Distribution of chlorophyll- and bacteriochlorophyll-derived photosensitizers in human blood plasma

Article abstract of DOI:10.1111/j.1751-1097.2009.00621.x

Chlorophylla and, in particular, bacteriochlorophyll a derivatives are promising candidates for photosensitizers in photodynamic therapy. The distribution of 21 (bacterio)chlorophyll derivatives among human blood plasma fractions was studied by iodixanol gradient ultracentrifugation and in situ absorption spectroscopy. Modifications of the natural pigments involved the central metal (Mg²⁺, Zn²⁺, Pd²⁺, none), the isocyclic ring (closed, open and taurinated), substituents at C-3 (vinyl, acetyl, 1-hydroxyethyl) and C-17³ (phytyl ester, free acid). Cellular blood components bound only a small fraction of the pigments. Distribution among low-density lipoproteins (LDL), high-density lipoproteins (HDL) and high-density proteins (HDP) of the plasma was influenced as follows: (1) application in Cremophor EL slightly altered pigment distribution by lipoprotein modification, (2) only very polar pigments with multiple hydrophilic substituents showed substantial HDP binding, (3) the presence of the esterifying alcohol at C-17³ caused enrichment in LDL, this was more pronounced with bacteriochlorophylls than with chlorophylls, (4) substituents at C-3 had only little influence on the distribution, (5) Zn²⁺-complexes were enriched in HDL compared to Mg²⁺ and Pd²⁺ complexes, indicating specific binding of the former. Equilibration of pigments among the different fractions was largely complete within 3 h.

Full text of DOI:10.1111/j.1751-1097.2009.00621.x

Photochemistry and Photobiology, 2010, 86: 182–193
Distribution of Chlorophyll- and Bacteriochlorophyll-derived
Photosensitizers in Human Blood Plasma
¨
Jorg Dandler, Brigitte Wilhelm and Hugo Scheer*
¨
Department Biologie I — Botanik, Universitat Munchen, Munich, Germany
¨
Received 23 May 2009, accepted 18 July 2009, DOI: 10.1111 ⁄ j.1751-1097.2009.00621.x
the latter type are chlorophyll (Chl) and bacteriochlorophyll
ABSTRACT
(BChl) derivatives (9–17). These pigments absorb intensely in
the red (Chls) and NIR spectral regions (BChls) where tissue is
relatively transparent to light (18). Their rapid clearance is
probably due to a combination of factors including an active
excretion system (19,20), the photolability of many such
derivatives (21–23) and possibly also to dark degradations by
oxidative ring opening (24,25). The pharmacokinetics of these
photosensitizers is therefore of considerable practical interest,
but little is known so far on how this depends on the pigments’
structures.
Details of the treatment will also depend on the intended
target. While originally a cellular action had been assumed that
involves ROS-initiated necrosis (26), more indirect actions
have subsequently been suggested, including apoptotic signal-
ing by ROS and a vascular action (6,27–33). Due to their rapid
turnover times, (B)Chl derivatives are particularly suited for
vascular targeting (34,35). Although the concept of such an
indirect action has been emphasized increasingly during recent
years, its translation into rational protocols is still difficult. It
has also been recognized that the ROS generated by these
pigments depend on their microenvironment (36,37). As a
contribution to a rational approach, we present here a study
on the partitioning of (B)Chl derivatives among the different
fractions of human blood plasma (BP), its dependence on the
structure of the pigments and the time required for equilibra-
tion. The pigments investigated range from the highly hydro-
phobic (B)Chls that are esterified by the diterpenoid alcohol,
phytol, to the water-soluble taurinated BChl derivative,
WST11, and HpD as a control. Almost all BChl derivatives
are mainly bound to the lipoprotein fractions, with varying
preferences for low-density lipoproteins (LDL) or high-density
lipoproteins (HDL), and only the most polar derivatives
showed significant accumulation in the high-density protein
(HDP) fraction.
Chlorophyll a and, in particular, bacteriochlorophyll a deriva-
tives are promising candidates for photosensitizers in photo-
dynamic therapy. The distribution of 21 (bacterio)chlorophyll
derivatives among human blood plasma fractions was studied by
iodixanol gradient ultracentrifugation and in situ absorption
spectroscopy. Modifications of the natural pigments involved the
central metal (Mg2+, Zn2+, Pd2+, none), the isocyclic ring
(closed, open and taurinated), substituents at C-3 (vinyl, acetyl,
1-hydroxyethyl) and C-173 (phytyl ester, free acid). Cellular
blood components bound only a small fraction of the pigments.
Distribution among low-density lipoproteins (LDL), high-density
lipoproteins (HDL) and high-density proteins (HDP) of the
plasma was influenced as follows: (1) application in Cremophorꢀ
EL slightly altered pigment distribution by lipoprotein modifi-
cation, (2) only very polar pigments with multiple hydrophilic
substituents showed substantial HDP binding, (3) the presence of
the esterifying alcohol at C-173 caused enrichment in LDL, this
was more pronounced with bacteriochlorophylls than with
chlorophylls, (4) substituents at C-3 had only little influence on
the distribution, (5) Zn2+-complexes were enriched in HDL
compared to Mg2+ and Pd2+ complexes, indicating specific
binding of the former. Equilibration of pigments among the
different fractions was largely complete within 3 h.
INTRODUCTION
Photodynamic therapy (PDT) uses reactive oxygen species
(ROS) to treat malignancies including cancer. ROS are
generated by irradiating photosensitizing dyes by visible or
near-infrared (NIR) light. Topical action of the generally
intravenously injected photosensitizers is achieved by two
factors. The first one involves concentration differences
between healthy and transformed tissues, which were noted
early (1,2) but whose origin is still only partly understood
(3–6). The second factor is a spatially controlled irradiation of
the transformed tissue (7,8). The strategy and details of the
treatment protocols depend, among other factors, on the
dynamics of pigment transport, its uptake and excretion,
which in turn depend on the structure of the photosensitizer
and its mode of injection. Protocols differ, for example,
between slowly excreted pigments like hematoporphyrin deriv-
ative (HpD), and more rapidly excreted pigments. Examples of
MATERIALS AND METHODS
Materials. Chemicals were analytical grade or better. Unless otherwise
noted, chemicals were obtained from Sigma-Aldrich (Taufkirchen,
Germany), Roth (Karlsruhe, Germany) or Merck (Darmstadt,
Germany). Diethyl ether was dried and peroxides removed by Alumina
B–Super I columns (ICN Biomedicals, Eschwege, Germany). Doubly
demineralized water was used for all aqueous solutions. Blood
products were obtained from the Bavarian Red Cross (Munich,
Germany) and stored at 4ꢁC for up to 4 weeks. A separation of
occasionally existent chylomicrons was not necessary with either
fractionation method used.
*Corresponding author email: hugo.scheer @lmu.de (Hugo Scheer)
ꢂ 2009 The Authors. JournalCompilation. The AmericanSocietyof Photobiology 0031-8655/10
182

