Poly Ethoxy Ethyl Glycinamide (PEE-G) Dendrimers: Dendrimers Specifically Designed for Pharmaceutical Applications

Article abstract of DOI:10.1002/cmdc.201600270

Poly ethoxy ethyl glycinamide (PEE-G) dendrimers have been specifically designed and synthesized with the aim of providing a readily available dendrimer scaffold that can be used to make products that can meet the stringent requirements of pharmaceutical applications. The synthesis has been refined to produce dendrimers that are of high HPLC purity. The suitability of PEE-G dendrimers for their designed use has been verified by subsequent measurements to demonstrate that they are of high stability, high aqueous solubility, low cytotoxicity, low immunogenicity and with low in vivo toxicity in an escalating-dose rat study. PEE-G dendrimers therefore provide a useful scaffold for researchers wanting to develop dendrimer-based drug candidates.

Full text of DOI:10.1002/cmdc.201600270

DOI: 10.1002/cmdc.201600270
Communications
Poly Ethoxy Ethyl Glycinamide (PEE-G) Dendrimers:
Dendrimers Specifically Designed for Pharmaceutical
Applications
Steven Toms,[b] Susan M. Carnachan,[a] Ian F. Hermans,[c] Keryn D. Johnson,[b]
Ashna A. Khan,[b] Suzanne E. O’Hagan,[d] Ching-Wen Tang,[c] and Phillip M. Rendle*[a]
Poly ethoxy ethyl glycinamide (PEE-G) dendrimers have been
specifically designed and synthesized with the aim of provid-
ing a readily available dendrimer scaffold that can be used to
make products that can meet the stringent requirements of
pharmaceutical applications. The synthesis has been refined to
produce dendrimers that are of high HPLC purity. The suitabili-
ty of PEE-G dendrimers for their designed use has been veri-
fied by subsequent measurements to demonstrate that they
are of high stability, high aqueous solubility, low cytotoxicity,
low immunogenicity and with low in vivo toxicity in an escalat-
ing-dose rat study. PEE-G dendrimers therefore provide
a useful scaffold for researchers wanting to develop dendri-
mer-based drug candidates.
the ability to consistently produce material with reproducible
characteristics. If material cannot be made with the same or
higher purity than previously, then safety data from toxicity
trials can be of little value.
Robust measurement of the purity of dendrimers is chal-
lenging.[5] Mass spectrometric data are useful for showing the
presence of impurities if a good method has been developed
that shows a strong resolved molecular ion, but is not a good
tool to accurately quantify impurities.[6] While size-exclusion
chromatography is more quantitative, it is not a good method
to detect subtle structural imperfections, for example, where
one terminal arm of the dendrimer is miss-capped. These im-
purities will be of very similar size to the product and so are
unlikely to be resolved. Because of the highly repetitive struc-
ture of dendrimers, NMR is not good for purity evaluation
when considering product-like impurities and is not a very sen-
sitive method for detecting low-level impurities, especially for
larger-generation dendrimers.
Dendrimers have been developed for many uses since the first
reports,[1] including for use in the pharmaceutical field.[2] De-
spite the many efforts in this area, there have only been a few
dendrimer products that have entered clinical trials,[3] and even
fewer have reached the market. We have developed a new
dendrimer scaffold that has been specifically designed to meet
the stringent requirements of pharmaceutical applications.
