Determination of nitrate and nitrite in rat brain perfusates by capillary electrophoresis

A fast and simple method for the direct, simultaneous detection of nitrite (NO(2) (-)) and nitrate (NO(3) (-)) in rat striatum has been developed using a capillary electrophoresis separation of low-flow push-pull perfusion samples. The method was optimized primarily for nitrite because nitrite is more important physiologically and is found at lower levels than nitrate. We obtained a complete separation of NO(2) (-) and NO(3) (-) in rat striatum within 1.5 min. Optimal CE separations were achieved with 20 mM phosphate, 2 mM cetyltrimethylammonium chloride (CTAC) buffer at pH 3.5. The samples were injected electrokinetically for 2 s into a 40 cm x 75 microm ID fused-silica capillary. The separation voltage was 10 kV (negative polarity), and the injection voltage was 16 kV (negative polarity). UV detection was performed at 214 nm. The limits of detection obtained at a signal-to-noise ratio (S/N) of 3 for nitrite and nitrate were 0.96 and 2.86 microM. This is one of the fastest separations of nitrite and nitrate of a biological sample ever reported. Interference produced by the high physiological level of chloride is successfully minimized by use of CTAC in the run buffer.     

GC-MS and HPLC methods for peroxynitrite (ONOO- and O15NOO-) analysis: a study on stability, decomposition to nitrite and nitrate, laboratory synthesis, and formation of peroxynitrite from S-nitrosoglutathione (GSNO) and KO2

Nitric oxide (˙NO) and superoxide (O(2)(-)˙) are ubiquitous in nature. Their reaction product peroxynitrite (ONOO(-)) and notably its conjugated peroxynitrous acid (ONOOH) are highly unstable in aqueous phase. ONOO(-)/ONOOH (referred to as peroxynitrite) isomerize and decompose to NO(3)(-), NO(2)(-) and O(2). Here, we report for the first time GC-MS and HPLC methods for the analysis of peroxynitrite in aqueous solution. For GC-MS analysis peroxynitrite in alkaline solution was derivatized to a pentafluorobenzyl derivative using pentafluorobenzyl bromide. O(15)NOO(-) was synthesized from H(2)O(2) and (15)NO(2)(-) and used as internal standard. HPLC analysis was performed on stationary phases consisting of Nucleosil? 100-5C(18)AB or Nucleodur? C(18) Gravity. The mobile phase consisted of a 10 mM aqueous solution of tetrabutylammonium hydrogen sulfate and had a pH value of 11.5. UV absorbance detection at 300 nm was used. HPLC allows simultaneous analysis of ONOO(-), NO(2)(-) and NO(3)(-). The GC-MS and HPLC methods were used to study stability, synthesis, formation from S-[(15)N]nitrosoglutathione (GS(15)NO) and KO(2), and isomerization/decomposition of peroxynitrite to NO(2)(-) and NO(3)(-) in aqueous buffer.     

Enhanced selectivity and sensitivity for inorganic anions using an ion-pairing reagent and sample stacking in capillary zone electrophoresis with direct UV detection

In this paper, the use of an ion-pairing reagent to improve the separation selectivity of inorganic anions in CZE was demonstrated by the addition of tetramethylammonium hydroxide (TMAOH) to the electrolyte. The separation of inorganic anions (Cl(-), I(-), Br(-), NO(2)(-), NO(3)(-) and SCN(-)) was performed using co-electroosmotic flow (EOF) with direct UV detection at 185 nm. The parameters affecting the mobility of the tested anions and the EOF such as the electrolyte pH and concentration of TMAOH in the electrolyte were examined to optimise the separation conditions. In addition, sample-stacking techniques were investigated to improve detection sensitivity. Detection sensitivities were improved 5-13-fold using electrokinetic sample stacking. The detection limits ranged from 1-3 micro mol L(-1). Finally, the proposed method was used for the separation of anions in groundwaters.     