Photochemistry and Photobiology, 2010, 86 183
Pigments. General: All preparations were carried out under green
light (<1 lE m)2 s)1) and argon. Solvents were gassed with argon
prior to and during use. Reactions were followed by absorption
spectroscopy. Sodium ascorbate (ꢀ10 mg) was added as an antioxi-
dant with all zinc insertion reactions. Zinc and palladium complexes
were purified under argon on silica gel plates containing 1.5% sodium
ascorbate. Pigments were stored dry in the dark and under argon at
)20ꢁC. Tetrapyrrole amounts are given in ODÆmL units, namely, the
product of the optical density (OD at the Qy absorption maximum,
1 cm path length) and the volume (mL) of the pigment solution unless
otherwise noted in diethyl ether (ODÆmLDE). The C-132 epimers were
not separated.
Esterified Mg-complexes: Chl and BChl were extracted with methanol
from spray-dried Spirulina platensis (Behr Import, Bonn, Germany) and
freeze-dried Rhodospirillum rubrum G9 (DSM No. 486; DSMZ,
Braunschweig, Germany), respectively. 3-Acetyl-3-devinyl-chlorophyll
([3Ac]-Chl) was synthesized from BChl by oxidation with 2,3-dichloro-
5,6-dicyano-p-benzoquinone (38). 3-Vinyl-3-deacetyl-bacteriochloro-
phyll ([3Vi]-BChl) was synthesized from BChl via 3-(1-hydroxy)
ethyl-3-deacetyl-bacteriochlorophyll ([3HE]-BChl) according to Struck
et al. (39). The pigments were purified by DEAE-Sepharose CL-6B
(Amersham-Pharmacia, Freiburg, Germany) chromatography (40).
Nonesterified Mg-complexes: Chlorophyllide (Chlid) and bacterio-
chlorophyllide (BChlid) were obtained from Chl and BChl, respec-
tively, by enzymatic hydrolysis of the phytyl ester moiety (41). The
vector-bearing chlorophyllase (Arabidopsis thaliana ecotype Columbia)
was provided by C. E. Benedetti (Brazilian Synchrotron Light
Laboratory, Campinas, Brazil), and overexpressed in Escherichia coli
(strain ER 2508) by U. Oster (University of Munich, Germany).
Metal-free derivatives: Pheophytin (Phe) and bacteriopheophytin
(BPhe) were obtained by demetalation of Chl and BChl, respectively,
with acetic acid (15 min, ambient temperature). The free acids,
pheophorbide (Pheid) and bacteriopheophorbide (BPheid) were
obtained by treating Chl and BChl, respectively, with trifluoroacetic
acid (15 min, ambient temperature). As shown by absorption spec-
troscopy and HPTLC, the hydrolysates are free of colored by-products
and did not require any additional purification.
Tidas UVNIR ⁄ 16-1024 ⁄ 100-1 diode array detector (J&M, Aalen,
Germany) and a CLH 20 W lamp (J&M). The applied software was
Spectralys 1.82 (J&M) based on Spectacle (LabControl). For pigment
quantifications, identical absorption coefficients were assumed in the
different BP fractions. Mass spectra were recorded with an LTQ
Orbitrap Mass spectrometer (Thermo Electron, Dreieich, Germany).
All samples were diluted 9:1 (vol ⁄ vol) with 1% (vol ⁄ vol) formic acid in
methanol.
Blood (buffy coat) fractionation. Buffy coat fractionation by
Ficollꢀ-Paque PLUS (GE Healthcare, Munich, Germany) was per-
formed according to the manufacturer’s instructions.
BP fractionations.(A) Lipoprotein fractionation by sequential density
gradient ultracentrifugation (SDGU): The first method used to separate
human BP was SDGU (46,47). Densities were determined by aerometry.
When sample volumes were insufficient to completely fill two centrifuge
tubes, adequate amounts of Tris-EDTA (TE)) buffer of identical density
were added. All centrifugation steps were carried out at 10ꢁC in No.
342414 Quick-SealTM centrifuge tubes (39.0 mL; 1.0 · 3.5 in; Beckman,
Palo Alto) and a Type 70 Ti fixed-angle rotor (8 · 39.0 mL; Beckman).
For sealing and slicing of the tubes, Tube Sealer and Tube Slicer,
respectively, were used (Beckman). The following TE buffers with
different densities were used: TE buffer I (q = 1.0 g mL)1): Tris-HCl
(10 mM, pH 7.5) containing EDTA (1 mM); TE buffer II
(q = 1.019 g mL)1): 27.81 g NaCl plus 1 L of TE buffer I; TE buffer
III (q = 1.063 g mL)1): 96.25 g NaCl plus 1 L of TE buffer I; TE
buffer H (q = 1.125 g mL)1; 94.24 g KBr plus 1 L of TE buffer III);
TE buffer IV (q = 1.21 g mL)1): 234.93 g KBr plus 1 L of TE buffer
III.
Tetrapyrroles (ꢀ250 ODÆmLDE) were dissolved in 110 lL of a
mixture of Cremophor EL (CrEL) and ethanol (1:1 vol ⁄ vol), which was
diluted to 1.1 mL with normal saline immediately before addition to
55.0 mL BP (modified from Ref. [48]). Exceptions to this procedure were
WST11 and HpD (ꢀ250 ODÆmLmethanol) which were dissolved in 50 mM
phosphate buffer (PB; pH 7.2). All tetrapyrrole-loaded plasma samples
were incubated under slow rotation for 3 h at ambient temperature.
After adjusting the density to 1.019 g mL)1 with TE buffer I, the
BP was centrifuged at 120 000 g for 20 h. The top 15 mL was removed
as the very low-density lipoprotein (VLDL) fraction. The subnatant
density was adjusted to 1.063 g mL)1 with solid NaCl and centrifuged
at 120 000 g for 24 h. The top 15 mL was removed as the LDL
fraction. The subnatant density was adjusted to 1.21 g mL)1 with solid
KBr and centrifuged at 120 000 g for 48 h. The top 15 mL was
Transmetalated chlorin derivatives: Zn-pheophytin (Zn-Phe) and
Pd-pheophytin (Pd-Phe) were synthesized from Phe by the acetate
method in acetic acid or methanol (42). Zn-pheophorbide (Zn-Pheid)
was synthesized from Pheid by reaction with Zn(OAc)2 and Cd(OAc)2
(43). Pd-pheophorbide (Pd-Pheid) was obtained from Pheid by
reaction with Pd(OAc)2 in acetic acid (44).
removed as the HDL fraction; the subnatant (q > 1.21 g mL)1
)
contained the HDP fraction. The amounts of NaCl or KBr to increase
the density of a solution from qold to qnew were calculated according to
the formula given in the supplement. For storage, the LDL and HDL
fractions were dialyzed overnight at 4ꢁC against several liters of TE
buffer III (Servaporꢀ tubing, MWCO: 12 000–14 000; Ø: 16 mm;
Serva, Heidelberg, Germany). Samples were filtered sterile with
PuradiscTM FP30 CA-S syringe filters (pore size: 0.45 lm; Whatman,
Dassel, Germany).
HDL subfractionation into HDL2 and HDL3: HDL fractions were
subfractionated into HDL2 and HDL3 fractions by dialyzing them
overnight at 4ꢁC against TE buffer H (Servaporꢀ tubing, MWCO
12,000–14,000; Ø: 16 mm; Serva). The samples were centrifuged in
sealed No. 342412 Quick-SealTM centrifuge tubes (5.1 mL; 0.5 · 2.0 in;
Beckman) in a VTi 80 vertical-tube rotor (Beckman) at ꢀ260 000 g for
4 h. The top 0.75 mL was removed as the HDL2 fraction; the lower
4.5 mL contained the HDL3.
Transmetalated bacteriochlorin derivatives: Zn-bacteriopheophytin
(Zn-BPhe) and Zn-bacteriopheophorbide (Zn-BPheid) were synthe-
sized from BPhe and BPheid, respectively, by reaction with Zn(OAc)2
in acetic acid (42). Pd-bacteriopheophytin (Pd-BPhe) was obtained by
transmetalation of Cd-BPhe (obtained from BPhe by the acetate
method) with PdCl2 in acetone (45). Pd-Bacteriopheophorbide
(Pd-BPheid; WST09; Tookadꢀ) was a gift of Negma-Lerads (Magny-
Les-Hameaux, France). Tookadꢀ as provided is designated WST09 in
this investigation, while the chromatographically purified preparation
(silica-60 column, gradient of 1–10% methanol in chloroform) is
referred to as Pd-BPheid.
Others: Palladium 31-oxo-15-methoxycarbonylmethyl-rhodobacte-
riochlorin 131-(2-sulfoethyl) amide dipotassium salt (WST11; Stakelꢀ,
Steba Biotech, Toussus-Le-Noble, France) and 31-oxo-15-methoxy-
carbonylmethyl-rhodobacteriochlorin 131-(2-sulfoethyl)amide dipotas-
sium salt ([H2]-WST11) (35) were provided by A. Scherz (Weizmann
Institute of Science, Rehovot, Israel). Pd-[3-(1-Hydroxy)ethyl]-bacte-
riopheophorbide ([3HE]-Pd-BPheid) was synthesized by reduction of
Pd-BPheid with sodium borohydride in methanol according to
(B) Lipoprotein fractionation by iodixanol gradient ultracentrifuga-
tion (IGU): The alternatively used IGU fractionation of BP is based
on the method of Graham (47) and Graham et al. (49). Most of the
tetrapyrroles investigated (2.0 ODÆmLDE) were dissolved in 40 lL of a
mixture of CrEL and ethanol (1:1 vol ⁄ vol), which was diluted to
400 lL with normal saline (48) immediately before addition to
3.6 mL BP. Exceptions were WST11, [H2]-WST11 and HpD (2.0
ODÆmLmethanol) which were dissolved in 50 mM PB (pH 7.2). [3HE]-
WST09 was dissolved in CrEL ⁄ ethanol ⁄ normal saline (CES), DMSO
or PB for comparative measurements. All tetrapyrrole-loaded plasma
samples were incubated for 3 h under slow rotation at ambient
temperature.
the method for [3HE]-BChl (39). MS: 717.19 ([M+H]+
,
12
12
C
H39O6N4106Pd), 739.17 ([M+Na]+
,
C
H38O6N4Na106Pd)
35
35
761.16 ([M-H+Na2]+
,
C
35
H37O6N4Na2106Pd) m ⁄ z. UV–VIS in
12
diethyl ether (DE) (ꢀrel.): 326 (0.555), 377 (0.508), 514 (0.127), 659
(0.130), 714 (1.000) nm. HpD (Photosanꢀ) was from Seelab (Seehof
Laboratorium, Wesselburenerkoog, Germany).
Spectroscopy. Absorption spectra were recorded with a UV 2401 PC
Spectrometer equipped with an ISR 240-A Integrating Sphere Assem-
bly (Shimadzu, Duisburg, Germany). Spectra were processed using
Hyper UV 1.5 (Shimadzu) based on Spectacle (LabControl, Cologne,
Germany). In situ absorption spectra of iodixanol gradients were
recorded with a diode array detection (DAD) system consisting of a
After incubation, each sample was mixed with 0.8 mL OptiPrepꢀ
(60% iodixanol solution; Sigma-Aldrich). No. 344075 Quick-Sealꢀ
Ultra-Clearꢀ centrifuge tubes (5.1 mL; 0.5 · 2.0 in; Beckman) were

