Improved peptide prodrugs of 5-ALA for PDT: Rationalization of cellular accumulation and protoporphyrin IX production by direct determination of cellular prodrug uptake and prodrug metabolization

Article abstract of DOI:10.1021/jm900224r

Twenty-seven dipeptide derivatives of general structure Ac-Xaa-ALA-OR were synthesized as potential prodrugs for 5-aminolaevulinic acid-based photodynamic therapy (ALA-PDT). Xaa is an R-amino acid, chosen to provide a prodrug with appropriately tailored l

Full text of DOI:10.1021/jm900224r

4026
J. Med. Chem. 2009, 52, 4026–4037
Improved Peptide Prodrugs of 5-ALA for PDT: Rationalization of Cellular Accumulation and
Protoporphyrin IX Production by Direct Determination of Cellular Prodrug Uptake and
Prodrug Metabolization
Francesca Giuntini,† Ludovic Bourre´,‡ Alexander J. MacRobert,‡ Michael Wilson,§ and Ian M. Eggleston*,†
Wolfson Laboratory of Medicinal Chemistry, Department of Pharmacy and Pharmacology, UniVersity of Bath, Bath BA2 7AY, U.K., National
Medical Laser Centre, DiVision of Surgical and InterVentional Sciences, UCL Medical School, UniVersity College London, London W1W 7EJ, U.K.,
DiVision of Microbial Diseases, UCL Eastman Dental Institute, UniVersity College London, London WC1X 8LD, U.K.
ReceiVed February 20, 2009
Twenty-seven dipeptide derivatives of general structure Ac-Xaa-ALA-OR were synthesized as potential
prodrugs for 5-aminolaevulinic acid-based photodynamic therapy (ALA-PDT). Xaa is an R-amino acid,
chosen to provide a prodrug with appropriately tailored lipophilicity and water solubility. Although no simple
correlation is observed between downstream production of protoporphyrin IX (PpIX) in PAM212 keratinocytes
and HPLC-derived descriptors of compound lipophilicity, quantification of prodrug uptake reveals that most
of the dipeptides are actually more efficiently accumulated than ALA in PAM212 and also A549 and Caco-2
cell lines. Subsequent ALA release is the limiting factor, which emphasizes the importance of decoupling
prodrug uptake and intracellular metabolization when assessing the efficacy of ALA derivatives for PDT.
In agreement with PpIX fluorescence studies, at a concentration of 0.1 mM, L-Phe derivatives 4m and 4o,
and L-Leu, L-Met, and L-Glu derivatives 4f, 4k, and 4u, exhibit significantly enhanced photoxicity in PAM212
cells compared to ALA.
Introduction
5-Aminolaevulinic acid-based photodynamic therapy (ALA-
PDTa) is gaining increasing acceptance in medicine as an
effective technique for the treatment of a variety of neoplastic
lesions and premalignant disorders.1 In mammalian cells, ALA
is metabolized to protoporphyrin IX (PpIX), the precursor of
heme and a potent photosensitizer (Figure 1).2-4 Under physi-
ological conditions, high intracellular concentrations of heme
cause the feedback inhibition of ALA biosynthesis in mito-
chondria, but this may be bypassed by external administration
of ALA, and together with the relatively slow downstream
transformation of PpIX into heme by ferrochelatase, this results
in the accumulation of the natural photosensitizer. Clinically,
when sufficient intracellular levels of PpIX are attained, the
targeted tissue is irradiated with visible light to activate the
sensitizer and trigger a chain of events that ultimately result in
cell death. At a molecular level, this involves the interaction of
the excited photosensitizer with molecular oxygen, leading to
the generation of electrophilic species (singlet oxygen and/or
radicals) that cause oxidative damage to cellular constituents
such as phospholipidic membranes, nucleic acids, and proteins.5
Figure 1. 5-Aminolaevulinic acid (ALA) and protoporphyrin IX
(PpIX).
ALA-PDT currently finds wide use in dermatology for the
treatment of actinic keratosis, squamous cell carcinoma, and
Bowen’s disease,6,7 as well as cutaneous microbial infections
such as acne, onychomycosis, and verrucae.8-10 Promising
results have also emerged from studies of photorejuvenation of
the skin after sun-induced damage.6 In gastroenterology and
urology, PpIX from administered ALA may be used not only
to directly treat conditions such as Barret’s esophagus, inflam-
matory bowel disease, and bladder cancer by PDT11,12 but also
as a diagnostic tool for the visualization of precancerous changes
in the mucosae by fluorescence spectroscopy.13,14 Generally,
PpIX generated by exogenous ALA administration presents
several advantages over other types of photosensitizers that have
been applied in PDT, such as synthetic porphyrin, chlorin, or
phthalocyanine derivatives. For example, the risk of overtreat-
ment with ALA-PDT is limited due to the fast clearance of ALA
from the body, while saturation of PpIX production at high ALA
doses also prevents excessive PpIX production in tissues.15 The
relatively rapid photobleaching of PpIX also means that the
phototoxic effects of ALA-PDT are nonpersistent.16
* To whom correspondence should be addressed. Phone: +441225383101.
Fax: +44 1225386114. E-mail: ie203@bath.ac.uk.
†
Wolfson Laboratory of Medicinal Chemistry, Department of Pharmacy
and Pharmacology, University of Bath.
‡
National Medical Laser Centre, Division of Surgical and Interventional
Sciences, UCL Medical School, University College London.
§
Division of Microbial Diseases, UCL Eastman Dental Institute,
University College London.
a Abbreviations: ALA, 5-aminolaevulinic acid; PDT, photodynamic
therapy; PpIX, protoporphyrin IX; FD, fluorescence diagnosis; Su, succin-
imido; DIPEA, diisopropylethylamine; TIPS, triisopropylsilane; PyBrOP,
bromo-tris-pyrrolidino phosphoniumhexafluorophosphate; DMAP, 4-dim-
ethylaminopyridine; PBS, phosphate-buffered saline; PAM212, murine
keratinocyte cell line; A549, human Caucasian lung carcinoma cell line;
Caco-2, human colonic adenocarcinoma cell line; MTT, 3-(4,5-dimeth-
ylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; R2, linear correlation; LOD,
limit of detection; LOQ, limit of quantification; SD, standard deviation,
RSD, relative standard deviation.
The main drawbacks associated with ALA-PDT and ALA
fluorescence diagnosis (ALA-FD) result from the hydrophilic
nature of ALA itself. At physiological pH, ALA is a zwitterion,
which severely impairs its ability to cross cell membranes and
results in poor penetration and nonhomogeneous distribution
10.1021/jm900224r CCC: $40.75  2009 American Chemical Society
Published on Web 06/03/2009

