Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

N-Methyl-2-pyrrolidone (NMP¹) is a widely and increasingly used solvent. It dissolves most monomers and polymers and catalyzes many polymerization reactions. One of its main uses is in the petrochemical industry to extract aromatic oils. It is also used in the microelectronics fabrication industry to manufacture electronic capacitors and batteries. It is a solvent and cosolvent in pharmaceuticals, cosmetics, insecticides, herbicides, or in paint preparations. Finally, NMP is increasingly being used to remove graffiti (Akesson, 1994).

A recent study of NMP use in the microelectronics industry indicates that severe eye irritation and headaches can be expected at concentrations as low as 0.7 ppm, even for a short period of exposure (30 min) (Beaulieu and Schmerber, 1991). After two days of working with NMP in a electrotechnical company, 10 of the 12 workers involved displayed acute irritant contact dermatitis of the hands (Leira et al., 1992).

Acute toxicity is low. LD₅₀ in rabbits was 4 to 8 g/kg and 1.5 to 7 g/kg in rats after topical application. The LD₅₀ in mice after i.v., intraperitoneal, and oral administration was 3.5, 4.3, and 7.5 ml/kg, respectively. In rats, the LD₅₀ was 2.2, 2.4, and 3.8 ml/kg, respectively (Bartsch et al., 1976).

In a subchronic feeding study on beagles (13 weeks old, 25 to 250 mg NMP/kg), no statistically significant treatment-related effects judged biologically meaningful were observed in any parameter of either the male or female animals exposed. However, a dose-dependent decrease in body weight and increase in platelet count correlated with increased megakaryocytes was observed. In addition, serum cholesterol in male rats decreased with increasing doses (Becci et al., 1983).

Studies on reproductive toxicity have shown that NMP causes developmental toxicity at doses causing no or mild maternal toxicity (Hass et al., 1994; Solomon et al., 1995). Cases of stillbirth after occupational exposure to NMP have been reported (Solomon et al., 1996; Bower, 1997).

Animal studies have shown efficient absorption of NMP through the skin or gastrointestinal tract (Midgley et al., 1992) and the respiratory tract (Ravn-Jonsen et al., 1992). After inhalation exposure (150 ppm for 8 h), the elimination of unchanged NMP in plasma was a zero order reaction both in nonpregnant and pregnant rats. The volume of distribution was estimated to be 0.7 l/kg (Ravn-Jonsen et al., 1992). After an intravenous administration of [¹⁴C]NMP (45 mg/kg), radioactivity was extensively distributed to all major organs (Wells and Digenis, 1988). About 80% of the radioactivity administered was excreted in the first day after dosing. The major metabolite, which represented 70 to 75% of the administered dose, was identified later as 5-hydroxy-N-methylpyrolidone (5-HNMP) (Wells et al., 1992). In the first hours after oral or dermal exposure, unchanged NMP accounted for most of the radioactivity in the plasma. The half-life of unchanged NMP in the plasma was estimated to be 9 to 12 h.

After experimental inhalation exposure (10 to 50 mg/m³, 8 h) in human volunteers, the concentration of unchanged NMP in the plasma was maximum at the end of the exposures and declined thereafter, following a nonlinear pattern (Akesson and Paulsson, 1997). The mean plasma half-life was estimated to be 4 h. Unchanged NMP excreted in the urine during and 44 h after exposure accounted for about 2% of the calculated inhaled dose, and the urinary mean half-life was 4.5 h. 5-HNMP (94 mg/l) and 2-hydroxy-methylsuccinimide (2-HNMS) (6.7 mg/l) were also present in the urine at the end of an 8-h exposure at 25 mg/m³ (Jonsson and Akesson, 1997). Plasma and urinary half-life was about 6 to 8 h (Akesson and Jonsson, 2000). Similar half-life was calculated for N-methylsuccinimide after an 8-h NMP exposure (Jonsson and Akesson, 2001). After oral administration of 100 mg to volunteers, unchanged NMP was only detected in the urine during the first day and represented 0.8% of the administered dose (Akesson and Jonsson, 1997). 5-HNMP was excreted in the urine during the first two days with a half-life of about 4 h and represented 44% of the dose. 2-HNMS was detected for 6 days and accounted for 20% of the dose with a half-life of 17 h. N-Methylsucinimide (NMS), which was excreted in the urine over the first 3 days, was a minor metabolite (0.4% of the dose). The increasing half-lives of NMP < 5-HNMP < NMS < 2-HNMS suggest the existence of a metabolic pathway in which NMP is first hydroxylated to 5-HNMP and then oxidized to NMS, which in turn is hydroxylated to 2-HNMS (Akesson and Jonsson, 1997). It has been shown from the radioactivity measured in excreta that between 73 and 82% of a topical application dose of NMP in mixture with 2-vinylpyrrolidone was absorbed through rat skin (Midgley et al., 1992). The percutaneous absorption rate was calculated by the authors to be 25.3 µg/cm²/h. In contrast, the percutaneous absorption rate was determined to be 3 orders of magnitude higher in human skin (Ursin et al., 1995). This latter value seems more realistic as NMP is known to enhance the absorption of different compounds through the skin (Priborsky et al., 1988).

