Generation effect of Newkome dendrimers on cellular uptake

Article abstract of DOI:10.1016/j.polymer.2017.02.040

Poly(amide)-based dendrimers can be used as delivery scaffolds in conjunction with the cell-penetrating peptide gH625 derived from the glycoprotein of the Herpes Simplex virus type 1. In this contribution, we aim to isolate the optimal dendrimer generation for cellular uptake for Newkome type dendrimers conjugated with gH625. For this study, we synthesized generations zero to three of the Newkome dendrimer-gH625 bioconjugate. Fluorescent microscopy experiments showed that the second and third generations are the most efficient for cellular uptake with the second generation having the synthetic advantage. The optimal second generation can be used as an improved material for a dendrimer based delivery scaffold for peptide therapeutics.

Full text of DOI:10.1016/j.polymer.2017.02.040

Accepted Manuscript
Generation effect of Newkome dendrimers on cellular uptake
Elizabeth A. Kaufman, Rossella Tarallo, Anarita Falanga, Stefania Galdiero, Marcus
Weck
PII:
S0032-3861(17)30163-5
10.1016/j.polymer.2017.02.040
JPOL 19441
DOI:
Reference:
To appear in: Polymer
Received Date: 9 December 2016
Revised Date: 6 February 2017
Accepted Date: 10 February 2017
Please cite this article as: Kaufman EA, Tarallo R, Falanga A, Galdiero S, Weck M, Generation effect of
Newkome dendrimers on cellular uptake, Polymer (2017), doi: 10.1016/j.polymer.2017.02.040.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Generation Effect of Newkome Dendrimers on
Cellular Uptake
Elizabeth A. Kaufmana, Rossella Taralloa, Anarita Falangab, Stefania Galdierob,* and
Marcus Wecka*
aMolecular Design Institute, New York University, New York, NY 10003, USA
bDipartimento di Farmacia, Università degli Studi di Napoli Federico II, Via
Mezzocannone 16, Napoli 80134, Italy
Abstract: Poly(amide)-based dendrimers can be used as delivery scaffolds in conjunction
with the cell-penetrating peptide gH625 derived from the glycoprotein of the Herpes
Simplex virus type 1. In this contribution, we aim to isolate the optimal dendrimer
generation for cellular uptake for Newkome type dendrimers conjugated with gH625. For
this study, we synthesized generations zero to three of the Newkome dendrimer-gH625
bioconjugate. Fluorescent microscopy experiments showed that the second and third
generations are the most efficient for cellular uptake with the second generation having
the synthetic advantage. The optimal second generation can be used as an improved
material for a dendrimer based delivery scaffold for peptide therapeutics.
Introduction: Peptide therapeutics have shown promise in various systems but suffer
from drawbacks such as protease susceptibility and size limitations common for large
therapeutic agents. 1, 2 Therapeutics with molecular weights greater than 500 g/mol show
3
potential for the treatment of a variety of diseases ranging from HIV to cancer but
delivery of these drugs from the aqueous extracellular matrix across the amphiphilic
bilayer of the cell membrane into cells has proven challenging. 4, 5 Additionally, peptides
are vulnerable to proteases and can lead to immune responses in the body. 2 Conjugation

