Synthesis, Hydrolysis, and Skin Retention of Amino Acid
Esters of a
a-Tocopherol
FABIO MARRA,1 CARMINE OSTACOLO,2 SONIA LANERI,2 ANTONIETTA BERNARDI,2 ANTONIA SACCHI,2
CRISTINA PADULA,1 SARA NICOLI,1 PATRIZIA SANTI1
1
Dipartimento Farmaceutico, University of Parma, Viale G.P. Usberti 27/A, 43100 Parma, Italy
2
Dipartimento di Chimica Farmaceutica e Tossicologica, UniversitÄ… di Napoli   Federico II  , Via D. Montesano 49,
80181 Napoli, Italy
Received 12 October 2007; accepted 16 September 2008
Published online 30 October 2008 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.21608
ABSTRACT: The aim of this work was to synthesize new pro-vitamins of a-tocopherol
(VE) able to release another moiety such as an amino acid, in order to obtain a combined
antioxidant and moisturizing effect upon topical application. The new derivatives were
characterized and tested for sensitivity to chemical and enzymatic hydrolysis. Lipophi-
licity was estimated using Log capacity factor and skin retention was determined
in vitro, using rabbit ear skin as barrier. Five molecules were synthesized using
L-proline, L-serine, L-tyrosine, L-asparagine, and L-citrulline as amino acidic moiety.
All pro-vitamins showed similar or lower lipophilicity than a-tocopheryl acetate (VEAc),
taken as reference, and similar stability in aqueous solutions. All pro-vitamins showed to
be sensitive to enzymatic hydrolysis. None of the pro-vitamins crossed the skin in
significant amounts, whereas they accumulated into the skin, in both the dermis and the
epidermis. They are more hydrophilic and more water-soluble than the currently
used acetate. ß 2008 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci
98:2364 2376, 2009
Keywords: tocopherol; pro-vitamin; amino acids; esterase; skin metabolism
INTRODUCTION effect, limiting trans-epidermal water loss, and
seems to be able to regulate keratin turnover. For
Tocopherols are a class of lipophilic compounds, stability reasons, a-tocopherol is normally used
globally known as vitamins E. They exist as four in the form of esters, a-tocopheryl acetate4,5 or
homologous, a, b, g, and d, which differ in the succinate, which once absorbed into the skin are
number and position of the methyl groups in the hydrolyzed to a-tocopherol, the active form of
chroman ring.1 a-Tocopherol (VE) is the most the vitamin.6 9 Recently, new derivatives of a,10
active lipophilic antioxidant in biological mem- d,11 and g12 tocopherol were proposed. The
branes2 and, together with ascorbic acid, constitu- derivative d-tocopherol glucoside was shown to
tes the   antioxidant network  that protect skin be metabolized to d-tocopherol in vitro and had a
against oxidative damage.3 Moreover a-tocopherol considerable reservoir effect, associated with
plays different roles in the maintenance of skin gradual delivery of free tocopherol.11 Topical
physiological conditions. It shows a moisturizing application of the derivative g-tocopherol-N,N-
dimethylglycinate hydrochloride in vivo in mice
had a protective effect from UV-induced skin
Correspondence to: Patrizia Santi (Telephone: 39-0521-
damage.12 We synthesized and tested in vitro some
905069; Fax: [phone]; E-mail: [e-mail])
amino acid derivatives of a-tocopherol10 with non-
Journal of Pharmaceutical Sciences, Vol. 98, 2364 2376 (2009)
ß 2008 Wiley-Liss, Inc and the American Pharmacists Association. functionalized side chain amino acids (L-glycine
2364 JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009
AMINO ACID ESTERS OF a-TOCOPHEROL 2365
and L-alanine) and pyroglutamic acid. The idea was (Fiorenzuola d Arda, Piacenza, Italy). All other
to combine the antioxidant effect of a-tocopherol chemicals used for buffer and HPLC solvent
with the moisturizing effect of amino acids or preparation were of analytical grade. Reagent
pyroglutamic acid. In fact, the natural moisturiz- grade chemicals and solvents were used for the
ing factor, which consists of 16 different amino synthesis.
acids (40%), pyroglutamic acid (12%) and other
water soluble molecules, acts to preserve the
optimal hydration status of stratum corneum.13
Synthesis and Identification of
The derivatives synthesized were demonstrated to
a-Tocopherol Derivatives
be sensitive to enzymatic hydrolysis and were able
to regenerate a-tocopherol in the skin in vitro.
VE pro-vitamins were synthesized as described in
The purpose of the present work was to
Figures 1 4.
synthesize and test other new a-tocopherol pro-
The course of reactions and purity of products
vitamins, via ester linkage, rapidly reconverted
were controlled by TLC, using pre-coated silica gel
by esterases in the skin, and able to release
plates (Merck 60 F254) and ethyl acetate/n-
tocopherol and an amino acid, so to play a
hexane and ethyl acetate/methanol mixtures as
combined antioxidant and moisturizing effect on
eluents. Preparative separations were performed
skin. The new derivatives were synthesized using
in columns containing Merck 60 silica gel.
L-proline, L-tyrosine, L-serine, L-asparagine, and
Elemental analysis results were within 0.4%
1 13
L-citrulline as hydrophilic moieties. These amino
of the theoretical values. H and C NMR were
acids are all components of the natural moisturiz-
recorded using a Varian Mercury 400 or a Varian
ing factor, moreover, some of them play an
Mercury 500 spectrometer equipped with a
important role in skin biochemistry. In particular
L-serine stimulates the activity of serine palmitoyl
transferase, that catalyzes the rate limiting step
of ceramide biosynthesis14 and L-tyrosine is well
known as stimulating factor for melanin biosynth-
esis.15 Proline improves skin texture and aids
collagen formation.16
A new HPLC method was set up, in order to
analyze simultaneously pro-vitamins and a-toco-
pherol. The derivatives were characterized and
tested for sensitivity to chemical and enzymatic
hydrolysis. The skin retention and metabolism of
the new pro-vitamins was determined in vitro,
using rabbit ear skin, which has a permeability
comparable to human skin17,18 (ratio rabbit/
human skin permeability 1.2 and 1.6 for lidocaine
and thiocolchicoside, respectively).
EXPERIMENTAL
a-Tocopherol (m.w. 430.72, yellow oil), a-tocopheryl
acetate, as well reactants and catalysts, were
purchased by Sigma (Sigma Chemical, St. Louis,
MO), while all amino acids and anhydrous solvents
were obtained by Fluka (Fluka Chemie, Buchs,
Switzerland). All other solvents were of analytical
grade and were purchased by Riedel-de Haen
(Seelze, Germany).
Dimethyl-b-cyclodextrin (DM-b-CD) was pur-
chased from Wacker-Chemie (Milan, Italy), while
polysorbate 80 was obtained from A.C.E.F. Figure 1. Synthesis of proline derivative (4).
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009
2366 MARRA ET AL.
Figure 3. Synthesis of tyrosine derivative (15).
stirred at room temperature for 2 h. Precipitated
dicyclohexylurea was filtered off, and the solution
Figure 2. Synthesis of citrulline (7) and asparagine
was evaporated in vacuo to dryness. The residue
(10) derivatives.
was dissolved in 10 mL of ethyl acetate and wash-
ed twice with water. The organic solution was
dried on anhydrous Na2SO4, evaporated and puri-
Varian VNMR software (Varian Inc., Palo Alto, fied by silica gel chromatography using n-hexane/
CA). Chemical shift values are reported in d units ethyl acetate 9.7/0.3 (v/v) as eluent. 1.1 g of BF, as
(ppm) relative to TMS used as internal standard. a brown oil, were obtained (yield 87.5%).