ImproVed Peptide Prodrugs of 5-ALA for PDT
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 4027
Scheme 1a
Figure 2. Generic structure of dipeptide ALA prodrugs.
in the target tissue.17 To overcome this obstacle, forms of
delivery involving the use of penetration enhancers have been
devised,18 and much effort has been put into the development
of ALA prodrugs with more favorable lipid solubility19,20 based
upon initial observations that administration of ALA esters
(methyl, ethyl, propyl) caused a quicker, more homogeneous,
and more intense porphyrin fluorescence in mice skin than
ALA.21-24 Numerous prodrugs of this type have been synthe-
sized with linear, branched, or cyclic alkyl ester groups, as well
as examples with aromatic, heteroaromatic, perfluorinated
hydrocarbon, or ethylene glycol-type moieties.19 Other ester-
based strategies for improving uptake include the incorporation
of ALA into dendrimeric structures,25-28 and the synthesis of
glycoside esters of ALA, designed to potentially target tumoral
cells that overexpress nutrient transporters.29 Most recently,
acyloxyalkyl esters have also been proposed as multifunctional
ALA prodrugs that may induce cell death by both PDT-
dependent and independent mechanisms.30
a Reagents and conditions: (a) Fmoc-Xaa-OSu, DIPEA, THF, 0 °C to
rt; (b) Boc-Xaa-OSu, DIPEA, THF, 0 °C to rt; (c) HCl, 1,4-dioxane, then
Ac2O, DIPEA, CH2Cl2, 0 °C to rt (4a-v); (d) side-chain deprotection step
(5r-t; see Table 1 for details).
Table 1. Preparation of N-Urethane-Protected and N-Acetyl Dipeptide
Derivatives
An attractive way to obtain ALA prodrugs that have both
improved physicochemical properties and can selectively release
ALA in specific cell lines is to incorporate ALA into a short
peptide derivative.31,32 Following this approach, upon cellular
uptake, ALA release is mediated by the action of cytoplasmic
esterases and/or proteases, and it may be possible to design ALA
prodrugs that target disease-dependent levels of a given
enzymatic activity. In this general context, we have identified
prodrugs of the general structure shown in Figure 2 as potential
candidate derivatives that provide enhanced intracellular ALA
delivery. Such molecules are stable at physiological pH, unlike
ALA, its esters, and dipeptide derivatives with a free amino
terminus,33-35 but they may be fine-tuned in terms of their
overall lipophilicity to favor passive uptake while still retaining
water solubility, by variation of R′ (the side chain of the amino
acid coupled to ALA) and the ester moiety R.
Results and Discussion
Synthesis. The desired N-acetylated dipeptides were obtained
according to the method previously described, with some
important modifications, as outlined in Scheme 1.
We have previously reported that the coupling of N-urethane-
protected amino acids to ALA can be successfully accomplished
without the use of excess reagents provided that a low overall
concentration of ALA relative to the acylating species is
maintained throughout the course of the reaction.36 Thus, when
ALA, as the hydrochloride salt, is slowly neutralized in the
presence of an equimolar quantity of the appropriate activated
amino acid derivative, the formation of the required coupled
product is favored over competing intermolecular Schiff base
formation.37,38 In this original procedure, the key acylation
reaction was followed by conversion of the resulting dipeptide
to the methyl ester by treatment with diazomethane.39 While
this provided efficient access to the versatile derivatives 3, it
was only compatible with relatively small scale syntheses. To
simplify the procedure, methyl aminolaevulinate hydrochloride
1a was used directly in the coupling reaction, and under carefully
controlled conditions with a fixed concentration of reagents (0.07
M for 1a and the amino acid succinimido ester derivative, 0.14
M for DIPEA; see Scheme 1), the desired N-urethane-protected
dipeptides were obtained in excellent yields on gram scales
a Coupling reaction was performed with ethyl-5-aminolaevulinate hy-
drochloride (1b). b Conditions for final side chain deprotection: H2, Pd/C,
CH3OH, rt, (5r, 89%). c Conditions for final side chain deprotection: TFA,
CH2Cl2, TIS, rt, (5s, 85%). d Conditions for final side chain deprotection:
H2, Pd/C, CH3OH, HCl, rt, (5t, 96%).
(Table 1). Use of an increased addition time of 7 h was found
to give optimum results for coupling with various Boc- (entries
3a-u) and Fmoc- (entries 2a and 2b) amino acids, in both L-
and D-forms, with a range of side chain protection and with
significant improvements in overall yield relative to our original
procedure. The procedure could also be readily adapted to other
simple esters of ALA (see entry 3o). Compound 1a and the
ethyl ester, 1b, were obtained on a multigram scale by a novel
esterification procedure by treatment of ALA with trimethyl-
orthoformate or triethylorthoformate in the corresponding

4028 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Table 2. Preparation of L-Phenylalanyl Ester Prodrugs
Giuntini et al.
compounds under study. The classic descriptor for the lipophi-
licity of a drug is the octanol/water partition coefficient (log
P), however experimental log P determinations by the “shake-
flask” method may be both lengthy and laborious for a large
compound series. Not only are relatively large amounts of
analytically pure material required, but quantification of com-
pounds that are sparingly soluble in one of the phases may not
be feasible without a suitably sensitive analytical technique. To
address these difficulties, several chromatographic descriptors
for lipophilicity have been suggested based upon the principle
that the behavior of a chemical entity in an RP-HPLC system
is mainly the result of partition equilibrium between a lipophilic
solid phase and a predominantly aqueous mobile phase. Such
chromatographic hydrophobicity indices are amenable to rapid
high-throughput determination, under either gradient or isocratic
elution RP-HPLC conditions, with isocratic elution-based
techniques showing good correlation to the traditional log P
values. In this context, the isocratic chromatographic hydro-
phobicity index, ꢀ0, developed by Valko´ et al., is a very useful
descriptor that characterizes compound hydrophobicity and has
been shown to correlate well with log P for a wide range of
drug molecules.43-45 ꢀo is derived from the relationship between
the percentage of organic solvent in the mobile phase, ꢀ, and
the isocratic retention factor, log k, expressed by the eqs 1-3:
alcohol in the presence of a catalytic amount of an acidic ion-
exchange resin at room temperature. This was found to give a
substantial improvement in both yield and ease of product
isolation compared to the widely used thionyl chloride esteri-
fication method.40
log k ) log(tR - t0)/t0
(1)
The desired N-acetylated dipeptides 4 and 5 were obtained
from 3 following cleavage of N-terminal Boc protection with 4
M HCl in dioxane and subsequent acetylation with acetic
anhydride (2 equiv) in the presence of DIPEA (2.2 equiv). As
shown in Table 1, this two-step transformation generally
proceeded with high efficiency (54-94%). Final side chain
deprotection for His derivative 4s was accomplished by aci-
dolysis of the trityl group with TFA/CH2Cl2/TIPS. For Tyr(Bn)
and Lys(Z) derivatives 5r and 5t, final deprotection was effected
by hydrogenolysis (10% Pd-C). In the case of the Lys
derivative, it proved essential to perform the hydrogenolysis
reaction in the presence of HCl in order to trap the ε-amino
group as the hydrochloride salt and prevent intra- or intermo-
lecular Schiff base formation with ALA keto functions.41
Interestingly, it was preferable to prepare Ser derivative 4l
without the use of side chain protection because Boc removal
was found to result in partial cleavage of Ser(Bn), thus leading
to a mixture of the desired N-acetylated dipeptide and N,O-
diacetylated material. This problem was circumvented by
coupling of Boc-L-Ser-OSu to 1a, subsequent Boc group
removal, and performing the acetylation reaction with just 1
equiv of acetic anhydride.
(where tR ) retention time and t0 ) dead time).
log k ) Sꢀ + log kw
(2)
(3)
ꢀ0 ) -log kw/S
(where log kw represents the retention factor on extrapolation
to pure aqueous buffer). ꢀ0 thus corresponds to the value of ꢀ
needed to obtain log k ) 0, i.e., a retention time that is double
the dead time (tR ) 2t0). To determine ꢀ0, the log k values for
a given compound are thus measured using mobile phases with
different solvent composition ꢀ, and the former are plotted as
a function of ꢀ according to eq 2. ꢀ0 is then obtained from the
slope and intercept of the straight line according to eq 3 (ꢀ0 )
-slope/intercept). Values of ꢀ0 depend on pH and the type of
organic solvent employed. Parallel determinations with more
than one solvent are generally recommended to confirm the
validity of any trend within a compound series.
Applying this method, we calculated the ꢀ0 for the com-
pounds of Table 1 under isocratic conditions with either CH3CN
or CH3OH as the organic component and 50 mM aq ammonium
formate (pH ) 6.8). The correlation between log k and ꢀ was
found to be linear for both solvents, provided that values of ꢀ
were chosen such that -1.5 < log k < 0.6. Under these
conditions, the linear range was greater for determinations with
CH3OH.
The experimentally determined ꢀ0 values obtained with both
solvents are reported in Table 3, alongside the calculated log P
(clogP). The compounds are listed in order of increasing values
of ꢀ0. Gratifyingly, for the prodrugs under study, an acceptable
linear correlation between the experimental and calculated
hydrophobicity descriptors was observed (R ) 0.966 and R )
0.865 for CH3CN and CH3OH, respectively), further validating
the use of ꢀ0 as an alternative to log P.
Enhancement of PpIX Fluorescence with ALA Peptide
Prodrugs. The production of PpIX after incubation with either
ALA or L-amino acid-containing prodrugs was evaluated in
PAM212 keratinocytes as a function of the dose and incubation
As reported previously,36 ALA-containing dipeptides such
as 4m could be readily converted into a variety of other ester
derivatives via the intermediate acid. Thus, saponification of
4m with LiOH in aqueous CH3OH, gave 6, which then
underwent PyBroP-mediated esterification42 with hexan-1-ol,
2-methoxyethanol, 2-(2-methoxy)-ethoxyethanol, geraniol, and
menthol to give the novel derivatives, 7a-e, in satisfactory
yields (Table 2).
Evaluation of Lipophilicity of the Dipeptides. We have
previously speculated that increasing the lipophilicity of pro-
drugs of the general structure described should result in
enhanced intracellular delivery of ALA and PpIX production,
assuming predominant uptake by passive diffusion. To inves-
tigate whether a correlation could be established between the
lipophilicity of the dipeptides 4a-v and 5r-t and the corre-
sponding efficiency of PpIX production, it was therefore
necessary to establish the relative lipophilicity of all the