ACCEPTED MANUSCRIPT
of peptides to polymers can mitigate some of these negatives while allowing for longer
circulation times in vivo and increased bioavailability. 6, 7
One promising class of polymer for peptide ligation for biolomedical applications is
dendrimers. 8, 9, 10 Dendrimer growth is defined by generations, counting each branching
point as a new generation. 9 Increasing the number of termini potentially allows for a
higher local concentration of drugs either adsorbed in the dendrimer core or attached to
the termini. Synthetic complexity, however, also increases with generation. 11, 12 Often,
higher generation dendrimers are less perfect and have a dispersity above 1. In contrast to
linear polymers of the same composition, the radius of gyration of a dendrimer grows
linearly with generation while the intrinsic viscosity has a maximum value and then
decreases when the dendrimer becomes globular. 13 These properties are an advantage
when used in biological applications as increasing the size of the scaffold does not greatly
effect the viscosity of the intracellular matrix upon delivery. 13 Branched carriers have
been shown to be cleared from the kidneys more slowly than their linear counterparts,
resulting in longer circulation times giving dendrimers another potential advantage. 7, 14
Biological systems have been shown to be sensitive to many aspects of polymeric
scaffolds. Polymer size, functional density and shape have all been shown to effect cell
interactions with polymers. 15 , 16, 17 Thus, optimization of a polymer-peptide conjugate
requires careful study of the polymer’s activity in a biological application.
Dendrimer generations in vivo have been shown to have a marked effect on the
behavior of a dendritic drug delivery scaffold. Different generations often show
18, 19
differences in cell uptake and cell toxicity.
For example, higher generations of
poly(propyleneimine) showed better release of the drug Melphalan, but also a large
increase in toxicity. 20 Differences in tumor growth were shown to be negligible between
the fourth and fifth generation poly(propyleneimine) dendrimer even as toxicity
increased, demonstrating that there is an optimal dendrimer generation for delivery
vehicles. 20
The most widely studied dendrimer for delivery applications is poly(aminoamide)
(PAMAM), due to its easy availability and low cytotoxicity when its cationic nature is
21, 22
mitigated.
PAMAM dendrimers have been shown to be taken up into cells without
the need of cell-penetrating peptides when free amines are present on the termini,

ACCEPTED MANUSCRIPT
Figure 1. Previously synthesized second generation azidodendrimer and structure of
dye-alkyne modified gH625.
23
however the dendrimer then shows higher toxicity. Obviously, each of the dendrimer
systems used as delivery scaffolds for therapeutics, must be optimized for the desired
properties. 24, 25, 26
The bare dendrimer scaffold is often not enough to deliver cargo into a cell. Building
on previous work, our strategy to increase cellular uptake of the scaffold takes advantage
of the dendrimer scaffold ligated to the peptide gH625 (Figure 1), derived from a
27
segment of the glycoprotein H from Herpes Simplex virus type 1.
This peptide
sequence is able to enter the cell and deliver various cargos, a proposed mechanism
suggests the amphiphilic nature of gH625’s α helical architecture allows the interaction
with cellular membranes. 1 , 27, 28
We have previously demonstrated that compared to free gH625, attachment to the
termini of a second generation (G2) Newkome-type dendrimer scaffold greatly increases
cellular uptake with low cell toxicity up to 20 ꢀM. 29, 30 At that time, we also performed
cell viability assays to determine the optimal concentration of peptidodendrimer based on

ACCEPTED MANUSCRIPT
the concentration of peptide using UV-vis analysis. This concentration was used in our
study to allow for comparison between these findings and previously reported studies.
The uptake was previously measured using both fluorescence microscopy and flow
cytometry and it was shown in both cases that the peptidodendrimer had an advantage
29, 30
over free gH625.
The use of the G2 dendrimer was wholly arbitrary in our prior
research; no comparison was made between various generations of dendrimer. The initial
results suggested a potentially new peptide scaffold based on Newkome-type dendrimers
functionalized with gH625 as a cell penetrating peptide to deliver payload into HeLa and
Vero cells. As a first step towards the optimization of our cell penetrating scaffold, this
contribution investigates the generation dependence of our gH625 functionalized
Newkome-type delivery scaffold to determine which generation is most suited for further
study. We show that there is an
ideal size for our scaffold based on
cellular uptake, ease of synthesis
and maximization of cargo.
Results
Our dendrimers of choice are
poly(amide)-based with a 1→3
branching unit structurally derived
from dendrimers first reported by
31, 32
Newkome, Scheme 1.
All
dendrimers are synthesized from
the commercially available
bifunctional dendrons di-t-butyl-4-
[2-(t-butoxycarbonyl) ethyl]-4-
aminoheptanedicarboxylate
(aminotriester) and 4-(2-
carboxyethyl)-4-nitroheptanedioic
acid (nitrotriacid). The dendrons
Vfkhp h# 41# V| qwkhwlf # vf khp h# iru# wkh# iluvw#
j hqhudwlrq#d}lgrghqgulp hu1
are
coupled
using

ACCEPTED MANUSCRIPT
Figure 2. Structure of the Newkome type dendrimers: A) zeroth, B) first, C) second, and D)
third generation azidodendrimers.
carbodiimide/DIPEA peptide-coupling schemes. The nitro group is reduced subsequently
to an amine in order to yield a reactive terminus. The tert-butyl esters are hydrolyzed
yielding multiple reactive termini. This strategy can be repeated multiple times to yield
the dendrimer of the desired generation (Figure 1). A 3-azidopropylamine linker can be
coupled to each terminus of the dendrimer of interest giving a handle to attach alkyne-