Microchip electrophoresis with amperometric detection for the study of the generation of nitric oxide by NONOate salts

Microchip electrophoresis (ME) with electrochemical detection was used to monitor nitric oxide (NO) production from diethylammonium (Z)-1-(N,N-diethylamino)diazen-1-ium-1,2-diolate (DEA/NO) and 1-(hydroxyl-NNO-azoxy)-L-proline disodium salt (PROLI/NO). NO was generated through acid hydrolysis of these NONOate salts. The products of acid hydrolysis were introduced into a 5-cm separation channel using gated injection. The separation was accomplished using reverse polarity and a background electrolyte consisting of 10 mM boric acid and 2 mM tetradecyltrimethylammonium bromide, pH 11. Electrochemical detection was performed using an isolated potentiostat in an in-channel configuration. Potentials applied to the working electrode, typically higher than +1.0 V vs. Ag/AgCl, allowed the direct detection of nitrite, NO, DEA/NO, and PROLI/NO. Baseline resolution was achieved for the separation of PROLI/NO and NO while resolution between DEA/NO and NO was poor (1.0 ± 0.2). Nitrite was present in all samples tested.     

A novel nitrite biosensor based on gold dendrites with egg white as template

Gold dendrites (AuD) were synthesized with egg white as the soft template and a novel nitrite (NO(2)(-)) biosensor was fabricated by assembly of a myoglobin (Mb)-L-cysteamine (Cys)-AuD biological hybrid. The results of Fourier transform infrared spectra and UV-visible spectra indicated that Mb retained its original structure in the resulting Mb-Cys-AuD. Electrochemical investigation of the biosensor showed a pair of well-defined, quasi-reversible redox peaks with E(pa) = -0.314 V and E(pc) = -0.344 V (vs. SCE) in 0.1 M, pH 7.0 sodium phosphate buffered saline at the scan rate of 200 mV/s. The transfer rate constant (k(s)) was 1.49 s(-1). The Mb-Cys-AuD showed a good electrochemical catalytic response for the reduction of NO(2)(-), with the linear range from 0.5 to 400 μM and the detection limit of 0.3 μM (S/N = 3). The apparent Michaelis-Menten constant (K(M)(app)) was estimated to be 0.2 mM. Therefore, the assembled bio-hybrid as a novel matrix opened up a further possibility for study on the design of enzymatic biosensors with potential applications.     

Method development and validation for the determination of various sulfur-containing anions and other anions in the corrosion process by capillary ion electrophoresis with indirect detection

A method of capillary ion electrophoresis with indirect detection is developed for the simultaneous determination of the sulfur-containing anions S2O4(2-), S2O3(2-), SO4(2-), SO3(2-), and S2- and other anions (Cl-, Br-, NO2-, NO3-, (COO)2(2-), F-, and PO4(3-)) in the corrosion process. The effects of pH, tetradecyltrimethylammonium hydroxide, chromate, 2-[n-cyclohexylamino]-ethane sulfonate, calcium gluconate, and acetonitrile on the migration and resolution of the anions and the stability of sulfur-containing anions are systematically investigated. The detection limits, repeatability, and linearity for the anions are comparatively studied at 374, 274, and 254 nm, and the results show that 374 nm is the optimal length. The simultaneous multiwavelength detection at 374, 254, 214, and 195 nm can assist in confirming the identification of UV-absorbing anions.     

Mechanism of nitrite formation by nitrate photolysis in aqueous solutions: the role of peroxynitrite, nitrogen dioxide, and hydroxyl radical

Photolysis of aqueous NO3(-) with lambda > or = 195 nm is known to induce the formation of NO2(-) and O2 as the only stable products. The mechanism of NO3- photolysis, however, is complex, and there is still uncertainty about the primary photoprocesses and subsequent reactions. This is, in part, due to photoisomerization of NO3(-) to ONOO(-) at lambda < 280 nm, followed by the formation of *OH and *NO2 through the decomposition of ONOOH (pKa = 6.5-6.8). Because of incomplete information concerning the mechanism of peroxynitrite (ONOOH/ONOO(-)) decomposition, previous studies were unable to account for all observations. In the present study aqueous nitrate solutions were photolyzed by monochromatic light in the range of 205-300 nm. It is shown that the main primary processes at this wavelength range are NO3(-) hv-->*NO2 + O*(-) (reaction 1) and NO3(-) hv--> ONOO(-) (reaction 2). Based on recent knowledge on the mechanisms of peroxynitrite decomposition and its reactions with reactive nitrogen and oxygen species, we determined Phi(1) and Phi(2) using different experimental approaches. Both quantum yields increase with decreasing the excitation wavelength, approaching Phi(1) = 0.13 and Phi(2) = 0.28 at 205 nm. It is also shown that the yield of nitrite increases with decreasing the excitation wavelength. The implications of these results on UV disinfection of drinking water are discussed.     