ImproVed Peptide Prodrugs of 5-ALA for PDT
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 4029
Table 3. ꢀ0 (CH3CN and CH3OH) and clogP Values for L-Amino
34.71; L-Met 4k, ꢀ0 ) 28.89; Table 3). However, high levels
of PpIX are also obtained after exposing the cells to compounds
that are notably more hydrophilic as demonstrated by their ꢀ0
values, such as L-Ala 4b (ꢀ0 ) 14.18) and L-Glu 4u (ꢀ0 )
23.14). In contrast, L-Trp derivative 4p (ꢀ0 ) 37.35) gives no
apparent increase in PpIX production at either concentration
relative to ALA.
Acid-Containing ALA Dipeptidesa
No PpIX production was detected in a parallel series when
PAM212 cells were exposed to the dipeptides 4c, 4e, 4g, 4j,
4n, and 4q, which contain D-amino acids (data not shown).
These results confirm that the release of ALA from the prodrugs
is enzyme-mediated, hence PpIX production cannot be elicited
using these compounds in cells that do not possess D-peptidase
activity.
The induction of PpIX production after incubation with
compounds 7a-e and 4o was also evaluated as above. As shown
in Figure 4, the results indicate that at 1 mM the dipeptides
and ALA induce PpIX production with similar efficiency, but
at lower doses, a marked enhancement of PpIX fluorescence
compared to that seen with ALA is observed after exposure to
compounds 4o and 7a-e. Still no direct correlation is evident
between the efficacy of the dipeptides studied in terms of
enhancing PpIX production and a simple increase in their
lipophilicity. As such, this alone is not a sufficient explanation
for the improved properties of some of the derivatives relative
to ALA.
Quantitative Determination of Intracellular Prodrug
Accumulation. Assessments of improvements in the cellular
uptake of a given ALA prodrug or formulation for PDT are
frequently based upon a direct measurement of PpIX fluores-
cence. While this provides a convenient readout of apparent
efficacy whenever different modalities of treatment need to be
compared (i.e., different prodrugs, different incubation times,
etc.), it should be borne in mind that the increase in fluorescence
measured in pharmacokinetics experiments in fact represents
the end result of a sequence of processes, namely the penetration
of the prodrug into the cells, the metabolization of the prodrug
and subsequent release of ALA, and ultimately the conversion
of ALA into PpIX. As pointed out by other authors,24,46 it is
particularly important to decouple the stages of prodrug uptake
and intracellular metabolization in order to properly rationalize
structure-function relationships. In the light of the apparent
lack of correlation between the lipophilicity of our prodrug
derivatives and improvements in PpIX production, we decided,
in parallel, to unequivocally determine their uptake in PAM212
cells. This would allow us to determine whether the lack of
induction of PpIX production displayed by several of our
dipeptides could be ascribed to poor cellular penetration or
instead to lack of release of ALA due to inefficient metabo-
lization of the prodrugs.
To compare the extent of intracellular accumulation of the
prodrugs studied, we chose to use a HPLC-fluorescence method
recently developed in our laboratories.47 The method, based on
a procedure originally reported by Tomokuni48 and later
optimized by Oishi,49 relies on the reaction of ALA with two
molecules of acetylacetone and one molecule of formaldehyde
to give 2,6-diacetyl-1,5-dimethyl-7-(2-carboxyethyl)-3-H-pyr-
rolizine, a stable fluorescent derivative that can be isolated and
quantified by HPLC (inset Figure 5). We have successfully
applied this method for the detection and quantification of free
ALA directly in cell lysates over a broad range of concentrations,
and we also showed that it could be successfully adapted to the
quantification of ALA esters following hydrolysis with dilute
aq HCl.
a The values for the corresponding D-amino acids are shown in
parentheses.
time. The results of these experiments are summarized in Figure
3. Panel (A) shows that at 0.1 mM drug concentration after 4 h
of incubation, six dipeptides (4b, 4d, 4f, 4m, 4k, and 4u)
displayed an enhanced PpIX fluorescence compared to equimo-
lar ALA, while dipeptides 5r and 4a showed comparable
efficacy to ALA. When the incubation time was extended to
6 h, the difference between the efficacy of the aforementioned
dipeptides and ALA increased, especially for derivatives 4f, 4m,
4k, and 4u. At 0.01 mM (Figure 3B), both after 4 and 6 h of
incubation, no difference was observed between the PpIX
production induced by 4a, 4d, 5r, and ALA, whereas dipeptides
4b, 4f, 4m, and 4k gave a marked enhancement. It is worth
underlining that at 0.01 mM, the PpIX fluorescence induced by
ALA alone is not significantly different from the background
fluorescence registered in unexposed cells.
In our previous study, compound 4m was found to be more
efficient than ALA in enhancing PpIX production in PAM212
keratinocytes, especially at low concentrations (0.01 mM and
0.1 mM).35 Besides giving further evidence of the efficacy of
compound 4m, the present experiments show that other dipep-
tides, such as 4f, 4k, and, to a lesser extent 4b and 4u, are
similarly able to induce PpIX production in the cells when
administered at a concentration at which ALA is ineffective.
Comparison of the results of Figure 3 with the experimentally
determined hydrophobicity descriptor for each compound
indicates that some of the dipeptides that are most efficient in
inducing PpIX production do indeed display correspondingly
higher values of ꢀ0 (L-Leu 4f, ꢀ0 ) 36.67; L-Phe 4m, ꢀ0 )

4030 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Giuntini et al.
Figure 3. PpIX fluorescence registered after exposure of PAM212 cells to 0.1 mM (A) and 0.01 mM (B) ALA and dipeptides. The results obtained
after 4 h (dark-gray) and 6 h (light-gray) of incubation are shown.
For the dipeptides evaluated previously, complete hydrolysis
to liberate ALA could be achieved upon refluxing in either 6
M HCl for 2 h, or 4 M HCl for 3 h, as judged by analytical
HPLC (data not shown). We had established during the
optimization of our analytical method above that the ALA
derivatization reaction does not proceed effectively in strongly
acidic media (pH < 3), hence neutralization of the hydrolyzed
samples prior to derivatization was required. After several
attempts, it was found that this could be best achieved by adding
solid NaHCO3 to the hydrolysis mixtures. This allows neutral-
ization without risking bringing the pH to alkaline, which could
lead to ALA degradation, and also as a solid base does not cause
dilution of the sample. The optimized conditions for the
quantitative determination of ALA after hydrolysis of the
dipeptide prodrugs thus consisted of 3 h reflux in 4 M HCl,
neutralization with solid NaHCO3, followed by the derivatization
test and HPLC analysis according to our procedure. This method
was validated over the range 0.6-65 µM for linearity, accuracy,
and precision (see Supporting Information).
Once the method was validated, the stage was set to measure
the accumulation of ALA and a selected group of dipeptides in
PAM212 keratinocytes. Based upon the measured ꢀ0 values,
we chose examples of lipophilic dipeptides that should be
expected to accumulate effectively if uptake by passive diffusion
is significant.46 Alongside these, more hydrophilic species were
examined whose membrane permeability should be correspond-
ingly poor. The following compounds were therefore included:
(1) 4m, 4f, and 4k (containing the amino acids Phe, Leu,
and Met) as examples of lipophilic dipeptides which, as
expected, efficiently induce PpIX production.