Of the polymeric or macromolecular materials available, den-
drimers are favorable for drug development because of their
discrete definable structure. However, to progress through the
regulatory process, regulators require demonstration of control
over the synthetic chemistry process backed up by informative
analytical methods.[4] Lack of control in the synthetic chemistry
can lead to variable efficacy and toxicity of the product. Lack
of ability to show purity and more importantly the level and
nature of impurities leads to concerns by the regulators over
HPLC is the key method for the measurement of purity of
dendrimer substances and should be used throughout the syn-
thesis to verify control of the chemical process. There are only
limited examples[7] of HPLC being used to measure dendrimer
product purity. Reymond et al. regularly synthesize peptide
dendrimers via solid-phase chemistry which they isolate to
high purity using preparative HPLC.[8]
We felt that there is a need for a readily available dendrimer
scaffold that fulfills the requirements for pharmaceutical appli-
cations and that can be made on scale.[3,9] This scaffold could
then be used in drug discovery with the knowledge that the
underlying dendrimer is suitable and unlikely to raise hurdles
when measuring the purity, stability, and toxicity properties of
the resulting active pharmaceutical ingredients (APIs). We
therefore report herein a new dendrimer scaffold that we have
specifically designed, synthesized, and evaluated as a tool for
researchers carrying out dendrimer-based drug discovery.
The criteria that we gave ourselves for a dendrimer scaffold
to be of use as a tool to make therapeutic drugs were that it
must have: low toxicity to allow for a suitable, safe dosing
window; high stability with measurable degradation, not pro-
ducing degradants likely to be toxic; low cost of goods and to
be able to be made by an efficient, scalable synthesis; high
HPLC-measurable purity; suitable solubility in biologically rele-
vant media; and amine or carboxylic acid termini available for
convenient further modification by subsequent coupling reac-
tions.
[a] Dr. S. M. Carnachan, Dr. P. M. Rendle
Victoria University of Wellington,
PO Box 33436, Petone 5046 (New Zealand)
E-mail: Phillip.rendle@vuw.ac.nz
[b] Dr. S. Toms, Dr. K. D. Johnson, Dr. A. A. Khan
Callaghan Innovation, PO Box 31310, Lower Hutt 5040 (New Zealand)
[c] Dr. I. F. Hermans, Dr. C.-W. Tang
Malaghan Institute of Medical Research,
PO Box 7060, Wellington 6242 (New Zealand)
[d] Dr. S. E. O’Hagan
Centre for Integrated Preclinical Drug Development,
Faculty of Medicine and Biomedical Sciences,
The University of Queensland, St. Lucia Campus, QLD 4072 (Australia)
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under http://dx.doi.org/10.1002/
cmdc.201600270.
ChemMedChem 2016, 11, 1 – 5
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
Scheme 1. Synthesis of PEE-G dendrimers. a) TsOH·Gly-OBn, nBu4NI, Et3N, dioxane, 908C, 58%; b) TsOH, THF, 80%; c) H2, Pd/C, MeOH, 96% for 4; d) 3, HBTU,
HOBT, DIPEA, DMF. Yield over two steps, d) then c): 5, 87%; 6, 84%; 7, 75%; e) 3m HCl, THF; f) succinic anhydride, py, DMSO, 76% over two steps.
The new form of dendrimer that we have developed is
shown in Scheme 1. It is built up from the readily available raw
materials of the amino acid glycine and 2-(2-aminoethoxy)e-
thanol to give scaffolds that we have called poly ethoxy ethyl
glycinamide (PEE-G) dendrimers. This dendrimer structure is
similar to that reported by Zanini and Roy,[10] except PEE-G
dendrimers have an extra methylene and ether oxygen in the
linker between branching points to increase water solubility.
PEE-G dendrimers are low-density dendrimers,[3] with long dis-
tances between branching points, thereby having a longer
‘reach’ than some other types of dendrimers of the same gen-
eration. These ‘dendrons’ are produced with orthogonal pro-
tection of the central carboxylic acid and the external amines
with commonly used, easily removed protecting groups to
allow flexibility in elaboration strategies. The preparation and
properties of PEE-G dendrimers are described below.
were investigated, but it was found for our purposes, the best
results were achieved with a benzyl protecting group for the
carboxy moiety and tert-butyloxycarbonyl (Boc) protecting
groups for the amines. Therefore, key building block 2 was
prepared from a mesylate derivative of 2-(2-aminoethoxy)etha-
nol (1)[11] by nucleophilic substitution with benzyl-protected
glycine[12] (Scheme 1). p-Toluenesulfonic acid was employed to
remove the Boc groups from building block 2. The use of p-tol-
uenesulfonic acid leads to poor atom economy; however, the
tosylate salt 3 produced is crystalline, resulting in a simple
crystallization purification to produce a high-purity solid. The
benzyl protecting group can be removed from building block
2 using standard hydrogenolysis conditions to afford carboxyl-
ic acid 4.