The following symbols were used to indicate the MS (m/z): 650.3 (M þ Naþ). Anal calcd
multiplicity of signals: s ź singlet, d ź doublet, for C39H65NO5 (627.92): C, 74.59; H, 10.43; N,
dd ź double doublet, t ź triplet, m ź multiplet. All 2.23; Found C, 74.51; H, 10.40; N, 2.22.
1
signals were determined in accordance with H NMR (CDCl3): d 4.58 4.62 (m, 1H, CH
literature. The mass spectra (MS) were obtained proline), 3.53 3.47 (m, 2H, CH2-5 proline), 2.55 (t,
using an API 2000 mass spectometer equipped 2H, J ź 11.9 Hz, CH2-4 tocopherol), 2.33 2.30 (m,
with Analyst 1.3 Data system software (Applied 2H, CH2-3 proline), 2.05 (s, 3H, CH3-7 toco-
Biosystems, Foster City, CA). pherol), 1.99 (s, 3H, CH3-8 tocopherol), 1.95 (s,
3H, CH3-5 tocopherol), 1.75 1.69 (m, 2H, CH2-3
tocopherol), 1.44 (s, 9H, Boc), 1.21 (s, 3H, CH3-2
N-tert Butoxycarbonyl-L-Proline
tocopherol), 0.85 0.80 (m, 12H, CHCH3 toco-
a-Tocopheryl Ester (3)
a
pherol).
To a solution of a-tocopherol (0.86 g, 2.0 mmol) in
5 mL of anhydrous dichloromethane were added
L-Proline a-Tocopheryl Ester (4)
a
0.41 g (2.0 mmol) of N-N0 didcyclohexylcarbodi-
imide (DCC), 0.049 g (0.4 mmol) of 4-dimethyla- To a solution of ester BF (1.1 g, 1.75 mmol) in 7 mL
minopyridine (Dmap) and 0.43 g (2.0 mmol) of of anhydrous dichloromethane were added
N-boc-L-proline (2). The resulting mixture was 1.32 mL of trifluoracetic acid and 0.10 mL of
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009 DOI 10.1002/jps
AMINO ACID ESTERS OF a-TOCOPHEROL 2367
13
C NMR (CDCl3): d 167.86 (CO ester), 149.86
(C-8a tocopherol), 139.84 (C-6 tocopherol), 126.12
(C-7 tocopherol), 124.69 (C-5 tocopherol), 123.29
(C-4a tocopherol), 119.89 (C-8a tocopherol), 75.21
(C-2 tocopherol), 59.17 (C-2 proline), 45.73 (C-5
proline), 39.32 (CH2-4 tocopherol), 39.27 (CH2-11
aliphatic chain tocopherol), 37.50 (CH2-7 a.c.),
37.34 (CH2-9 a.c.), 37.23 (CH2-5 a.c.), 32.73 (CH-8
a. c.), 32.65 (CH-4 a.c.), 30.90 (CH2-3 tocopherol),
28.83 (CH2-3 a.c.), 27.93 (CH2-2 a.c.), 24.77 (CH-
12 a.c.), 24.75 (C-3 proline), 24.40 (CH2-6 a.c.),
23.53 (C-4 proline), 22.67 (CH2-10 a.c.), 22.57
(CH3-3 tocopherol), 20.97 (CH3-12 a.c.), 20.48
(CH3-8 a.c.), 19.70 (CH3-8 a.c.), 19.63 (CH3-4
a.c.),12.68 (CH3-5 tocopherol), 11.84 (CH3-8),
11.73 (CH3-7).
N-tert Butoxycarbonyl-L-Citrulline
a-Tocopheryl Ester (6)
a
To a stirred solution of a-tocopherol (5.0 g, 11.60
mmol) and DCC (3.2 g, 11.60 mmol) in 15 mL of
anhydrous pyridine were added 3.2 g (11.60 mmol)
of N-boc-L-citrulline (5). The reaction mixture was
allowed to stir for 2.5 h. Dicyclohexylurea was
filtered off and the resulting solution was evapo-
rated to dryness. The residue was dissolved in
anhydrous diethyl ether, in order to allow
precipitation of residual dicyclohexylurea and
then filtered again. The solution obtained was
Figure 4. Synthesis of serine derivative (19).
evaporated to dryness, the residue was dissolved
in ethyl acetate and then extracted twice
triethylsilane. The resulting mixture was stirred
with water. The organic phase was dried on
at room temperature for 45 min. The solvent was
anhydrous Na2SO4, evaporated in vacuo and
removed under reduced pressure and the result-
purified by silica gel chromatography (ethyl
ing oil was dissolved in ethyl acetate and extracted
acetate). 5.5 g of BF, as a white wax, were
twice with an aqueous solution of NaOH (2 N). The
obtained (yield 68.9%).
organic solution was dried on anhydrous Na2SO4,
MS (m/z): 696.1 (M þ Naþ). Anal calcd
evaporated and purified by silica gel chromato-
for C39H67N3O6 (673.95): C, 69.50, H, 10.02, N,
graphy (n-hexane/ethyl acetate 5/5, v/v). 0.85 g of
6.24: Found C, 69.36, H, 10.01; N, 6.22.
BF, as a yellow-orange wax, were obtained (yield
1
H NMR (CD3OD): d 4.23 (t, 1H, J ź 12.1 Hz, CH
92%).
citrulline), 3.08-3.22 (m, 2H, CH2-4 citrulline),
MS (m/z): 528.5 (Mþ). Anal calcd for C34H57NO3
2.50 (t, 2H, J ź 11.9 Hz, CH2-4 tocopherol), 2.08 (s,
(527.43): C, 77.43; H, 10.89; N, 2.65; Found C,
3H, CH3-7 tocopherol), 2.05 1.98 (m, 2H, CH2-3
77.23; H, 10.88, N, 2.64.
citrulline), 1.95 (s, 3H, CH3-8 tocopherol), 1.92 (s,
1
H NMR (CDCl3): d 4.72 4.67 (m, 1H, CH
3H, CH3-5 tocopherol), 1.75 1.62 (m, 2H, CH2-3
proline), 3.51 3.49 (m, 2H, CH2-5 proline), 2.65
tocopherol), 1,37 (s, 9H, Boc), 0.81 0.70 (m, 12H,
2.55 (m, 1H, CH2-3 proline), 2.53 (t, 2H, J ź
CHCH3 tocopherol).
12.0 Hz, CH2-4 tocopherol), 2.45 2.35 (m, 1H,
CH2-3 proline), 2,21 2.09 (m, 2H, CH2-4 proline),
L-Citrulline a-Tocopheryl Ester (7)
a
2.06 (s, 3H, CH3-7 tocopherol), 1.97 (s, 3H, CH3-8
tocopherol), 1.90 (s, 3H, CH3-5 tocopherol), 1.80 The title compound 7 was synthesized using the
1.69 (m, 2H, CH2-3 tocopherol), 1.20 (s, 3H, same method described for 4. 2.5 g (3.63 mmol) of
CH3-2 tocopherol), 0.91 0.80 (m, 12H, CHCH3 intermediate 6 were used as starting material.
tocopherol). Purification of the crude product was carried out
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009
2368 MARRA ET AL.
using silica gel chromatography (ethyl acetate/ L-Asparagine a-Tocopheryl Ester (10)
a
methanol 9/1, v/v). 1.94 g of BF, as a white wax,
The title compound 10 was synthesized according
were obtained (yield 91%).
to the procedure previously described for 4. 2.5 g of
MS (m/z): 574.25 (Mþ). Anal calcd
9 were dissolved in 14 mL of anhydrous dichlor-
for C34H59N3O4 (573.84): C, 71.16, H, 10.36, N,
omethane and reacted together with 2.92 mL of
7.32; Found C, 71.02; H, 10.35; N, 7.31.