ImproVed Peptide Prodrugs of 5-ALA for PDT
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 4031
Figure 4. PpIX fluorescence registered after exposing the cells to 0.01 mM (dark-gray), 0.1 mM (gray), and 1 mM (white) ALA and compounds
7a-e, 4m, and 4o after 4 h of incubation.
each drug, PpIX production was quantified by fluorescence
spectroscopy and total ALA content determined following
hydrolysis and derivatization according to the methods
described above. Figures 6 and 7 summarize the data obtained
in terms of ng/mg or µg/mg of protein, respectively, for ALA
and PpIX.
The results displayed in Figure 6 now reveal that the majority
of the dipeptides do in fact penetrate PAM212 cells more
efficiently than ALA itself. This is particularly true for the
species with high ꢀ0, such as 4m (Phe) and the corresponding
hexyl-esters 7a, 4f (Leu), 4h (Ile), 4k (Met), and 4p (Trp).
Strikingly, the more hydrophilic dipeptides 5t, 4l, 4b, and 4v
(Lys, Ser, Ala, Asp, respectively) although unable to elicit
elevated levels of PpIX, appear to be able to penetrate the cells
just as efficiently as ALA. The values of intracellular ALA
measured in PAM212 keratinocytes after incubation with 0.1
mM drug for 4 h show that compound 4m, the corresponding
Phe hexyl esters 7a, 4f, 4h, and 4k are all 2-5 times more
effectively accumulated than equimolar ALA. The same dif-
ference is found when the cells are exposed to a higher dose of
drugs (1 mM), where in addition a larger number of dipeptides
seem to penetrate more efficiently than ALA.
Figure 5. Analytical HPLC following derivatization of 10 µM aq ALA.
Conditions (see Experimental Section): Phenomenex C18 column;
mobile phase: 0.1% aq acetic acid (solvent A), 0.1% acetic acid in
CH3OH (solvent B); fluorescence detection: λexc ) 370 nm, λem ) 460
nm. (Inset: 2,6-diacetyl-1,5-dimethyl-7-(2-carboxyethyl)-3-H-pyrroliz-
ine).
Alongside the general finding that most of the dipeptides
accumulate more effectively than ALA, the results obtained
under these conditions also appear consistent with some
contribution to uptake from diffusion mechanisms. Compari-
son of total accumulated ALA and measured hydrophobicity
do indeed show a generally good correlation, although for
some molecules other uptake mechanisms must also come
into play. For example, compound 4n gives a lower value of
intracellular ALA compared to its enantiomer 4m. On the
basis of ꢀ0, the two enantiomeric prodrugs would be expected
to show comparable values of intracellular ALA if a
nonspecific phenomenon such as passive diffusion was the
predominant uptake mechanism involved. This is consistent
with our previous observation of the involvement of active
transport in the uptake of 4m in PAM212 cells.35 Similarly,
the relatively high efficacy of accumulation of Asp derivative
4v in spite of its low lipophilicity also suggests the role of
some active transport mechanism. Notwithstanding the
precise mechanism of uptake in these and other cases, it
(2) 5t and 4v (Lys, Asp), which being charged species at
physiological pH are likely to show poor accumulation and
hence not to promote PpIX production.
(3) 4h and 4p (Ile, Trp) as examples of dipeptides that on
the basis of their value of ꢀ0 would be expected to induce PpIX
production but fail to do so.
(4) 4l and 4b (Ser, Ala), which display intermediate values
of ꢀ0 and do not induce PpIX production.
(5) 4u (Glu), hydrophilic but nevertheless efficient in inducing
PpIX production.
Besides ALA itself as the reference drug, we also evaluated
ALA-Hex, the n-hexyl ester of ALA, as a standard for an ALA
prodrug, which, although chemically different from the com-
pounds of our study, has been extensively characterized with
respect to efficacy of PpIX production and mechanisms of
cellular uptake.49
PAM212 cells were incubated for 4 h at 37 °C, with either
0.1 or 1 mM of drug. The cells were then lysed, and for

4032 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Giuntini et al.
Figure 6. PpIX production and intracellular ALA concentration after incubation of PAM12 keratinocytes (A) and A549 cells (B) with 0.1 mM
concentrations of selected prodrugs.
should be noted that for D-Phe derivative 4n, a simple
assessment of PpIX fluorescence might have led to the
conclusion that no drug was internalized at all.
for APN/M aminopeptidases, which are expressed in this cell
line.34 In preliminary studies (data not shown) it emerged
that our dipeptide prodrugs are not able to induce the
production of PpIX in A459 cells. This result was not
completely unexpected because our N-acetylated prodrugs
are rather potential substrates for acylpeptide hydrolase,
which is sparingly expressed in the A549 line.50 When we
exposed A549 cells to compounds 4f, 4m, 4p, and 6a and
subsequently quantified ALA according to the HPLC-
fluorescence method described above, the uptake observed
was, although reduced compared to PAM212, still signifi-
cantly higher for the dipeptides than ALA (see Supporting
Information and Figure 6B). These data again prove conclu-
sively that the absence of PpIX production may be attributed
to poor metabolization rather than simply inefficient uptake
Because all the dipeptides examined are at least as efficient
as ALA in gaining access to PAM212 cells, it is reasonable
to assume that the poor performance of certain derivatives
in terms of PpIX production reflects a slower metabolization
and a corresponding lower affinity for the esterases and/or
peptidases involved in the release of ALA from the prodrug.
To confirm this hypothesis and further define the general
utility of the above prodrugs, we examined a selection of
derivatives in two other cell lines. Berger et al. have shown
that in A549 human lung carcinoma cells, PpIX synthesis
can be efficiently induced by ALA prodrugs such as L-Phe-
ALA-Me and L-Ala-ALA-Me, which are potential substrates

ImproVed Peptide Prodrugs of 5-ALA for PDT
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 4033
Figure 7. PpIX production and intracellular ALA concentration after incubation of PAM12 keratinocytes (A) and Caco-2 cells (B) with 1 mM
concentration of selected drugs.
in this cell line. The same dipeptides also accumulated 3-7
times more efficiently in Caco-2 cells compared to ALA (see
Supporting Information and Figure 7B), notwithstanding the
lower level of PpIX fluorescence observed relative to
equimolar ALA.
Phototoxicity. The phototoxicities of the prodrugs that gave
rise to the most significant enhancements in PpIX fluores-
cence were also evaluated in PAM212 keratinocytes. As
shown in Figure 8, the cells were irradiated with blue light
(2.5 J/cm2) after 4 h of incubation with three doses of the
selected prodrugs, as well as ALA and ALA-Hex, for
comparison. In agreement with the data from the PpIX
fluorescence experiments, the dipeptides that exhibited a
marked enhancement of PpIX fluorescence also exhibited
high phototoxicity. At 1 mM, the phototoxicity of the
dipeptides ALA-Hex and ALA were comparable. However,
at 0.1 mM, the cell survival observed after exposure to ALA
was 11%, while dipeptides 4b (Ala), 4f (Leu), 4k (Met), 4m
(Phe), 4u (Glu), and the hexyl ester 7a were as effective as
ALA-Hex, reducing the cell viability to only 3%. Further-
more, when the cells were exposed to 0.01 mM ALA, no
decrease in cell viability was observed, but compounds 4k,
4m, and 7a still retained high phototoxicity. It is important
to underline that under the conditions adopted in all these
experiments, the dipeptides displayed no dark toxicity (data
not shown).
Conclusions
We have synthesized and characterized a series of ALA-
containing peptide prodrugs of general structure Ac-Xaa-ALA-

