ARTICLE IN PRESS

Biomaterials 28 (2007) 4779–4787

www.elsevier.com/locate/biomaterials

In vivo biological evaluation of high molecular weight

hyperbranched polyglycerols
Rajesh K. Kainthana, Donald E. Brooksa,b,
a
Department of Pathology and Laboratory Medicine, Centre for Blood Research, University of British Columbia, Vancouver, BC, Canada V6T 2B5
b
Department of Chemistry, University of British Columbia, Vancouver, BC, Canada V6T 2B5
Received 8 June 2007; accepted 24 July 2007

Abstract

Hyperbranched polyglycerols (HPGs) are water-soluble polyether polyols that can be synthesized in a controlled manner with low
polydispersity. Recently we reported the synthesis and characterization of very high molecular weight and narrowly polydispersed HPGs
that could be used as potential alternatives to high generation dendrimers, their advantage being the relative simplicity of synthesis.
Reported in this article are the pharmacokinetic properties of these polymers. Two polymers of number average molecular weights
106,000 and 540,000 were tested in mice for their pharmacokinetic behavior. The plasma half-life for the lower molecular weight polymer
was around 32 h whereas that of the higher molecular weight HPG was 57 h. Our results show that these high molecular weight HPGs,
which can be prepared in a single step reaction, are potential candidates for drug delivery and imaging applications where a long
circulating polymer is highly desirable. A detailed tissue distribution profile of these polymers as a function of molecular weight is
described. These polymers were also found to be hydrolytically stable and the concentration dependence of solution viscosity
measurements suggested the absence of any aggregation.
r 2007 Published by Elsevier Ltd.

Keywords: Hyperbranched polyglycerol; Animal toxicity; Biocompatibility; Drug delivery; Biodistribution; Pharmacokinetics

1. Introduction reported the synthesis of very high molecular weight HPGs

Because hyperbranched polymers can be prepared in a
convenient single step they are potential alternatives to
dendrimers in applications where a precise structure is not
necessary [1]. However, until recently the large polydisper-
sity remained a major drawback and the synthesis of
hyperbranched polymers with narrow polydispersity was a
significant synthetic challenge [2,3]. Polyglycerol is one of
the first hyperbranched polymers which has been synthe-
sized with low polydispersities and predetermined mole-
cular weights. Hyperbranched polyglycerols (HPGs) are
synthesized by anionic ring opening multibranching poly-
merization of glycidol with slow monomer addition and
partial deprotonation of the initiator [4,5]. Recently we
Corresponding author. Centre for Blood Research, Life Sciences
Centre, 2350 Health Sciences Mall, The University of British Columbia,
Vancouver, BC, Canada V6T 1Z3. Fax: +1 604 822 7742.
E-mail address: don.brooks@ubc.ca (D.E. Brooks).
with low polydispersity using high monomer to initiator
ratios in the presence and absence of emulsifying non-
solvents such as dioxane or diglyme [6]. We prepared a
series of polymers with various molecular weights and the
polydispersity index (PDI) of most of these polymers was
relatively low (PDI ¼ 1.1–1.8) with Mw values up to
1.48  106. As the number of functional hydroxyl groups
approximately equals the degree of polymerization, synth-
esis of higher molecular weight HPGs is of significant
interest. Such materials have numerous applications and
likely have properties unattainable by other approaches,
given the difficulty and cost associated with the synthesis of
higher generation dendrimers. The HPGs, with low
intrinsic viscosity, narrow polydispersity and a very large
number of derivatizable hydroxyl groups hold great
potential for applications in nanobiotechnology and in
nanomedicine [7,8].
The large number of reactive functional groups in the
HPG class of polyether polyols provides an advantage in

0142-9612/$ - see front matter r 2007 Published by Elsevier Ltd.