TFA and 0.17 mL of triethylsilane. The crude
1
H NMR (DMSO-d6): d 4.37 (t, 1H, J ź 13.1 Hz,
product was purified by chromatography using
CH citrulline), 3.22 3.08 (m, 2H, CH2-4 citrul-
ethyl acetate as eluent. 1.96 g of a white wax were
line), 2.53 (t, 2H, J ź 12.0 Hz, CH2-4 tocopherol),
obtained (yield 93%).
2.17 2.12 (m, 2H, CH2-3 citrulline), 2.04 (s, 3H,
MS (m/z): 545.1 (Mþ). Anal calcd for
CH3-7 tocopherol), 1.95 (s, 3H, CH3-8 tocopherol),
C33H56N2O4 (544.79): C, 72.75; H, 10.36; N,
1.88 (s, 3H, CH3-5 tocopherol), 1.80 1.62
5.14; Found C, 72.60; H, 10.34; N, 5.15.
(m,2H,CH2-3 tocopherol), 0.80 0.70 (m, 12H,
1
H NMR (DMSO-d6): d 4.68 (t, 1H, 13.8 Hz, CH
CHCH3 tocopherol).
asparagine), 3.15 3.02 (m, 2H, CH2 asparagine),
13
C NMR (DMSO-d6): d 170.14 (CO ester),
2.60 (t, 2H, 12.1 Hz, CH2-4 tocopherol), 2.01 (s,
158.69 (CO citrulline), 148.57 (C8-a tocopherol),
3H, CH3-7 tocopherol), 1.94 (s, 3H, CH3-8
140.12 (C-6 tocopherol), 126.34 (C-7 tocopherol),
tocopherol), 1.90 (s, 3H, CH3-5 tocopherol),
124.91 (C-5 tocopherol), 121.77 (C-4a tocopherol),
1.83 1.72 (m, 2H, CH2-3 tocopherol), 0.91 0.77
117.35 (C-8 tocopherol), 74.80 (C-2 tocopherol),
(m, 12H, CHCH3 tocopherol).
54.02 (CH citrulline), 38.53 (CH2-4 tocopherol),
13
C NMR (DMSO-d6): d 172.04 (CO ester),
36.69 (CH2-11 aliphatic chain), 36.59 (CH2-7 a.c.),
168.84 (CO asparagine), 148.70 (C8-a tocopherol),
32.03 (CH2-4 citrulline), 31.99 (CH2-9 a.c.), 31.92
139.92 (C-6 tocopherol), 126.75 (C-7 toco-
(CH2-9 a.c.), 31.75 (CH2-3 tocopherol), 30.60
pherol), 125.01 (C-5 tocopherol), 120.99 (C4a
(CH2-3 a.c.), 27.37 (CH2-3 citrulline) 26.58
tocopherol), 118.35 (C-8 tocopherol), 75.81 (C-8
(CH2-2 a.c.), 24.13 (CH-12 c.a.c.), 23.68 (CH2-6
tocopherol), 50.61 (CH asparagine), 40.01 (CH2-4
a.c.), 22.55 (CH2-10 a.c.), 22.47 (CH3-3 toco-
tocopherol), 36.99 (CH2-11 aliphatic chain toco-
pherol), 21.03 (CH3-12 a.c.), 20.37 (CH3-12 a.c.),
pherol), 36.69 (CH2-7 a.c.), 32.01 (CH2-9 a.c.),
19.63 (CH3-8 a.c.), 19.57 (CH3-4 a.c.), 12.72 (CH3-
31.89 (CH2-9 a.c.), 31.65 (CH2-3), 30.50 (CH2-3
5 tocopherol), 11.86 (CH3-8 tocopherol), 11.57
a.c.), 29.35 (CH2 asparagine) 26.64 (CH2-2 a.c.),
(CH3-7 tocopherol).
24.22 (CH-12 a.c.), 23.87 (CH2-6 a.c.), 22.45 (CH2-
10 a.c.), 22.32 (CH3-3 tocopherol), 21.26 (CH3-12
a.c.), 20.67 (CH3-12 a.c.), 19.55 (CH3-8 a.c.), 19.43
N-tert Butoxycarbonyl-L-Asparagine (CH3-4 a.c.), 13.76 (CH3-5 tocopherol), 12.59
a-Tocopheryl Ester (9) (CH3-8 tocopherol), 11.85 (CH3-7 tocopherol).
a
The title compound 9 was synthesized according
N-tert Butoxycarbonyl-O-
to the procedure previously described for 6. 5.0 g
Tbdms-L-Tyrosine-OH (12)
(11.60 mmol) of a-tocopherol, 2.69 g (11.60 mmol)
of N-boc-L-asparagine (8) and 2.39 g (11.60 mmol) tert-Butyldimethylsylil chloride (TbdmsCl, 4.02 g,
of DCC were allowed to react in 15 mL of 26.65 mmol) was dissolved in 8 mL of anhydrous
anhydrous pyridine. The crude product was dimethylformamide. This solution was slowly
purified by chromatography (ethyl acetate/n- added to a cooled mixture (08C) of N-boc-L-tyrosine
hexane 7/3, v/v). 4.86 g of 9, as a yellow wax, (5.0 g, 17.77 mmol) and imidazole (2.42 g, 35.45
were obtained (yield 65%). mmol) in 12 mL of anhydrous dimethylformamide.
MS (m/z): 667.3 (M þ Naþ). Anal calcd The resulting solution was warmed to room
for C38H64N2O6 (644.91): C, 70.77; H, 10.00; N, temperature and stirred for 18 h. Ten milliters
4.34; Found C, 70.63; H, 9.99; N, 4.33. of water were, then, added and the resulting
1
H NMR (DMSO-d6): d 4.82 (t, 1H, J ź 11.9 Hz, solution was extracted three times with diethyl
CH asparagine), 3.13 3.04 (m, 2H, CH2 aspar- ether and three times with chloroform. The
agine), 2.63 (t, 2H, 12.1 Hz, CH2-4 tocopherol), organic phases were collected, dried over
2.06 (s, 3H, CH3-7 tocopherol), 2.0 (s, 3H, CH3-8 anhydrous Na2SO4, evaporated and purified by
tocopherol), 1.95 (s, 3H, CH3-5 tocopherol), 1.85 chromatography (dichloromethane/ethyl acetate
1.74 (m, 2H, CH2-3 tocopherol), 1.45 (s, 9H, Boc), 8/2, v/v). 6.4 g of 12, as a colorless oil, were
0.94 0.80 (m, 12H, CHCH3 tocopherol). obtained (yield 91.5%).
