 |
Introduction |
N-Methyl-2-pyrrolidone
(NMP) is a limpid liquid, which is soluble in water and a wide range of
organic solvents. Its volatility is low (vapor pressure 32 Pa at
25°C). The use of NMP as a solvent is increasing, in particular as a
substitute for methylene chloride in paint strippers. One of the major
uses of NMP is the extraction of aromatics from lubricating oils. NMP
is also used as a vehicle for drugs or to facilitate their percutaneous absorption.
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 intravenous (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
(Bartsh et al., 1976
). 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).
NMP is well absorbed through gastrointestinal, pulmonary tract, and
skin (Midgley et al., 1992; Akesson and Paulsson,
1997; Ursin et al., 1999). Metabolites of NMP
are intensively excreted in urine mainly as 5-hydroxypyrrolidone
(5-HNMP) (Wells et al., 1992; Akesson and
Jonsson, 1997; Payan et al., 2002).
A percutaneous absorption rate of 25.3 µg/cm2/h for NMP
has been calculated from an in vivo experiment in rats with a
co-exposition of NMP and 2-vinylpyrrolidone (Midgley et al.,
1992). This value is about 3 orders of magnitudes lower than
the flux determined (17 mg/cm2/h) in human skin
(Ursin et al., 1999). The low percutaneous flux determined in rats is incompatible with the use of NMP as a
percutaneous absorption enhancer for drugs (Barry and Bennett,
1987; Sugibayashi et al., 1989).
Thus, this work was carried out to determine in vivo and in vitro the
percutaneous absorption of neat [14C]NMP in rats.
Additional experiments were also conducted in vitro with aqueous
solutions of NMP (1:2-1:32, v/v).
| |
Materials and Methods |
Chemicals.
Radiolabeled N-[14C]methyl-2-pyrrolidone
([14C]NMP) was supplied by Amersham International Plc
(Buckinghamshire, England). Radiochemical purity exceeding 99% was
determined by HPLC before each experiment. The specific activity was
1.04 GBq/mmol (28 mCi/mmol).
Animals.
Male haired Sprague-Dawley rats (Iffa Credo,
Saint-Germain-sur-l'Arbresle, France) weighing 250 to 300 g were
used for all studies. The animals were acclimatized to laboratory
conditions for at least 4 days prior to initiating the studies in rooms
with a 12-h light/dark cycle, that were designed to control relative humidity at 50 ± 5% and temperature at 22 ± 1°C.
Commercial food pellets (UAR Alimentation-Villemoison, Epinay-sur-Orge,
France) and tap water were available ad libitum.
In Vivo Percutaneous Penetration and Absorption of
[14C]NMP by Sacrifice.
One day before dosing, the middle of the back of the rats was clipped
with electric clippers, and a circular ring (10 cm2) was
glued. After topical application of neat [14C]NMP, 20 or
40 µl/cm2, the skin was covered by a perforated circular
plastic cap to allow aeration. Batches of three to five hairy male rats
were sacrificed at different times (0.5, 0.75, 1, and 2 h) after
dosing by bleeding the abdominal aorta under mild ether anesthesia. The blood was collected on heparin. After sacrifice, the skin area of the
application site was washed five times with 200 µl of water to remove
the unabsorbed fraction of NMP. A preliminary study showed that 5 min
after topical application of [14C]NMP, more than 95% of
the radioactivity was removed by this washing process
(n = 3). The radioactivity of the application skin area
and the skin area around the ring (about 30 cm2) was
measured after digestion in KOH solution. Radioactivity in the carcass
and excreta (urine and feces) was also analyzed.
In Vivo Percutaneous Penetration and Absorption of
[14C]NMP by Catheterism.