doi:10.1016/j.biomaterials.2007.07.046
 ARTICLE IN PRESS
4780 R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787
pharmaceutical and medical fields compared to linear
polymers such as PEG. An additional advantage is that
they are thermally and oxidatively more stable than PEG
[9,10]. Recently we reported the biocompatibility of HPGs
using several in vitro techniques and animal toxicity studies
where HPGs were found to be as biocompatible as PEG
[10,11]. As well, oligoglycerols are approved as food and
pharma additives by the FDA [12]. These observations
support a role for these robust polymers as promising
biomaterials.
The use of biocompatible polymers as drug or protein
delivery vehicles is a subject of considerable interest
[7,8,13–20]. Conjugation of a long circulating polymer to
an anticancer drug increases the probability of drug
accumulation in tumor tissues due to their enhanced
permeability to the leaky tumor vasculature and the
limited lymphatic drainage, a phenomenon known as
enhanced permeation and retention (EPR) effect [21]. As
highlighted in a number of recent reviews, there is need for
long circulating polymers with multiple end groups for
drug attachment; dendrimers are being extensively studied
for this purpose [7,13–19]. However, dendrimers suffer
from generally short circulation half-lives and have to be
prepared by tedious multi step reaction/purification cycles
[22]. There are evidently not many dendrimer families
which are suitable for parenteral use [23]. Therefore the
narrowly distributed high molecular weight HPGs, with
more than 10,000 hydoxyl groups available for functiona-
lization, could be very interesting scaffolds for drug
delivery applications providing a huge capacity for
chemical modification with drugs, targeting moieties,
pegylation or to add anionic charges. The latter two
strategies are known to be successful against RES
recognition [1,24–25]. The use of a low molecular weight
HPG has been reported for making drug conjugates [26].
Because of their compact structure they can be used at high
concentration without increasing solution viscosity drama-
tically, making handling easy and reducing deleterious
blood flow effects to low levels, particularly in comparison
to high molecular weight linear polymers [6,27].
The biodistribution characteristics of these polymers
have to be evaluated in order to further their applications
in drug delivery and other areas of nanomedicine.
Reported in this article are the circulation longevity,
tissue/organ uptake and urine and stool clearance of HPGs
of two different molecular weights and hydrodynamic sizes
following intravenous administration in Balb/C mice. The
degradation properties of these polymers are also dis-
cussed.
2. Materials and methods
2.1. Materials
Two polymers of number average molecular weight 106,000 and
540,000 were prepared as previously reported and are designated as HPG-
106K and HPG-540K, respectively. Tritiated methyl iodide was obtained
from ARC Radiochemicals (St. Louis, MO) as a solution in toluene and
used without further purification after dilution in dimethylsulphoxide.
2.2. Viscometry
An Ubbelohde viscometer immersed in a temperature controlled water
bath was used for viscosity measurements. The viscosities of aqueous
polymer solutions of concentrations ranging from 10 to 100 mg/mL were
measured at various temperatures. The solutions were filtered (0.45 mm
pore size) prior to measurement. Each measurement was repeated at least
five times to calculate the average viscosity. The concentrations were
verified by determination of the solid content after evaporation of the
solvent. The intrinsic viscosities were determined from plots of Zsp/C vs. C,
where Zsp ¼ (ZZo)/Zo, Z ¼ solution viscosity, Zo ¼ solvent viscosity and
C ¼ concentration (g/mL).
2.3. Hydrolysis
HPG-540K was dissolved in buffer solutions of pH 5 (30 mM acetate
buffer with 70 mM NaNO3) and 7.4 (30 mM phosphate buffer with 70 mM
NaNO3) at a concentration of 10 mg/mL and incubated at 37 1C. Samples
were withdrawn at 1, 2 and 4 week intervals and analyzed by size exclusion
chromatography for molecular weight characteristics. The GPC system
used consists of a Waters 2690 separation module fitted with a DAWN
EOS multiangle laser light scattering (MALLS) detector from Wyatt
Technology Corp. with 18 detectors placed at different angles (laser
wavelength ¼ 690 nm) and a refractive index detector (Optilab DSP from
Wyatt Technology Corp.). An ultrahydrogel linear column with bead size
6–13 mm (elution range 103–5  106 Da) and an ultrahydrogel 120 with
bead size 6 mm (elution range 150–5  103 Da) from Waters were used. An
aqueous 0.1 N NaNO3 solution was used as the mobile phase at a flow rate
of 0.8 mL/min. The dn/dc value for polyglycerol was determined to be 0.12
in aqueous 0.1 N NaNO3 solutions and was used for molecular weight
calculations. The data were processed using Astra software provided by
Wyatt Technology Corp.
2.4. Radiolabelling of HPGs
Radiolabelling was done by partial conversion of hydroxyl groups to
methoxides using tritiated methyl iodide. Briefly, 1 gm of polymer was
dissolved in 10 mL dimethylsulfoxide (DMSO) and approximately 5% of
the hydroxyl groups were converted in to potassium alkoxides by reaction
with potassium hydride (35 mg). To this 20 mL of tritiated methyl iodide
(toluene solution) dissolved in 1 mL DMSO was added, the reaction
mixture stirred at room temperature for 15 h then 5 mL water was
added and the reaction mixture was acidified with dilute HCl. The
polymer was purified by dialysis against water using dialysis membrane of
MWCO 1000 until the dialyzate contained low amounts of radioactivity;
this normally took 24 h. The polymer solutions were then filtered through
a syringe filter (0.2 mm) and the total polymer weight was determined
from the total volume and the dry weight of a known volume (100 mL)
of solution after freeze drying (measured in duplicate). The polymer
solution was then concentrated by evaporating off the water in a fume
hood and the final concentration of 100 mg/mL in 150 mM saline solution
was made by appropriate dilution with an aqueous NaCl solution. This
allowed the specific activity to be determined by counting an aliquot of
solution.
2.5. Animal study protocol
The protocol was reviewed and approved by the Institutional Animal
Care Committee (IACC) at UBC prior to conducting the studies. Female
Balb/C mice (7–11 weeks) were injected intravenously (bolus) via lateral
tail vein with the polymer solution of concentration 100 mg/mL (four mice
per group) to a prescribed dose of 1 g/kg. The injected volume was 200 mL/
20 g mouse. All animals were observed post administration, and at least
 ARTICLE IN PRESS
R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787 4781