4034 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Giuntini et al.
results in an in vivo setting. We are currently applying the
synthetic and analytical methods described to fine-tune the
structures of our ALA prodrugs to target specific or elevated
expressed protease activities51 and effect selective ALA release
in normal or malignant tissue for diagnostic or therapeutic
purposes.
Experimental Section
General Remarks. Chemical reagents were purchased from
Sigma-Aldrich, Fluka, Acros, Lancaster, and Novabiochem. An-
hydrous CH2Cl2 was obtained by distillation from calcium hydride,
and anhydrous THF and Et2O were distilled from sodium/
benzophenone. All other solvents were purchased from Fisher
Scientific and used as received. Analytical TLC was performed
using silica gel 60 F254 precoated on aluminum sheets (0.25 mm
thickness). Column chromatography was performed on silica gel
60 (35-70 µ) from Fisher Scientific. Melting points were recorded
on a Reichert-Jung Thermo Galen Kofler block and are uncor-
rected. 1H and 13C NMR were recorded using a JEOL Delta
spectrometer at 270 MHz (1H) and 68 MHz (13C) or on a Varian
Mercury-VX spectrometer at 400 MHz (1H) and 100 MHz (13C).
Chemical shift values are given in ppm (δ). J values are given in
Hz. HPLC analyses were performed on a Dionex Ultimate 3000
system (Dionex, UK). The system consisted of a LPG-3400 pump
fitted with an internal vacuum degasser, a WPS-300SL semi-
preparative autosampler equipped with a 130 µL loop, a TCC-3000
column compartment, a VWD-3400 variable wavelength detector,
and a RF-2000 fluorescence detector. The separations were
performed on a Gemini 5 µ C18 110A column, 150 mm × 4.6
mm (Phenomenex, UK), equipped with a SecurityGuard C18 (ODS)
4 mm × 3.0 mm ID guard column (Phenomenex, UK), at 35 (
0.1 °C, with a flow rate of 0.7 mL/min. All compounds submitted
to biological analysis had a purity g95% as judged by HPLC. High
resolution mass spectrometry was performed using a Bruker
MicroTOF autospec ESI mass spectrometer. Elemental analyses
were performed on an EAI CE-440 instrument. Linear regression
analysis was performed using the software package Origin 8
(OriginLab Corporation). clogP values of the compounds were
obtained using the software package ChemDraw 11.0.1, (Cam-
bridgeSoft).
Figure 8. Phototoxicity after incubation with 0.01 mM (white), 0.1
mM (light-gray), and 1 mM (gray) of ALA, ALA-Hex, 4b, 4f, 4k,
4m, 4u, and 7a in PAM212 keratinocytes. Incubation time was 4 h.
Irradiation was performed with blue light (2.5 J/cm2). Cell viability
was assessed by MTT assay (see Experimental Section for details).
Scheme 2a
a Reagents and conditions: (a) LiOH, CH3OH/H2O (2:3), rt; (b) ROH,
PyBroP, DIPEA, DMAP, CH2Cl2, rt.
OR, where Xaa is an R-amino acid, that may be selected so as
to provide an appropriate balance of lipophilicity and water
solubility. The compounds were obtained in excellent yields
and display significantly higher stability in aqueous solution
compared to ALA itself. The lipophilicities of the new
compounds were quantified by RP-HPLC methods, and the
metabolization of the former to produce PpIX upon uptake in
PAM212 keratinocytes was evaluated by fluorescence spectros-
copy. This revealed a lack of apparent correlation between
prodrug lipophilicity and efficiency of PpIX production, which
could be rationalized by evaluating the intracellular accumula-
tion of the compounds independently from their metabolic fate.
Our method of chemical derivatization and quantification of
ALA from intact/partially metabolized prodrugs allows for the
first time a distinction to be drawn between prodrugs, which
are not internalized in a given cell line, and those which are
efficiently internalized but poorly metabolized to PpIX. This
should provide a valuable tool for rationalizing the efficiency
of other ALA-containing prodrugs. The results obtained herein
across three different cell lines confirm that the incorporation
of ALA into a short peptide derivative is an effective general
approach for increasing cellular delivery of ALA. Several
derivatives show significantly elevated cellular accumulation in
PAM212 keratinocytes compared to ALA and are indeed able
to induce PpIX production at concentrations where ALA is not
effective. Future studies will attempt to validate these promising
General Procedure for the Synthesis of Urethane-Protected
Dipeptides. tert-Butoxycarbonyl-L-phenylalanyl-5-aminolaevulinic
Acid Methyl Ester (3m).36 A solution of Boc-L-Phe-OSu (4.00
g, 11.00 mmol) in dry THF (100 mL) was treated with 1a (2.00
g, 11.00 mmol), and the resulting suspension was cooled at 0 °C under
N2. A solution of DIPEA (2.00 mL, 11.00 mmol) in dry THF (40
mL) was added by slow infusion with a syringe pump at a rate of
1.49 mL/h. Stirring was continued overnight, and then the solvent
was evaporated. The crude product was dissolved in ethyl acetate,
preadsorbed onto silica, and purified by flash chromatography (3:1
ethyl acetate/hexanes). Recrystallization from CH2Cl2/hexane gave
a white solid (4.19 g, 95%); mp 82-85 °C (lit.,36 83-85 °C).
General Procedure for the Synthesis of N-Acetylated Dipep-
tides. Acetyl-L-phenylalanyl-5-aminolaevulinic Acid Methyl Ester
(4m).36 Compound 3m (1.50 g, 3.80 mmol) was treated with 4 M
HCl in 1,4-dioxane (19 mL), and the resulting solution was stirred
at room temperature for 40 min. The solvent was evaporated, and
the residue was dried under vacuum. A suspension of the crude
hydrochloride salt thus prepared in dry CH2Cl2 (20 mL) was cooled
to 0 °C under a N2 atmosphere and DIPEA (1.45 mL, 8.36 mmol)
was added, followed by acetic anhydride (0.77 mL, 7.65 mmol).
The mixture was allowed to attain room temperature with stirring
overnight and then the solvent was evaporated. The crude product
was dissolved in CH3OH, preadsorbed onto silica, and purified by
flash chromatography (30:1 ethyl acetate/CH3OH). Recrystallization
from ethyl acetate gave a white solid (900 mg, 70%); mp 127-130
°C (lit.,36 126-128 °C).