For the sequential collection of urine and blood, a catheter was
introduced into the carotid artery and bladder, respectively, 1 week
before topical administration of [14C]NMP. The catheters
(internal diameter, 0.58 mm; external diameter, 0.96 mm) were
introduced subcutaneously, exteriorized through the back of the neck,
and inserted into a protective stainless tubing (about 2 g in
weight) ligatured firmly to the skin. Urine was excreted by injecting
saline solution (2 ml) into the bladder. Blood was collected on
heparin. The rats were clipped 1 day before dosing. Neat
[14C]NMP (20 µl/cm2) was applied on the
skin (10 cm2). The application area was washed with water
24 h after dosing. Blood and urine were collected at different
times until the animal was sacrificed (72 h).
In Vitro Percutaneous Absorption of NMP.
In vitro percutaneous absorption was assessed with static
diffusion cells using fresh full-thickness skin of male rats (0.9-1.5 mm, 1.3 ± 0.0, n = 73). The rats were sacrificed
with Pentobarbital. The whole dorsal region was shaved, and the excess
of subcutaneous tissue carefully removed. The skin section was cut into
circular sections (four per rat, 1.76 cm2) and placed,
stratum corneum side up, in diffusion cells. The diffusion cells were
maintained at a temperature of 36°C with a circulating water bath,
yielding a skin surface temperature of 32 ± 1°C. The dermis
side was kept in contact with the RPMI receptor fluid (Life
Technologies, Paisley, Scotland) containing 2% bovine albumin and 1%
penicillin-streptomycin. The fluid receptor was previously filtered
through a sterile Millex (Millipore, Bedford, MA) 0.22-µm pore size
filter and degassed with a vacuum pump. Preliminary experiments had
shown that absorption flux was not significantly different when the
receptor fluid was NaCl 0.9%. The integrity of the skin samples was
assessed by determining the trans-epidermal water loss
(Tewameter, TM210, Courage + Khazaka) after an equilibrium time of
1 h.
Neat or aqueous solutions of [14C]NMP were applied on a
skin surface area of 1.76 cm2. The cells were non-occluded.
An aliquot (200 µl) of receptor fluid (5.15 ml) was collected at
regular intervals from 24 to 52 h after sacrifice with an
automatic fraction collector (Gilson FC 204, Middleton, WI). The same
volume of fresh receptor fluid was automatically introduced into the
cell to maintain the volume of the receptor fluid constant. At the end
of the experiment, the unabsorbed dose of [14C]NMP was
removed with water (1 × 500 µl) and cotton swabs. The skin was
digested in KOH (25%, w/v). The radioactivity contained in the
receptor fluid samples, the washing water, and the skin homogenates was
measured by adding 10 ml of liquid scintillation solution (Pico Fluor
30; Packard, St. Louis, MO). Counting efficiency was determined by
quenching correction curves for the various addition and scintillation
fluids with a liquid scintillation spectrophotometer (CA 1900; Packard).
Metabolism of NMP.
Neat NMP was applied (200 µl/cm2) on fresh skin samples
of one rat. After 2 h of exposure, the radioactivity contained in
the receptor fluid, and skin homogenates in water were analyzed by the
HPLC method described below.
Reproducibility.
Maximal percutaneous absorption rate (Fmax) and
the time to reach it (Tmax) were determined in
three independent experiments (three rats, two skin samples per rat)
with neat NMP (400 µl/cm2) for 24 h of exposure.
Fmax was normalized for a skin thickness of 1.3 mm according to eq. 1.
Effect of Neat NMP Doses.
Fmax Nor and Tmax were
determined after topical application of neat NMP (25-400
µl/cm2) for a 24-h exposure period.
Effect of Hydration on NMP Absorption Flux.
Different doses of neat [14C]NMP (100, 200, and 400 µl/cm2) or 3H2O (400 µl/cm2) were applied on skin samples, and aliquots of the
receptor fluid were collected for 4 h. At the end of the
experiment, the volumic radioactive concentration of unabsorbed NMP was
determined. In a second experiment, neat [14C]NMP was
applied on the skin of male rats (400 µl/cm2) aliquot of
unabsorbed dose (10 µl) and of the receptor fluid (160 µl) were
collected at different times for 30 h to determine the evolution
of the volumic radioactive concentration of the unabsorbed dose and
percutaneous absorption fluxes, respectively. Additionally, neat NMP
(400 µl/cm2) was introduced on a glass watch, and
aliquots of NMP were collected to measure the radioactive concentration.