once a day, more if deemed necessary, during the pre-treatment and
treatment periods for mortality and morbidity. In particular, signs of
clinical ill health was based on body weight loss, change in appetite, and
behavioral changes such as altered gait, lethargy and gross manifestations
of stress.
Blood was collected at various time intervals (30 min, 2, 4, 8, 24, 48 h, 7,
14 and 30 days) by cardiac puncture after termination by CO2 inhala-
tion. Plasma was separated by centrifuging samples at 2500 rpm for
15 min. Aliquots of plasma were placed in scintillation vials and analyzed
by a Packard 1900 scintillation counter after adding 5 mL scintillation
cocktail.
Mice in one group (time point 48 h) were housed in a metabolic cage.
Urine and stools were collected as pooled samples at 1, 4, 8, 24 and
48 h post injection. Aliquots of urine were analyzed by scintillation
counting. Stool samples were pooled and made into a 30% homogenate in
a known amount of water prior to obtaining aliquots for scintillation
counting.
Upon termination, liver, spleen, kidney, lungs and heart were removed
from all the mice, weighed and processed for scintillation counting. Livers
(one lobe) were made into a 30% homogenate in a known amount of
water using a polytron tissue homogenizer. Aliquots (in triplicate) of
200 mL homogenate were transferred to scintillation vials. All other organs
were dissolved in 500 mL Solvables. Vials were incubated at 50 1C
overnight, then cooled prior to addition of 50 mL 200 mM EDTA, 25 mL
10 M HCl and 200 mL 30% H2O2. This mixture was incubated at room
temperature for one hour prior to addition of 5 mL scintillation cocktail.
Samples were analyzed by scintillation counting.
3. Results and discussion
Two high molecular weight HPGs were prepared as
described previously [6]. The structure of the polymers is
given in Fig. 1. The molecular characteristics of the two
polymers, designated as HPG-106K (Mn ¼ 106,000) and
HPG-540K (Mn ¼ 540,000) are given in Table 1.
3.1. Viscometry
The solution properties of these polymers were studied in
detail using a gel permeation chromatography system
equipped with a multi angle laser light scattering detector,
Viscotek triple detector array and a quasi elastic light
scattering detector (QELS). The results were described in
an earlier manuscript [6] and suggested the absence of any
polymer aggregation in solution. The absence of aggrega-
tion is critical for drug delivery use as it affects the
pharmacokinetics and cellular uptake. However the GPC
analyses were done in aqueous 0.1 N NaNO3 solutions, and
therefore the properties are not those of simple aqueous
solutions. Moreover the GPC results correspond to very

OH OH
HO
HO HO
OH OH

O HO
O O O OH
OH
O O
OH
O OH
O
OH HO
O O OH
HO OH
O
O
O O O
HO
HO O OH
OH OH
O
O OH
HO O
O OH
O O OH OH
O
O O O OH
O
O O
O
HO O OH
OH
O
OH O
HO OH
O
OH
O OH
HO
O OH
OH
O O

HO O
OH
O

HO OH

Fig. 1. Structure of hyperbranched polyglycerol.

 ARTICLE IN PRESS
4782 R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787

Table 1
Polymer characterization data

Polyglycerol Mn Mw/Mn Intrinsic viscosity Rg (nm)a RZ(nm)b Rh (nm)c

[Z] (mL/g)

HPG-106K 106 2.9 3.8 6.8 5.2 4.8

HPG-540K 540 1.1 3.0 9.2 7.1 6.8
a
Radius of gyration determined by multi angle laser light scattering detector.
b
Viscocity radii determined from Viscotek triple detector.
c
Hydrodynamic radii determined by quasi elastic light scattering (QELS) detector.

dilute polymer solutions. One of the potential uses of this

class of polymers is as plasma expanders which are 0.0065
typically infused at higher concentrations (60–100 mg/ 0.0060
mL). Therefore lack of aggregation at these concentrations 0.0055
is a prerequisite for such applications. Since aggregation is
0.0050
expected to be concentration and temperature dependent
we checked the possibility of aggregation by measuring the 0.0045
viscosities of aqueous solutions of various concentrations
ηsp/C
0.0040
at different temperatures. Aggregation is known to 0.0035
produce non-linearities at higher concentrations in plots 0.0030
of Zsp/C vs. C. In such plots the intrinsic viscosity [Z] is 12 °C [η] = 3.2 mL/g
0.0025 22 °C [η] = 2.5 mL/g
given by
0.0020 37 °C [η] = 2.2 mL/g
½Z ¼ lim ½Zsp =C. 0.0015
50 °C [η] = 1.8 mL/g
C!0
0.0010
A series of polymer solutions up to 100 mg/mL solutions 0 20 40 60 80 100
of HPG-540K were analyzed in this way and the results are Concentration (mg/ml)
shown in Fig. 2. The intrinsic viscosities at each tempera-
ture were calculated as the intercepts of the plot of Zsp/C Fig. 2. Effect of concentration and temperature on viscosities of aqueous
solutions of HPG-540K.
against C. The linearity of these plots implies the absence
of detectable aggregation in the concentration range
0.5
studied. The intrinsic viscosities calculated were below t=0
4 mL/g similar to the Viscotek triple detector data reported pH 7.4, 14 days
pH 5.0, 14 days
earlier [6]. The intrinsic viscosity was found to decrease 0.4
pH 5.0, 30 days
with increasing temperature (T). Since [Z] is a direct pH 7.4, 30 days
RI Detector Response