A zero order and a nonlinear elimination of unchanged NMP in rats and human plasma, respectively, suggests a saturable NMP elimination process. Thus, this study was carried out to confirm this hypothesis. To achieve this, the toxicokinetics of NMP and its main metabolites in plasma and urine were determined at different intravenous doses (0.1 to 500 mg/kg).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. Radiolabeled N-[¹⁴C]methylpyrrolidone ([¹⁴C]NMP) was supplied by Amersham Biosciences UK, Ltd. (Little Chalfont, Buckinghamshire, UK). Radiochemical purity exceeding 99% was determined by HPLC before each dosing. The specific activity was 1.04 GBq/mmol (28 mCi/mmol). Unlabeled NMP, 2-HNMS, and NMS with a purity of >98% were purchased from Sigma-Aldrich Chimie S.A.R.L (Saint-Quentin Fallavier, France). All the other reagents and chemicals were obtained from commercial sources at the highest purity available. 5-HNMP was synthesized according to Hubert et al. (1975).

Animals. Male Sprague-Dawley rats (Iffa Credo, Saint-Germain-sur-l'Arbresle, France) weighing 250 to 300 g were used for all the studies. The animals were acclimatized to laboratory conditions for at least 4 days prior to initiating the studies in a room with a 12-h light/dark cycle designed to control relative humidity 50 ± 5% and temperature 22 ± 1°C. Commercial food pellets (UAR Alimentation-Villemoison, Epinay-sur-Orge, France) and tap water were available ad libitum.

Toxicokinetics after Intravenous Administration. [¹⁴C]NMP in saline solution was administered intravenously (1 ml/kg) into the dorsal vein of the penis of lightly etherized rats. Individual doses of [¹⁴C]NMP (0.1, 1, 10, 100, and 500 mg/kg) were determined by weighing the syringe before and after each administration. After dosing, the animals (n = 3-6) were immediately placed in individual metabolism plastic cages for urine and feces collection. Additionally, exhaled air from three rats per NMP dosing was collected every 24 h for 72 h. [¹⁴C]NMP or its volatile [¹⁴C] metabolites and ¹⁴CO₂ were collected by extracting the air from the individual metabolism glass cages through a series of two water traps and a trap containing 100 ml of Carbosorb (PerkinElmer Life Sciences, Boston, MA) respectively.

In Vitro Metabolism of [¹⁴C]NMP. A first experiment was conducted to determine the Michaelis-Menten parameters of NMP hydrolysis in microsomes. The microsomes were prepared from the liver of control male rats (250 g). The liver (12 g) was homogenized in an ice-cold mixture (1:2, w/v) of phosphate buffer (0.01 M, pH 7.4) and KCl (0.15 M) and immediately centrifuged at 9000g for 15 min at 5°C. The S9 fraction was centrifuged at 100,000g for 1 h. The microsomal fraction was stored at 80°C for up to 8 weeks. Protein concentration (Lowry method; Lowry et al., 1951) was 16.5 mg/ml and about 1.2 g of fresh liver/ml. The metabolization step used a mixture of microsomal fraction (200 µl, 3.3 mg of protein) and phosphate buffer (50 µl, 0.01 M, pH 7.4), NADPH (100 µl, 40 mM). The reactions were initiated by adding 150 µl of [¹⁴C]NMP (5500 Bq, 0.625 to 20 mM). The tubes were closed with screw plugs. After vortex mixing for 5 s, the mixture was incubated at 37°C for 15 min with constant shaking. Enzymatic reactions were stopped by adding 10 µl H₂SO₄ at 10% (v/v). The incubates were analyzed by HPLC.