Effect of NMP on the Transfer of 3H2O
from the Receptor Fluid.
Four groups of excised skin (n = 3) were treated as follows.
Group 1: 400 µl of water were deposited on skin sample and
3H2O was introduced into the receptor fluid to
measure the transfer of water from receptor fluid to the deposited
water on skin.
Group 2: 400 µl of 3H2O were
deposited on skin to measure the transfer of water from the deposited
area to the receptor fluid.
Group 3: 400 µL of NMP were deposited on skin and
3H2O was introduced into receptor fluid to
measure the transfer of water from the receptor fluid to deposited skin
NMP.
Group 4: 400 µl of [14C]NMP were deposited
on skin to measure the absorption flux of NMP.
At the end of the exposure (4 h), unabsorbed water or NMP was collected
to measure 3H2O content for groups 1 and 3. For
groups 2 and 4, radioactivity present in the receptor fluid was analyzed.
Effect of Aqueous Dilution of NMP.
An identical volume of aqueous dilutions of NMP (400 µl/cm2) at different concentrations (1:2 to 1:32, v/v)
was applied on the skin. Percutaneous absorption was determined for 24 or 52 h.
Effect of Occlusion or Desquamation.
Percutaneous absorption fluxes of neat NMP (400 µl/cm2)
were compared between skin samples of haired rats, with or without occlusion. Percutaneous absorption fluxes of neat NMP (100 µl/cm2) were compared between skin samples of hairless
rats after desquamation (20 stripping, 100 g/cm2) or
without desquamation.
HPLC Analysis.
An HPLC method to analyze unchanged NMP and its main metabolites has
been previously described (Payan et al., 2002). Briefly, proteins from plasma were precipitated in methanol. Unchanged NMP and
its main metabolites (5-HNMP, 2-hydroxy-N-methylsuccinimide, and N-methylsuccinimide) contained in methanolic phase from
plasma or aliquot of urine are analyzed by HPLC with a mixture of
H2SO4 0.001 N and acetonitrile (85:15, v/v).
The radioactivity contained in the HPLC eluates was measured with a
liquid scintillation spectrophotometer.
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; Packard). 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.
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 gram (%Qo/g) of
fresh tissue. The one-way ANOVA test was used to determine the
significance of the means. The level of significance was set at
p < 0.05.
The terminal elimination rate (
) of NMP and its main metabolites in
plasma were obtained by log-linear concentration time data. The area
under the plasma curves of NMP and its metabolites from time 0 to the
end of the experiment (AUC0-t) were calculated
by the linear trapezoidal rule. The AUC from infinity was estimated by
the calculated concentration at t divided by . The sum of
both areas was AUC0-inf.
The percentage of the absorbed dose was calculated from:
| |
a) the radioactivity content in the excreta and
carcass
|
(1)
|
|
| |
b) the ratio of the AUC of NMP or 5-HNMP after
topical application versus intravenous administration (500 mg/kg)
|
(2)
|
|
| |
c) the ratio of the total radioactivity, NMP, or
5-HNMP excreted in the urine after topical application versus
intravenous administration (500 mg/kg)
|
(3)
|
Intravenous values are from Payan et al. (2002).
|
| |
Results |
Percutaneous Absorption of NMP.
The in vivo percutaneous absorption of [14C]NMP was
determined by sacrificing the animals at different times after topical
application of neat [14C]NMP (20 and 40 µl/cm2). After topical application of 20 µl/cm2 of neat NMP, the percentage of the absorbed dose
increased rapidly with exposure time and accounted for about 60% of
the dose after 2 h of exposure (Table
1). The absorption flux was maximal after 30 min of exposure (9.7 mg/cm2/h) and then decreased (Fig.
