-nitrosohaemoglobin: a dynamic activity of blood involved in vascular control, Nature 380 (1996) 221– 226.