¨
184 Jorg Dandler et al.
filled with 0.5 mL TE buffer I (see SDGU). Four milliliters of the
sample was carefully layered under this solution, and finally 0.6 mL of
pure OptiPrepꢀ was layered at the bottom. The rest of the sample
(0.8 mL) was kept at 4ꢁC for further measurements. The sealed tubes
were centrifuged in a VTi 80 vertical-tube rotor (8 · 5.1 mL; Beck-
man) at 260 000 g for 4 h. After centrifugation, several bands were
clearly visible in the self-generated iodixanol gradients (Fig. S1).
The gradients were scanned in situ from top to bottom with a home-
made DAD scanner, which allowed the recording of absorption
spectra between 300 and 900 nm with a spatial resolution of ꢀ0.33 mm
(see Figs. S1–S3). The tubes were inserted vertically in a hole (Ø
14 mm) in a gray plastic block (L 30, W 30, H 92 mm). The top of the
tube was positioned just below two opposite horizontal holes into
which two optical fibers from the lamp and the DAD detector were fit.
Using the tube as a cuvette with 12 mm path length, it was raised at a
speed of ꢀ20 mm min)1, and a full spectrum was recorded every
second. For calculation of pigment concentrations in BP fractions, the
average Qy absorption maximum of the pigment (arithmetic mean of
the values for LDL and HDL) in each 3D densitogram was
determined. A 2D densitogram was extracted at this wavelength out
of the 3D data file, as well as a baseline at a position 100 nm to the red
of this maximum. The difference 3D densitogram was resolved into
influences the position of the Qy absorption band in the red
(chlorins) and NIR (bacteriochlorins) spectral region (52), the
substituent at C-3 that has a modulating effect, e.g. on
the redox potential (53) and optical properties (39), and the
substituent at C-13 which in most cases forms the isocyclic ring
E, but is open in the very polar taurinated WST11 pigments
(35). HpD, the oldest licensed PDT photosensitizer (16), has
been included for comparison.
Pigment distribution among blood fractions
In a first test, thrombocyte-enriched human blood (buffy
coats) was incubated with WST09. Approximately three-
quarters of the pigment were found after centrifugation in
the topmost layer, a plasma–phosphate buffer saline–mixture
(Fig. 2). The gradient fractions comprising cellular compo-
nents contained only about one-tenth of the pigment, the
remainder was located in the Ficoll layer, most probably
because the pigment molecules were swept along by the large
number of polymorphonuclear neutrophils (PMNs) and ery-
throcytes which crossed the Ficoll layer. In view of the small
amounts of pigment in the cellular fraction, all subsequent
experiments were focused on the pigment distribution among
BP components, that is, the lipoprotein fractions and the HDP
fraction that contains the serum albumin (HSA).
Gaussians by
a computer-generated fit (Origin 7.0; OriginLab,
Northampton). Each curve was assigned to the different BP fractions,
and the pigment distribution among the fractions was calculated by
correlating the curves’ areas.
Fractionation of gradient samples: The density of the collected
samples was determined by measurement of the refractive index with
an Abbe refractometer (Atago, Tokyo, Japan). The refractive indices
(RI) were converted into densities by the following formula, generated
from the best-fit line (R = 0.99983) of 11 comparison measurements
with different mixtures of OptiPrepꢀ and BP:q[mg ⁄ mL] = (RI –
1.06463) ⁄ 0.27446.
Gel filtration. Tetrapyrrole binding analyses were performed by gel
filtration with a Sephadexꢀ G-25 Medium (GE Healthcare) column in
a Pasteur pipette. The outlet of the column was connected to a flow-
through cuvette (No. 178.010; Hellma, Mullheim, Germany). To
achieve a constant flow in every run exactly 5.0 mL TE buffer III (see
SDGU) was used as the mobile phase. Absorption spectra were
recorded with the standard DAD system.
Lipoprotein analyses. Protein concentrations were determined with
the Flukaꢀ Advanced Protein Assay Reagent (Sigma-Aldrich). Phos-
pholipid concentrations were determined with the Rocheꢀ Phospho-
lipids Enzymatic colorimetric test (No. 691844; Roche, Mannheim,
Germany). Cholesterol (ester) concentrations were determined with the
Rocheꢀ Cholesterol Enzymatic colorimetric test (No. 2016630;
Roche). Triglyceride concentrations were determined with the Rocheꢀ
Triglycerides Enzymatic colorimetric test (No. 2016648; Roche). All
analyses were performed in triplicate.
Pigment distribution among BP fractions: nonpolar tetrapyrroles
Human BP was incubated with the different pigments and
subsequently fractionated by density gradient ultracentrifuga-
tion. The standard preparative fractionation method, namely
SDGU (46,47), requires long periods (5 days) and is very
demanding of material (55 mL BP, ꢀ4 lmol pigment) and is
thus unsuitable for a pigment distribution survey. It was,
therefore, replaced in most cases by IGU (47,49). This method
is much faster (1 day) and requires only very small quantities
of materials (4 mL BP; ꢀ30 nmol pigment). The IGU method
is still considerably slower than electrophoretic methods (see
Ref. [54]), but it avoids high dilution and is suitable for
micropreparative purposes, even though the fractionation is
somewhat less complete than with SDGU. For the current
analytical study, however, this was compensated by the ability
to analyze iodixanol gradients in situ with high spatial and
spectral resolution (see Figs. S1–S3), which is facilitated by the
high stability of these gradients (55): even after 18 h, there
were no significant changes detectable in the band patterns. A
further advantage is that the IGU method retains the
osmolarity of the BP, whereas increasingly hyperosmotic salt
concentrations are used in the SDGU method (47). The latter
cause dehydration of the lipoproteins, as indicated by signif-
icantly increased densities of the SDGU fractions (LDL:
1.019–1.063 g mL)1; HDL: 1.063–1.21 g mL)1) relative to the
respective IGU fractions in their native hydrated state in
iodixanol gradients (LDL: 1.008–1.026 g mL)1; HDL: 1.026–
1.12 g mL)1, see Fig. S2). The latter values, determined by
refractive index measurements, are nearly identical with those
given in the original publication of the IGU method (LDL:
1.010–1.030 g mL)1; HDL: 1.030–1.14 g mL)1) (49).
Transfer between lipoprotein fractions. Lipoprotein suspensions with
the same density as BP (q ꢁ 1.028 g mL)1) were generated by mixing
of the following solutions obtained by SDGU (see above): 700 lL
VLDL fraction, 700 lL dialyzed LDL fraction, 700 lL dialyzed HDL
fraction, 1525 lL TE buffer I. In each experiment, only one of the
lipoprotein fractions (LDL or HDL) was loaded with exogenous
tetrapyrroles. The mixtures (3625 lL) were incubated for 3 or 24 h at
ambient temperature, and analyzed by IGU (see above) after addition
of 725 lL OptiPrepꢀ solution.
RESULTS AND DISCUSSION
Pigments
The present study was aimed at investigating the influence of
structural factors of Chl and BChl derivatives on their
distribution among blood components. The following struc-
tural elements were varied (see Fig. 1 for structures and
abbreviations): the central metal that governs excited state
dynamics and ligand coordination (45,50,51), the presence or
absence of the phytyl residue at C-173 that profoundly affects
the polarity, the oxidation state of the macrocycle that
The separation was monitored by optical spectroscopy.
Very reproducible 3D densitograms were obtained by scanning
of the cylindrical part of the iodixanol gradients (central

Photochemistry and Photobiology, 2010, 86 185
Pigment abbreviation
HpD a
Chl
[3Ac]-Chl
Zn-Phe
Pd-Phe
Phe
Chlid
Zn-Pheid
Pd-Pheid
Pheid
Formula
M
-
Mg
Mg
Zn
Pd
2 H
Mg
Zn
Pd
2 H
Mg
Mg
Zn
Pd
2 H
Mg
Zn
Pd
Pd
2 H
Pd
2 H
R1
-
vinyl
acetyl
vinyl
vinyl
vinyl
vinyl
vinyl
vinyl
R2
-
[nm]
626*
659
675
653
634
667
660
652
635
667
769
744
763
755
749
770
763
1
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
Phy
Phy
Phy
Phy
Phy
H
H
H
H
Phy
Phy
Phy
Phy
Phy
H
H
H
H
H
vinyl
acetyl
vinyl
BChl
[3Vi]-BChl
Zn-BPhe
Pd-BPhe
BPhe
acetyl
acetyl
acetyl
acetyl
acetyl
acetyl
1-hydroxyethyl
acetyl
-
BChlid
Zn-BPheid
Pd-BPheid (WST09)
[3HE]-Pd-BPheid
BPheid
WST11
[H2]-WST11
755 (754)
714
749
-
-
747*
749*
-
Figure 1. Molecular structures and abbreviations of pigments. Numbering is according to IUPAC-IUB (96). The recommended but cumbersome
omission-resubstitution nomenclature is replaced by placing the modified group in square brackets; thus, for example 3-devinyl-3-acetyl-Chl
becomes [3Ac]-Chl. kmax refers to the Qy absorption maximum in diethyl ether (* in methanol); further spectral information of all pigments is given
in Table S1. Pigments with an isocyclic ring (formulas 2 and 3) are generally present as 132-epimer mixtures; but only the 132R-epimers are shown,
which are usually present in large excess over the 132S-epimers. aHpD is a mixture of di- and oligomeric compounds (17); only the basic porphyrin
structure is shown.
40 mm). In the two hemispherical areas at the top and bottom
of the tubes, quantification was less reliable and only quali-
tative absorption spectra were obtained, due to the increased
scattering of the walls and the shortening of the optical
pathway. The pigment concentrations in VLDL fractions,
which were located completely in the upper hemisphere could,
therefore, not be measured satisfactorily in situ. The estimation
of the pigment concentrations in the HDP fractions was
further complicated by the presence of an overwhelming
amount of heme in these fractions.
preferential accumulation in LDL and HDL emphasizes their
relevance as transport vehicles for lipophilic to moderately
hydrophilic photosensitizers (like WST09) in the vascular
The results obtained by IGU for all pigments with an
intact isocyclic ring E are shown in Fig. 3. In average BP, the
combined pigment content in the VLDL and HDP fractions
of each of these pigments was £10% as determined by the
SDGU method; therefore, results are only given for the LDL
and HDL fractions. In extremely high-fat BP charges, the
pigment content in the VLDL fraction amounted to >40%.
However, the relative pigment distribution among the respec-
tive LDL, HDL and HDP fractions remained unaltered. The
Figure 2. Distribution of WST09 among human blood fractions.
Buffy coats were fractionated by underlayering with Ficollꢀ-Paque
Plus and subsequent centrifugation. PBMC = peripheral blood
mononuclear cells.