1). A similar result was obtained with a
40 µl/cm2 dose, the maximal percutaneous absorption flux
being 23.4 ± 3.4 mg/cm2/h after 45 min of exposure.
View this table:
[in this window]
[in a new window]
|
TABLE 1
Mass balance of 14C after a topical application of neat
[14C]NMP in male Sprague-Dawley rats
Batches of rats were sacrificed at different times after a topical
application of neat [14C]NMP (20 µl/cm2, 10 cm2). Values are expressed as mean ± S.E.M.
(n = 4).
|
|
The time course of [14C]NMP, unchanged NMP, and 5-HNMP in
plasma and urinary excretion after topical application of 20 µl/cm2 are presented in Figs.
2 and 3.
The unchanged NMP plasma levels were maximal 4 h after the
beginning of exposure (0.29 ± 0.01 percent of the applied dose;
0.58 mg/ml) and represented more than 90% of the total radioactivity
in the plasma. The 5-HNMP plasma level was more or less constant
between 6 and 24 h (0.039% of the applied dose per milliliter,
0.08 mg/ml). The NMP urinary excretion rate is peaked between 3 and
4 h after exposure (0.65 ± 0.1% of the dose; 1.3 mg/h). The
total urinary excretion of NMP was 6.9 ± 0.3 percent of the
applied dose. The urinary excretion rate of 5-HNMP reached a peak
between 9 and 24 h (1.7% of the applied dose/h, 3.4 mg/h).
For a 24-h dermal exposure to neat [14C]NMP (20 µl/cm2), the percentage of the absorbed dose calculated
from the content of radioactivity in excreta and carcass was 85 ± 4% of the applied dose. An estimation of the absorbed dose, calculated
from the ratio of the AUC of NMP and 5-HNMP in plasma after topical
application versus an intravenous administration of NMP (500 mg/kg) was
overestimated (102%) (Payan et al., 2002). In contrast,
the percentage of the absorbed dose calculated from the ratio of
urinary excretion of 5-HNMP after topical application versus
intravenous administration, was well estimated (87% of the applied
dose, result not shown).
In Vitro Skin Metabolism of NMP.
After a skin exposure of neat [14C]NMP for 2 h, HPLC
profiles of radioactivity of receptor fluids or skin homogenates
(n = 4) were similar to a standard of
[14C]NMP. Radioactivity eluted at a retention time of
5-HNMP standard is barely detectable (result not shown).
Effect of Skin Thickness and Reproducibility on in Vitro
Percutaneous Absorption Flux.
Skin thickness from shoulder area is significantly higher than that of
the skin from haunches area (Table 2). In
contrast, the maximal percutaneous absorption fluxes
(Fmax) of neat NMP (25-400
µl/cm2) are significantly lower in the skin samples from
the shoulders than those from the haunches. When the maximal fluxes are
normalized for a mean thickness of 1.3 mm
(FmaxNor), no significant difference can be
observed between the fluxes determined with sample skin from the
shoulder or the haunches area.
View this table:
[in this window]
[in a new window]
|
TABLE 2
Effect of the skin thickness of the male rat on the in vitro
percutaneous absorption of neat [14C]NMP
Values are given as mean ± S.E.M. (n = 13 rats).
The skin of the back of each rat was cut into four parts. Parts 1 and 2 correspond to the skin from the shoulders and parts 3 and 4 correspond
to the skin from the haunches. For each animal, parts 1 and 4 or 2 and
3 were tested with the same dose of [14C]NMP (25 to 400 µl/cm2).
|
|
For the same dose of neat NMP (400 µl/cm2), the
coefficients of intra-assay and inter-assay variation for
Tmax and FmaxNor are less
than 30% (Table 3). No significant
differences are observed between the three independent experiments.