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009 DOI 10.1002/jps
AMINO ACID ESTERS OF a-TOCOPHEROL 2369
1
MS (m/z): 418.5 (M þ Naþ). Anal calcd H NMR (CDCl3): d 7.14 (d, 2H, aromatic ring),
for C20H33NO5Si (395.50): C, 60.74; H, 8.41; 6.77 (d, 2H, aromatic ring), 4.82 (t, 1H, J ź 13.5
3.54 (N); Found C, 60.70; H, 8.42; N, 3.53. Hz, CH tyrosine), 3.32 3.28 (dd,1H, J ź
1
H NMR (CDCl3): d 7.16 (d, 2H, aromatic ring), 9.1 Hz CH2 tyrosine), 3.08 3.04 (dd, 1H, J ź
6.76 (d, 2H, aromatic ring), 4.35 (t, 1H, J ź 13.8 9.6 Hz CH2 tyrosine), 2.56 (t, 2H, J ź 11.9
Hz, CH tyrosine), 3.52-3.47 (dd, 1H, J ź 8.9 Hz, Hz, CH2-4 tocopherol), 2.07 (s,3H,CH3-7 toco-
CH2 tyrosine), 3.26 3.21 (dd, 1H, J ź 9.3 Hz, CH2 pherol), 1.95 (s, 3H, CH3-8 tocopherol), 1.91
tyrosine), 1.45 (s, 9H, Boc), 0.98 (s, 9H, Tbdms), (s, 3H, CH3-5 tocopherol), 1.80-1.74 (m, 2H,
0.2 (s, 6H,Tbdms). CH2-3 tocopherol), 1.41 (s, 9H, Boc), 1.22 (s,
3H, CH3-2 tocopherol), 0.87 0.83 (m, 12H, CH-
N-tert Butoxycarbonyl-O-Tbdms-L-Tyrosine-a CH3 tocopherol).
a-
Tocopheryl ester (13)
L-Tyrosine a-Tocopheryl Ester (15)
a
The title compound 13 was synthesized according
to the procedure described for 4. a-Tocopherol The title compound 15 was synthesized according
(3.15 g 7.33 mmol), compound 12 (2.9 g, to the procedure described for 4. The crude
7.33 mmol), Dmap (0.18 g, 1.46 mmol) and DCC product was purified by chromatography using
(1.51 g, 7.33 mmol) were reacted in 15 mL of ethyl acetate as eluent. The final product was
anhydrous dichloromethane for 1 h. The crude isolated as a white wax (yield 95%).
product was purified by chromatography using MS (m/z): 616.44 (M þ Naþ). Anal calcd for
n-hexane/ethyl acetate (8/2, v/v) as eluent. 5.27 g C38H59NO4 (593.86): C, 76.85; H, 10.01; N, 2.36;
of BF, as a yellow oil, were obtained (yield 89%). Found C, 76.70; H, 10.00; N, 2.37.
1
MS (m/z): 830.3 (M þ Naþ). Anal calcd H NMR (DMSO-d6): d 7.06 (d, 2H, aromatic
for C49H81NO6Si (807.3): C, 72.83; H, 10.11; N, ring), 6.66 (d, 2H, aromatic ring), 3.8 (t, 1H,
1.72; Found C, 72.45; H, 10.08; N, 1.71. J ź 14.1 Hz, CH tyrosine), 3.04 2.98 (dd, 1H,
1
H NMR (CDCl3): d 7.16 (d, 2H, aromatic ring), J ź 9.0 Hz CH2 tyrosine), 2.76 2.70 (dd, 1H,
6.81 (d, 2H, aromatic ring), 4.87 (t, 1H, J ź 13.7 J ź 9.3 Hz CH2 tyrosine), 2.50 (t, 2H, J ź 11.9
Hz, CH tyrosine), 3.49 3.44 (dd,1H, J ź 8.9 Hz, Hz, CH2-4 tocopherol), 1.97 (s, 3H CH3-7 toco-
CH2 tyrosine), 3.17 3.12 (dd,1H, J ź 9.3 Hz CH2 pherol), 1.94 (s, 3H, CH3-8 tocopherol), 1.84 (s,
tyrosine), 2.56 (t, 2H, J ź 12.1 Hz, CH2-4 toco- 3H, CH3-5 tocopherol), 1.74-1.70 (m, 2H, CH2-3
pherol), 2.07 (s, 3H CH3-7 tocopherol), 1.95 (s, 3H, tocopherol), 1.21 (s, 3H, CH3-2 tocopherol), 0.85
CH3-8 tocopherol), 1.93 (s, 3H, CH3-5 tocopherol), 0.83 (m, 12H, CH-CH3 tocopherol).
13
1.89 1.77 (m, 2H, CH2-3 tocopherol), 1.41 (s, 9H, C NMR (CDCl3): 170.60 (CO ester), 155.74 (C-
Boc), 1.22 (s, 3H, CH3-2 tocopherol), 1.06 (s, 9H, 40 aromatic ring), 149.95 (C8-a tocopherol), 140.37
Tbdms), 0.87 0.83 (m, 12H, CH-CH3 tocopherol), (C-6 tocopherol), 130.92 (aromatic ring), 130.61
0.19 (s, 6H, Tbdms). (aromatic ring), 126.74 (C-7 tocopherol), 125.11
(C-5 tocopherol), 123.45 (C-4a tocopherol), 117.83
N-tert Butoxycarbonyl-L-Tyrosine (C-8 tocopherol), 116.35 (aromatic ring), 77.53 (C-
a-Tocopheryl Ester (14) 8 tocopherol), 55.4 (CH tyrosine), 39.59 (CH2-4
a
tocopherol), 37.81 (CH2-11 aliphatic chain), 37.68
2.4 mL of a 1.1 M solution of tetrabutylammonium
(CH2-7 a.c.), 37.51 (CH2 tyrosine), 33.01 (CH2-9
fluoride (0.71 g, 2.7 mmol) in anhydrous tetra-
a.c.), 32.94 (CH2-9 a.c.), 31.28 (CH2-3 tocopherol),
hydrofurane were slowly added to a solution of of
28.20 (CH2-3 a.c.), 25.03 (CH2-2 a.c.), 24.67 (CH-
BF (1.7 g, 2.1 mmol) in 20 mL of anhydrous
12 a.c.), 22.94 (CH2-6 a.c.), 21.26 (CH2-10 a.c.),
tetrahydrofurane at 08C. The resulting mixture
20.79 (CH3-3 tocopherol), 19.96 (CH3-12 a.c.),
was stirred at room temperature for 30 min.
19.90 (CH3-12 a.c.), 19.84 (CH3-8 a.c.), 19.78 (CH3-
Fifteen milliters of water were, then, added and
4 a.c.), 13.34 (CH3-5 tocopherol), 12.51 (CH3-8
the organic phase was extracted, dried on Na2SO4,
tocopherol), 12.03 (CH3-7 tocopherol).
evaporated to dryness and, finally, purified by
silica gel chromatography (n-hexane/ethyl acetate
N-tert Butoxycarbonyl-O-Benzyl-L-Serine
7/3, v/v). 1.23 g of BF, as white wax, were obtained
a-Tocopheryl Ester (17)
a
(yield 85%).
MS m/z: 693.46 (Mþ). Anal calcd for C43H67NO6 The title compound 17 was synthesized according
(693.46): C, 74.41; H, 9.73; N, 2.02; Found C, to the general procedure described for 3. 5.0 g
74.26; H, 9.74; N, 2.01. (11.60 mmol) of a-tocopherol, 3.42 g (11.60 mmol)
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009
2370 MARRA ET AL.
of N-boc-O-Bz-L-serine (16), 0.28 g (2.32 mmol) of MS (m/z): 540.78 (M þ Naþ). Anal calcd
Dmap and 2.39 g (11.60 mmol) of DCC were for C32H55NO4 (517.77): C, 74.23; H, 10.71; N,
allowed to react in anhydrous dichloromethane. 2.71; Found C, 74.08; H, 10.69; N, 2.70.