Can the ebselen derivatives catalyze the isomerization of peroxynitrite to nitrate?

The reaction of ebselen and its derivatives (1-7) with peroxynitrite anion (ONOO(-); PN) has been studied in gas phase and in aqueous, dichloromethane, benzene, and cyclohexane solutions using B3LYP/6-311+G(d,p)//B3LYP/6-311G(d,p) and PCM-B3LYP/6-311+G(d,p)//B3LYP/6-311G(d,p) approaches, respectively. It was shown that the reaction of 2 (R=H) with PN proceeds via 2 + PN --> 2-PN --> 2-TS1 (O-O activation) --> 2-O(NO(2)(-)()) --> 2-SeO + NO(2)(-) pathway with a rate-determining barrier of 25.3 (14.8) kcal/mol at the NO(2)(-) dissociation step (numbers presented without parentheses are enthalpies, and those in parentheses are Gibbs free energies). The NO(3)(-) formation process, starting from the complex 2-O(NO(2)(-)()), requires by (7.9) kcal/mol more energy than the NO(2)(-) dissociation process and is unlikely to compete with the latter. Thus, in the gas phase, the peroxynitrite --> nitrate isomerization catalyzed by complex 2 is unlikely to occur. It is shown that the NO(3)(-) formation process is slightly more favorably than the NO(2)(-) dissociation process for complex 4, with a strongest electron-withdrawing ligand R=CF(3). Therefore, complex 4 (as well as complex 6 with R=OH) is predicted to be a good catalyst for peroxynitrite <--> nitrite isomerization in the gas phase. Solvent effects (a) change the rate-determining step of the reaction 2 + PN from NO(2)(-) dissociation in the gas phase to O-O activation, which occurs with barriers of (13.9), (8.4), (8.4), and (8.2) kcal/mol in water, dichloromethane, benzene, and cyclohexane, respectively, and (b) significantly reduce the NO(2)(-) dissociation energy, while only slightly destabilizing the NO(3)(-) formation barrier, and make the peroxynitrite <--> nitrate isomerization process practically impossible, even for complex 4.     

Separation of arsenic species by capillary electrophoresis with sample-stacking techniques

A simple capillary zone electrophoresis procedure was developed for the separation of arsenic species (AsO(2)(2-), AsO(4)(2-), and dimethylarsinic acid, DMA). Both counter-electroosmotic and co-electroosmotic (EOF) modes were investigated for the separation of arsenic species with direct UV detection at 185 nm using 20 mmol L(-1) sodium phosphate as the electrolyte. The separation selectivity mainly depends on the separation modes and electrolyte pH. Inorganic anions (Cl(-), NO(2)(-), NO(3)(-) and SO(4)(2-)) presented in real samples did not interfere with arsenic speciation in either separation mode. To improve the detection limits, sample-stacking techniques, including large-volume sample stacking (LVSS) and field-amplified sample injection (FASI), were investigated for the preconcentration of As species in co-CZE mode. Less than 1 micromol L(-1) of detection limits for As species were achieved using FASI. The proposed method was demonstrated for the separation and detection of As species in water.     

Determination of anions using monolithic capillary column ion chromatography with end-to-end differential contactless conductometric detectors under resonance approach

An end-to-end differential measurement approach with capacitively coupled contactless conductivity detection (C(4)D) was applied to anion-exchange monolithic capillary column ion chromatography. The column was prepared by thermally initiated radical polymerization of poly(glycidyl methacrylate) in a fused-silica capillary of 320 μm i.d. and modified by quaternary ammonium latex surface coating. Two C(4)Ds were placed near both ends of the capillary column and the output difference between them was measured. With 15 mM potassium hydrogen phthalate used as the eluent, good separation of a mixture of inorganic anions (F(-), Cl(-), NO(2)(-), NO(3)(-)) was achieved. The detection limits of conventional C(4)D are 1.6, 0.28, 0.53, and 0.47 mg L(-1) for F(-), Cl(-), NO(2)(-), and NO(3)(-), respectively. To further enhance the sensitivity, the capacitive impedance from C(4)D was neutralized by an inductive impedance from a piezoelectric resonator. An increase in sensitivity by a factor of 7-8 was achieved in the resonating C(4)D in comparison with the conventional C(4)D. The detection limits of the resonating C(4)D are 0.23, 0.041, 0.065, and 0.059 mg L(-1) for F(-), Cl(-), NO(2)(-), and NO(3)(-), respectively. The response of the resonating C(4)D was analyzed based on an equivalent circuit model.     