View this table:
[in this window]
[in a new window]
|
TABLE 3
Intra- and inter-assay reproducibility of the in vitro percutaneous
absorption of [14C]NMP in full-thickness rat skin
Values are given as mean ± S.D. (n = 3 rats, two
skin samples per rat shoulder and haunch). The skin of the back of each
rat was cut into four parts. For each animal neat [14C]NMP
(400 µl/cm2) was deposited on a piece of skin from the
shoulder (Sh) and from the haunch (Ha) position.
|
|
Effect of the Topical Dose of Neat NMP on in Vitro Percutaneous
Absorption Flux.
Figure 4 shows the change in
percutaneous absorption fluxes with exposure time after different
topical doses of NMP (25-400 µl/cm2). Whatever the dose
of neat NMP, the percutaneous absorption flux increases with time up to
a maximum and then decreases gradually following an exponential
function. Fmax Nor and
Tmax increase as the dose of NMP increases. At
the highest dose, Fmax Nor is 7.7 ± 0.2 mg/cm2/h (Table 4). Variation
of Fmax Nor with the applied dose is well fitted
by the classical saturable kinetic model (Fig.
5). The maximal calculated flux for
infinite NMP dose was 10.7 ± 0.1 mg/cm2/h.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
Time curve of the in vitro percutaneous
absorption flux of neat solutions of NMP.
Values are expressed as mean ± S.E.M. (n = 4-12). Neat solutions of NMP (25 to 400 µl/cm2) were
applied on full thickness skin of male rats. Percutaneous absorption
fluxes are calculated from the radioactivity contained in the receptor
fluid after different exposure times. Absorption flux curves are fitted
by two exponential functions. The decrease in absorption flux is given
by the following equations. 400 µl/cm2 ( ), flux = 8.5 ± 0.3 × exp[( 0.039 ± 0.002 × (t 0.73 ± 0.08)], n = 12; 200 µl/cm2 ( ), flux = 6.9 ± 0.3 × exp[(0.065 ± 0.003 × (t 0.90 ± 0.05)], n = 7; 100 µl/cm2 ( ),
flux = 5.3 ± 0.3 × exp[(0.064 ± 0.007 × (t 0.92 ± 0.07)], n = 4; 50 µl/cm2 ( ), flux = 3.4 ± 0.2 × exp[(0.095 ± 0.011 × (t 0.87 ± 0.09)], n = 6; and 25 µl/cm2 ( ),
flux = 2.1 ± 0.0 × exp[(0.11 ± 0.01 × (t 0.78 ± 0.01)], n = 6.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 4
In vitro percutaneous absorption of different doses of neat
[14C]NMP into full-thickness skin of male rats
Values are given as mean ± S.E.M. Increasing volumes of neat
[14C]NMP were deposited on full thickness skin of male rats.
For each rat, a piece of haunch and shoulder was treated with the same
volume of NMP.
|
|
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5.
Variation of maximal absorption fluxes with
the applied doses.
Values are expressed as mean ± S.E.M. Values were fitted with a
non-linear regression model following the classical equation of a
saturable kinetic model. The parameters are flux max 10.7 ± 0.1 mg/cm2/h and K (NMP dose that leads to have a
flux equal to half of flux max = 104 ± 5 µl/cm2.
|
|
Effect of Hydration of the Unabsorbed Dose of NMP on in Vitro
Percutaneous Absorption Flux.
Neat [14C]NMP with no initial volumic radioactive
concentration were applied on skin samples (400 µl/cm2).
Figure 6 represents the change with time
of the ratio of volumic radioactivity of non-absorbed NMP versus
initial volumic radioactive concentration of NMP and of the absorption
flux. The volumic radioactive concentration of unabsorbed NMP decreased
with time, which indicated a progressive dilution of the unabsorbed NMP
fraction. As previously indicated, the flux of NMP reaches a maximum
then gradually declines. In contrast, when the flux is related to the
dilution factor of unabsorbed NMP, the flux tends towards a constant
value as time goes on.