measure of the dimension of the polymer molecules in

solutions this implies the polymer coil contracts as T 0.3
increases. This shrinkage at higher temperatures is well
known for polymers whose aqueous solubility decreases
0.2
with increasing T due to disruption of hydrogen bonding
between polymer segments and water molecules [28,29].
0.1

3.2. Hydrolysis
0.0
HPGs are reported to be thermally and oxidatively more
stable than poly(ethylene glycol) [9,10]. The stability of the 14 16 18 20
Retention Time (min)
high molecular weight polymer (HPG-540K) in buffer
solutions of pH 5 and 7.4 (physiological pH) incubated at Fig. 3. Time dependent GPC chromatograms of HPG-540K after
37 1C were tested by monitoring molecular weight over incubation in buffer solutions of pH 5 and 7.4 at 37 1C.
time by GPC. The chromatograms are shown in Fig. 3. It is
seen that the polymer is very stable under these conditions degradation shows that these polymers are very stable.
with no observable change in molecular weight. No Since these polymers consist of –C–C– and C–O–C– bonds
degradation products were observed during the 30-day only, no rapid enzymatic degradation is anticipated in the
monitoring period. This data along with that of thermal circulation [30].
 ARTICLE IN PRESS
R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787 4783

3.3. Animal studies HPG-106 k

20
We have found that HPGs were non toxic to mice even

Polymer levels in plasma (mg/mL)

when injected at a dose of 1 g/kg [11]. However the
pharmacokinetic properties of these polymers were not 15
reported. The two HPGs studied here were radiolabeled
Half life = 32.4 ± 2.6 h
with tritium after converting approximately 1% of hydro-
xyl groups to methoxides by reaction with CT3I. Balb/c
10
mice (36 animals, 4 per time point) were injected
intravenously (bolus) with the polymer solutions at a dose
of 1 g/kg (body weight). The high dose testing was done to
5
mimic one of the intended uses of this class of polymers as
an albumin substitute, after proper derivatization, where a
high dose of infusion is needed [31]. No signs of toxicity
(eg: lethargy, dry eyes, change in appetite, weight loss, 0
0 100 200 300 400 500 600 700 800
altered gait, scruffy coats) were observed for any of the
Time (h)
animals and they all grew normally.
Mice were terminated at each time point (30 min, 2, 4, 8, 30 HPG-540 k
24, 48 h, 7, 14 and 30 days) for tissue distribution studies.
The blood collected by cardiac puncture was centrifuged
Polymer levels in plasma (mg/mL)
25
and the plasma was assessed for levels of polymer by
scintillation counting. The liver, spleen, heart, lungs and
20
kidney were collected, weighed and also assessed for levels half life = 57.5 ± 9.1 h
of polymers by scintillation counting. Since the residual
blood in organs cannot be removed completely by washing, 15
radioactivity due to the plasma present was calculated
using the plasma radioactivity data and the literature 10
values of plasma volumes in the mouse organs where
sufficiently consistent values were available [32].
5
The circulating levels of both the polymers are shown in
Fig. 4 as a function of time expressed as mg polymer per
mL of plasma. The data were analyzed by a standard two- 0
compartment open model [33]. The system consists of a 0 100 200 300 400 500 600 700 800
Time (h)
central compartment (C) that is mainly plasma, and the
tissue compartment (T). The elimination process is Fig. 4. Plot of polymer concentration in plasma with respect to time.
represented by compartment E. The central compartment
is considered open since elimination occurs from this
compartment by excretion. It is assumed that a reversible HPG-540K
Liver
exchange occurs between the blood and the tissue HPG-106K
compartments. The notations k12 and k21 represent the 12
rate constants for transfer between compartments C and T,
the plasma and tissues, and k2 is the rate constant for 11
elimination.
When the polymer solutions are injected intravenously 10
% dose/g tissue

into mice, the macromolecules will undergo simultaneous

elimination and distribution processes that include rever- 9
sible diffusion into the tissues. The polymer concentration
8
in plasma decreases with time in two distinguishable
exponential phases. During the distribution phase, the
7
concentration decreases due to diffusion to tissue and
elimination. During the elimination phase the concentra-
6
tion decreases both in tissue and blood compartments as a
function of time. Such a system can be expressed by the
5
equation P(t) ¼ A1eat+A2ebt where P is the concentra-
0 100 200 300 400 500 600 700
tion of polymer in plasma and t is time. Curve fitting of the
time (h)
data (Fig. 4) gave t ¼ 0 values for the plasma concentration
of the material associated with the distribution and Fig. 5. Levels of polymer accumulated over time in liver.
 ARTICLE IN PRESS
4784 R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787