PEE-G dendrimer-based compounds 5–7 were synthesized
by the repeated sequence of amide coupling followed by hy-
drogenolysis (Scheme 1). For the amide formation, our initial
attempts included using N,N,N’,N’-tetramethyl-O-(1H-benzotria-
zol-1-yl)uroniumhexafluorophosphate (HBTU) and N,N-diisopro-
pylethylamine (DIPEA) in N,N-dimethylformamide (DMF). It is
difficult to separate subtle dendrimer structural imperfections
by chromatography, and so with the aim of increasing purity
and chemical control of the process, each reaction product
Synthesis: A convergent synthesis was planned to aid in ob-
taining high-purity products. In convergent dendrimer synthe-
sis, incomplete coupling reactions give contaminants which
are significantly structurally different from the desired product
and so are easier to separate. Orthogonal protecting groups
were desired that could be cleanly and simply deprotected in
very high yield in a scalable manner. Several protecting groups
&
ChemMedChem 2016, 11, 1 – 5
www.chemmedchem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
Table 1. Forced degradation of PEE-G dendrimer 9.
Conditions
Degradation [%]
Degradation observed[a]
24 h neat and in PBS at 408C
24 h in aqueous 1m HCl at 208C
24 h in aqueous 1m NaOH at 208C
24 h in aqueous 1% H2O2 at 208C
24 h neat at 808C
0
None
12
30
63
41
Amide hydrolysis, loss of succinic acid
Amide hydrolysis, loss of succinic acid
N-oxide formation, N-alkyl bond cleavage
Anhydride formation (dehydration)
[a] As determined by LC–MS analysis.
was interrogated by LC–MS. One of the higher-level impurities
from the coupling reactions was the addition of a guanidine to
the product; HBTU is a known guanidylation agent.[13] Addition
of 1-hydroxybenzotriazole (HOBT) in stoichiometric amounts
minimized the formation of this by-product, thereby increasing
the yield and purity of the products.
Stability. An active pharmaceutical ingredient ideally re-
quires high measurable stability at ambient conditions. PEE-G
dendrimer 9 was exposed to forced degradation conditions to
determine the magnitude of conditions required to effect deg-
radation. The amount of degradation was determined by HPLC
analysis and degradants determined by LC–MS. Detection of
expected degradants demonstrates that the HPLC method is
a good indicator of stability and that purities measured by this
method are valid. The material was separately exposed to in-
creasing acid, base, oxidation and temperature until significant
degradation was observed (Table 1). The results show that the
predictable degradants of PEE-G dendrimer 9 can be detected
and also that this material has good stability with 59% still
intact after 24 h at 808C. Any API made from the PEE-G scaf-
fold would need to undergo a full separate stability trial.
Solubility. PEE-G dendrimers generally have good water sol-
ubility with PEE-G dendrimer 9 for example being soluble in
phosphate-buffered saline (PBS) at ambient temperature at
>400 mgmLÀ1.
Each protected dendrimer generation can be converted into
an amine-terminated dendrimer and then to a carboxylic acid
terminated dendrimer as shown for dendrimer 7 (Scheme 1).