Different doses of neat [14C]NMP (100, 200, and 400 µl/cm2) with the same volumic radioactive concentration
where applied on skin. After 4 h of exposure, the volumic
radioactive concentrations of unabsorbed NMP and the
Fmax Nor were proportional to the dose (results
not shown). In contrast, the FmaxNor divided by
the dilution factor of the initial concentration of the unabsorbed dose
were not significantly different for the three NMP doses tested.
Effect of NMP on the Transfer of 3H2O from
the Receptor Fluid.
Transfer of 3H2O from receptor fluid or
deposited on skin were similar and accounted 2.5 mg/cm2 for
4 h of experiment (Table 5). When
skin is exposed with neat NMP, the transfer of
3H2O from the receptor fluid was more than
10-fold higher (36 mg/cm2).
View this table:
[in this window]
[in a new window]
|
TABLE 5
Effect of NMP on the transfer of 3H2O from the receptor
fluid
Values are expressed as mean ± S.E.M. (n = 4).
Transfer of 3H2O from the receptor fluid was determined
from the radioactivity content recovered in the unabsorbed fraction of
H2O or NMP deposit on skin after 4 h of exposure.
|
|
Effect of Desquamation and Occlusion on in Vitro Percutaneous
Absorption Flux.
Occlusion does not affect the percutaneous absorption of neat NMP
(Table 6). Desquamation increases
Fmax slightly (+17%) but does not modify
Tmax.
View this table:
[in this window]
[in a new window]
|
TABLE 6
Occlusion and desquamation effects on in vitro percutaneous absorption
of neat NMP into full thickness skin of male rats
Values are expressed as mean ± S.E.M. (n = 3 rat,
two skin samples per rat and per modality). Neat NMP was deposited on
full thickness skin of male rats for 24 h. Fmax
normalized to a skin thickness of 1.3 mm divided by the concentration
of NMP.
|
|
Effect of Aqueous Dilution of NMP on in Vitro Percutaneous
Absorption Flux.
For different volumes of neat or diluted NMP aqueous solution (1:1 to
1:32, v/v), the relative constant of permeability
(FmaxNor divided by the NMP concentration) and
Tmax were determined after 24 or 52 h of
exposure (Table 7). Except for the lower
volume of diluted NMP (25 µl/cm2),
Kpmax is independent of the volume of diluted
NMP applied on the skin. In contrast, the time to reach maximal
percutaneous absorption flux increases as the NMP dilution increases.
View this table:
[in this window]
[in a new window]
|
TABLE 7
In vitro percutaneous absorption of aqueous dilutions of NMP into full
thickness skin of male rats
Values are expressed as mean ± S.E.M. (two skin samples per rat).
A volume of 400 µl of neat or aqueous dilution was deposited on full
thickness skin of male rats for 24 h or 52 h.
Kpmax (10 3 cm/h) is calculated from
Fmax normalized to a skin thickness of 1.3 mm
divided by the concentration of NMP.
|
|
Modeling of NMP Transfer.
To take into account the transfer of water from the receptor fluid, a
simplified mathematical model has been developed (Appendix 1). The
"non-dimensional" evolution of the flux
(
a) with a non-dimensional time (
) is
defined as a sum of two exponential functions µ and µ0
are constants without dimension.
The second exponential term corresponds to the decrease of the
absorption flux with time after that maximum absorption flux was
reached. For a long time, the evolution of the slope of ln = P(e) with the dose of NMP(e) should be
able to be expressed by a relationship of the type
P(e) = A/e; where
A is a constant. However, the relation obtained from the
experimental data gave a relationship
ln(P(e) = ln(0.07) 0.34 ln(e) which correspond to the approximative relation of
P(e) = P(e) = K/3
(Fig.