elimination phases (A1 and A2, respectively) and the values
for a and b. The apparent rate constants k12, k21, and k2
can be calculated using the following equations:
P0 ¼ A1 þ A2 ,
k21 ¼ ðA1 b þ A2 aÞ=P0 ,
k2 ¼ ab=k21 ,
k12 ¼ a þ b  k2  k21 .
P0 is the concentration of polymer in the central
compartment immediately after injection, before distribu-
tion to the tissue compartments and elimination. Vc the
apparent volume of the central compartment is calculated
as D/P0 where D is the dose of polymer injected. The
plasma half-life was calculated as (ln 2)/b and the area
under the curve from t ¼ 0 to infinity (AUC0N) was
calculated as (A1/a+A2/b). The values calculated from the
graphs are given in Table 2.
It can be seen that the high molecular weight HPGs have
long plasma half-lives. HPG-106K which has a number
average molecular weight of 106,000 has a half-life of
32.472.7 h whereas the HPG-540K with Mn of 540,000 has
a half-life of 57.579.1 h. The corresponding AUC0N
values were 525746 and 16817277 mg/mL/h, respectively,
indicating that exposure of the animals was over three fold
greater for the higher molecular weight material. The half-
lives measured for these polymers are significantly longer
than that reported for other polymers [21,34–36] and even
exceed values for most stealth liposome reported in the
literature [37,38]. In general branched polymers have
longer plasma residence times than linear polymers of
equivalent Rh, as discussed below. N-(2-hydroxypropyl)-
methacrylamide (HPMA) polymer has been extensively
studied because of its use for anticancer drug delivery and
its circulation half-life is only 25 h for a molecular weight of
556 K [21]. Regular dendrimers of low generations have
very low plasma half-lives [1]. To enhance their residence
times in vivo, other classes of branched polymers such as
linear-dendritic hybrids, bow-tie dendrimers and dendro-
nised polymers have been specifically designed [1,39,40],
the highest value reported is 50710 h for a pegylated
polyester ‘‘bow-tie’’ dendrimer of molecular weight 160,000
[40]. The long circulation half-life of high molecular weight
HPG is a property of great interest for drug delivery,
particularly in cancer therapeutics [15,21]. Drug molecules
can be encapsulated inside the polymer or attached
covalently to the surface functionalities using degradable
linkers.
It is also evident from the data in Table 2 (k2 values) that
the low molecular weight polymer is eliminated from the
system two and half times faster than the higher molecular
weight polymer. According to the model employed here the
diffusion of the lower molecular weight polymer to tissues
(k12) is almost three times faster and the reverse process
(k21) is at a comparable rate compared to the higher
molecular weight one.
The experimental values for clearance into the urine and
feces are given in Table 3. Substantial renal clearance was
observed for the lower molecular weight polymer with 7%
eliminated within 4 h whereas it was 1% in the case of
HPG-540K. Approximately 28% of HPG-106K was
eliminated within 24 h whereas the corresponding amount
was 2% for HPG-540K. The data also shows that most
of the urinary excretion was over within 24 h.

Table 2
Rate constants determined from pharmacokinetic analysis of polymer concentrations in plasma over time

Polymer P0 (mg/mL Vc (mL a (h1) b (h1) t1/2 (h) k21 (h1) k2 (h1) k12 (h1) AUC0N mg/
plasma) plasma) mL/h

HPG- 21.370.7 0.9470.03 0.6470.08 0.02170.002 32.472.7 0.3470.04 0.04170.008 0.28370.093 525746
106K
HPG- 26.372.6 0.7670.08 0.4970.29 0.01270.002 57.579.1 0.3870.22 0.01570.013 0.10870.362 16817277
540K

Table 3
Polymer levels in urine and feces at the times indicated

HPG-540K HPG-106K

Time (h) % dose in feces % dose in urine % dose in feces % dose in urine

1 0.01 Not collected 0.06 Not collected

4 0.16 0.90 0.11 7.36
8 0.05 0.23 0.19 6.37
24 1.80 0.37 9.06 14.54
48 3.96 0.17 8.16 1.68

(Error bars are not shown as the data are obtained from the pooled samples).
 ARTICLE IN PRESS
R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787 4785

The glomerular filtration rate depends on the size [41],

charge [42], shape, and deformability of a polymer [43]. For 10
neutral polymers, it has been reported that the equivalent HPG-540K
pore radius of the glomerular filter in terms of the HPG-106K
8 Kidney
hydrodynamic radius for a linear polymer such as dextran