1
The crude product was purified by chromatogra- H NMR (DMSO-d6): d 4.60 (t, 1H, J ź 13.4 Hz,
phy (n-hexane/ethyl acetate 8/2, v/v). 7.39 g of BF, CH serine), 4.10 4.05 (dd, 1H, J ź 8.5 Hz CH2
as a yellowish oil, were isolated (yield 90%). serine), 4.03 3.98 (dd, 1H, J ź 8.9 Hz, CH2 serine),
MS (m/z): 708.06 (Mþ). Anal calcd for 2.57 (t, 2H, J ź 12.0 Hz CH2-4 tocopherol), 2.09 (s,
C44H69NO6 (708.00): C, 74.64; H, 9.82; N, 1.98; 3H CH3-7 tocopherol), 2.04 (s, 3H, CH3-8 toco-
Found C, 74.78; H, 9.81; N, 1.97. pherol), 2.01 (s, 3H, CH3-5 tocopherol), 1.83 1.75
1
H NMR (CDCl3): d 7.39 7.35 (m, 5H aromatic (m, 2H, CH2-3 tocopherol), 1.25 (s, 3H, CH3-2
ring), 4.79 (d, 2H, J ź 6.1 Hz, CH2 benzyl group), tocopherol), 0.99 0.81 (m, 12H, CH-CH3 toco-
4.69 (t, 1H, J ź 12.9 Hz, CH serine), 4.10 4.05 (dd, pherol).
13
1H, J ź 8.5 Hz CH2 serine), 3.95 3.90 (dd, 1H, C NMR (DMSO-d6): d 167.80 (CO ester),
J ź 8.9 Hz, CH2 serine), 2.57 (t, 2H, J ź 12.0 152.23 (C8-a tocopherol), 142.20 (C-6 tocopherol),
Hz, CH2-4 tocopherol), 2.13 (s, 3H CH3-7 toco- 127.50 (C-7 tocopherol), 126.01 (C-5 tocopherol),
pherol), 2.05 (s, 3H, CH3-8 tocopherol), 1.96 (s, 124.07 (C4a tocopherol), 118.23 (C-8 tocopherol),
3H, CH3-5 tocopherol), 1.84 1.75 (m, 2H, CH2-3 76.82 (C- tocopherol 8), 60.08 (CH2 serine), 55.18
tocopherol), 1.39 (s, 9H, Boc), 1.25 (s, 3H, CH3-2 (CH serine), 39.89 (CH2-4 tocopherol), 38.01 (CH2-
tocopherol), 0.92 0.86 (m, 12H, CH-CH3 toco- 11 aliphatic chain), 37.88 (CH2-7 a.c.), 34.06 (CH2-
pherol). 9 a.c.), 31.78 (CH2-9 a.c.), 30.95 (CH2-3 toco-
pherol), 25.63 (CH2-2 a.c.), 25.06 (CH-12 a.c.),
N-tert Butoxycarbonyl-L-Serine 23.41 (CH2-6 a.c.), 22.62 (CH2-10 a.c.), 21.14 (CH3-
a-Tocopheryl Ester (18) 3 tocopherol), 20.35 (CH3-12 a.c.), 20.05 (CH3-12
a
a.c.), 19.98 (CH3-8 a.c.), 19.89 (CH3-4 a.c.), 14.74
A solution of 17 (2.5 g, 3.53 mmol) in a mixture of
(CH3-5 tocopherol), 13.61 (CH3-8 tocopherol),
n-butanol/methanol/acetic acid 30/15/10 (v/v/v)
12.98 (CH3-7 tocopherol).
was hydrogenated at 3 atm using 5% palladium
over activated carbon as catalyst. After 3.5 h the
catalyst was filtered off and the resulting solution
was dried at reduced pressure. The residue was HPLC Analysis
dissolved in ethyl acetate and extracted three
A new HPLC method was developed in order to
times with water. The organic solution was dried
simultaneously analyze both pro-vitamins and
on Na2SO4, evaporated and purified by chromato-
a-tocopherol. A reversed-phase column (Supelco
graphy (n-hexane/ethyl acetate 5/5, v/v). The final
RP-amide C16, 3.9 mm 150 mm, Supelco,
product was obtained in quantitative yield.
Bellefonte, PA) on a Perkin Elmer HPLC system
MS (m/z): 640.90 (M þ Naþ). Anal calcd
(Norwalk, CT) was used. Detection wavelength
for C37H63NO6 (617.88): C, 71.92; H, 10.28; N,
was 215 nm. The mobile phase was a mixture of
2.27; Found C, 71.70; H, 10.24; N, 2.26.
acetonitrile and distilled water 95/5 (v/v), pumped
1
H NMR (CDCl3): d 4.70 (t, 1H, J ź 12.8 Hz CH
at a flow rate of 2.0 mL/min. These conditions were
serine), 4.25 4.19 (dd, 1H, J ź 8.5 Hz CH2 serine),
adopted to analyze serine (retention time 4.4 min,
4.15 4.09 (dd, 1H, J ź 8.9 Hz, CH2 serine), 2.60 (t,
linearity range 0.97 90.7 mg/mL; LOQ 0.97),
2H, CH2-4 tocopherol), 2.14 (s, 3H, CH3-7
tyrosine (retention time 5.0 min, linearity range
tocopherol), 2.07 (s, 3H, CH3-8 tocopherol), 1.98
1.0 80.0 mg/mL; LOQ 1.0), asparagine (retention
(s, 3H, CH3-5 tocopherol), 1.87 1.70 (m, 2H, CH2-3
time 4.0 min, linearity range 1.2 97.0 mg/mL;
tocopherol), 1.41 (s, 9H, Boc), 1.28 (s, 3H, CH3-2
LOQ 1.2) and citrulline (retention time 3.5 min,
tocopherol), 0.95 0.83 (m, 12H, CH-CH3 toco-
linearity range 0.83 102 mg/mL; LOQ 0.83)
pherol).
derivatives, simultaneously with vitamin E
(retention time 5.5 min., linearity range 0.29
L-Serine a-Tocopheryl Ester (19)
a
23.7 mg/mL; LOQ 0.29). In order to analyze the
The title compound 19 was synthesized as proline derivative, the mobile phase was slightly
described for 4. Chromatographic purification modified, keeping 95% of acetonitrile, but adding
was performed on silica gel using ethyl acetate/ 3% of methanol and only 2% of distilled water
n-hexane 8/2 (v/v). The final product was isolated (retention time 9.9 min, linearity range 0.26
as an orange wax (yield 90%). 21.8 mg/mL; LOQ 0.26). Vitamin E (retention time
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009 DOI 10.1002/jps
AMINO ACID ESTERS OF a-TOCOPHEROL 2371
3.5 min) was analyzed also using this mobile detector at 215 nm. Each HPLC method was
phase and adequately separated from proline validated according to USP 30 as described above.
derivative, so to be detected simultaneously. K0 values were calculated using the following
The injection volume was 50 mL in all cases. relationship:
The analytical method was assessed according to
tr t0
K0 ź (1)
USP 30. The specificity (absence of interfering
t0
peaks derived from skin samples) was assessed as
well. where tr is the retention time of the product and t0
The HPLC analysis of acetyl salicylic and is retention time of the nonretained solvent
salicylic acid was performed using the method (methanol).