ImproVed Peptide Prodrugs of 5-ALA for PDT
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 4035
General Procedure for the Synthesis of N-Acetyl-L-phenylala-
nyl-5-aminolaevulinic Acid Esters. Acetyl-L-phenylalanyl-5-ami-
nolaevulinic Acid Hexyl Ester (7a). A solution of 4m (50.0 mg,
0.15 mmol) in water (3 mL) and CH3OH (2 mL) was treated with
LiOH (11.0 mg, 0.45 mmol), and the resulting mixture was stirred
at room temperature for 1 h. Dowex (H+) ion-exchange resin was
then added to the mixture until pH 7 was attained. The resin was
filtered off, and the solvent was evaporated. The crude acid 6
was dissolved in dry CH2Cl2 (5 mL), and the resulting solution
was treated with PyBroP (70 mg, 0.15 mmol), DIPEA (78 µL, 0.45
mmol), and DMAP (2 mg, 0.01 mmol), followed by hexan-1-ol
(28 µL, 0.22 mmol). The mixture was allowed to stir at room
temperature overnight and then the solvent was evaporated.
Purification by flash chromatography (30:1 ethyl acetate/CH3OH
) 30/1) gave 7a as a white solid (19.5 mg, 31%); mp 94-96 °C.
1H NMR (400 MHz, CD3OD) δ: 7.29-7.18 (5H, m, Ph), 4.66 (1H,
dd, J 8.8, 5.2, NHCH(R′)CO), 4.05 (2H, t, J 6.4, OCH2), 4.02 (2H,
br, NHCH2CO, OCH2), 3.17 (1H, dd, J 13.8, 5.2, CHAHBPh), 2.87
(1H, dd, J 13.8, 8.8, CHAHBPh), 2.69 (2H, t, J 6.9, COCH2CH2-
CO2R), 2.56 (2H, t, J 6.9, COCH2CH2CO2R), 1.89 (3H, s, CH3CO),
1.63-1.59 (2H, m, CH2 hexyl), 1.38-1.29 (6H, m, CH2 hexyl),
0.91 (3H, s, CH3 hexyl). 13C NMR (100 MHz, CD3OD) δ: 204.7,
174.4, 174.1, 173.2, 138.5, 130.2, 129.4, 127.7, 65.8, 56.1, 38.9,
35.1, 32.6, 29.7, 28.7, 26.7, 23.6, 22.4, 14.3. ESI-HRMS+: calcd
for C22H33N2O5, 405.2384; found, 405.2371 [M + H]+.
Determination of Chromatographic Lipophilicity. The chro-
matographic descriptor for lipophilicity ꢀ0 was determined in
isocratic conditions, using mixtures of either CH3CN or CH3OH
and 50 mM aqueous ammonium formate. The isocratic retention
factor (log k) was calculated using the equation log k ) (tR - t0)/
t0. The dead time (t0) was measured by injecting a solution of uracil
according to the recommendations of the manufacturer. The
retention time (tR) was measured in duplicate, and the average value
was used to calculate log k. For each compound, six values of log
k were determined using six different mobile phase compositions,
whose percentages of organic solvent (ꢀ) were chosen so that -1.5
< log k < 0.6. The isocratic runs consisted of 5 min of equilibration
with the chosen ꢀ prior to the injection, followed by 15 min of
isocratic flow, a 0.1 min gradient ramp to 95% of organic solvent,
a 5 min wash in the same conditions, a 0.1 min ramp back to the
initial conditions, and finally 2 min of re-equilibration. The flow
rate was 1.0 mL/min. The values of log k were plotted versus the
percentage of organic solvent in the mobile phase, and linear
regression was applied to calculate the values of the slope (S), the
intercept (log kw), and the linear correlation (R2).
Cell Culture. The spontaneously transformed murine kerati-
nocyte cell line, PAM212 (obtained from Prof. R. Groves, King’s
College, London) was cultured in RPMI-1640 medium (Gibco BRL,
Life Technologies Ltd., Paisley, UK) containing L-glutamine (2
mM) and phenol red, supplemented with 10% fetal calf serum
(Sigma-Aldrich Ltd.), and penicillin and streptomycin (500 units/
mL, and 0.5 mg/mL, Gibco BRL). Human Caucasian lung
carcinoma, A549 cells were cultured in Ham’s F12K (Sigma-
Aldrich Ltd.) containing L-glutamine (2 mM) and phenol red,
supplemented with 10% fetal bovine serum (Sigma-Aldrich Ltd.),
and penicillin and streptomycin (500 units/mL, and 0.5 mg/mL,
Gibco BRL). Human colonic adenocarcinoma, Caco-2 cells were
cultured in EMEM (Sigma-Aldrich Ltd.) containing L-glutamine
(2 mM) and phenol red, supplemented with 10% fetal bovine serum,
1% nonessential amino acids (Sigma-Aldrich Ltd. Sigma-Aldrich
Ltd. Sigma-Aldrich Ltd.) and penicillin and streptomycin (500 units/
mL, and 0.5 mg/mL, Gibco BRL). The cells were routinely grown
as monolayers in 75 cm2 culture flasks (TPP, Helena Bioscience,
Gateshead, UK), at 37 °C, under a 5% CO2 atmosphere until
confluent.
0.1 mL of serum-free medium containing varying prodrug concen-
trations was added to a designated series of wells. Each plate
contained control wells with cells but without added drug for
determination of the background reading, and reference wells
containing cells incubated with the same ALA concentrations. For
drug incubation, serum-free medium was used because serum is
known to cause release of PpIX from cells, thus resulting in loss
of fluorescence signal.52
The fluorescence signal from each well was measured with a
Perkin-Elmer LS 50B fluorescence spectrometer (Perkin-Elmer,
Beaconsfield, UK) coupled to an automated plate reader, using 410
nm excitation and 635 nm emission wavelengths with slit widths
set to 10 nm and the internal 530 nm long-pass filter used on the
emission side; spectral scans were recorded between 600 and 750
nm to check for presence of any porphyrins other than PpIX.26
PpIX concentration was calculated from a standard curve of PpIX.
The mean fluorescence intensity (expressed in arbitrary units) was
calculated after subtraction of the control values. Intensity calibra-
tions were performed using rhodamine B embedded in a Perspex
disk as a standard.
Photodynamic Treatment. Cells were seeded into 96-well plates
at a density of approximately 5 × 104 cells per well. Following
incubation for 48 h, the cells were washed with PBS, and 0.1 mL
of solutions containing the appropriate drug at a concentration were
added to the wells and incubated for 4 h. The plates were then
irradiated using a LumiSource lamp (PCI Biotech, Oslo, Norway),
emitting a uniform field of low-power blue light (peak output
centered on 420 nm) over an area of 14 cm × 32 cm at a fluence
of 1.25 J/cm2. Immediately following irradiation, the medium was
replaced and cells were incubated for a further 24 h. Cytotoxicity
was determined using the MTT assay: cells were incubated with
medium containing 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-
zolium bromide (MTT) (1 mg/mL dissolved in full RPMI-1640
medium) for 3 h. The insoluble end product (formazan derivatives)
was dissolved in DMSO (0.1 mL) after removing the medium. UV
absorption was quantified at 570 nm using a 96-well plate reader
(MR 700 Dynatech, Dynex, Worthing, UK). The mean cell survival
was calculated for each prodrug at every concentration tested and
expressed as a percentage of control (incubated with the compounds
but not irradiated) cell survival values. Dark toxicity was determined
by assessing the cell survival after incubation with the drugs without
exposing the cells to irradiation.
Determination of Intracellular PpIX Content. Cells were
seeded into 100 mm Petri-dishes at a density of 5 × 104 cells per
well for 48 h at 37 °C. The culture medium was removed and the
cells were washed with PBS. The cells were then incubated with
the appropriate dose of ALA or of the selected dipeptides for 4 h
and washed with PBS. CelLytic (Sigma-Aldrich) (1 mL) was added
and incubated for 15 min at room temperature, and the cells were
mechanically scraped. Cell extracts were centrifuged at 1800 g for
10 min to remove the cell debris, and the supernatant containing
the PpIX was collected. The content of PpIX in the cell lysates
was determined by fluorescence as reported above. The protein
content of the cells was determined using a bicinchoninic acid
protein determination kit (Sigma-Aldrich).53
Determination of Intracellular ALA Accumulation.
Solutions. Stock solutions of ALA and ALA esters in deionized
water were prepared daily from powders stored at 4 °C. Working
solutions were obtained after dilution of the stock solutions using
class A glassware. The two ranges of concentrations investigated
were 0-5 µg/mL (0-30 µM) and 5-100 µg/mL (30-603 µM).
Acetylacetone reagent was prepared by mixing water, absolute
ethanol, and acetylacetone in a ratio of 11/6/3 by volume, and 10%
formaldehyde reagent was obtained by dilution of commercial 37%
v/v aqueous solution in water. Both the solutions were stored at 4
°C.
Preparation of Calibration Samples. For the determination of
ALA after hydrolysis of the prodrugs, the calibration samples were
prepared by spiking 370 µL of cell lysate (prepared as reported
above; see porphyrin extraction) with 30 µL of the appropriate
working solution, and then 200 µL of this mixture were diluted to
Fluorescence Pharmacokinetics. Cells were seeded into γ-s-
terilized 96-well plates (Orange Scientific, Triple Red Laboratory
Technologies, Long Crendon, UK) at a density of 5 × 104 cells
per well for 48 h. After removing the culture medium, the wells
were washed with phosphate buffered saline (PBS) and incubated
with freshly prepared solutions of ALA or ALA peptide derivatives:

4036 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Giuntini et al.
2 mL with 4 M aqueous HCl in a 16 mm Ø Greenhouse reaction
tube. The tubes were transferred into the Greenhouse reactor
preheated at 100 °C and refluxed for 3 h under stirring. The samples
were then cooled in an ice bath and were subsequently neutralized
by addition of equal quantities of solid NaHCO3 until neutral to
litmus. Then 50 µL of each sample were used for the derivatization
test. For each range of concentrations, two samples were prepared
for recovery evaluation purposes.
Derivatization Procedure. First, 50 µL of calibration sample
were added to 3500 µL of acetylacetone reagent and 450 µL of
10% formaldehyde solution in a Greenhouse reaction tube equipped
with a magnetic stirrer. The tubes were placed in the Greenhouse
reactor preheated at 100 °C and stirred for 10 min. The reactor
chamber was wrapped with foil in order to protect the tubes with
the reaction mixture from light. The samples were then cooled in
an ice bath in the dark for 2 h, transferred into HPLC vials, and
kept in the autosampler at room temperature until the analysis was
performed.
parameters for the HPLC-fluorescence method. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Krammer, B.; Plaetzer, K. ALA and its clinical impact, from bench
to bedside. Photochem. Photobiol. Sci. 2008, 7, 283–289.
(2) Peng, Q.; Berg, K.; Moan, J.; Kongshaug, M.; Nesland, J. M.
5-Aminolevulinic acid-based photodynamic therapy: principles and
experimental research. Photochem. Photobiol. 1997, 65, 235–251.
(3) Peng, Q.; Warloe, T.; Berg, K.; Moan, J.; Kongshaug, M.; Giercksky,
K. E.; Nesland, J. M. 5-Aminolevulinic acid-based photodynamic
therapy. Cancer 1997, 79, 2282–2308.
(4) Kennedy, J. C.; Pottier, R. H.; Pross, D. C. Photodynamic therapy
with endogenous protoporphyrin IX: basic principles and present
clinical experience. J. Photochem. Photobiol., B 1990, 6, 143–148.
(5) Brown, S. B.; Brown, E. A.; Walker, I. The present and future role of
photodynamic therapy in cancer treatment. Lancet Oncol. 2004, 5,
497–508.
(6) Blume, J. E.; Oseroff, A. R. Aminolevulinic acid photodynamic therapy
for skin cancers. Dermatol. Clin. 2007, 25, 5–14.
HPLC-Fluorescence Method. The mobile phase consisted of
0.1% acetic acid in water (solvent A) and 0.1% acetic acid in
CH3OH (solvent B). The composition of the mobile phase was as
follows: -5.0 to 0.0 min at 60% solvent B, 0.0 to 6.0 min at 60%
solvent B, 6.0 to 6.1 min 60 to 95% solvent B, 6.1 to 12.0 min at
95% solvent B, 12.0 to 12.1 min 95 to 60% solvent B. The flow
rate was 0.7 mL/min. λexc ) 370 nm and λem ) 460 nm were used
for the detection, and the fluorescence detector was set on high
sensitivity. The peak corresponding to 2,6-diacetyl-1,5-dimethyl-
7-(2-carboxyethyl)-3-H-pyrrolizine eluted at 4.71 min. The injection
volume was 20 µL. The total time required for the analysis was
12.1 min.
Method Validation. The methods developed were validated for
linearity, accuracy, and precision, limits of detection, and quanti-
fication (see Supporting Information). The linearity was verified
over the range 60-600 µM. For every curve, nine calibration
standards (in duplicate) were generated as described above on six
consecutive days. The samples were then analyzed by HPLC (in
duplicate) as reported above, and the detector response (peak area)
was plotted against the concentration of the analyte. Linear
regression analysis was applied to calculate the slope, the intercept,
and the linear correlation (R2). The intraday precision was evaluated
by analyzing three different samples within the concentration range
for six times in the same day. Interday precision was determined
at the same concentrations over three days. The limit of detection
(LOD) and the limit of quantification (LOQ) were determined
mathematically from the standard curve equations. The LOD and
LOQ were obtained by multiplying the standard deviation (SD) of
the intercepts by 3.3 and 10, respectively, and dividing by the slope.
Determination of ALA Concentration in Cell Lysates. Cell
lysates were prepared as reported above (see porphyrin extraction).
Then 200 µL of lysate were added to 1800 µL of 4 M aqueous
HCl. The resulting solution was refluxed for 3 h and then it was
cooled in an ice bath. The solutions were neutralized as described
above, and 50 µL were submitted to the derivatization reaction and
HPLC detection. The concentration of ALA in the sample was
calculated from the detector response (peak area) and the calibration
curve obtained as described above.
(7) Schleyer, V.; Szeimies, R.-M. ALA/MAL-PDT in dermatology. Comp.
Ser. Photochem. Photobiol. Sci. 2006, 7, 79–125.
(8) Pollock, B.; Turner, D.; Stringer, M. R.; Bojar, R. A.; Goulden, V.;
Stables, G. I.; Cunliffe, W. J. Topical aminolaevulinic acid-
photodynamic therapy for the treatment of acne vulgaris: a study of
clinical efficacy and mechanism of action. Br. J. Dermatol 2004, 151,
616–622.
(9) Gold, M. H.; Moiin, A. Treatment of Verrucae Vulgaris and Molluscum
contagiosum with photodynamic therapy. Dermatol. Clin. 2007, 25,
75–80.
(10) Kamp, H.; Tietz, H. J.; Lutz, M.; Piazena, H.; Sowyrda, P.; Lademann,
J.; Blume-Peytavi, U. Antifungal effect of 5-aminolevulinic acid PDT
in Trichophyton rubrum. Mycoses 2005, 48, 101–107.
(11) Ackroyd, R.; Kelty, C. J.; Brown, N. J.; Stephenson, T. J.; Stoddard,
C. J.; Reed, M. W. R. Eradication of dysplastic Barrett’s oesophagus
using photodynamic therapy: long-term follow-up. Endoscopy 2003,
35, 496–501.
(12) Barr, H.; Messmann, H.; Endlicher, E. ALA-PDT in gastroenterology.
Comp. Ser. Photochem. Photobiol. Sci. 2006, 7, 225–248.
(13) Kennedy, J. C.; Marcus, S. L.; Pottier, R. H. Photodynamic therapy
(PDT) and photodiagnosis (PD) using endogenous photosensitization
induced by 5-aminolevulinic acid (ALA): mechanisms and clinical
results. J. Clin. Laser Med. Surg. 1996, 14, 289–304.
(14) Marcus, S. L.; Sobel, R. S.; Golub, A. L.; Carroll, R. L.; Lundahl, S.;
Shulman, D. G. Photodynamic therapy (PDT) and photodiagnosis (PD)
using endogenous photosensitization induced by 5-aminolevulinic acid
(ALA): current clinical and development status. J. Clin. Laser Med.
Surg. 1996, 14, 59–66.
(15) Webber, J.; Kessel, D.; Fromm, D. Plasma levels of protoporphyrin
IX in humans after oral administration of 5-aminolevulinic acid. J.
Photochem. Photobiol., B 1997, 37, 151–153.
(16) Kennedy, J. C.; Pottier, R. H. New trends in photobiology: Endogenous
protoporphyrin IX, a clinically useful photosensitizer for photodynamic
therapy. J. Photochem. Photobiol., B 1992, 14, 275–292.
(17) Peng, Q.; Warloe, T.; Moan, J.; Heyerdahl, H.; Steen, H. B.; Nesland,
J. M.; Giercksky, K. E. Distribution of 5-aminolevulinic acid-induced
porphyrins in noduloulcerative basal cell carcinoma. Photochem.
Photobiol. 1995, 62, 906–913.
(18) Donnelly, R. F.; McCarron, P. A.; Woolfson, A. D. Drug delivery of
aminolevulinic acid from topical formulations intended for photody-
namic therapy. Photochem. Photobiol. 2005, 81, 750–767.
(19) Fotinos, N.; C, M. A.; Popowycz, F.; Gurny, R.; Lange, N. 5-Ami-
nolevulinic acid derivatives in photomedicine: characteristics, applica-
tion and perspectives. Photochem. Photobiol. 2006, 82, 994–1015.
(20) Lopez, R. F. V.; Lange, N.; Guy, R.; Bentley, M. V. L. B.
Photodynamic therapy of skin cancer: controlled drug delivery of
5-ALA and its esters. AdV. Drug DeliVery ReV. 2004, 56, 77–94.
(21) Gaullier, J.-M.; Berg, K.; Peng, Q.; Anholt, H.; Selbo, P. K.; Ma,
L.-W.; Moan, J. Use of 5-aminolevulinic acid esters to improve
photodynamic therapy on cells in culture. Cancer Res. 1997, 57, 1481–
1486.
(22) Kloek, J.; Beijersbergen van Henegouwen, G. M. J. Prodrugs of
5-aminolevulinic acid for photodynamic therapy. Photochem. Photo-
biol. 1996, 64, 994–1000.
(23) Peng, Q.; Moan, J.; Warloe, T.; Iani, V.; Steen, H. B.; Bjørseth, A.;
Nesland, J. M. Build-up of esterified aminolevulinic-acid-derivative-
induced porphyrin fluorescence in normal mouse skin. J. Photochem.
Photobiol., B 1996, 34, 95–96.
(24) Uehlinger, P.; Zellweger, M.; Wagnie`res, G.; Juillerat-Jeanneret,
L.; van den Bergh, H.; Lange, N. 5-Aminolevulinic acid and its
derivatives: physical chemical properties and protoporphyrin IX
Acknowledgment. We thank the Biotechnology and Bio-
logical Sciences Research Council for financial support
BBD0127831 (I.M.E.) and BBD0113291 (A.J.R., M.W.).
Note Added after ASAP Publication. This paper was published
on the web on June 3, 2009 with errors in the caption of Figure 3.
The revised version published on June 10, 2009.
Supporting Information Available: Characterization data for
compounds 2a, 2b; 3a-l, 3n-3u, 4a-l, 4n-v, 7b-7e. Detailed
experimental procedures and characterization data for compounds
5r-5t. Measured cellular PpIX and ALA content in PAM212,
A549, and Caco-2 cells after incubation with prodrugs. Validation