% dose /g organ
is 4.8–6.0 nm, for a branched polymer such as Ficoll is
4.5–5.2 nm and for a protein is 3.7–3.8 nm [44]. These differ 6
presumably because linear polymers are able to pass
through pores smaller than the unconstrained dimension
of the polymer in solution via reptation, an option not 4
available to proteins or dendritic polymers, which are much
less deformable, even when subjected to strong elonga-
2
tional shear [45,46]. We have previously reported that these
polymers have spherical conformations in water [6]. The
average hydrodynamic radius of HPG-540K is about 0
6.8 nm, well above the filtration threshold for even linear 0 100 200 300 400 500 600 700 800
polymers. The higher renal filtration of HPG-106K with time (h)
Rh ¼ 4.8 nm might be due to its much broader molecular
Fig. 7. Levels of polymer accumulated over time in kidney.
weight distribution. The 30% of the polymer which was
eliminated within the 24 h is likely enriched in the lower
molecular weight fraction.
Lungs
Both the polymers were excreted via the feces with about
10% and 18% of HPG-106K excreted after 24 and 48 h,
20 HPG-540K without correction
respectively, whereas only 2% and 6% were excreted in the HPG-540K with correction
case of HPG-540K. HPG-106K without correction
Upon termination liver, gall bladder, spleen, lung, HPG-106k with correction
% dose /g organ

kidney, heart, intestine, lymph nodes and bladder were 15

examined macroscopically for gross pathology and no
unusual findings observed. The liver, spleen, heart, kidney
and lungs were removed, weighed and examined for 10
radioactivity (Figs. 5–9). The data was not corrected for
the plasma present in these organs except lungs and heart,
5
however, since we were not able to identify reliable plasma
volumes for them. Different values are reported for
different radiolabeled markers such as red blood cells and
0
albumin [32]. Hence, over the first 100–200 h the majority
0 100 200 300 400 500 600 700
of the initial radioactivity could be due to plasma counts.
time (h)

Fig. 8. Levels of polymer accumulated over time in lungs.

HPG-540K
Spleen
18 HPG-106K
However, the data after 7 days can be compared safely as
16 the polymer levels in plasma by then are negligible. The
major polymer accumulations were observed in the liver
14
and spleen. Organ accumulation of the high molecular
% dose /g organ