Analysis of Radioactivity. Samples of urine (500-1000 µl) and plasma (500 µl) were accurately weighed and added directly to liquid scintillation vials containing 10 ml of liquid scintillation solution (Pico Fluor 30; PerkinElmer Life Sciences). Samples of fresh feces were weighed and homogenized in water (1:5, w/v) in glass vials. Tissues (liver or kidney) were homogenized in water (1:5, w/v). Aliquots of feces or tissue homogenates (250-500 mg) were mixed with 10 ml of Pico Fluor 30. The radioactivity of all the samples was measured in a Packard liquid scintillation spectrophotometer model 1900. Counting efficiency was determined by quenching the correction curves of the various additions and scintillation fluids.

HPLC Analysis of [¹⁴C]NMP. Proteins from an aliquot of plasma (50 to 200 µl) were precipitated in methanol (2000 µl) at 20°C for 1 h. After centrifugation (2,500g, 10 min), the precipitate was washed twice with 500 µl of methanol. The methanolic phases were pooled and concentrated to about 100 to 200 µl under a nitrogen flux at 80°C. The radioactivity contained in 100 µl of the methanolic concentrate from plasma or of the supernatant from urine (12,000g, 1 min) were analyzed by HPLC. The HPLC system used comprised a Waters 717⁺ autosampler (Waters, Milford, MA), a Waters 501 isocratic pump, a Waters 490E UV detector set at 263 nm, and a styrene divinylbenzene column (4.5 × 250 mm, Sugar SH 1011, Shodex; Showa Denko K.K., Tokyo, Japan).

Expression of Data and Statistical Analysis. Values were expressed as the percentage of [¹⁴C]NMP dose per organ (% Qo/organ) or as the percentage of [¹⁴C]NMP dose per g (% Qo/g) of fresh tissue. The one-way analysis of variance test was used to determine the significance of the means. The level of significance was set at p < 0.05. The toxicokinetics parameters in plasma and urine were calculated by an urinary clearance method (InnaPhase software, Philadelphia, PA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Toxicokinetics Parameters after Intravenous Administration. Table 1 shows the mass balance radioactivity 72 h after a single intravenous administration of [¹⁴C]NMP (0.1, 1, 10, 100, or 500 mg/kg). The total radioactivity excreted in the urine over 3 days (85.4 ± 1.2% of the dose, n = 27) or in the feces (3.4 ± 0.2% of the dose) was independent of the administered dose. However, urinary excretion rate was about 2- and 4-fold lower over the 4 to 8 h period after administration of the 100 and 500 mg/kg doses, respectively, compared with the 0.1 to 10 mg/kg doses. Pulmonary excretion of ¹⁴C was low at the lowest dose but increased gradually with increasing doses and accounted for more than 2% of the two highest doses.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Plasma time-curve of 14C, unchanged NMP, and its two main metabolites after a single intravenous administration of 0.1 mg/kg [14C]NMP in male Sprague-Dawley rats. Values are expressed as mean ± S.E.M. (n = 5). Symbols without bars indicate that the S.E.M. is within the symbol. LOD and LOQ are the limits of detection and quantification, respectively. , 14C; , NMP; , 5-HNMP; ×, NMS.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Urinary excretion rate of 14C, unchanged NMP, and its two main metabolites after a single intravenous administration of 0.1 mg/kg [14C]NMP in male Sprague-Dawley rats. Values are expressed as mean ± S.E.M. (n = 5). LOD and LOQ are the limits of detection and quantification, respectively. , 14C; , NMP; , 5-HNMP; ×, NMS.

Metabolic Rate of NMP. The metabolic rate of NMP was determined from the disappearance of NMP from the plasma after the distribution phase minus the urinary excretion rate. The estimation of the Michaelis-Menten constants K_m and V_m is 2.0 mM and 634 nmol/min (3.8 mg/h) for a rat weighing 250 g (Fig. 3a).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   In vivo and in vitro [14C]NMP metabolism rate in male Sprague-Dawley rats. a, the rate of in vivo NMP metabolism was estimated from the disappearance rate of unchanged NMP in the plasma (dCp/dt × Vd) after intravenous administration of [14C]NMP minus the urinary excretion rate. The enzymatic parameters determined by a nonlinear regression method were Km = 2.0 mM and Vm = 634 nmol/min. b, [14C]NMP from 0.625 to 20 mM were incubated for 15 min in the microsome fraction of rat liver containing NADPH. The radioactivity contained in the incubate was analyzed by an HPLC method. The enzymatic parameters for the NMP metabolism were Km = 5.10 ± 0.4 mM and Vm = 0.28 ± 0.02 nmol/min/mg of protein.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study confirm that NMP is extensively metabolized and rapidly excreted in urine. Thus, about 86% of the administered [¹⁴C] activity was excreted in the urine 72 h after a single intravenous administration (0.1 mg/kg to 500 mg/kg of [¹⁴C]NMP), and 90% of the total radioactivity recovered in the urine was excreted within 1 day of being administered. The volume of distribution was 70% of body weight, which corresponds to the total aqueous volume of the animal. These results are in accordance with earlier animal studies (Wells and Digenis, 1988; Ravn-Jonsen et al., 1992).