ImproVed Peptide Prodrugs of 5-ALA for PDT
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 4037
formation in cultured cells. J. Photochem. Photobiol., B 2000, 54,
72–80.
(25) Battah, S.; Balaratnam, S.; Casas, A.; O’Neill, S.; Edwards, C.; Batlle,
A.; Dobbin, P.; MacRobert, A. J. Macromolecular delivery of
5-aminolaevulinic acid for photodynamic therapy using dendrimer
conjugates. Mol. Cancer Ther. 2007, 6, 876–885.
(26) Battah, S.; O’Neill, S.; Edwards, C.; Balaratnam, S.; Dobbin, P.;
MacRobert, A. J. Enhanced porphyrin accumulation using dendritic
derivatives of 5-aminolaevulinic acid for photodynamic therapy: an
in vitro study. Int. J. Biochem. Cell Biol. 2006, 38, 1382–1392.
(27) Battah, S. H.; Chee, C. E.; Nakanishi, H.; Gerscher, S.; MacRobert,
A. J.; Edwards, C. Synthesis and biological studies of 5-aminolevulinic
acid-containing dendrimers for photodynamic therapy. Bioconjugate
Chem. 2001, 12, 980–988.
(28) Di Venosa, G. M.; Casas, A. G.; Battah, S.; Dobbin, P.; Fukuda, H.;
MacRobert, A. J.; Batlle, A. Investigation of a novel dendritic
derivative of 5-aminolaevulinic acid for photodynamic therapy. Int.
J. Biochem. Cell Biol. 2006, 38, 82–91.
(29) Vallinayagam, R.; Schmitt, F.; Barge, J.; Wagnieres, G.; Wenger, V.;
Neier, R.; Juillerat-Jeanneret, L. Glycoside esters of 5-aminolevulinic
acid for photodynamic therapy of cancer. Bioconjugate Chem. 2008,
19, 821–839.
(30) Berkovitch, G.; Doron, D.; Nudelman, A.; Malik, Z.; Rephaeli, A.
Novel multifunctional acyloxyalkyl ester prodrugs of 5-aminolevulinic
acid display improved anticancer activity independent and dependent
on photoactivation. J. Med. Chem. 2008, 51, 7356–7369.
(31) Casas, A.; Batlle, A. M. d. C.; Butler, A. R.; Robertson, D.; Brown,
E. H.; MacRobert, A.; Riley, P. A. Comparative effect of ALA
derivatives on protoporphyrin IX production in human and rat skin
organ cultures. Br. J. Cancer 1999, 80, 1525–1532.
(32) Schneider, R.; Tirand, L.; Frochot, C.; Vanderesse, R.; Thomas, N.;
Gravier, J.; Guillemin, F.; Barberi-Heyob, M. Recent improvements
in the use of synthetic peptides for a selective photodynamic therapy.
Anti-Cancer Agents Med. Chem. 2006, 6, 469–488.
(33) Berger, Y.; Greppi, A.; Siri, O.; Neier, R.; Juillerat-Jeanneret, L.
Ethylene glycol and amino acid derivatives of 5-aminolevulinic acid
as new photosensitizing precursors of protoporphyrin IX in cells.
J. Med. Chem. 2000, 43, 4738–4746.
(34) Berger, Y.; Ingrassia, L.; Neier, R.; Juillerat-Jeanneret, L. Evaluation
of dipeptide-derivatives of 5-aminolevulinic acid as precursors for
photosensitizers in photodynamic therapy. Bioorg. Med. Chem. 2003,
11, 1343–1351.
(35) Bourre, L.; Giuntini, F.; Eggleston, I. M.; Wilson, M.; MacRobert,
A. J. 5-Aminolaevulinic acid peptide prodrugs enhance photosensiti-
zation for photodynamic therapy. Mol Cancer Ther 2008, 7, 1720–
1729.
(36) Rogers, L. M. A.; McGivern, P. G.; Butler, A. R.; MacRobert, A. J.;
Eggleston, I. M. An efficient synthesis of 5-aminolaevulinic acid
(ALA)-containing peptides for use in photodynamic therapy. Tetra-
hedron 2005, 61, 6951–6958.
(37) Jaffe, E. K.; Rajagopalan, J. S. Nuclear magnetic resonance studies
of 5-aminolevulinate demonstrate multiple forms in aqueous solution.
Bioorg. Chem. 1990, 18, 381–394.
(38) Butler, A. R.; George, S. The nonenzymatic cyclic dimerisation of
5-aminolevulinic acid. Tetrahedron 1992, 48, 7879–7886.
(39) Diazomethane can be replaced with the safer trimethylsilyldiaz-
omethane without any decrease of yields: Presser, A.; Huefner, A.
TrimethylsilyldiazomethanesA mild and efficient reagent for the
methylation of carboxylic acids and alcohols in natural products.
Monats. Chem. 2004, 135, 1015–1022.
(40) Dabrowski, Z.; Kwasny, M.; Kaminski, J.; Beldowicz, M.; Lewicka,
L.; Obukowicz, B.; Kaliszewski, M.; Pirozynska, E. The synthesis
and applications of 5-aminolevulinic acid (ALA) derivatives in
photodynamic therapy and photodiagnosis. Acta Pol. Pharm. 2003,
60, 219–224.
(41) Attempted preparation of Ser derivative 4m and Lys 5u via the
corresponding Fmoc precursors 2a and 2b resulted in significant
decomposition upon base-mediated deprotection of the amino
function.
(42) Davies, J. S.; Howe, J.; Lebreton, M. A model reaction for assessing
the coupling and chiral efficiency of reagents in depside bond
formation. J. Chem. Soc., Perkin Trans. 2 1995, 2335–2339.
(43) Du, C. M.; Valko, K.; Bevan, C.; Reynolds, D.; Abraham, M. H. Rapid
gradient RP-HPLC method for lipophilicity determination: a solvation
equation based comparison with isocratic methods. Anal. Chem. 1998,
70, 4228–4234.
(44) Valko´, K. Application of high-performance liquid chromatography
based measurements of lipophilicity to model biological distribution.
J. Chromatogr., A 2004, 1037, 299–310.
(45) Valko´, K.; Bevan, C.; Reynolds, D. Chromatographic hydrophobicity
index by fast-gradient RP-HPLC: a high-throughput alternative to log
P/log D. Anal. Chem. 1997, 69, 2022–2029.
(46) Perotti, C.; Fukuda, H.; DiVenosa, G.; MacRobert, A. J.; Batlle, A.;
Casas, A. Porphyrin synthesis from ALA derivatives for photodynamic
therapy. In vitro and in vivo studies. Br. J. Cancer 2004, 90, 1660–
1665.
(47) Giuntini, F.; Bourre´, L.; MacRobert, A. J.; Wilson, M.; Eggleston,
I. M. Quantitative determination of 5-aminolaevulinic acid and its
esters in cell lysates by HPLC-fluorescence. J. Chromatogr., B 2008,
875, 562–566.
(48) Tomokuni, K.; Ichiba, M.; Hirai, Y.; Hasegawa, T. Optimized liquid-
chromatographic method for fluorometric determination of urinary
δ-aminolevulinic acid in workers exposed to lead. Clin. Chem. 1987,
33, 1665–1667.
(49) Oishi, H.; Nomiyama, H.; Nomiyama, K.; Tomokuni, K. Fluorometric
HPLC determination of δ-aminolevulinic acid (ALA) in the plasma
and urine of lead workers: biological indicators of lead exposure. J.
Anal. Toxicol. 1996, 20, 106–110.
(50) Scaloni, A.; Jones, W.; Pospischil, M.; Sassa, S.; Schneewind, O.;
Popowicz, A. M.; Bossa, F.; Graziano, S. L.; Manning, J. M.
Deficiency of acylpeptide hydrolase in small-cell lung carcinoma cell
lines. J. Lab. Clin. Med. 1992, 120, 546–552.
(51) Bourre´, L.; Giuntini, F.; Eggleston, I. M.; Wilson, M.; MacRobert,
A. J. Protoporphyrin IX enhancement by 5-aminolaevulinic acid
peptide derivatives and the effect of RNA silencing on intracellular
metabolism. Br. J. Cancer 2009, 100, 723–731.
(52) Kloek, J.; Akkermans, W.; Beijersbergen Henegouwen, G. M. J.
Derivatives of 5-aminolevulinic acid for photodynamic therapy:
enzymatic conversion into protoporphyrin. Photochem. Photobiol.
1998, 67, 150–154.
(53) Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia, A. K.; Gartner,
F. H.; Provenzano, M. D.; Fujimoto, E. K.; Goeke, N. M.; Olson,
B. J.; Klenk, D. C. Measurement of protein using bicinchoninic acid.
Anal. Biochem. 1985, 150, 76–85.
JM900224R