12 weight polymer was clearly greater than that of the lower

molecular weight one. While these measurements appear to
10 be inconsistent with the corresponding rate constants (k12
and k21) obtained from the plasma data, they are relevant
8 to times after the rate constants were determined so cannot
6
be related to them. The data illustrate the role of molecular
weight in organ accumulation. Related molecular weight
4 dependence has been reported for other polymers as
well [47].
2 The accumulations were approximately 10% of the
0 100 200 300 400 500 600 700 800
injected dose (ID) per gram of liver (this value corresponds
time (h)
to the accumulation in the whole liver assuming 5 wt% of
Fig. 6. Levels of polymer accumulated over time in spleen. the total 20 g body weight) and were found to decrease in
 ARTICLE IN PRESS
4786 R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787
HPG-540K without correction
12
heart HPG-540K with correction
HPG-106K without correction
10 HPG-106k with correction
% dose /g tissue
8
6 the lack of biodegradable groups in these polyethers allows
4 Further work is required to shed light on this issue. There
2 the animals over this period, however, and we have seen no
0 100 200 300 400 500 600 700
time (h)
Fig. 9. Levels of polymer accumulated over time in heart.
the case of HPG-106K during the 7–30 days time period
whereas they remained constant for the higher molecular
weight polymer. Up to 17% and 8% ID/g were present in
spleen for the high and low molecular weight polymer
respectively. This corresponds to 1.85% and 0.82% ID per
spleen, respectively, for the two polymers. The reason for
the high RES accumulation of polymer (10% of the ID in
liver and 2% in the spleen) is not clear at present. Similar
levels of accumulation have been reported for dextran and
hydroxyethyl starch [30,36,47].
Tissue accumulation was found to decrease with time in
the case of kidney, lungs and heart and again comparison
of the data at 7, 14 and 30 days time points show that
higher amounts were detected for the higher molecular
weight polymer. In the case of lungs and heart plasma
corrections were applied as comparatively similar values
for plasma volumes were obtained for these organs using
different markers [32]. The values after plasma volume
correction were below 10% ID/g for both polymers. The
amount of HPG-540K present in the lungs decreased from
10% to 1.5% ID/g in lungs whereas it was roughly
constant between 3% and 5% ID/g in heart. The amounts
of HPG-106K in lungs decreased from 8% to 1.5% ID/g
and from 6% to 2% ID/g in heart during the 30 days
period.
4. Conclusions
The use of multifunctional nanocarriers for imaging,
drug delivery and other applications in polymer therapeu-
tics is a topic of considerable interest. The high molecular
weight hyperbranched HPGs examined here exhibit nu-
merous advantages for these applications. Their synthesis
is simple and inexpensive, particularly in comparison with
dendrimers or modified dendrimers that have attracted
considerable interest as drug delivery vehicles in recent
years. The data in this contribution establishes the long
circulation half-life of high molecular weight HPG, a
property of great interest for drug delivery, particularly in
cancer therapeutics. Urinary excretion is very low due to
the molecular size. Accumulation by the RES is significant
over the course of the experiments, the counts originating
on the labeled HPGs remaining in the liver and spleen for
at least 30 days. We have no information on the chemical
structures with which these counts are associated although
for the possibility that the polymers degrade very slowly.
was no indication of any untoward effects on the health of
indication of liver toxicity from blood work in rat exposure
800 studies currently being completed so the significance of
these observations is not clear. It may also prove valuable
to have biodegradable analogues of these HPGs, the
development of which we are currently pursuing.
Acknowledgments
We thank the Canadian Institutes of Health Research,
Natural Science and Engineering Research Council,
Canada, the Canada Foundation for Innovation, Cana-
dian Blood Services, Bayer Canada Inc., Michael Smith
Foundation for Health Research and BC Knowledge
Development Fund for financial support.
References
[1] Lee CC, MacKay JA, Frechet JMJ, Szoka FC. Designing dendrimers
for biological applications. Nat Biotechnol 2005;23:1517–26.
[2] Gao C, Yan D. Hyperbranched polymers: from synthesis to
applications. Prog Polym Sci 2004;29:183–275.
[3] Yates CR, Hayes W. Synthesis and applications of hyperbranched
polymers. Eur Polym J 2004;40:1257–81.
[4] Sunder A, Hanselmann H, Frey H, Mulhaupt R. Controlled synthesis
of hyperbranched polyglycerols by ring-opening multibranching
polymerization. Macromolecules 1999;32:4240–6.
[5] Frey H, Haag R. Dendritic polyglycerol: a new versatile biocompatible
material. Rev Mol Biotechnol 2002;90:257–67.
[6] Kainthan RK, Muliawan EB, Hatzikiriakos SG, Brooks DE.
Synthesis, characterization, and viscoelastic properties of high
molecular weight hyperbranched polyglycerols. Macromolecules
2006;39:7708–17.
[7] Svenson S, Tomalia DA. Dendrimers in biomedical applications-
reflections on the field. Adv Drug Delivery Rev 2005;57:2106–29.
[8] Torchilin VP. Multifunctional nanocarriers. Adv Drug Delivery Rev
2006;58:1532–55.
[9] Siegers C, Biesalski M, Haag R. Self-assembled monolayers of
dendritic polyglycerol derivatives on gold that resist the adsorption of
proteins. Chem-A Eur J 2004;10:2831–8.
[10] Kainthan RK, Hester SR, Levin E, Devine DV, Brooks DE. In vitro
biological evaluation of high molecular weight hyperbranched
polyglycerols. Biomaterials, in press, doi:10.1016/j.biomaterials.
2007.07.011.
[11] Kainthan RK, Janzen J, Levin E, Devine DV, Brooks DE.