ACCEPTED MANUSCRIPT
33, 34, 35
functionalized peptides using copper catalyzed 1,3-dipolar cycloaddition.
Alternatively, a dendrimer can be directly functionalized using peptide bond coupling
strategies. All dendrons are synthesized using iterations of the coupling and deprotecting
reactions and detailed syntheses and characterizations can be found in the supporting
1
13
information. The dendrimers are characterized by H and C NMR spectroscopies and
mass spectrometry.
To visualize the dendrimer in cell culture, a fluorescent tag with a butynyl handle was
synthesized in a single step by reacting 4-chloro-7-nitrobenzofuran with 1-amino-3-
butyne. The dendrimers were functionalized quantitatively with the fluorescent tag using
microwave-assisted copper catalyzed 1,3-dipolar cycloaddition. The gH625 peptide, was
synthesized and fluorescently labeled using standard Fmoc solid phase synthesis. An
alkynyl handle was added to the C terminus to allow for copper catalyzed 1,3-dipolar
cycloaddition. The dendrimers were functionalized with gH625 using again microwave
assisted copper catalyzed
1,3-dipolar cycloaddition.
The conversions ranged
from 72-100% depending
on dendrimer generation
as calculated by MALDI
(Figure 3).
With the functionalized
dendrimers of generations
0-3 in hand, we targeted
the effect of generation
Figure 3. MALDI of G0-G3 Peptidodendrimers Top
size on cellular uptake
left) G0 dendrimer with two peptides: 100% conversion;
using
HeLa
cells.
top right) G1 dendrimer functionalized with six and five
peptides: 83.3-100% conversion; bottom left) G2
dendrimer with 18 and 17 peptides: 94.4-100%
conversion; bottom right) G3 dendrimer with 54, 48, 44
and 39 peptides: 72.2-100% conversion.
Dynamic light scattering
(DLS) was run to probe
the solubility of the
dendrimer dye conjugates
at
the
targeted

ACCEPTED MANUSCRIPT
concentration to insure that there was not aggregation at experimentally relevant time
scales. The larger generations (G2-G3) showed no aggregation in 1% DMSO in buffer
and thus 20 ꢀM solution was used as in previous studies of the dendrimer dye conjugate.
DLS had previously been run on the G2 peptitodendrimer and thus solubility was not
retested. The cellular uptake of each generation of dye-dendrimer conjugate and
peptidodendrimer was explored by incubating a solution of the peptidodendrimer of
interest in cell medium and HeLa cells for 30 minutes, one hour, two hours, and four
hours with a 20 ꢀM solution of each generation of dendrimer relative to the dye. This was
calculated by using the extinction coefficient of the 7-nitrobenzofurazan (NBD) dye to
determine the concentration of the solution and the desired amount of each dendrimer dye
conjugate and peptidodendrimer. The sample was then freeze dried and a stock solution
for each cell study was made in 1% DMSO in Dulbecco’s modified Eagle’s medium.
After incubation, excess peptidodendrimer was washed away and the cells were imaged
using fluorescent microscopy (60× PlanFluor objective, NA 0.3, Eclipse TE 2000-U;
Nikon) and the fluorescent uptake was calculated (Figures 4 and 5).
The zero generation (G0) dendrimer-dye conjugate had full conversion with two dye
1
molecules per scaffold as calculated by H NMR. The G0 peptidodendrimer was fully
functionalized with two peptides per scaffold as calculated by MALDI (Figure 3). The
G0 dendrimer with dye was taken up more readily by the cells than the G0
peptidodendrimer likely due to the low molecular weight of the compound. The sum of
all fluorescent uptake for the zero generation dendrimer-dye conjugate was 996.2 +/- 92.1
rfus (relative fluorescence units) compared to 348.4 +/- 39.0 rfus for the
peptidodendrimer (Figure 6). The G0 dendrimer-dye conjugate was able to cross the
cellular membrane easily due to the small size of the scaffold. The result shows that the
G0 peptidodendrimer is not more readily taken up by cells and is therefore not a viable
scaffold for efficient cellular uptake.
The first generation (G1) dendrimer-dye conjugate yielded complete conversion with
1
six dye molecules per scaffold as calculated by H NMR. The G1 peptidodendrimer was
functionalized with four to six peptides per scaffold as calculated by MALDI. The G1
dendrimer-dye conjugate is taken up less readily by the cells than the G0 dendrimer-dye
conjugate. The G1 peptidodendrimer, however, is taken up by the cells more readily than