Liquid core waveguide spectrophotometry for the sensitive determination of nitrite in river water samples.

A flow injection analysis system was built with a liquid core waveguide spectrophotometric detector using an 80 cm Teflon AF-1600 capillary tube (2,2-bistrifluoromethyl-4,5-difluoro-1,3-dioxole/tetrafluoroethylene). The system was applied to determine nitrite ion in river water samples. The lower limit of detection for nitrite was 2.1 nmol dm(-3) (0.1 ng dm(-3) as NO(2)(-)) and the relative standard deviation of measurements was typically 0.56% (n = 5) at 0.21 micromol dm(-3).     

Argon cluster-mediated isolation and vibrational spectra of peroxy and nominally D(3h) isomers of CO3(-) and NO3(-)

Vibrational predissociation spectra are reported for two isomeric forms of the gas-phase ions, CO(3)(-) and NO(3)(-). The peroxy forms, (OOCO(-) and OONO(-)) were isolated using an Ar-mediated synthetic scheme involving exchange of CO and NO for the more weakly bound Ar ligands in O(2)(-)Ar(m) clusters, while the forms based on a central heteroatom (CO(3)(-) and NO(3)(-)) were generated by electron impact on CO(2) and HNO(3) vapor. The simple two-band spectrum of OOCO(-) indicates that it is best described as the O(2)(-) x CO ion-molecule complex, whereas the covalently bound CO(3)(-) form yields a much more complicated vibrational spectrum with bands extending out to 4000 cm(-1). In contrast, the NO(3)(-) ion yields a simple spectrum with only one transition as expected for the antisymmetric NO stretching fundamental of a species with D(3h) structure. The spectrum of the peroxynitrite isomer, OONO(-), displays intermediate complexity that can be largely understood in the context of fundamentals associated with its cis and trans structures previously characterized in an Ar matrix.     

Experimental and density functional theoretical investigations of linkage isomerism in six-coordinate FeNO(6) iron porphyrins with axial nitrosyl and nitro ligands

A critical component of the biological activity of NO and nitrite involves their coordination to the iron center in heme proteins. Irradiation (330 < lambda < 500 nm) of the nitrosyl-nitro compound (TPP)Fe(NO)(NO(2)) (TPP = tetraphenylporphyrinato dianion) at 11 K results in changes in the IR spectrum associated with both nitro-to-nitrito and nitrosyl-to-isonitrosyl linkage isomerism. Only the nitro-to-nitrito linkage isomer is obtained at 200 K, indicating that the isonitrosyl linkage isomer is less stable than the nitrito linkage isomer. DFT calculations reveal two ground-state conformations of (porphine)Fe(NO)(NO(2)) that differ in the relative axial ligand orientations (i.e., GS parallel and GS perpendicular). In both conformations, the FeNO group is bent (156.4 degrees for GS parallel, 159.8 degrees for GS perpendicular) for this formally {FeNO}(6) compound. Three conformations of the nitrosyl-nitrito isomer (porphine)Fe(NO)(ONO) (MSa parallel, MSa perpendicular, and MSa(L)) and two conformations of the isonitrosyl-nitro isomer (porphine)Fe(ON)(NO(2)) (MSb parallel and MSb perpendicular) are identified, as are three conformations of the double-linkage isomer (porphine)Fe(ON)(ONO) (MSc parallel, MSc perpendicular, MSc(L)). Only 2 of the 10 optimized geometries contain near-linear FeNO (MSa(L)) and FeON (MSc(L)) bonds. The energies of the ground-state and isomeric structures increase in the order GS < MSa < MSb < MSc. Vibrational frequencies for all of the linkage isomers have been calculated, and the theoretical gas-phase absorption spectrum of (porphine)Fe(NO)(NO(2)) has been analyzed to obtain information on the electronic transitions responsible for the linkage isomerization. Comparison of the experimental and theoretical IR spectra does not provide evidence for the existence of a double linkage isomer of (TPP)Fe(NO)(NO(2)).     