¨
186 Jorg Dandler et al.
Such a model is supported by the location of bacteriochlorin
a, which possesses three negative charges at rings C and D,
near the membrane surface of liposomes (59). It also explains
the similar pigment distribution pattern of all tested tetra-
pyrroles without a phytyl residue, that are all Chlid and
BChlid derivatives (with the exception of the Zn-complexes,
see Specific interaction of Zn-complexes with HDL).
Specific interaction of Zn-complexes with HDL
The (relative) enrichment of Zn-complexes in the HDL
fraction (see Pigment distribution among BP fractions:
nonpolar tetrapyrroles), compared to other metal complexes
and the free bases, was accompanied by a redshift of the Qy
absorption maximum, relative to its position in LDL (see
Fig. S4). This effect was least pronounced with Zn-BPhe,
which was also the Zn-complex that showed the lowest
accumulation in the HDL fraction. In the Pd-derivatives, the
respective spectral shifts were very weak, and of variable
direction, and the shifts were negligible for the metal-free
derivatives. Intermediate redshifts were observed for the
Mg-derivatives, the least pronounced again with the ‘‘bacte-
riopheophytin derivative,’’ BChl. By analogy with the respec-
tive Zn-complex, the latter showed the lowest enrichment in
the HDL fraction among all Mg-complexes. The extent of all
observed redshifts, therefore, seems to be directly related to the
pigment distribution. This preferential binding prompted us
to further subfractionate the HDL, using again Zn- and
Pd-BPheid as test pigments. HDL2 and HDL3 subfractions
were obtained by SDGU. The ratio of the pigment contents in
Figure 3. Pigment distribution of selected tetrapyrroles between LDL
and HDL fractions. BP was incubated with selected pigments and
fractionated by IGU. The pigment contents of the respective LDL and
HDL bands in the formed iodixanol gradients were analyzed spectro-
scopically by in situ scanning (see Materials and Methods, BP
fractionations and Supporting Information for details). Mean-
s
SEM, n = 2–3.
system. Similar results have been reported for a series of
porphyrinic PDT sensitizers with up to 8 carboxyl groups
(56–58).
The distribution of pigments between LDL and HDL
fractions of human BP shows some clear patterns. The phytyl
residue resulted in a preferential pigment accumulation in the
LDL fraction; this effect was much more pronounced in BChl
compared with Chl derivatives. The two extremes were the Zn-
Phe ⁄ Zn-Pheid pair that showed only a minor difference and
the BPhe ⁄ BPheid pair where the LDL ⁄ HDL ratios were nearly
reversed (21:1 for BPhe vs 1:2 for BPheid). By contrast, no
significant difference was detectable between the Chlid and
BChlid derivatives that have a free propionic acid residue at C-
17. There was also a distinct effect of the central metal: the Zn-
derivatives accumulated more in the HDL fraction than their
respective Mg-, Pd- and H2-analogs. The latter three pigments
exhibited in most cases a similar pigment distribution between
HDL and LDL, with the notable exception of BPhe, which
was almost completely located in the LDL fraction (see
above). The in situ scanning showed that there was no pigment
aggregation in any of these fractions, as judged by the absence
of redshifted absorption bands (see Fig. S3).
The pigments esterified with phytol are probably anchored
by this long hydrocarbon chain to the lipoprotein surface. This
might be facilitated by the less charged membrane surface
and ⁄ or the higher cholesterol content of the LDL compared to
the HDL membrane, thereby rationalizing the higher accu-
mulation of these esterified pigments in LDL. Due to steric or
electronic effects, the BChl derivatives may be anchored more
deeply into the membrane than the respective Chl derivatives
(the C-3 residue contributes only little to pigment binding, see
Influence of structural details: Chl vs BChl).
HDL3 vs HDL2 was 1.13
0.01 (normalized to the same total
HDL2 and HDL3 lipoprotein mass), irrespective of the applied
pigment. The specific binding of Zn-BPheid must therefore
occur in both subfractions at comparable levels. The bound
pigment content in these two HDL subfractions seems to be
directly correlated with the total lipoprotein surface: The
surface-to-mass ratio of an average HDL3 particle (B: 8 nm;
m: 200 kDa [60]) is 1.15 times higher than that of an average
HDL2 particle (B: 10 nm; m: 360 kDa [60]).
Further support for a specific binding of Zn-complexes to
HDL comes also from a comparison of the pigment–lipopro-
tein ratios of Zn- and Pd-BPheid. Molar lipoprotein concen-
trations in HDL and LDL were estimated by summing up the
protein, phospholipid, total cholesterol and triglyceride deter-
minations, while molar pigment concentrations were measured
by their absorptions using an integrating sphere. The precision
of the method is limited by the need to estimate the average
lipoprotein particle mass, and by the assumption of the same
extinction coefficients of the pigments in the lipoproteins as in
CES (Table 1). According to this analysis, HDL binds
approximately three times more Zn-BPheid than Pd-BPheid.
By contrast, the number of tetrapyrrole molecules bound to
individual LDL particles was almost identical for Zn- and
Pd-BPheid. The coordination ability of the central metal
therefore only influences HDL binding, while LDL can serve
as a control.
At the near-neutral pH of the experiments, the noneste-
rified pigments carry a negatively charged carboxylate group
at C-173, which probably imposes a different orientation of
the pigment in which the now less polar macrocycle (relative
to the C-17 substituent) penetrates into the membrane, and
the carboxylate interacts with the positively charged choline
head groups of the phosphatidylcholins and sphingomyelins.
This differential behavior of pigments with Zn2+ and Mg2+
as central metal parallels observations of Szczygiel et al. (61)
on the tissue distribution of Chlid and its Zn-derivative,
Zn-Pheid, where the Zn-complex showed slower uptake and
loss from mice after interperitoneal injection. While aggrega-

Photochemistry and Photobiology, 2010, 86 187
Table 1. Molar ratios of bound pigment per lipoprotein particle.
Pd- or Zn-BPheid were added to BP before SDGU fractionation.
Extinction coefficients were determined in CES relative to the known
ones in DE (45). The average lipoprotein particle masses used for LDL
and HDL were 2500 and 250 kDa, respectively (60).
environment in the latter. The coordination state also influ-
enced the stability of pigment binding as investigated by gel
filtration experiments. LDL samples lost more pigment (3–6%)
than HDL samples (1.5–2.5%), and the pigment loss was
always less for Zn-BPheid than for Pd-BPheid. The Qx
absorption band of the Pd-derivatives was buried under the
red-most carotenoid band and its position could not be
accurately determined; however, Pd usually does not bind any
extra ligands (67), which in tetrapyrroles resembles the binding
situation of the free bases.
Pigment per lipoprotein
particle (mol mol)1
)
)1
)1
ꢀ (M cm)1
in DE
)
ꢀ (M cm)1
)
in CES
Pigment
LDL
HDL
Pd-BPheid
Zn-BPheid
92 000
67 700
56 000
46 000
3.57
3.43
0.07 0.65
0.07 1.84
0.01
0.02
The differential effects of Zn- compared to Mg-complexes
may relate to the fact that Mg binds extra ligands very strongly
(68); it may therefore carry a water molecule irrespective of the
environment. There are, by contrast, "naked" Zn-tetrapyrroles
in which the central metal lacks extra ligands; they are,
therefore, more responsive to environment changes. A specific
binding of the Zn-complexes may also explain the observation
that the Mg-derivatives showed nearly identical distributions
as the respective Pd- and H2-derivatives, in spite of their
different coordination state. This is reminiscent of the binding
of (bacterio)chlorophylls in photosynthetic pigment proteins,
where binding is rescued in mutants where enough space is
provided, in a hydrophobic pocket, to bring along a small
ligand (52). This difference between the Zn- and
Mg-derivatives is further support of a special binding situation
in the HDL fraction that favors Zn over Mg as the central
metal, which may be related to their difference in hardness
(69). Apolipoproteins might be binding partners that control
this selectivity for HDL. Further support for such a special
binding was the occurrence of a Krasnovskii reaction, a
photoreduction requiring an amino group in the immediate
vicinity of the tetrapyrrole (70). This study will be published
separately together with the photochemistry of the sensitizers
in the different BP fractions.
Pd-BPheid = Pd-bacteriopheophorbide;
Zn-BPheid = Zn-bacteri-
opheophorbide; BP = blood plasma; SDGU = sequential density
gradient ultracentrifugation; CES = CrEL ⁄ ethanol ⁄ normal saline;
DE = diethylether; LDL = low-density lipoprotein; HDL = high-
density lipoprotein.
tion, active excretion and degradation are involved in the
whole animal, these processes can be excluded in the BP
fractions. This, therefore, points to a distinct binding differ-
ence among the two central metals.
Coordination of central metal
In the bacteriochlorins, additional information on the binding
situation can be obtained by an analysis of the Qx absorption
which is very sensitive to the electron densities of their central
metals, and thereby to their coordination state (62). Studies in
solution (45,63,64), with synthetic (65) and natural proteins
(66) indicate characteristic regions for four-, five- and six-fold
coordinated central metals that, in spite of additional influ-
ences (e.g. solvent polarity), do not overlap. According to this
classification, the Mg- and Zn-derivatives had five-fold coor-
dinated central metals in all lipoproteins (Fig. 4). The exact
position within the range defining five-fold coordination allows
furthermore a rough estimate of the average polarity prevail-
ing in the pigment environment. The position of the Qx
absorption band was nearly identical in HDL and LDL
fractions for both Mg-complexes while, for both Zn-com-
plexes, it was significantly redshifted in HDL fractions
compared to LDL fractions, thus indicating a less polar
Influence of structural details: Chl vs BChl
There was a remarkable difference in the pigment distribution
between LDL and HDL fractions for the Chl ⁄ BChl pair (see
Fig. 3). These two pigments differ in two details of their
chemical structure, viz. the oxidation state of the macrocycle
and the substituent at C-3, which is a vinyl group in Chl and an
acetyl group in BChl. Therefore [3Ac]-Chl and [3Vi]-BChl were
tested as well, in which the substituents at C-3 are reversed.
This reverses the polarities, too, if judged by their chromato-
graphic mobilities, because the acetyl group is considerably
more polar than the vinyl group (38), while the oxidation state
has only little influence on the polarity (71).
Surprisingly, the pigment distribution was only slightly
influenced by the side chain at C-3, but much more so by the
oxidation state of the macrocycle (Fig. 5). Irrespective of the
C-3 residue, the presence of a chlorin macrocycle more than
tripled the pigment content of the HDL fraction relative to a
bacteriochlorin macrocycle, while the presence of a 3-acetyl
group led only to a slight increase in the pigment content in
the HDL fraction relative to a 3-vinyl group. These findings
support the above-mentioned assumption (see Pigment
distribution among BP fractions: nonpolar tetrapyrroles) that
the bacteriochlorin derivatives are more deeply anchored into
the lipid membrane than the respective chlorin derivatives.
The reason for this distinction is presently unclear. During
Figure 4. Positions of the Qx absorption maxima of Mg- and
Zn-bacteriochlorins in HDL and LDL fractions. Values in diethyl
ether and pyridine are given for comparison. The horizontal lines
indicate the ranges for the different coordination states of the central
metal (nc = 4, 5 or 6), corresponding to 0, 1 or 2 extra ligands.