ACCEPTED MANUSCRIPT
the control dendrimer but also appears to deposit on the cell membrane leading to cell
death (Figure 4). The uptake of the G1 dendrimer also decreases from 194.6 rfus at one
hour to 134.1 rfus at two hours as the cells begin to contract. The sum of all fluorescent
uptake for the G1 dendrimer-dye conjugate was 125.7 +/- 15.6 rfus compared to 541.8 +/-
71.1 rfus for the peptidodendrimer (Figure 6). The aggregation of the G1
peptidodendrimer, despite the improved uptake, is therefore not a viable scaffold for
cellular uptake.
Figure 4. Microscopy images of cells at 30 min, 1 h, 2 h, and 4 h of incubation with the G0 or G1
dendrimer-dye and peptidodendrimer conjugates. Scale bars represent 50 µM.

ACCEPTED MANUSCRIPT
Figure 5. Microscopy images of cells at 30 min, 1 h, 2 h, and 4 h of incubation with the second
or third generation dendrimer-dye and peptidodendrimer conjugates. Scale bars represent 50 µM.
The G2 dendrimer-dye conjugate was functionalized with eighteen dye molecules per
1
scaffold as calculated by H NMR spectroscopy. The G2 peptidodendrimer was
functionalized with an seventeen to eighteen peptides per scaffold. The G2 dendrimer-
dye conjugate shows similar uptake to the G1. The peptidodendrimer, however, shows
significantly higher uptake than the G0 or G1 dendrimers. The uptake is shown to
continue to increase as the kinetic study progresses through the four hour time point
(Figure 5). The fluorescence of the G2 peptidodendrimer appears diffuse through the

ACCEPTED MANUSCRIPT
cytoplasm indicating that it is a viable delivery scaffold. The sum of all fluorescent
uptake for the G2 dendrimer-dye conjugate was 579.4 +/- 80.1 rfus compared to 2053.0
+/- 340.0 rfus for the peptidodendrimer (Figure 6). These results show that the G2
peptidodendrimer is a viable scaffold for efficient cellular uptake.
The third generation (G3) dendrimer-dye conjugate was functionalized with fifty-four
dye molecules per scaffold as calculated by 1H NMR spectroscopy. The G3
peptidodendrimer was functionalized with thirty nine to fifty three peptides per scaffold
as calculated by MALDI. The G3 dendrimer-dye conjugate is not readily taken up by the
cells. The G3 peptidodendrimer is taken up at a slower rate than the G2 dendrimer,
(Figure 5). The fluorescence of the G3 peptidodendrimer appears diffuse through the
cytoplasm. The sum of all fluorescent uptake for the G3 dendrimer-dye conjugate was
Figure 6. Raw fluorescent uptake dendrimer-dye conjugates and peptidodendrimer of all
dendrimer generations over three biological and nine technical replicates at time points 30 min, 1
h, 2 h, and 4 h.

ACCEPTED MANUSCRIPT
373.5 +/- 61.4 rfus compared to 1855.3 +/- 245.0 rfus for the peptidodendrimer (Figure
6). These results demonstrate that the G3 peptidodendrimer is also a viable scaffold for
efficient cellular uptake.
The raw relative fluorescence of each dendrimer-dye conjugate and peptidodendrimer
conjugate was measured over nine technical replicates, (Figure 6). The trend that emerges
with the raw cellular uptake date shows the previously studied G2 peptidodendrimer has
the highest uptake overall of peptide. The cell uptake results suggest that the optimal
generation for cell delivery is either the G2 or G3 generation dendrimer due to the high
uptake into cells. The different generations however have an exponential difference in the
number of peptides per scaffold thus the result indicates the number of peptides taken up
and not the number of dendrimers.
The maximum number of peptides on each dendrimer scaffold can normalize the raw
analysis, which allows to calculate the relative number of dendrimers taken up in each
cellular assay (Figure 7). In this analysis, the G0 peptidodendrimer appears to outperform
every
other
generation
of
dendrimer. This normalization,
however does not take into account
the difference in dendrimer loading
since it is normalized to peptide.
There is a nine times higher
concentration of G0 dendrimer in
solution compared to the G2
dendrimer. When the data are
again normalized to the loading of
G2 dendrimers, we return to the
original relative uptake and the G2
dendrimer again outperforms each
generation of peptidodendrimer
conjugate. This suggests that the
G2 dendrimer is the most efficient
for cellular uptake and that there is
Figure 5. Fluorescent uptake of dendrimer-dye
conjugates and peptidodendrimer for all dendrimer
generations over three biological and nine technical
replicates at time points 30 min, 1 h, 2 h, and 4 h
normalized by number of peptides per scaffold.