described in the USP 24 on a Perkin Elmer HPLC
system, using a Novapak1 C18 column (Waters,
Milford, MA). In Vitro Hydrolysis
Pro-vitamins sensitivity to enzymatic hydrolysis
was determined using porcine liver esterase,
Solution Stability
according to Ostacolo et al.10 Each derivative
Solution stability was determined by dissolving was dissolved in methanol reaching a final
pro-vitamins in methanol: 5 mM Tween 80 solution concentration of 1 mg/mL. The resulting solution
(1:10, v/v). The solutions were stored at room was diluted 1:10 (v/v) with 5% DM-b-CD phos-
temperature, 37 and 708C for 48 h. Aliquots of phate buffer solution at pH 8, and added with
solutions were withdrawn at pre-determined 5 IU/mL of porcine esterase. The final solution was
intervals and analyzed with HPLC in order to incubated at 378C and stirred. An aliquot of
evaluate the residual concentration of pro-vitamin. 100 mL was withdrawn at pre-determined inter-
vals and diluted with methanol to 1 mL, in order to
precipitate salts and inactivate enzymes. After
Solubility and log Capacity Factor (log K() filtration the fractions were analyzed by HPLC, to
determine the residual concentration of pro-
A 5% solution of DM-b-CD in phosphate buffer at
vitamins and of VE produced.
pH 8, obtained by mixing 50 mL of 0.2 M
Preliminary experiments were performed using
potassium dihydrogen phosphate with 46.8 mL
acetyl salicylic acid (concentration 20 mg/mL) as
of 0.2 M sodium hydroxide and adding water to
substrate to evaluate the effect of 5% DM-b-CD
200 mL, was prepared. The same solution was
and 10% methanol on the enzymatic activity of
used in enzymatic hydrolysis experiments. A large
porcine liver esterase.
excess of product was suspended in the buffer and
magnetically stirred at room temperature for 24 h.
Then the suspension was filtered with 0.45 mm
In Vitro Skin Accumulation and Metabolism
nylon filter (Lida, Kenosha, WI) and diluted 1:10
(v/v) in methanol. This solution was analyzed by Skin accumulation and metabolism were deter-
HPLC in order to evaluate pro-vitamin concen- mined using Franz-type cells (0.6 cm2 area) and
trations. Stability at 378C in 5% DM-b-CD freshly excised skin from rabbit ear as barrier.
phosphate buffer solution was determined as well. Receptor compartment contained 4 mL of a 5%
Lipophilicity indices of VE and pro-vitamins DM-b-CD in phosphate buffer at pH 8.0, while
were estimated by reverse-phase chromato- donor compartment contained 1 mL of pro-
graphic retention times, since previous works vitamin saturated solution (without excess solid).
demonstrated a good correlation between log P In analogy with our previous work, the donor
and log K0,19 21 and both pro-vitamins analyzed solution solvent system was a mixture of ethanol,
and VE are almost insoluble in water. log K0 was propylene glycol and water 0.5:0.1:0.4 by weight.
calculated using a ODS Hypersil column (5 mm; The concentrations of the saturated solutions
25 cm 4.6 mm i.d.; Thermo Hypersil, Bellefonte, used as donor are reported in Table 1. Rabbit skin
PA) and a mixture of methanol:water 95:5 (v/v) (obtained from a local slaughter house) was
with 0.0075 0.03% (v/v) of 2-amino heptane as removed from the inner side of ears by blunt
mobile phase. Different concentrations of organic dissection and mounted with corneal side facing
amine were used to exclude an influence on the donor compartment. Receptor compartment
results. Flow rate was set at 1.5 mL/min and was thermostatted at 378C and magnetically
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009
2372 MARRA ET AL.
Table 1. Physico-Chemical Properties of the Pro-Vitamins (Mean Values SEM; n ź 4)
Molecular Aqueous Donor Solutionb
Derivative Weight (Da) Solubilitya (mg/mL) Solubility (mg/mL) log K0c Half-Lifed (min)
VEe 430.7 429 3  0.911 0.009 
VEAce 472.8 182 6  1.185 0.004 48 8
Proline (4) 527.8 891 74 10.4 1.0 1.188 0.010 103 9
Citrulline (7) 587.7 7115 251 7.7 0.7 0.678 0.015 1013 250
Asparagine (10) 562.8 1361 18 5.4 0.4 0.753 0.002 341 10
Tyrosine (15) 593.7 1322 190 7.6 0.7 0.862 0.015 54 7
Serine (19) 517.8 374 23 5.4 0.4 0.806 0.017 223 9
a
In pH 8.0 buffer with 5% (w/v) dimethyl b cyclodextrin at room temperature.
b
Ethanol:propylene glycol:water (0.5:0.1:0.4 by volume).
c
Value determined according to Eq. (1).
d
In the presence of 5 IU/mL of porcine liver esterase.
e
From Ref. 10.
Significantly different from acetate.
stirred to avoid any boundary effects. At pre- 24 h. Then the suspension was filtered with
determined intervals (2, 4, 6 h) the experiment 0.45 mm nylon filter (Lida) and diluted 1:10 (v/v) in
was stopped, the receptor solution was sampled methanol. This solution was used as donor for
(and analyzed) and the recovered skin washed permeation experiments as well as analyzed by
with phosphate buffer. A disc of tissue, fitting the HPLC in order to evaluate pro-vitamin solubility.
area covered by donor compartment, was cut,
heated with hot air for 20 s and separated in
epidermis and dermis, in correspondence of the
basal membrane, with the use of a spatula. The Statistical Analysis
two skin layers were separately weighted in
The results were expressed as the mean
plastic tubes and extracted with 2 mL of methanol
standard error of the mean (SEM), and statistical
for 1 h at room temperature. The solutions
differences were determined by Student s t-test.
obtained were centrifuged for 10 min at 11,000
rpm, filtered over nylon filters (0.45 mm) and
injected directly in the HPLC system. All samples
RESULTS AND DISCUSSION
of skin retention of pro-vitamins and VE were
above the LLOQ.
Solubility, Stability, and Lipophilicity
The extraction method was validated in blank
experiments and by spiking the skin with a known Since the solubility of pro-vitamins and VE in pH
amount of the derivatives. No interference peaks, 8.0 PBS, at room temperature, resulted very low
deriving from skin samples, were detected in (lower than the limit of quantification of the
blank experiments. The amount of derivatives analytical method), DM-b-CD 5% was added as
recovered was in the range of 98 106% (VE), 93 solubilizing agent, to guarantee sink conditions in
104% (VEAc), 95-103% (citrulline derivative), 96 permeation experiments.22 The solubility values
105% (asparagine derivative), 94 104% (tyrosine obtained in the presence of cyclodextrin are
derivative), 97 104% (serine derivative), and 99 reported in Table 1. The citrulline derivative (7)
103% (proline derivative). was the most soluble, showing a value more than
All accumulation experiments were repeated 15 times higher that VE and almost 40 times
four to six times. higher than VEAc. Serine derivative (19) was the
less soluble, with solubility similar to VE. All new
derivatives were significantly more soluble than
the reference compound, VEAc.