7).
| |
Discussion |
In vivo and in vitro experiments have shown that NMP penetrates
the skin of rats rapidly and intensively. Fifteen minutes after topical
administration of neat NMP (20 µl/cm2), the skin
contained 2 mg of NMP per square centimeter. This value remained more
or less constant for 2 h. The absorbed dose increased linearly
with time for 30 min and then declined, although more than 50% of the
applied dose was not absorbed. The maximal flux was 10 mg/cm2/h. A similar decline in flux was also observed
1 h after topical application of 40 µl/cm2 of neat
NMP. The flux calculated was 20 mg/cm2/h, which is similar
to the flux determined in vitro with human skin (Ursin et al.,
1995). A very low percutaneous absorption flux (25.3 µg/cm2/h) was calculated after topical application of a
mixture of NMP and 2-pyrrolidone in isopropanol (Midgley et al.,
1992). Differences in the dosing conditions could explain the
low flux previously reported.
Twenty-four hours after topical application of neat NMP (20 µl/cm2), 80% of the dose had penetrated into the skin
and less than 2% of the applied dose was recovered in the
H2O and CO2 traps. This result indicates that
NMP is intensively absorbed and weakly evaporated. In contrast, 24 h after topical application of neat NMP (10 µl/cm2) in
Sprague-Dawley rats, only 32% of the dose had been absorbed (Bourds et al., 1999). However, in this latter
experiment, a large fraction of radioactivity (67%) was found in the
charcoal filter covering the skin application site.
The AUC ratio of plasma of total radioactivity, unchanged NMP, or
5-HNMP after topical administration versus an intravenous administration overestimates the absorbed dose. This is probably the
result of metabolic saturation of NMP (J.-P. Payan et al., 2002). The maximal plasma concentration of unchanged NMP was
higher after topical application than after intravenous administration of the highest NMP dose tested. Similarly, the AUC ratio data of plasma
radioactivity after topical administration versus oral dosing an
coadministration of NMP and 2-pyrrolidone in rats gave a poor
estimation of the absorbed dose (Midgley et al., 1992). In contrast, the total absorbed dose was accurately estimated from the
ratio of the urinary excretion of this compound after topical versus
intravenous administration.
The decline in flux with time of exposure and the increase of flux with
the dose were also observed in vitro. Whatever the dose of neat NMP
applied on samples of rat skin, no steady-state percutaneous absorption
flux was obtained. In particular, at the highest dose tested (400 µl/cm2), the percutaneous flux reached a maximum about
6 h after the beginning of exposure and then declined. The
decrease in percutaneous flux with exposure time was not the result of
a lack of product on the skin. The dose used can be considered as "an
infinite dose." At the end of the experiment, 47% of the dose was
unabsorbed. It is unlikely the fall in percutaneous was a result of a
macroscopic change of the characteristic of the skin as verified by
microscopy (results not shown). In contrast, a decrease in the
concentration of unabsorbed dose with time correlated well with the
decrease in percutaneous absorption flux. The dilution of the
unabsorbed NMP resulted from the hygroscopic properties of NMP. The
dilution of unabsorbed dose was mainly the consequence of a transfer of the receptor fluid through the skin.
From the evolution of the radioactive concentration of unabsorbed NMP
fraction with time of exposure, the transfer rate of water through the
skin was calculated to be about 10 mg/cm2/h for the first
4 h of exposure at the highest dose. A similar value was obtained
with the two other doses of neat NMP tested (100 and 200 µl/cm2). A counter-experiment with
3H2O introduced in the receptor fluid confirmed
these results. When neat NMP or water was deposited on skin, the
transfer rates of 3H2O from the receptor fluid
were 9 mg/cm2/h and 0.6 mg/cm2/h, respectively.
This finding indicates that the progressive dilution of the unabsorbed
dose of NMP with time was a consequence of the hygroscopic properties
of NMP. These results explained the increase in maximal absorption flux
and in the time to reach it when the dose increased.
A simplified semiquantitative model of the transfer of NMP through the
skin and its dilution over time is proposed. The evolution of the
absorption flux with time can be represented by the sum of two
exponential functions, which is compatible with the experimental data.