¨
188 Jorg Dandler et al.
Figure 6. Influence of pigment insertion system (CES, DMSO and PB)
on the distribution of [3HE]-Pd-BPheid among LDL, HDL and HDP
fractions. BP was incubated with [3HE]-Pd-BPheid dissolved in either
CES, DMSO or PB and fractionated by IGU. The pigment contents of
the respective LDL, HDL and HDP bands in the formed iodixa-
nol gradients were analyzed spectroscopically by in situ scanning.
Figure 5. Pigment distribution of Chl, [3Ac]-Chl, BChl and [3Vi]-BChl
between LDL and HDL fractions. BP was incubated with selected
pigments and fractionated by IGU. The pigment contents of the
respective LDL and HDL bands in the formed iodixanol gradients
were analyzed spectroscopically by in situ scanning. Means
n = 2–3.
SEM,
Means
SEM, n = 2–3.
chromatography on reverse phase material, bacteriochlorins
moved consistently slower than chlorins with otherwise
identical structure, indicating a less polar structure of the
former, which may relate to changes in the conjugated
p-system (Fig. 1). However, steric factors and a differential
flexibility at ring B may also contribute. This is another
lower than when it was added in CES, while the contents in
all other fractions increased slightly (VLDL: +1.5%; HDL:
+4.5%; HDP: +4.5%). Comparable results have also been
reported for tin ethyl etiopurpurin (SnET2); it accumulated
preferentially in the LDL fraction if added in CrEL relative
to its application in ethanol, DMSO, cyclodextrins or
liposomes (74–77). The pigment distribution among the two
lipoprotein fractions, LDL and HDL, was almost identical
for [3HE]-Pd-BPheid (HDL: ꢀ66%) and Pd-BPheid (HDL:
ꢀ64%), indicating again the small impact of the C-3
substituent on the distribution patterns.
This preferential pigment accumulation in LDL has been
attributed to a preferential fusion of CrEL with an LDL
subfraction (78). Alternatively, a buoyant density shift in HDL
particles is induced by CrEL, which would then mimic an LDL
subfraction (79); this is supported by a slightly increased
carotenoid content in the LDL fraction of BP incubated with
CES (data not shown). Further support for a lipoprotein
reorganization came from a small, but distinct density shift in
the maximum absorption of the pigment in the LDL fraction
by ꢀ0.01 g mL)1 (corresponding to ꢀ1 mm in the iodixanol
gradient), relative to the maximum of the carotenoid absorp-
tion (Fig. 7); the effect was absent if the pigment was added in
DMSO or PB. In the absorption spectra at these two positions,
the intensity ratio of the BChl and the carotenoid main bands
was changed accordingly. This effect was particularly apparent
for all pigments, like BChl, that partition preferentially in the
LDL fraction (see Fig. 3). It was detectable in all BP charges
used in this study, possibly because individual LDL subfrac-
tion patterns, which can vary widely, tend to equilibrate within
a short time (80). These data indicate that the (B)Chl
derivatives and carotenoids partition differently among the
subfractions in the presence of CrEL. The CrEL-based CES
was used in this work as the standard application medium. It
has been developed as an alternative to DMSO that is
principally an excellent solvent but has multiple adverse effects
(81,82), yet CrEL should not be seen as an inert solubilizer
(83).
indication for
a specific interaction of pigments with
lipoproteins that relies on structural details rather than on
gross changes in their lipophilicity. The interaction of
pigments with the lipid surface of lipoproteins is often
treated as a phase distribution between an aqueous and a
lipid phase (72). Pigment lipophilicity correlates, however,
only for structural homologs with their measured binding
constants on membranes or liposomes; in other cases the
correlation is poor (73).
Influence of the solubilizer
As all the aforementioned pigments are only poorly soluble in
aqueous media, they have to be dissolved in organic solvents
or detergents for intravenous injection. To analyze the effect
of the application method on the pigment distribution,
Pd-[3-(1-hydroxy)ethyl]-bacteriopheophorbide
([3HE]-Pd-
BPheid) was synthesized that has intermediate solubility in
aqueous media between WST09 and the very hydrophilic
WST11. Judged from the position and shape of the Qy
absorption band (not shown), [3HE]-Pd-BPheid was mono-
meric in aqueous solutions at pH ꢀ8, in contrast to its
precursor WST09. When [3HE]-Pd-BPheid was applied in
CES (a mixture of CrEL, ethanol and normal saline, see
Materials and Methods), ꢀ50% was found in the HDL
fraction, and ꢀ25% each in the LDL and HDP fractions
(Fig. 6). If, however, [3HE]-Pd-BPheid was added in DMSO
or PB, its content in the HDL and HDP fractions was
increased to ꢀ55% and ꢀ35%, respectively, while only
ꢀ10% remained in the LDL fraction. The application system
also affected the partitioning of WST09: when applied in
DMSO, the pigment content in the LDL fraction was 10.5%

Photochemistry and Photobiology, 2010, 86 189
Distribution of polar bacteriochlorin derivatives among BP
fractions
Recently, two extraordinarily hydrophilic bacteriochlorin
derivatives have become accessible: the taurinated Pd-complex
WST11 and its metal-free analog, [H2]-WST11. Their distri-
bution among BP fractions was compared to that of the much
less polar WST09, and to the classical PDT photosensitizer
HpD (in the form of Photosanꢀ). Approximately 50% of HpD
was found in the HDP fraction, and over 80% and 90% of
WST11 and [H2]-WST11, respectively, with serum albumin
being the most likely carrier (Fig. 8). In contrast only ꢀ15% of
WST09 were located in this fraction (see above).
As in the case of [3HE]-Pd-BPheid (see Influence of the
solubilizer), the pigment concentration of the three polar
pigments in the HDP fraction was high enough for a reliable
determination by IGU; three of the above-mentioned pigments
(WST09, WST11, HpD) were, therefore, used to compare the
IGU with the classical SDGU for BP fractionation. When
using the SDGU method, reduced pigment content in the HDP
fraction was consistently observed relative to the IGU method.
This effect was most pronounced for WST11, where the
pigment content in the HDP fraction was halved, in favor of
the LDL and HDL fractions. This effect, which appears to
increase with increasing hydrophilicity (water solubility) of the
pigment, may be due to a less preferred pigment binding to
lipoproteins. This is supported by the distribution pattern of
[3HE]-Pd-BPheid, which is slightly more water soluble than
WST09. If applied as a CES solution, the pigment content in
HDP was ꢀ9% higher than for WST09, but still ꢀ18% less
than for HpD if dissolved in PB. The correlation of increasing
pigment polarities with increasing pigment concentrations in
the respective HDP fraction has already been reported for
other pigments (58,88). Preferential binding of lipophilic
photosensitizers to plasma lipoproteins and of hydrophilic
photosensitizers to serum albumin was shown for a series of
sulfonated derivatives of tetraphenylporphine (89).
Figure 7. Spectroscopic indication for lipoprotein reorganization by
Cremophorꢀ EL. BP was incubated with BChl and fractionated by
IGU. Densitograms (a) of the LDL band at the Qy absorption
maximum of BChl (775 nm) and at the carotenoid absorption
maximum (463 nm) obtained by scanning of the tube along the axis,
and absorption spectra (b) at the respective densitogram maxima (9.0
and 10.0 mm, respectively).
Interestingly, the ratios of all tested polar (see Fig. 8) and
nonpolar (see Fig. S5) tetrapyrroles in the LDL and HDL
Influence of other parameters
A significantly higher concentration of WST09 in HDL has been
reported for fetal calf serum (35) than was observed here for
human plasma, indicating that the pigment distribution in blood
serum is species dependent. Similar results have been reported
for other sensitizers (84) which relate to a different chemical
composition of the lipoproteins. It should be noted, however,
that such comparisons require otherwise identical conditions. It
has been shown, for example, that the pH (85) or concentrations
of pigment and solvent applied are critical (86). These factors
were kept constant throughout this study to obtain internal
consistency. We furthermore verified, using WST09, that the
partitioning among the plasma fractions was not affected within
an eight-fold concentration range (quarter to double standard).
These variations are relatively small, however, when compared
with the more than 10-fold higher concentrations used by
Brandis et al. (35). Aggregation at higher concentrations may
contribute to rapid excretion of the pigment in PDT (A. Scherz,
private communication, 2009) (87).
Figure 8. Influence of fractionation method on the distribution of
WST09, WST11 and HpD among LDL, HDL and HDP fractions. BP
was incubated with selected pigments and fractionated by IGU. The
pigment contents of the respective LDL, HDL and HDP bands in the
iodixanol gradients were analyzed spectroscopically by in situ scan-
ning. [H2]-WST11 was fractionated by IGU only. IGU = plain
columns; SDGU = hatched columns.