Deprotection under acidic conditions provides polyamine 8,
which was reacted with succinic anhydride to produce the sev-
enteen-carboxylate species 9. Stability, solubility and toxicity
experiments were then carried out to see if the PEE-G dendri-
mer scaffold met the criteria we had set for it to be a suitable
tool in the synthesis of APIs. PEE-G dendrimer 9 was used in
toxicity assays, as primary amines are often protonated at
physiological pH, forming in the case of polyamine 8, a polycat-
ionic species which can have inherent toxicity unrelated to the
dendrimer scaffold. Tertiary amines are generally observed to
be less toxic.[9,14]
Cytotoxicity. The cytotoxicity of PEE-G dendrimer 9 was
measured by observing the viability of sheep spleen cells at
36.58C after 24- and 48-hour exposure to PEE-G dendrimer 9
at 0.01, 0.1, and 1 mgmLÀ1 in PBS. No significant change in cell
viability was observed relative to the PBS vehicle alone for all
three concentrations, showing that PEE-G dendrimer 9 has low
cytotoxicity and compares favorably with other dendrimers.[15]
Immunogenicity. Immune activation properties of PEE-G
dendrimer 9 were measured by intravenous injection of 20 mg
of the compound in 200 mL PBS into the tail vein of C57BL/6
mice. This corresponds to ~1000 mgkgÀ1. No activation of anti-
gen-presenting cells was observed as measured by upregula-
tion of CD86 on dendritic cells or B cells compared with the
vehicle. No increase in cytokines (IL-1b, IL-10, GM-CSF, IL-5,
TNF-a, IL-4, IFNg, and IL-2) was observed in the blood as com-
pared with vehicle. The adjuvant a-GalCer[16] was used as a pos-
itive control. Therefore, PEE-G dendrimer 9 has low immunoge-
nicity at 1000 mgkgÀ1 in mice.
Purity. Protected PEE-G dendrimer cores are routinely syn-
thesized to >90% purity (HPLC with charged aerosol detector
(CAD)). The HPLC chromatograms are shown for PEE-G den-
drimers 4, 5, 6, and 7 in Figure 1. Each generation is resolved,
and minor imperfections in the structure could be readily de-
tected as demonstrated in the stability tests below and the ob-
servance of single-point guanidylation as mentioned above.
These products can potentially bind metals, and so PEE-G den-
drimer 7 was analyzed for palladium by ICPMS after acid diges-
tion. The level of palladium in this material was below the de-
tection limit of 1 ppm.
In vivo toxicity. Further possible toxicity properties of PEE-G
dendrimer 9 were measured by carrying out an escalating-
dose study in male and female Sprague–Dawley rats using in-
traperitoneal administration. Doses of PEE-G dendrimer 9 at
17.5, 55, 175, 550, and 1000 mgkgÀ1 in PBS were administered
at 48-hour intervals starting with the lowest dose on day 0.
Subsequent increasing doses were administered to the same
animals on study days 2, 4, 6, and 8. Clinical observations, food
consumption, and body weights were recorded prior to treat-
Figure 1. Purity of PEE-G dendrimers by HPLC-CAD (see the Supporting In-
formation for extra information on conditions used).
ChemMedChem 2016, 11, 1 – 5
www.chemmedchem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
Keywords: dendrimers · drug discovery · purity · stability ·
toxicity
[1] a) E. Buhleier, W. Wehner, F. Vçgtle, Synthesis 1978, 155–158; b) D. A.
Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S. Martin, J. Roeck, J.
Ryder, P. Smith, Polym. J. 1985, 17, 117–132; c) C. Hawker, J. M. J. Fre-
chet, J. Chem. Soc. Chem. Commun. 1990, 1010–1013.
[2] a) L. Rçglin, E. H. M. Lempens, E. W. Meijer, Angew. Chem. Int. Ed. 2011,
50, 102–112; Angew. Chem. 2011, 123, 106–117; b) S. Mignani, S. El Kaz-
zouli, M. Bousmina, J.-P. Majoral, Adv. Drug Delivery Rev. 2013, 65, 1316–
1330; c) R. M. Kannan, E. Nance, S. Kannan, D. A. Tomalia, J. Intern. Med.
2014, 276, 579–617; d) C. C. Lee, J. A. MacKay, J. M. J. Frechet, F. C.