Donor Solution Preparation
Stability tests were performed in aqueous
A large excess of pro-vitamin was suspended in solution to evaluate the sensitivity of the pro-
the donor solution (mixture of ethanol, propylene vitamins to chemical hydrolysis and degradation.
glycol and water 0.5:0.1:0.4 by weight) and Initially the stability of the pro-vitamins in
magnetically stirred at room temperature for the donor and receptor solution was tested: the
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009 DOI 10.1002/jps
AMINO ACID ESTERS OF a-TOCOPHEROL 2373
pro-vitamins were shown to be stable for at least liver esterase according to Laneri et al.23 The pro-
24 h (percentage remaining higher than 98%). vitamins were dissolved in 10% methanol,
Since a-tocopherol and the synthesized pro- 5% DM-b-CD in pH 8.0 phosphate buffer to
vitamins are not water soluble, Tween 80 (5 mM) overcome their low solubility in pH 8.0 phosphate
and methanol 10% were added as cosolvents. The buffer alone.
addition of DM-b-CD was avoided because the Preliminary experiments were performed using
formation of an inclusion compound can mask acetyl salicylic acid as substrate of the enzyme to
the instability of the molecule. The concentration evaluate the effect of cosolvents on the activity of
of the four pro-vitamins in 10% methanol, 5 mM porcine liver esterase. The results obtained are
Tween 80 resulted unchanged (percentage reported in Figure 5, as fraction of salicylic acid
remaining higher than 98%) after 8 h at room remaining. The experiments were performed
temperature and at 378C. Stability tests were using pH 8.0 phosphate buffer in the presence
also performed at 708C, since this is a typical and absence of 10% methanol and 5% DM-b-CD.
temperature used for cream preparation. At this Appropriate controls without esterase were car-
temperature the amount lost after 2 h was lower ried out to exclude the presence of chemical
than 10%. hydrolysis to a significant extent. The results
Since VE and its derivatives are almost obtained demonstrate that the presence of metha-
insoluble in water, the use of the shake-flask nol and DM-b-CD does not influence the activity of
method in order to assess their lipophilicity was the enzyme.
avoided. This physico-chemical parameter was An exponential decrease of pro-vitamin concen-
estimated with the use of log capacity factor tration and an equivalent exponential increase in
(log K0), which has been shown to correlate with VE concentration are observed, whereas the pro-
the octanol/water distribution coefficient.21 Log vitamins were stable in the absence of the enzyme.
capacity factor values are often calculated using The half-life values of pro-vitamins were calcu-
traces of both organic acids or bases inside the lated according to Laneri et al.,23 assuming a first-
mobile phase.21 This method is reliable for order kinetics (the average regression coefficient
eliminating interactions between aminic or car- was typically equal or higher than 0.95), and are
boxylic group of molecules analyzed and silanolic reported in Table 1.
groups. In this way, the retention time is only Citrulline derivative (7) was the less sensitive
function of partition coefficient between mobile to enzymatic hydrolysis, showing a half-life of
and stationary phases. It is not known if varia- approx. 16 h, while tyrosine conjugate (15) had a
tions in primary amine concentrations could alter half-life comparable to VEAc, approximately 1 hr.
the results, due to the different functionalized
chains of the synthesized derivatives. For this
reason we used different concentration of organic
primary amine (2-aminoheptane), namely from
0.0075% to 0.03%. Lower and upper limits of this
range were chosen validating each HPLC method
according to USP 30. Results are reported in
Table 1 as average of all elution methods.
Statistical analysis of results (using t-test) indi-
cates that variable concentration of 2-aminohep-
tane did not influence log K0 values and relative
hydrophobicity of derivatives ( p < 0.01). Moreover
the results, reported in Table 1, indicate that
VEAc is more hydrophobic than VE, while the pro-
vitamins synthesized in the present work show
a lower (7, 10, 15, and 19) or comparable
Figure 5. Fraction of acetyl salicylic acid remaining
(4) hydrophobicity compared to VE.
in the presence (squares) and absence (circles) of por-
cine liver esterase. Full symbols refer to solutions con-
taining 10% methanol and 5% DM-b-CD in pH 8
In Vitro Hydrolysis
phosphate buffer, empty symbols to solutions in pH 8
Since the pro-vitamins are esters, the sensitivity
phosphate buffer. Each value is the average of three
to enzymatic hydrolysis was verified using porcine experiments ( SEM).
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009
2374 MARRA ET AL.
The other molecules (4, 10, and 19) showed With the aim of evaluating possible differences
intermediate half-life values. These results could in the localization of the pro-vitamins or VE, the
be explained in terms of sterical dimension of side data were analyzed individually for the epidermis
chain and lipophilicity of the pro-vitamin: it has and for the dermis. The raw data were normalized
been reported24 that the hydrolysis rate of esters by the weight of the tissue, to avoid a possible bias
in the presence of porcine esterases is enhanced due to inter individual variability of the weight of
with increasing their lipophilicity, unless they are each specimen, and are reported in Figure 7 as
too large to efficiently enter the catalytic pocket. nmol of derivative and VE per mg of tissue. Each
Figure 6 reports the relationship between half-life column contains a gray portion, which indicates
in the presence of esterase and lipophilicity of the the amount of pro-vitamin recovered, and a white
various derivatives. From the plot it is evident part, indicating the amount of VE recovered. Both
that the half-life decreases exponentially as the pro-vitamins and VE were found in the epidermis
lipophilicity increases. (Panel a) and dermis (Panel b), the concentration
found being much higher for epidermis compared
to dermis (note the different scale in the two
Skin Retention and Metabolism
graphs).
Changing the application time does not seems to
Skin accumulation and metabolism experiments
affect the skin retention of the pro-vitamins in a
were performed using freshly excised rabbit ear
skin, which has been shown to be a reasonable
model for human skin in terms of permeabil-
ity.17,18 Additionally, the same skin model has
been used to test other a-tocopherol derivatives,10
that is, esters with alanine, glycine and pyroglu-
tamic acid.
In analogy with the previous work, the newly
synthesized derivatives were tested in permeation
experiments across the skin, using a vehicle
composed of water/propylene glycol/ethanol (40/
10/50, w/w/w). The vehicle was saturated with the
pro-vitamins to achieve the same thermodynamic
activity for all of them. This solvent mixture has
been chosen since it is a typical cosolvent system
used to realize water-based formulations and
guarantees a relatively high concentration of
the pro-vitamins.
Figure 6. Relationship between enzymatic half-life Figure 7. Pro-vitamin and derived vitamin E (white
and log K0 for the derivatives of a-tocopherol (mean bars) recovered in the epidermis (Panel a) and in the
values SEM). For some data points the error bar is dermis (Panel b). Each value is the average of four to six
smaller than the size of the symbol. experiments ( SEM).
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009 DOI 10.1002/jps
AMINO ACID ESTERS OF a-TOCOPHEROL 2375
clear, non-equivocal way, probably because three
phenomena, with different kinetics, are taking
place at the same time: diffusion into the tissue,
metabolism and diffusion out of the tissue.
Comparing the different derivatives, it seems
that proline, asparagine and serine derivatives
were the ones giving rise to higher skin accumula-
tion. The amount of VE recovered was very
variable in both epidermis and dermis and did
not show any particular trends. None of the
derivatives, or vitamin E, was ever found in the
receptor compartment.
The minimum flux detectable was calculated
from the LOQ of each derivative. Assuming to
have a concentration of the pro-vitamin (or VE) in
Figure 8. Extent of metabolism of the pro-vitamins
the receptor compartment after 6 h equal to the
synthesized, calculated according to Eq. (4).