¨
190 Jorg Dandler et al.
fractions were identical within the error limits for both
applied fractionation methods. Both methods separate BP
components according to their buoyant densities, but while
IGU conditions are isotonic with BP, those of SDGU are
highly hypertonic. Similar fractionation methods utilizing
high-salt solutions did not alter the pigment distribution
among lipoproteins (90). This suggests that the pigment
affinities of some HDP components (probably serum albu-
min) are reduced at the high salt concentrations used in
SDGU, while those of the different lipoproteins are either
unaltered or at least modified in a similar way compared to
isotonic conditions.
(Fig. 9). The partitioning was identical to that observed after
incubation of the original BP with the respective pigment,
and no significant difference could be measured whether the
LDL or HDL fraction was pigment-loaded prior to incuba-
tion. Even after incubation for 3 h (plus ꢀ15 min for sample
preparation), the pigments provided originally in the HDL
fractions were entirely equilibrated. The reverse pigment
transfer from LDL to HDL was not fully completed at the
end of this period; however, in the case of Pd- and Zn-
BPheid ꢀ94% and ꢀ97%, respectively, of the transfer had
already taken place. This gives for both pigments an upper
limit for the kinetic constants for equilibration of less than
1 h. Fast exchange has also been reported for tin etio-
purpurin (93) and zinc phthalocyanine (94). On the other
hand, only moderately polar carotenoids like lutein exchange
readily between human VLDL and HDL, while nonpolar
ones like carotenes or lycopene gave no measurable inter-
lipoprotein transfer (95).
Transfer between lipoprotein fractions
In our experience, the exchange of chlorophylls between
vesicular or micellar compartments can be rather slow. It was
of interest, therefore, to study the exchange kinetics of
pigments among the BP fractions. The experiments were
carried out with two pigments. One was Pd-BPheid, which is
currently scrutinized for vascular targeted phototherapy with
very short intervals between injection and irradiation, or
between multiple irradiations, which may cause depletion in
individual compartments (91,92). The other was Zn-BPheid,
which partitioned differently among LDL and HDL fractions
than the Mg-, Pd- and H2-derivatives (see Pigment distribution
among BP fractions: nonpolar tetrapyrroles and Specific
interaction of Zn-complexes with HDL). Pigment-loaded
LDL or HDL fractions were prepared by SDGU, and mixed
with the complementary empty lipoprotein complement. The
mixtures were incubated for 3 or 24 h at ambient temperature,
and then separated by IGU and analyzed for their pigment
contents.
CONCLUSIONS
Iodixanol-gradient ultracentrifugation is, in combination
with in situ scanning, a versatile and comparably rapid
method for investigating the distribution of photosensitizers
among the different BP fractions. Using standardized
conditions and Cremophorꢀ EL as solubilizing medium,
the structural effects among the 21 (bacterio)chlorophyllous
sensitizers can be summarized as follows: (1) Only the most
polar derivatives that are readily soluble in water accumulate
to a significant degree in the HDP fraction. (2) Decreasing
polarity results in an increasing accumulation in the LDL
relative to the HDL fraction. However, specific binding has
been observed for Zn2+-complexes, and when comparing
Chl and BChl derivatives with identical substituents. (3) The
latter difference is absent in derivatives with a free propionic
acid at C-17, Chlids and BChlids. (4) The speed of the
fractionation allowed for estimating an upper limit for
the equilibration periods among HDL and LDL fractions in
the range of 1 h. Specific binding will therefore probably not
affect drug delivery and action on longer time scales. Faster
analytical methods (54) can possibly reduce this time scale,
which may be relevant, in particular in vascular targeted
PDT using very short accumulation and irradiation times.
(5) Aggregation of pigments has, to our surprise, never been
observed in BP fractions at the concentrations used in this
study. Concerns that the photophysics in situ may be
influenced by aggregations therefore seem irrelevant. (6)
Dark oxidation, which occurs readily in solution, seems to
be impeded in lipoproteins.
Both of the tetrapyrroles tested were fully equilibrated
between the respective LDL and HDL fractions within 24 h
The results should be helpful in two ways for the rational
design of photosensitizers for accumulation in the different
blood components. One aspect is that the highly polar WST11
is set apart from all other (B)Chl-derived pigments, including
the moderately polar [3HE]-Pd-BPheid, as the only pigment
that is not mainly carried by the lipoprotein fractions of
human BP. The damage to these fractions, as part of the
photodynamic action, or as an unwanted side-reaction, will,
therefore, also be different. The second aspect is that although
the two main lipoproteins, LDL and HDL, discriminate
among most pigment derivatives tested only on the basis of
their bulk polarity, the photosensitizer distribution can be
Figure 9. Redistribution of Pd- and Zn-BPheid between LDL and
HDL fractions. BP: pigment distribution in the same BP batch,
determined by standard IGU (see Fig. 3). LDL and HDL indicate the
pigment-loaded BP fraction at the onset of the experiment (Pd:
Pd-BPheid; Zn: Zn-BPheid). The fractions were incubated for either 3
or 24 h. Arrows indicate the increase or decrease in the respective
pigment concentration, relative to the pigment concentrations at the
onset of the experiment (see Materials and Methods, Transfer between
lipoprotein fractions for details).

Photochemistry and Photobiology, 2010, 86 191
influenced by certain structural features, in particular the
central metal. Both aspects should be considered in fine-tuning
PDT, and warrant further studies to correlate the pigment
distributions in human BP with intracellular localization and
photosensitizing efficacy.
REFERENCES
1. Ackroyd, R., C. Kelty, N. Brown and M. Reed (2001) The history
of photodetection and photodynamic therapy. Photochem. Pho-
tobiol. 74, 656–669.
2. Moan, J. and Q. Peng (2003) An outline of the history of PDT. In
Comprehensive Series in Photochemistry and Photobiology, 2:
Photodynamic Therapy (Edited by T. Patrice), pp. 3–17. Royal
Society of Chemistry, Cambridge, UK.
3. Kessel, D. (2004) Delivery of photosensitizing agents. Adv. Drug
Deliv. Rev. 56, 7–8.
4. Solban, N., I. Rizvi and T. Hasan (2006) Targeted photodynamic
therapy. Lasers Surg. Med. 38, 522–531.
Acknowledgements—This work was supported by the Hans-Fischer-
Gesellschaft e. V. Munchen (Munich, Germany), by a PhD grant to
J.D. We are indebted to A. Scherz (Weizmann Institute of Science,
Rehovot) for providing WST11, and to R.J. Porra (CSIRO, Canberra)
for help in preparing this manuscript. WST09 was provided by Steba
N.V. through collaboration with Negma-Lerads.
5. Juzeniene, A., Q. Peng and J. Moan (2007) Milestones in the
development of photodynamic therapy and fluorescence diagnosis.
Photochem. Photobiol. Sci. 6, 1234–1245.
6. Hamblin, M. R. and E. L. Newman (1994) On the mechanism of
the tumour-localising effect in photodynamic therapy. J. Photo-
chem. Photobiol. B, Biol. 23, 3–8.
7. MacDonald, I. J. and T. J. Dougherty (2001) Basic principles of
photodynamic therapy. J. Porphyr. Phthalocyan. 5, 105–129.
8. Wilson, B. C. and M. S. Patterson (2008) The physics, biophysics
and technology of photodynamic therapy. Phys. Med. Biol. 53,
R61–R109.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online
version of this article:
Figure S1. Home-made DAD scanner (schematic) for
iodixanol gradients of tetrapyrrole-loaded human blood
plasma.
9. Rosenbach-Belkin, V., L. Chen, L. Fiedor, I. Tregub, F. Pavlot-
sky, V. Brumfeld, Y. Salomon and A. Scherz (1996) Serine con-
jugates of chlorophyll and bacteriochlorophyll: Photocytotoxicity
in vitro and tissue distribution in mice bearing melanoma tumors.
Photochem. Photobiol. 64, 174–181.
10. Spikes, J. D. and J. C. Bommer (1991) Chlorophyll and related pig-
ments as photosensitizers in biology and medicine. In Chlorophylls
(Edited by H. Scheer), pp. 1181–1204. CRC Press, Boca Raton, FL.
11. Chen, Y., G. Li and R. K. Pandey (2004) Synthesis of bacterio-
chlorins and their potential utility in photodynamic therapy
(PDT). Curr. Org. Chem. 8, 1105–1134.
12. Brandis, A., Y. Salomon and A. Scherz (2006) Bacteriochlorophyll
sensitizers in photodynamic therapy. In Chlorophylls and Bacte-
riochlorophylls: Biochemistry, Biophysics, Functions and Applica-
tions (Edited by B. Grimm, R. Porra, W. Rudiger and H. Scheer),
pp. 485–494. Springer, Dordrecht.
13. Brandis, A., Y. Salomon and A. Scherz (2006) Chlorophyll sen-
sitizers in photodynamic therapy. In Chlorophylls and Bacterio-
chlorophylls: Biochemistry, Biophysics, Functions and Applications
(Edited by B. Grimm, R. Porra, W. Rudiger and H. Scheer),
pp. 461–483. Springer, Dordrecht.
Figure S2. Iodixanol gradient fractionation of blood plasma
loaded with Pd-BPheid. Photograph (left) and schematic band
pattern (center, colored according to the isolated fractions)
after centrifugation (see Materials and Methods). The colors of
the faint pink HDL fraction and the milky white VLDL
fraction are not adequately reproduced by the digital camera.
The density distribution along the vertical axis (right) before
and after centrifugation, has been obtained from the refractive
index. For refractive index measurements the tube content
(5.1 mL) was portioned into 51 fractions of 100 lL. A linear
calibration curve for conversion of refractive indices into
densities was created separately.
Figure S3. Iodixanol gradient fractionation of blood plasma
loaded with Pd-BPheid. Color-coded spectral profile (top)
along the vertical axis of the tube shown in Fig. S2 (top of
tube = 0 mm). Pigment distribution (bottom left) along the
vertical axis obtained as the difference of the absorption
profiles at 757 nm (Qy absorption maximum of Pd-BPheid)
and 857 nm (scattering), and its Gaussian deconvolution
(bottom right). Ery = erythrocytes.
14. Moser, J. G. (1998) Photodynamic Tumor Therapy: 2nd and 3rd
Generation Photosensitizers. OPA, Amsterdam.
15. Pandey, R. K., S. Constantine, T. Tsuchida, G. Zheng, C. J.
Medforth, M. Aoudia, A. N. Kozyrev, M. A. J. Rodgers, H. Kato,
K. M. Smith and T. J. Dougherty (1997) Synthesis, photophysical
properties, in vivo photosensitizing efficacy, and human serum
albumin binding properties of some novel bacteriochlorins.
J. Med. Chem. 40, 2770–2779.
16. Dougherty, T. J., C. J. Gomer, B. W. Henderson, G. Jori, M.
Kessel, J. Korbelik, J. Moan and Q. Peng (1998) Photodynamic
therapy. J. Natl Cancer Inst. 90, 889–905.
17. Bonnett, R. (2000) Chemical Aspects of Photodynamic Therapy.
Gordon and Breach, Amsterdam.
18. Anderson, R. R. and J. A. Parrish (1981) The optics of human
skin. J. Invest. Dermatol. 77, 13–19.
Figure S4. Wavelength shifts of the Qy absorption maxima
between LDL and HDL fractions. Positive values correspond
to a redshift in the HDL vs the LDL fraction. k-scatter (not
shown) <1 nm (n = 2–3).
Figure S5. Pigment distribution of BChl, Pd-BPheid
(WST09) and Zn-BPheid between the LDL and HDL fractions
depending on the fractionation method. Selected pigments are
representative of strong, intermediate and weak accumulation
in the LDL fraction. Means
SEM; IGU (all pigments):
n = 2, SDGU (BChl): n = 2, SDGU (Pd-BPheid): n = 7,
SDGU (Zn-BPheid): n = 4.
19. Jonker, J. W., S. Musters, M. L. Vlaming, T. Plosch, K. E.
Gooijert, M. J. Hillebrand, H. Rosing, J. H. Beijnen, H. J.
Verkade and A. H. Schinkel (2007) Breast cancer resistance pro-
tein (Bcrp1 ⁄ Abcg2) is expressed in the harderian gland and
mediates transport of conjugated protoporphyrin IX. Am. J.
Physiol. Cell Physiol. 292, C2204–C2212.
20. Jonker, J. W., M. Buitelaar, E. Wagenaar, M. A. Van der Valk,
G. L. Scheffer, R. J. Scheper, T. Plosch, F. Kuipers, R. P. J. O.
Elferink, H. Rosing, J. H. Beijnen and A. H. Schinkel (2002) The
breast cancer resistance protein protects against a major chloro-
phyll-derived dietary phototoxin and protoporphyria. Proc. Natl
Acad. Sci. USA 99, 15649–15654.
Table S1. Spectral properties of used tetrapyrroles in diethyl
ether (* in methanol). Maxima of main absorption bands (nm)
and (in brackets) their absorptions normalized to the intensity
of the (0-0) transition of the respective Qy absorption band (for
HpD to the intensity of the Soret band).
Please note: Wiley-Blackwell are not responsible for the
content or functionality of any supporting materials supplied
by the authors. Any queries (other than missing material)
should be directed to the corresponding author for the
article.
21. Bonnett, R. and G. Martinez (2001) Photobleaching of sensitizers
used in photodynamic therapy. Tetrahedron 57, 9513–9547.