Szoka, Nat. Biotechnol. 2005, 23, 1517–1526; e) A. R. Menjoge, R. M.
Kannan, D. A. Tomalia, Drug Discovery Today 2010, 15, 171–185; f) W.
Wijagkanalan, S. Kawakami, M. Hashida, Pharm. Res. 2011, 28, 1500–
1519.
Figure 2. Body weight and food consumption of male and female Sprague–
Dawley rats administered five escalating intraperitoneal bolus doses on days
0, 2, 4, 6, and 8 of PEE-G dendrimer 9 in PBS followed by a two-day observa-
tion period. A) Absolute mean (ÆSD) body weight. B) Absolute average food
intake with administration days marked.
[3] S. Svenson, Chem. Soc. Rev. 2015, 44, 4131–4144.
[4] a) R. Gaspar, R. Duncan, Adv. Drug Delivery Rev. 2009, 61, 1220–1231;
b) B. S. Zolnik, N. Sadrieh, Adv. Drug Delivery Rev. 2009, 61, 422–427.
[5] K. Vetterlein, U. Bergmann, K. Bꢁche, M. Walker, J. Lehmann, M. W. Linsc-
heid, G. K. E. Scriba, M. Hildebrand, Electrophoresis 2007, 28, 3088–
3099.
[6] a) J. C. Hummelen, J. L. J. Van Dongen, E. W. Meijer, Chem. Eur. J. 1997,
3, 1489–1493; b) T. Felder, C. A. Schalley, H. Fakhrnabavi, O. Lukin,
Chem. Eur. J. 2005, 11, 5625–5636; c) R. Giordanengo, M. Mazarin, J.
Wu, L. Peng, L. Charles, Int. J. Mass Spectrom. 2007, 266, 62–75; d) B.
Baytekin, N. Werner, F. Luppertz, M. Engeser, J. Brꢁggemann, S. Bitter, R.
Henkel, T. Felder, C. A. Schalley, Int. J. Mass Spectrom. 2006, 249–250,
138–148.
[7] a) R. Rupp, S. L. Rosenthal, L. R. Stanberry, Int. J. Nanomed. 2007, 2,
561–566; b) B. Huang, S. Tang, A. Desai, K.-H. Lee, P. R. Leroueil, J. R. Ba-
ker Jr, Polymer 2011, 52, 5975–5984; c) I. Teo, S. M. Toms, B. Marteyn,
T. S. Barata, P. Simpson, K. A. Johnston, P. Schnupf, A. Puhar, T. Bell, C.
Tang, M. Zloh, S. Matthews, P. M. Rendle, P. J. Sansonetti, S. Shaunak,
EMBO Mol. Med. 2012, 4, 866–881; d) X. Zhang, Z. Zhang, X. Xu, Y. Li, Y.
Li, Y. Jian, Z. Gu, Angew. Chem. Int. Ed. 2015, 54, 4289–4294; Angew.
Chem. 2015, 127, 4363–4368; e) S. van der Wal, Y. Mengerink, J. C.
Brackman, E. M. M. de Brabander, C. M. Jeronimus-Stratingh, A. P. Bruins,
J. Chromatogr. A 1998, 825, 135–147.
[8] a) J.-L. Reymond, T. Darbre, Org. Biomol. Chem. 2012, 10, 1483–1492;
b) P. Geotti-Bianchini, T. Darbre, J.-L. Reymond, Org. Biomol. Chem. 2013,
11, 344–352; c) R. U. Kadam, M. Bergmann, D. Garg, G. Gabrieli, A.
Stocker, T. Darbre, J.-L. Reymond, Chem. Eur. J. 2013, 19, 17054–17063;
d) M. Stach, T. N. Siriwardena, T. Kçhler, C. van Delden, T. Darbre, J.-L.
Reymond, Angew. Chem. Int. Ed. 2014, 53, 12827–12831; Angew. Chem.
2014, 126, 13041–13045.
ment and daily during the treatment period. Terminal examina-
tions at the end of the study included clinical pathology (hem-
atology and clinical chemistry), gross pathology, and organ
weight analyses.