LOQ, it is possible to calculate the corresponding
flux according to the following equations:
CONCLUSIONS
Qt
J ź (2)
t
Five new conjugates of a-tocopherol with amino
acids were prepared and tested. The amino acids
CtVr
Qt ź (3)
used were L-proline, L-citrulline, L-asparagine,
A
L-tyrosine, and L-serine.
where J ź pro-vitamin flux (mgcm 2 h 1); t ź time
The new pro-vitamins synthesized were sensi-
(h). The value considered was 6 h; Qt ź amount
tive to enzymatic hydrolysis, in particular
permeated per unit area at time t (mgcm 2);
their half-life was inversely proportional to their
Ct ź LOQ (mgcm 3); Vr ź receptor compartment
lipophilicity as estimated by the log capacity
volume (cm3); A ź permeation area (cm2).
factor. The in vitro skin retention and metabolism
From the results obtained, it is possible to
of the pro-vitamins demonstrated that all of them
conclude that fluxes equal or lower than 0.05
accumulated into the skin in significant amounts
0.25 mgcm 2 h 1 were not detectable (the result is
and were able to generate vitamin E. The
given as interval because of the different LOQ for
synthesized pro-vitamins resulted more water-
each derivative).
soluble than the currently used acetate, thus
Finally, from the amount of VE and derivative
allowing the use of more hydrophilic vehicles for
accumulated into the skin (epidermis þ dermis)
skin application. The highest skin retention was
following the application of each derivative, the
obtained with the proline derivative, which shows
extent of skin metabolism (E) was calculated as:10
the highest log capacity factor, comparable with
vitamin E acetate, whereas the lowest was
VE
E ź % (4)
obtained with tyrosine and citrulline derivatives.
VE þ PV
The reason for differing performances of the
where VE is a-tocopherol and PV represents synthesized pro-vitamins is unknown. It is
a-tocopherol derivative accumulated in the skin theorized that performance is affected by the
(epidermis plus dermis). Figure 8 illustrates the simultaneous presence of the following phenom-
percentage of derivative that has been meta- ena: partitioning into the stratum corneum,
bolized in the skin (originating vitamin E) after 2, diffusion across the skin and skin metabolism.
4, 6 h, for the 5 pro-vitamins. The extent of
metabolism found for the asparagine and tyrosine
derivatives were approx. 40%, comparable to
the result previously obtained with VEAc,10 while
REFERENCES
the other derivatives were metabolized less. The
extent of metabolism does not seem to be clearly
1. Yoshida E, Watanabe T, Takata J, Yamazaki A,
correlated with time nor with the enzymatic
Karube Y, Kobayashi S. 2006. Topical application of
half-life in vitro. a novel, hydrophilic gamma-tocopherol derivative
DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009
2376 MARRA ET AL.
reduces photo-inflammation in mice skin. J Invest 14. Perry DK. 2002. Serine palmitoyltransferase: Role
Dermatol 126:1633 1640. in apoptotic de novo ceramide synthesis and other
2. Wang X, Quinn PJ. 2000. The location and function stress responses. Biochim Biophys Acta 1585:146
of vitamin E in membranes (review). Mol Membr 152.
Biol 17:143 156. 15. van Nieuwpoort F, Smit NP, Kolb R, van der Meu-
3. Thiele JJ, Schroeter C, Hsieh SN, Podda M, Packer len H, Koerten H, Pavel S. 2004. Tyrosine-induced
L. 2001. The antioxidant network of the stratum melanogenesis shows differences in morphologic
corneum. Curr Probl Dermatol 29:26 42. and melanogenic preferences of melanosomes from
4. Rumelin A, Humbert T, Fauth U. 2004. Determina- light and dark skin types. J Invest Dermatol
tion of alpha-tocopherol in plasma by high perfor- 122:1251 1255.
mance liquid chromatography with fluorescence 16. Clemmons JJ. 1966. Proline metabolism, collagen
detection and stability of alpha-tocopherol under formation, and lathyrism. Fed Proc 25:1010
different conditions. Arzneimittelforschung 54: 1015.
376 381. 17. Artusi M, Nicoli S, Colombo P, Bettini R, Sacchi A,
5. Mahamongkol H, Bellantone RA, Stagni G, Plako- Santi P. 2004. Effect of chemical enhancers and
giannis FM. 2005. Permeation study of five formu- iontophoresis on thiocolchicoside permeation across
lations of alpha-tocopherol acetate through human rabbit and human skin in vitro. J Pharm Sci 93:
cadaver skin. J Cosmet Sci 56:91 103. 2431 2438.
6. Norkus EP, Bryce GF, Bhagavan HN. 1993. Uptake 18. Nicoli S, Cappellazzi M, Colombo P, Santi P. 2003.
and bioconversion of alpha-tocopheryl acetate to Characterization of the permselective properties of
alpha-tocopherol in skin of hairless mice. Photo- rabbit skin during transdermal iontophoresis.
chem Photobiol 57:613 615. J Pharm Sci 92:1482 1488.
7. Beijersbergen van Henegouwen GM, Junginger 19. Mirrlees MS, Moulton SJ, Murphy CT, Taylor PJ.
HE, de Vries H. 1995. Hydrolysis of RRR-alpha- 1976. Direct measurement of Octanol-Water parti-
tocopheryl acetate (vitamin E acetate) in the skin tion coefficient by high-pressure liquid chromato-
and its UV protecting activity (an in vivo study with graphy. J Med Chem 19:615 618.
the rat). J Photochem Photobiol B 29:45 51. 20. Brent DA, Sabatka JJ, Minick DJ, Henry DW. 1983.
8. Baschong W, Artmann C, Hueglin D, Roeding J. A simplified method for determining lipophilicity
2001. Direct evidence for bioconversion of vitamin E for structure-activity relationships. J Med Chem
acetate into vitamin E: An ex vivo study in viable 26:1020 1040.
human skin. J Cosmet Sci 52:155 161. 21. Unger SH, Chiang GH. 1981. Octanol-physiological
9. Nabi Z, Tavakkol A, Dobke M, Polefka TG. 2001. buffer distribution coefficients of lipophilic
Bioconversion of vitamin E acetate in human skin. amines by reversed-phase high-performance
Curr Probl Dermatol 29:175 186. liquid chromatography and their correlation
10. Ostacolo C, Marra F, Laneri S, Sacchi A, Nicoli S, with biological activity. J Med Chem 24:262
Padula C, Santi P. 2004. Alpha-tocopherol pro-vita- 270.
mins: Synthesis, hydrolysis and accumulation in 22. Huang D, Ou B, Hampsch-Woodill M, Flanagan JA,
rabbit ear skin. J Control Release 99:403 413. Deemer EK. 2002. Development and validation of
11. Mavon A, Raufast V, Redoules D. 2004. Skin oxygen radical absorbance capacity assay for lipo-
absorption and metabolism of a new vitamin E philic antioxidants using randomly methylated
prodrug, delta-tocopherol-glucoside: In vitro eva- beta-cyclodextrin as the solubility enhancer.
luation in human skin models. J Control Release J Agric Food Chem 50:1815 1821.
100:221 231. 23. Laneri S, Sacchi A, di Frassello EA, Luraschi E,
12. Yasuoka S, Takata J, Karube Y, Katoh E, Tsuzuki Colombo P, Santi P. 1999. Ionized prodrugs of
T, Kizu J, Tsuchiya M, Kobayashi S. 2005. Topical dehydroepiandrosterone for transdermal ionto-
application of a novel, water-soluble gamma-toco- phoretic delivery. Pharm Res 16:1818 1824.
pherol derivative prevents UV-induced skin 24. Rojas-Garbanzo RE, Mata-Segreda JF. 1998.
damage in mice. Photochem Photobiol 81:908 913. Assessment of the steric tolerance of the P sector
13. Marty JP. 2002. NMF and cosmetology of cutaneous in the catalytic site of porcine liver esterase. Bioorg
hydration. Ann Dermatol Venereol 129:131 136. Med Chem Lett 8:7 10.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 7, JULY 2009 DOI 10.1002/jps