¨
192 Jorg Dandler et al.
22. Limantara, L., P. Koehler, B. Wilhelm, R. J. Porra and H. Scheer
(2006) Photostability of bacteriochlorophyll a and derivatives.
Photochem. Photobiol. 82, 770–780.
23. Fiedor, J., L. Fiedor, N. Kammhuber, A. Scherz and H. Scheer
(2002) Photodynamics of the bacteriochlorophyll-carotenoid sys-
tem. 2. Influence of central metal, solvent and b-carotene on
photobleaching of bacteriochlorophyll derivatives. Photochem.
Photobiol. 76, 145–152.
24. Krautler, B. and S. Hortensteiner (2006) Chlorophyll catabolites
and the biochemistry of chlorophyll breakdown. In Chlorophylls
and Bacteriochlorophylls: Biochemistry, Biophysics, Functions and
Applications (Edited by B. Grimm, R. Porra, W. Rudiger and
H. Scheer), pp. 237–260. Springer, Dordrecht.
41. Benedetti, C. E. and P. Arruda (2002) Altering the expression of
the chlorophyllase gene ATHCOR1 in transgenic Arabidopsis
caused changes in the chlorophyll-to-chlorophyllide ratio. Plant
Physiol. 128, 1255–1263.
42. Hynninen, P. H. (1991) Chemistry of chlorophylls: Modifications.
In Chlorophylls (Edited by H. Scheer), pp. 145–210. CRC Press,
Boca Raton, FL.
43. Helfrich, M., S. Schoch, U. Lempert, E. Cmiel and W. Rudiger
(1994) Chlorophyll synthetase cannot synthesize chlorophyll a¢.
Eur. J. Biochem. 219, 267–275.
44. Strell, M. and T. Urumow (1977) New methods for introduction
of metals into derivatives of chlorophyll. Justus Liebigs Ann.
Chem., 970–974.
25. Ortiz de Montellano, P. R. and K. Auclair (2002) Heme oxygenase
structure and mechanism. In The Porphyrin Handbook. Vol. 12,
The Iron and Cobalt Pigments: Biosynthesis, Structure, Degrada-
tion (Edited by K. M. Kadish, K. M. Smith and R. Guilard),
pp. 183–210. Academic Press, Amsterdam.
26. Dougherty, T. J. (1987) Photosensitizers: Therapy and detection of
malignant tumors. Photochem. Photobiol. 45, 879–889.
27. Posen, Y., V. Kalchenko, R. Seger, A. Brandis, A. Scherz and
Y. Salomon (2005) Manipulation of redox signaling in mamma-
lian cells enabled by controlled photogeneration of reactive oxy-
gen species. J. Cell Sci. 118, 1957–1969.
45. Hartwich, G., L. Fiedor, I. Simonin, E. Cmiel, W. Schaefer, D.
Noy, A. Scherz and H. Scheer (1998) Metal-substituted bacterio-
chlorophylls. 1. Preparation and influence of metal and coordi-
nation on spectra. J. Am. Chem. Soc. 120, 3675–3683.
46. Havel, R. J., H. A. Eder and J. H. Bragdon (1955) Distribution
and chemical composition of ultracentrifugally separated lipo-
proteins in human serum. J. Clin. Invest. 34, 1345–1353.
47. Graham, J. (2001) Biological Centrifugation. BIOS Scientific
Publishers, Oxford.
48. Whitacre, C. M., D. K. Feyes, T. Satoh, J. Grossmann, J. W.
Mulvihill, H. Mukhtar and N. L. Oleinick (2000) Photodynamic
therapy with the phthalocyanine photosensitizer Pc 4 of SW480
human colon cancer xenografts in athymic mice. Clin. Cancer Res.
6, 2021–2027.
28. Kessel, D., M. G. H. Vicente and J. J. J. Reiners (2006) Initiation
of apoptosis and autophagy by photodynamic therapy. Autophagy
2, 289–290.
29. Plaetzer, K., T. Kiesslich, C. B. Oberdanner and B. Krammer
(2005) Apoptosis following photodynamic tumor therapy: Induc-
tion, mechanisms and detection. Curr. Pharm. Design 11, 1151–
1165.
49. Graham, J. M., J. A. Higgins, T. Gillott, T. Taylor, J. Wilkinson,
T. Ford and D. Billington (1996) A novel method for the rapid
separation of plasma lipoproteins using self-generating gradients
of iodixanol. Atherosclerosis 124, 125–135.
30. Piette, J., C. Volanti, A. Vantieghem, J.-Y. Matroule, Y. Habra-
ken and P. Agostinis (2003) Cell death and growth arrest in re-
sponse to photodynamic therapy with membrane-bound
photosensitizers. Biochem. Pharmacol. 66, 1651–1659.
31. Oleinick, N. L., R. L. Morris and I. Belichenko (2002) The role of
apoptosis in response to photodynamic therapy: What, where,
why, and how. Photochem. Photobiol. Sci. 1, 1–21.
32. Moor, A. C. E. (2000) Signaling pathways in cell death and sur-
vival after photodynamic therapy. J. Photochem. Photobiol. B,
Biol. 57, 1–13.
33. Henderson, B. W. and T. J. Dougherty (1992) How does photo-
dynamic therapy work? Photochem. Photobiol. 55, 145–157.
34. Mazor, O., A. Brandis, V. Plaks, E. Neumark, V. Rosenbach-
Belkin, Y. Salomon and A. Scherz (2005) WST11, a novel
water-soluble bacteriochlorophyll derivative; cellular uptake,
pharmacokinetics, biodistribution and vascular-targeted photo-
dynamic activity using melanoma tumors as a model. Photochem.
Photobiol. 81, 342–351.
35. Brandis, A., O. Mazor, E. Neumark, V. Rosenbach-Belkin, Y.
Salomon and A. Scherz (2005) Novel water-soluble bacteriochlo-
rophyll derivatives for vascular-targeted photodynamic therapy:
Synthesis, solubility, phototoxicity and the effect of serum pro-
teins. Photochem. Photobiol. 81, 983–993.
36. Vakrat-Haglili, Y., L. Weiner, V. Brumfeld, A. Brandis, Y.
Salomon, B. McIlroy, B. C. Wilson, A. Pawlak, M. Rozanowska,
T. Sarna and A. Scherz (2005) The microenvironment effect on the
generation of reactive oxygen species by Pd-bacteriopheophor-
bide. J. Am. Chem. Soc. 127, 6487–6497.
37. Ashur, I., R. Goldschmidt, I. Pinkas, Y. Salomon, G. Szewczyk,
T. Sarna and A. Scherz (2009) Photocatalytic generation of oxy-
gen radicals by the water-soluble bacteriochlorophyll derivative
WST11, noncovalently bound to serum albumin. J. Phys. Chem.
113, 8827–8837.
50. Noy, D., L. Fiedor, G. Hartwich, H. Scheer and A. Scherz (1998)
Metal-substituted bacteriochlorophylls. 2. Changes in redox
potentials and electronic transition energies are dominated by
intramolecular electrostatic interactions. J. Am. Chem. Soc. 120,
3684–3693.
51. Buchler, J. W. (1975) Static coordination chemistry of metallo-
porphyrins. In Porphyrins and Metalloporphyrins (Edited by K. M.
Smith), pp. 157–232. Elsevier, Amsterdam.
52. Scheer, H. (2006) Overview. In Chlorophylls and Bacteriochloro-
phylls: Biochemistry, Biophysics, Functions and Applications (Edi-
ted by B. Grimm, R. Porra, W. Rudiger and H. Scheer), pp. 1–26.
Springer, Dordrecht.
53. Watanabe, T. and M. Kobayashi (1991) Electrochemistry of
chlorophylls. In Chlorophylls (Edited by H. Scheer), pp. 287–316.
CRC Press, Boca Raton, FL.
54. Madorin, M., P. van Hoogevest, R. Hilfiker, B. Langwost, G. M.
Kresbach, M. Ehrat and H. Leuenberger (1997) Analysis of
drug ⁄ plasma protein interactions by means of asymmetrical flow
field-flow fractionation. Pharm. Res. 14, 1706–1712.
55. Davies, I. G., J. M. Graham and B. A. Griffin (2003) Rapid
separation of LDL subclasses by iodixanol gradient ultracentrif-
ugation. Clin. Chem. 49, 1865–1872.
56. Reyftmann, J. P., P. Morliere, S. Goldstein, R. Santus, L.
Dubertret and D. Lagrange (1984) Interaction of human serum
low density lipoproteins with porphyrins: A spectroscopic and
photochemical study. Photochem. Photobiol. 40, 721–729.
57. Jori, G., M. Beltramini, E. Reddi, B. Salvato, A. Pagnan, L.
Ziron, L. Tomio and T. Tsanov (1984) Evidence for a major role
of plasma lipoproteins as hematoporphyrin carriers in vivo. Cancer
Lett. 24, 291–297.
58. Jori, G. and E. Reddi (1993) The role of lipoproteins in the
delivery of tumour-targeting photosensitizers. Int. J. Biochem. 25,
1369–1375.
38. Lindsay Smith, J. R. and M. Calvin (1966) Chemical and photo-
chemical oxidation of bacteriochlorophyll. J. Am. Chem. Soc. 88,
4500–4506.
59. Hoebeke, M. (2000) ESR associated to spin label method in the
study of photosensitization in liposomal solution. Bull. Soc. Roy.
Sci. Liege 69, 103–110.
39. Struck, A., E. Cmiel, I. Katheder, W. Schaefer and H. Scheer
(1992) Bacteriochlorophylls modified at position C-3: Long-range
intramolecular interaction with position C-132. Biochim. Biophys.
Acta 1101, 321–328.
60. Kostner, G. M. and W. Marz (2001) Zusammensetzung und
Stoffwechsel der Lipoproteine. In Handbuch Der Fettstoffwech
selstorungen (Edited by P. Schwandt, W. O. Richter and K. G.
¨
Parhofer), pp. 1–57. Schaffauer, Stuttgart.
40. Omata, T. and N. Murata (1986) A rapid and efficient method to
prepare chlorophyll a and b from leaves. Photochem. Photobiol.
31, 183–185.
61. Szczygiel, M., K. Urbanska, P. Jurecka, I. Stawoska, G. Stochel
and L. Fiedor (2008) Central metal determines pharmacokinetics
of chlorophyll-derived xenobiotics. J. Med. Chem. 51, 4412–4418.