Biocompatibility testing of branched and linear polyglycidol.
Biomacromolecules 2006;7:703–9.
[12] Friberg SE, Goubran RF, Kayali IH. In: Food emulsions. New York:
Marcel Dekker; 1990. p. 1.
 ARTICLE IN PRESS
R.K. Kainthan, D.E. Brooks / Biomaterials 28 (2007) 4779–4787
[13] Duncan R. Polymer conjugates as anticancer nanomedicines. Nat
Rev Cancer 2006;6:688–701.
[14] Khandare J, Minko T. Polymer – drug conjugates: progress in
polymeric prodrugs. Prog Polym Sci 2006;31:359–97.
[15] Duncan R, Ringsdorf H, Satchi-Fainaro R. Polymer therapeutics:
polymers as drugs, drug and protein conjugates and gene delivery
systems: past, present and future opportunities. Adv Polym Sci
2006;192:1–8.
[16] Satchi-Fainaro R, Duncan R, Barnes CM. Polymer therapeutics for
cancer: current status and future challenges. Adv Polym Sci
2006;193:1–65.
[17] Haag R, Kratz F. Polymer therapeutics: concepts and applications.
Angew Chem Int Ed 2006;5:1198–215.
[18] Duncan R. The dawning era of polymer therapeutics. Nat Rev Drug
Discovery 2003;2:347–60.
[19] Gillies ER, Frechet JMJ. Dendrimers and dendritic polymers in drug
delivery. Drug Discovery Today 2005;10:35–43.
[20] Caliceti P, Veronese FM. Pharmacokinetic and biodistribution
properties of poly(ethylene glycol)-protein conjugates. Adv Drug
Delivery Rev 2003;55:1261–77.
[21] Maeda H, Seymour LW, Miyamoto Y. Conjugates of anticancer
agents and polymers: advantages of macromolecular therapeutics
in vivo. Bioconjugate Chem 1992;3:351–62.
[22] Frechet JMJ, Tomalia D, editors. Dendrimers and other dendritic
polymers. Chichester and New York: Wiley; 2001.
[23] Duncan R. Designing polymer conjugates as lysosomotropic
nanomedicines. Biochem Soc Trans 2007;35:56–60.
[24] Yamamoto Y, Nagasaki Y, Kato Y, Sugiyama Y, Kataoka K. Long-
circulating poly(ethylene glycol)-poly(D,L-Lactide) block copolymer
micelles with modulated surface charge. J Controlled release
2001;77:27–38.
[25] Okuda T, Kawakami S, Maeie T, Niidome T, Yamashita F. Hashida
biodistribution characteristics of amino acid dendrimers and their
PEGylated derivatives after intravenous administration. J Controlled
release 2006;114:69–77.
[26] Kolhe P, Khandare J, Pillai O, Kannan S, Lieh-Lai M, Kannan R.
Hyperbranched polymer-drug conjugates with high drug payload for
enhanced cellular delivery. Pharm Res 2004;21:2185–95.
[27] Chien S. Biophysical behavior of red cells in suspensions. In:
Surgenor DM, editor. Red blood cell. New York: Academic Press;
1975. p. 1032–121.
[28] Briscoe B, Luckham P, Zhu S. Rheological Study of Poly(ethylene
oxide) in Aqueous Salt Solutions at High Temperature and Pressure.
Macromolecules 1996;29:6208–11.
[29] Briscoe B, Luckham P, Zhu S. The effects of hydrogen bonding upon
the viscosity of aqueous poly(vinyl alcohol) solutions. Polymer
2000;41:3851–60.
[30] Leuschner J, Opitz J, Winkler A, Scharpf R, Bepperling F. Tissue
storage of 14C-labelled hydroxyethyl starch (HES) 130/0.4 and HES
200/0.5 after repeated intravenous administration of rats. Drugs
R&D 2003;4:331–8.
4787
[31] Kainthan RK, Janzen J, Kizhakkedathu JN, Devine DV, Brooks DE.
Polymer based human serum albumin substitute. Polymer preprints
2006;47:119–20.
[32] Bally MB, Mayer LD, Hope MJ, Nayar R. Pharmacodynamics of
loposomal drug carriers: methodological considerations. In: Gregor-
iadis G, editor. Liposome technology. Boca Raton, Florida: CRC
press, Inc.; 1999. p. 27–42.
[33] Notari RE. Biopharmaceutics and clinical pharmacokinetics, an
introduction. New York: Marcel Dekker, Inc.; 1980.
[34] Tabata Y, Murakami Y, Ikada Y. Tumor accumulation of poly(vinyl
alcohol) of different sizes after intravenous injection. J controlled
release 1998;50:123–33.
[35] Yamaoka T, Tabata Y, Ikada Y. Distribution and tissue uptake of
poly(ethylene glycol) with different molecular weights after intrave-
nous administration to mice. J Pharm Sci 1994;83:601–6.
[36] Klotz U, Kroemer H. Clinical pharmacokinetic considerations in the
use of plasma expanders. Clin Pharmacokin 1987;12:123–35.
[37] Gabizon A, Shmeeda H, Barenholz Y. Pharmacokinetics of pegylated
liposomal doxorubicin. Clin Pharmacokinet 2003;42:419–36.
[38] Lasic D, Martin F. Stealth liposomes. Boca Raton, FL: CRC Press;
1995.
[39] Lee CC, Yoshida M, Frechet JMJ, Dy EE, Szoka FC. In vitro and
in vivo evaluation of hydrophilic dendronized linear polymers.
Bioconjugate Chem 2005;16:535–41.
[40] Gillies ER, Dy E, Frechet JMJ, Szoka FC. Biological evaluation of
polyester dendrimer: poly(ethylene oxide) ‘‘bow-tie’’ hybrids with
tunable molecular weight and architecture. Mol. Pharmaceutics
2005;2:129–38.
[41] Knauf MJ, Bell DP, Hirtzer P, Luo Z, Young JD, Katre NV.
Relationship of effective molecular size to systemic clearance in rats
of recombinant interleukin-2 chemically modified with water-soluble
polymers. J Biol Chem 1988;263:15064–70.
[42] Rennke HG, Patel Y, Venkatachalam MA. Glomerular filtration of
proteins: clearance of anionic, neutral, and cationic horseradish
peroxidase in the rat. Kidney Int 1978;13:278–88.
[43] Rennke HG, Venkatachalam MA. Effect of molecular configuration
on the fractional clearance of uncharged dextran and neutral horse
radish peroxidase in the rat. J Clin Invest 1979;63:713–7.
[44] Oliver JD, Anderson ST, Troy JL, Brenner BM, Deen WM.
Determination of glomerular size-selectivity in the normal rat with
Ficoll. J Am Soc Nephrol 1992;3:214–28.
[45] Brochard-Wyart F, de Gennes PG. Injection threshold for a star
polymer inside a nanopore. C.R. I’Acad Sci Ser II Univ
1996;323:473–9.
[46] Ogston AG, Preston BN, Wells JD. On the transport of compact
particles through solutions of chain-polymers. Proc R Soc London B
Biol Sci A 1973;333:297–309.
[47] Mehvar R. Recent trends in the use of polysaccharides for
improved delivery of therapeutic agents: pharmacokinetic and
pharmacodynamic perspectives. Curr Pharm Biotechnol 2003;4:
283–302.