No mortality or morbidity was observed during the study,
and the treatments with PEE-G dendrimer 9 were well tolerat-
ed. There were findings in the female rats of minor and transi-
ent decreases in body weight from day 2 onward with corre-
sponding decreases in average food intake (Figure 2). However,
the transient decreases did not correlate well with the adminis-
tration cycle, and there was an overall increase in body weight
for all animals from day 0 to the termination on day 10. All
other findings were considered incidental and not related to
treatment. The acute tolerated dose for PEE-G dendrimer 9 in
this study with intraperitoneal administration to adult Spra-
gue–Dawley rats is identified to be at least 1000 mgkgÀ1
.
In summary, we have developed a new dendrimer scaffold
that has been specifically designed with pharmaceutical appli-
cations in mind. A range of generations have been prepared
on a multi-gram scale to high HPLC purity, and the scaffold
has been shown to have high stability, aqueous solubility, and
low inherent toxicity. We now wish to develop variations of
this scaffold and also demonstrate its use as a building block
in real drug applications.
[9] U. Boas, P. M. H. Heegaard, Chem. Soc. Rev. 2004, 33, 43–63.
[10] D. Zanini, R. Roy, J. Org. Chem. 1996, 61, 7348–7354.
[11] Y.-S. Kim, K. M. Kim, R. Song, M. J. Jun, Y. S. Sohn, J. Inorg. Biochem.
2001, 87, 157–163.
[12] M. A. Brimble, N. S. Trotter, P. W. R. Harris, F. Sieg, Bioorg. Med. Chem.
2005, 13, 519–532.
[13] C. A. G. N. Montalbetti, V. Falque, Tetrahedron 2005, 61, 10827–10852.
[14] D. Fischer, Y. Li, B. Ahlemeyer, J. Krieglstein, T. Kissel, Biomaterials 2003,
24, 1121–1131.
Acknowledgements
[15] a) H. B. Agashe, T. Dutta, M. Garg, N. K. Jain, J. Pharm. Pharmacol. 2006,
58, 1491–1498; b) K. Jain, P. Kesharwani, U. Gupta, N. K. Jain, Int. J.
Pharm. 2010, 394, 122–142.
[16] T. Kawano, J. Cui, Y. Koezuka, I. Toura, Y. Kaneko, K. Motoki, H. Ueno, R.
Nakagawa, H. Sato, E. Kondo, H. Koseki, M. Taniguchi, Science 1997, 278,
1626–1629.
We thank the Industrial Research Ltd. Capability Fund and the
Ministry of Business, Innovation, and Employment Pre-Seed Accel-
erator Fund for funding. We thank Angela Raboczyj, Kelly Swee-
ney, Yuen Chiew Dieu, and Dr. Cora Lauat at TetraQ, the Centre
for Integrated Preclinical Drug Development, University of
Queensland for their contribution to the experimental work re-
quired for the escalating dose study.
Received: May 27, 2016
Revised: June 12, 2016
Published online on && &&, 0000
&
ChemMedChem 2016, 11, 1 – 5
www.chemmedchem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

COMMUNICATIONS
Scaffolds to build on: Poly ethoxy ethyl
glycinamide (PEE-G) dendrimers have
been made to high HPLC purity and are
shown to have high stability and low
cytotoxicity, immunogenicity, and in
vivo toxicity. This makes them the ideal
tools for dendrimer-based pharmaceuti-
cal drug discovery.
S. Toms, S. M. Carnachan, I. F. Hermans,
K. D. Johnson, A. A. Khan, S. E. O’Hagan,
C.-W. Tang, P. M. Rendle*
&& – &&
Poly Ethoxy Ethyl Glycinamide (PEE-G)
Dendrimers: Dendrimers Specifically
Designed for Pharmaceutical
Applications
ChemMedChem 2016, 11, 1 – 5
www.chemmedchem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