Photochemistry and Photobiology, 2010, 86 193
62. Yerushalmi, R., I. Ashur and A. Scherz (2006) Metal-substituted
bacteriochlorophylls: Novel molecular tools. In Chlorophylls and
Bacteriochlorophylls: Biochemistry, Biophysics, Functions and
Applications (Edited by B. Grimm, R. Porra, W. Rudiger and H.
Scheer), pp. 495–506. Springer, Dordrecht.
63. Evans, T. A. and J. J. Katz (1975) Evidence for 5 and 6-coordi-
nated magnesium in bacteriochlorophyll a from visible absorption
spectroscopy. Biochim. Biophys. Acta 396, 414–426.
64. Noy, D., R. Yerushalmi, V. Brumfeld, I. Ashur, H. Scheer, K. K.
Baldridge and A. Scherz (2000) Optical absorption and compu-
tational studies of [Ni]-bacteriochlorophyll a. New insight into
charge distribution between metal and ligands. J. Am. Chem. Soc.
122, 3937–3944.
80. Griffin, B. A., M. J. Caslake, B. Yip, G. W. Tait, C. J. Packard
and J. Shepherd (1990) Rapid isolation of low density lipoprotein
(LDL) subfractions from plasma by density gradient ultracen-
trifugation. Atherosclerosis 83, 59–67.
81. Jacob, S. W. and R. Herschler (1986) Pharmacology of DMSO.
Cryobiology 23, 14–27.
82. Santos, N. C., J. Figueira-Coelho, J. Martins-Silva and C.
Saldanha (2003) Multidisciplinary utilization of dimethyl sulfox-
ide: Pharmacological, cellular, and molecular aspects. Biochem.
Pharmacol. 65, 1035–1041.
83. Webster, L. K., D. M. Woodcock, D. Rischin and M. J. Millward
(1997) Cremophor: Pharmacological activity of an ‘‘inert’’ solu-
bilizer. J. Oncol. Pharm. Pract. 3, 186–192.
65. Haehnel, W., D. Noy and H. Scheer (2008) De novo designed
bacteriochlorophyll-binding helix-bundle proteins. In The Purple
Phototrophic Bacteria (Edited by C. N. Hunter, F. Daldal, M. C.
Thurnauer and J. T. Beatty), pp. 895–912. Springer, Dordrecht.
66. Fiedor, L. (2006) Hexacoordination of bacteriochlorophyll in
photosynthetic antenna LH1. Biochemistry 45, 1910–1918.
67. Broring, M. and C. D. Brandt (2003) A five coordinate Pd(II)
complex stable in solution and in the solid state. Chem. Commun.,
2156–2157.
68. Cotton, T. M., P. A. Loach, J. J. Katz and K. H. Ballschmiter
(1976) Studies of chlorophyll–chlorophyll and chlorophyll–ligand
interactions. Photochem. Photobiol. 27, 735.
69. Kania, A. and L. Fiedor (2006) Steric control of bacteriochloro-
phyll ligation. J. Am. Chem. Soc. 128, 454–456.
84. Haylett, A. K. and J. V. Moore (2002) Comparative analysis of
fetal calf and human low density lipoprotein: Relevance for
pharmacodynamics of photosensitizers. J. Photochem. Photobiol.
B, Biol. 66, 171–178.
85. Cunderlikova, B., M. Kongshaug, L. Gangeskar and J. Moan
(2000) Increased binding of chlorin e6 to lipoproteins at low pH
values. Int. J. Biochem. Cell Biol. 32, 759–768.
86. Hopkinson, H. J., D. I. Vernon and S. B. Brown (1999) Identifi-
cation and partial characterization of an unusual distribution of
the photosensitizer meta-tetrahydroxyphenyl chlorin (temoporfin)
in human plasma. Photochem. Photobiol. 69, 482–488.
87. Trachtenberg, J., R. A. Weersink, S. R. H. Davidson, M. A.
Haider, A. Bogaards, M. R. Gertner, A. Evans, A. Scherz, J.
Savard, J. L. Chin, B. C. Wilson and M. Elhilali (2008) Vascular-
targeted photodynamic therapy (padoporfin, WST09) for recur-
rent prostate cancer after failure of external beam radiotherapy: A
study of escalating light doses. BJU Int 102, 556–562.
70. Scheer, H. and J. J. Katz (1974) Structure of the Krasnovskii
photoreduction product of chlorophyll a. Proc. Natl Acad. Sci.
USA 71, 1626–1629.
71. Scheer, H. (1988) Chlorophylls: Chromatographic methods for the
separation of chlorophylls. In Chlorophylls (Edited by H.-P. Kost),
pp. 235–307. CRC Press, Boca Raton.
88. Kongshaug, M., J. Moan and S. B. Brown (1989) The distribution
of porphyrins with different tumor localizing ability among hu-
man plasma proteins. Br. J. Cancer 59, 184–188.
72. Beltramini, M., P. A. Firey, F. Ricchelli, M. A. J. Rodgers and G.
Jori (1987) Steady-state and time-resolved spectroscopic studies on
the hematoporphyrin-lipoprotein complex. Biochemistry 26, 6852–
6858.
73. Kepczynski, M., R. P. Pandian, K. M. Smith and B. Ehrenberg
(2002) Do liposome-binding constants of porphyrins correlate
with their measured and predicted partitioning between octanol
and water? Photochem. Photobiol. 76, 127–134.
89. Kessel, D., P. Thompson, K. Saatio and K. D. Nantwi (1987)
Tumor localization and photosensitization by sulfonated deriva-
tives of tetraphenylporphine. Photochem. Photobiol. 45, 787–790.
90. Woodburn, K. and D. Kessel (1995) Effect of density-gradients on
the binding of photosensitizing agents to plasma proteins. Int. J.
Biochem. Cell Biol. 27, 499–506.
91. Vilensky, J., N. V. Koudinova, A. Harmelin, A. Scherz and Y.
Salomon (2005) Vascular-targeted photodynamic therapy (VTP)
of a canine-transmissible venereal tumour in a murine model with
Pd-bacteriopheophorbide (WST09). Vet. Comp. Oncol. 3, 182–
193.
74. Garbo, G. M. (1990) The use of liposomes, emulsions or inclusion
complexes may potentiate in vivo effects of SnET2. Proc. SPIE
1203, 118–125.
75. Kessel, D., A. Morgan and G. M. Garbo (1991) Sites and efficacy
of photodamage by tin Etiopurpurin in vitro using different
delivery systems. Photochem. Photobiol. 54, 193–196.
76. Polo, L., E. Reddi, G. M. Garbo, A. R. Morgan and G. Jori
(1992) The distribution of the tumor photosensitizers zinc(II)-
phthalocyanine and tin (IV)-etiopurpurin among rabbit plasma
proteins. Cancer Lett. 66, 217–223.
92. Koudinova, N. V., J. H. Pinthus, A. Brandis, O. Brenner, P.
Bendel, J. Ramon, Z. Eshhar, A. Scherz and Y. Salomon (2003)
Photodynamic therapy with Pd-bacteriopheophorbide (TOO-
KAD): Successful in vivo treatment of human prostatic small cell
carcinoma xenografts. Int. J. Cancer 104, 782–789.
93. Kessel, D. and K. Woodburn (1993) Biodistribution of photo-
sensitizing agents. Int. J. Biochem. 25, 1377–1383.
77. Kongshaug, M., J. Moan, L. S. Cheng, G. M. Garbo, S. Kolboe,
A. R. Morgan and C. Rimington (1993) Binding of drugs to human
plasma proteins, exemplified by tin(IV)-etiopurpurin dichloride
delivered in Cremophor and DMSO. Int. J. Biochem. 25, 739–760.
78. Kongshaug, M., L. S. Cheng, J. Moan and C. Rimington
(1991) Interaction of Cremophor EL with human plasma. Int. J.
Biochem. 23, 473–478.
94. Polo, L., G. Bianco, E. Reddi and G. Jori (1995) The effect of
different liposomal formulations on the interaction of Zn(II)-
phthalocyanine with isolated low and high density lipoproteins.
Int. J. Biochem. Cell Biol. 27, 1249–1255.
95. Tyssandier, V., G. Choubert, P. Grolier and P. Borel (2002)
Carotenoids, mostly the xanthophylls, exchange between plasma
lipoproteins. Int. J. Vitam. Nutr. Res. 72, 300–308.
79. Woodburn, K. and D. Kessel (1994) The alteration of plasma
lipoproteins by Cremophor EL. J. Photochem. Photobiol. B, Biol.
22, 197–201.
96. Moss, G. P. (1988) IUPAC-IUB Joint Commission Biochemical
Nomenclature (JCBN). Nomenclature of Tetrapyrroles. Recom-
mendation 1986. Eur. J. Biochem. 178, 277–328.