In the first hour after a single i.v. administration, unchanged NMP represented more than 90% of the plasma radioactivity, as observed after oral and dermal administration (Midgley et al., 1992). For the three lowest doses (0.1, 1, and 10 mg/kg), after an initial fast distribution phase, unchanged NMP declined linearly with time until 3 h after administration. It then declined exponentially with a half-life of 0.8 h. This value is about 10 times lower than the value determined after intravenous administration of 45 mg/kg (6 to 9 h) (Wells and Digenis, 1988). This discrepancy may be the result of the high dosage and the protocol design in the previous study. This author measured the NMP level in the plasma of the rats until 6 h after administration of the toxicant. During this period, if the drop in unchanged NMP of a 10, 100, or 500 mg/kg dose is fitted by an exponential function, the slope will give also an apparent half-life of about 7 h. The slow decline in unchanged NMP was attributed to release from a deep compartment or a storage depot, such as fat. However, the decline in unchanged NMP was approximately linear with time until 24 h at the two highest doses tested in the present paper (100 and 500 mg/kg). This decline in unchanged NMP can be explained by a NMP metabolic saturation step. The values of Michaelis-Menten constant calculated from the plasma concentrations of unchanged NMP and determined in vitro from liver microsomes are very similar, corresponding to 2.0 and 5.1 mM, respectively. At the highest dose, the concentration of unchanged NMP was 5 mM, roughly the same as the K_m values. Moreover, the rate of disappearance of unchanged NMP was 2.6 mg/h, which is in accordance with the V_m value determined from the in vivo results (3.8 mg/h). However, this V_m value is about 20 times higher than that estimated from the in vitro experiment (0.2 mg/h). The 20-fold differences between the in vitro and in vivo calculated V_m could be the consequence of an extra-hepatic metabolism. The metabolic saturation step could explain the zero order reaction for the elimination of unchanged NMP in the plasma of nonpregnant and pregnant rats after inhalation exposure (150 ppm over 8 h) (Ravn-Jonsen et al., 1992).

A metabolic saturation step does not explain the linear decline in unchanged NMP observed during the first 3 h after administration of the 0.1 to 10 mg/kg doses. For these three doses, all the toxicokinetics parameters are very similar. Additionally, the unchanged NMP concentrations in the plasma were very low (0.001 to 0.095 mM) compared with the value of K_m (> 2 mM). Moreover, for the first 3 h after administering the NMP, 5-HNMP levels in plasma were very low and then increased to a maximum, after which the decrease in unchanged NMP followed an exponential function. From the ex vivo experiment, it is unlikely that these findings were the consequence of an enzyme induction. Indeed, the metabolic rates of 5-HNMP formation in the S9 fraction of liver from rats pretreated or not with NMP were not significantly different.

Unchanged NMP was not bound to plasma proteins and only 4 to 5% of the dose was excreted in the urine for the lowest administered doses of NMP. The low urinary clearance of NMP indicates intensive tubular reabsorption of the compound. In contrast, clearance of 5-HNMP (1 ml/min), similar to inulin clearance, suggests a simple glomerular filtration of the metabolite. 5-HNMP is the main urinary metabolite of NMP. Total urinary excretion accounted for 42 to 45% of the administered doses (0.1 to 10 mg/kg), which is similar to the value obtained in human volunteers after oral administration of 100 mg/kg and after inhalation exposure (Akesson and Jonsson, 1997; 2000). For the two highest doses, more than the half of the dose administered was excreted as 5-HNMP in the urine. This latter value is lower than the total excretion rate observed after intravenous administration in rats, in which 75% of the dose was excreted as 5-HNMP (Wells et al., 1992).

In conclusion, unchanged NMP is intensively reabsorbed by the glomerule, and its metabolism follows a saturable process.