On-column complexation of metal ions using 2,6-pyridinedicarboxylic acid and separation of their anionic complexes by capillary electrophoresis with direct UV detection

On-column complexation of metal ions with 2,6-pyridinedicarboxylate (2,6-PDC) to form anionic complexes enabled their separation by capillary zone electrophoresis with direct UV detection at 214 nm. Nine metal ions, Cu2+, Zn2+, Ni2+, Cd2+ Mn2+, Pb2+, Fe3+, Al3+ and Ca2+, were determined in less than 7 min using 10 mM 2.6-PDC solution containing 0.75 mM tetradecyltrimethylammonium bromide at pH 4.0. Satisfactory working ranges (20-300 microM), detection limits (3-10 microM) and good repeatability of the peak areas (RSD 2.1-4.2%, n=5) were obtained using hydrodynamic injection (30 s). The proposed method was used successfully for the determination of Mn2+, Fe3+, Al3+ and Ca2+ in groundwaters.     

NO* release from MbFe(II)NO and HbFe(II)NO after oxidation by peroxynitrite

In this work, we showed that the reaction of peroxynitrite with MbFe(II)NO, in analogy to the corresponding reaction with HbFe(II)NO (Herold, S. Inorg. Chem. 2004, 43, 3783-3785), proceeds in two steps via the formation of MbFe(III)NO, from which NO* dissociates to produce iron(III)myoglobin (Mb = myoglobin; Hb = hemoglobin). The second-order rate constants for the first steps are on the order of 10(4) and 10(3) M(-1) s(-1), for the reaction of peroxynitrite with MbFe(II)NO and HbFe(II)NO, respectively. For both proteins, we found that the values of the second-order rate constants increase with decreasing pH, an observation that suggests that HOONO is the species responsible for oxidation of the iron center. Nevertheless, it cannot be excluded that the pH-dependence arises from different conformations taken up by the proteins at different pH values. In the presence of 1.2 mM CO2, the values of the second-order rate constants are larger, on the order of 10(5) and 10(4) M(-1) s(-1), for the reaction of peroxynitrite with MbFe(II)NO and HbFe(II)NO, respectively. The pH-dependence of the values for the reaction with MbFe(II)NO suggests that ONOOCO2- or the radicals produced from its decay (CO3*-/NO2*) are responsible for the oxidation of MbFe(II)NO to MbFe(III)NO. In the presence of large amounts of nitrite (in the tens and hundreds of millimoles range), we observed a slight acceleration of the rate of oxidation of HbFe(II)NO by peroxynitrite. A catalytic rate constant of 40 +/- 2 M(-1) s(-1) was determined at pH 7.0. Preliminary studies of the reaction between nitrite and HbFe(II)NO showed that this compound also can oxidize the iron center, albeit at a significantly slower rate. At pH 7.0, we obtained an approximate second-order rate constant of 3 x 10(-3) M(-1) s(-1).     

Ultra-rapid analysis of nitrate and nitrite by capillary electrophoresis

Rapid analysis of nitrate and nitrite by capillary electrophoresis (CE) has been limited by the ions' very similar electrophoretic mobilities. With a pKa of 3.15, the mobility of nitrite can be selectively reduced using a low pH buffer in CE. A much shorter capillary can be used and separation voltages can be increased. With this method, nitrate and nitrite are separated in just over 10 s. This is roughly 20 times faster than current separation methods. Direct UV detection at 214 nm was employed and offered sub microM detection limits. Total analysis time (pre-rinse, injection, and separation) was less than 1 min, making this method ideal for high-throughput analysis.     

Complexes of HNO3 and NO3 - with NO2 and N2O4, and their potential role in atmospheric HONO formation