However, in this model it is postulated that for a short time of
exposure the increase of the absorption flux with time was independent
of the doses, which is not fully supported by the experimental data.
Moreover, the decline phases of the experimental absorption fluxes were
not inversely proportional to the doses as predicted by the
mathematical model. This discrepancy can be explained by several
reasons that are not taken into account in the mathematical model
(e.g., the diffusion coefficient is independent of the skin
thickness, no interaction between (water and NMP), and exponential
dilution, etc. Also, the transfer of NMP was assumed independent of
other substances that migrate. Moreover, it could be possible to make
the hypothesis of a value of diffusion coefficient dependent on NMP solvation.
The maximal absorption flux with skin of rats (7.7 mg/cm2/h) is lower than the absorption flux determined in
vitro in human skin (17 mg/cm2/h) (Ursin et al.,
1997). The difference in flux between the two species may
result from skin thickness differences. The thickness of the skin was
0.2 to 0.4 mm and 1.1 to 1.5 mm for the human and rat skin experiments,
respectively. It has been shown that, in the rat, the percutaneous
absorption flux depends greatly on the thickness of the skin. The
absorption flux determined in vitro is lower than that in vivo. The
difference in flux between the in vivo and in vitro experiment in rats
may be due to the microcirculation, which affects the delivery of drug
to the systemic circulation.
From in vitro results, it seems that NMP is not significantly
metabolized during its transfer through the skin. Indeed, radioactivity contained in fresh skin or in the receptor fluid after a 2-h exposure of [14C]NMP was predominantly unchanged NMP. Moreover,
the absorption flux is dependent on the thickness of the skin and on
the concentration of the applied dose of aqueous solution of NMP.
Additionally, skin desquamation only slightly increased the
percutaneous absorption flux of NMP. All these findings strongly
suggest that the percutaneous absorption of NMP is a passive diffusion process.
In conclusion, NMP is intensively absorbed through the skin in vitro
and in vivo in the rat, as indicated in vitro in humans. The
hygroscopic properties of this compound affect the process of its
absorption drastically.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
![&psgr;<SUP>a</SUP>=<FR><NU>&mgr;<SUB>0</SUB></NU><DE>&mgr;−&mgr;<SUB>0</SUB></DE></FR> [<UP>exp</UP>[<UP>−</UP>&mgr;<SUB>0</SUB>(&tgr;−&tgr;<SUB>0</SUB>)]−<UP>exp</UP>[<UP>−</UP>&mgr;(&tgr;−&tgr;<SUB>0</SUB>)]].](/content/vol31/issue5/fulltext/659/img002.gif)
= 1 (x = L).
![&psgr;<SUP>o</SUP><SUB>1</SUB>=1+2 <LIM><OP>∑</OP><LL>n=1</LL><UL>∞</UL></LIM> (<UP>−</UP>1)<SUP>n</SUP> <UP>exp</UP>[<UP>−</UP>n<SUP>2</SUP>&pgr;<SUP>2</SUP>&tgr;]](/content/vol31/issue5/fulltext/659/img015.gif)
is a constant.
Taking into account the change in the preceding variables, µ is
defined by
1 is defined by the expression:
![&psgr;<SUP>a</SUP>=<FR><NU>&mgr;<SUB>0</SUB></NU><DE>&mgr;−&mgr;<SUB>0</SUB></DE></FR>[<UP>exp</UP>[<UP>−</UP>&mgr;<SUB>0</SUB>(&tgr;−&tgr;<SUB>0</SUB>)]−<UP>exp</UP>[<UP>−</UP>&mgr;(&tgr;−&tgr;<SUB>0</SUB>)]]](/content/vol31/issue5/fulltext/659/img036.gif)
of these curves for
different values of thickness e of NMP deposited. If the
transfer flux of the water is more or less constant, in principle an
even more rapid dilution of NMP in the water can be expected as
e becomes smaller. In this case P(e)
should be able to be expressed by a relationship of the type:
0.07 × e