Calculations were performed to determine the structures, energetics, and spectroscopy of the atmospherically relevant complexes (HNO(3)).(NO(2)), (HNO(3)).(N(2)O(4)), (NO(3)(-)).(NO(2)), and (NO(3)(-)).(N(2)O(4)). The binding energies indicate that three of the four complexes are quite stable, with the most stable (NO(3)(-)).(N(2)O(4)) possessing binding energy of almost -14 kcal mol(-1). Vibrational frequencies were calculated for use in detecting the complexes by infrared and Raman spectroscopy. An ATR-FTIR experiment showed features at 1632 and 1602 cm(-1) that are attributed to NO(2) complexed to NO(3)(-) and HNO(3), respectively. The electronic states of (HNO(3)).(N(2)O(4)) and (NO(3)(-)).(N(2)O(4)) were investigated using an excited state method and it was determined that both complexes possess one low-lying excited state that is accessible through absorption of visible radiation. Evidence for the existence of (NO(3)(-)).(N(2)O(4)) was obtained from UV/vis absorption spectra of N(2)O(4) in concentrated HNO(3), which show a band at 320 nm that is blue shifted by 20 nm relative to what is observed for N(2)O(4) dissolved in organic solvents. Finally, hydrogen transfer reactions within the (HNO(3)).(NO(2)) and (HNO(3)).(N(2)O(4)) complexes leading to the formation of HONO, were investigated. In both systems the calculated potential profiles rule out a thermal mechanism, but indicate the reaction could take place following the absorption of visible radiation. We propose that these complexes are potentially important in the thermal and photochemical production of HONO observed in previous laboratory and field studies.     

Nitrite and nitrate measurement in human urine by capillary electrophoresis.

Nitrite and nitrate have been widely used as markers for nitric oxide (NO) formation in vivo and represent the major NO oxidation products in biological fluids. In the present study, the use of capillary electrophoresis (CE) in the measurement of nitrite and nitrate in human urine is described. Urine samples were electrophoresed in an extended light path fused-silica capillary (104 cm; 75 microm ID) at an applied negative potential of 30 kV, and UV detection at 214 nm. Using electrokinetic sample injection (-6 kV x 20 s), we found that urine concentration, pH, sodium and chloride interfered with nitrite and nitrate detection. The detection of nitrite and nitrate was decreased when hydrodynamic sample injection was used (30 mbar x 60 s). However, basal levels of urinary nitrite (0.25 +/- 0.05 microM) and nitrate (591 +/- 115 microM) were detected and no interference by variations in urine concentration and pH was noted when hydrodynamic sample injection was used. Thus, hydrodynamic sample injection is convenient for the measurement of urinary nitrite and nitrate and avoids the effect of variations in urine matrices and pH on nitrite and nitrate detection.     

Single-column ion-chromatographic determination of chromium(VI) in aqueous soil and sludge extracts

Single-column ion-chromatography (SCIC) was investigated as a routine, rapid, precise and selective analytical method for the determination of chromium(VI) in aqueous extracts of soil and sewage sludge. Chromatographic parameters were optimized for determination of Cr(VI), NO(-)(3) and SO(2-)(4). A low-capacity resin-based column was used for the separation and the anions were determined by conductometric detection. p-Hydroxybenzoic acid (5mM) at pH 8.5 was used as the eluent. The limit of detection, defined as S/N = 3, was 92 mug/l. The resolution between Cr(VI) and SO(2-)(4) was 2.8, the precision ranged from 0.9% for NO(-)(3) to 2.0% for Cr(VI) with a 500-mul injection. The SCIC results for Cr(VI) agreed closely with those obtained by inductively coupled argon-plasma emission and spectrophotometry.     

Analysis of prostaglandins in SD rats by capillary zone electrophoresis with indirected UV detection

This paper reports an investigation on optimal separation and detection for animal prostaglandins, i.e. prostaglandin E(2)(PGE(2)) and thromboxane B(2)(TXB(2)) by capillary zone electrophoresis with indirect detection at 254 nm. It was found that the addition of 3 mmol l(-1) phosphomolybdic acid in 5 mmol l(-1) chromate buffer can lower the detection limits of PGE(2) and TXB(2), i.e. the calculated detection limits (k=3) with phosphomolybdic acid is 0.05 mug ml(-1), only a 1/5-fold improvement compared to that without phosphomolybdic acid. Also adding phosphomolybdic acid, the mechanism for improving detection limits was explained properly. The separation and detection of PGE(2) and TXB(2) can be completed in 4 min. The linear ranges for PGE(2) and TXB(2) were the same, i.e. 5 approximately 80 mug ml(-1), analytical precision (n=8) was 1.2 approximately 1.4% and 3.5% for the measurement of migration times and determination of peak height, respectively. The approach was demonstrated in the lung tissue of SD rats, the measurement results were in good agreement with previous investigations.