ACCEPTED MANUSCRIPT
not a significant advantage to the third generation dendrimer.
The difference in the analysis is dependent on what is considered the desirable cargo
within the study. If the dendrimer would be considered the active agent, then further
studies might be warranted at equal molar concentration of G0 and G2 dendrimer.
However, in our system, the peptide itself is the biologically active agent and a cell
penetrating peptide. For this reason, the G2 dendrimer still holds an advantage over the
G0 dendrimer. With the confirmation that the G2 peptidodentimer is the optimal
generation, the cell viability experiments previously run with the G2 dendrimer were not
29, 30
repeated.
The flow cytommetry studies reported in past works have shown the
dendrimer peptidodendrimer is able to fully cross the cellular membrane and not simply
deposit on the cell as is the free peptide.
The G2 dendrimer is significantly easier to synthesize and isolate than the third
generation. Additionally, the second generation has a slightly higher raw cell uptake
(albeit within the error range), thus making it the optimal generation Newkome-type
dendrimer for cellular uptake. Taking into account both ease of synthesis and biological
activity, the second generation outperforms the third across all criteria.
Conclusions
In this study, the cellular uptake of Newkome-type dendrimers as a function of
dendrimer generation was probed. All generations except for the G0 showed improved
uptake when conjugated to the cell-penetrating peptide gH625. The G1 peptidodendrimer
was able to enter the cells but also deposited on the cell membrane causing cell death.
The G2 and G3 peptidodendrimers were found to have significantly improved cellular
uptake over their dendrimer-dye controls. This optimization of the peptidodendrimer
allows further research into cellular uptake using only the G2 dendrimer with the addition
of cargo to be delivered into the cells.
Acknowledgement
EAK acknowledges a Margaret Strauss Kramer fellowship. We thank Professor Lara K.
Mahal, Jessica Gruskos and Christopher Viana for help with the cellular uptake studies.

ACCEPTED MANUSCRIPT
References
1. Galdiero, S.; Russo, F.; Falanga, A.; Cantisani, M. Biochemistry 2012, 51, 3121-
3128.
2. Veronese, F.; Paust, G. . Drug Discov. Today 2005, 10, 1451-1458.
3. Duncan, R. Nature 2003, 2, 347-360.
4. Torchilin, V.; Lukyanov, A. Therapeutic Focus 2003, 8, 259-266.
5. Allen, T. M.; Cullis, P. R. Science 2004, 303, 1818-1822.
6. Patri, A.; Majoros, I.; Baker, J. Curr. Opin. Chem. Biol. 2002, 6, 466-471.
7. Maedea, H. J. Controlled Release 2012, 164, 138-144.
8. Newkome, G. R.; Moorefield, C. N.; Vögtle, F. In Dendrimers and Dendrons:
Concepts, Synthesis, Applications; Wiley-VCH: Weinheim, 2001.
9. Vögtle, F.; Richardt, G.; Werner, N. Functional Dendrimers; Wiley-VCH Verlag
GmbH & Co. KGaA, 2009; pp 49-79.
10. Gillies, E.; Fréchet, J. Drug Discov. Today 2005, 10, 35-43.
11. Ornelas, C.; Weck, M. Chem. Commun. 2009, 5710-5712.
12. Dykes, G. M. J. Chem. Technol. Biotechnol. 2001, 76, 903-918.
13. Mourey, T. H.; Turner, S. R.; Rubinstein, M.; Fréchet, J. M. J.; Hawker, C. J.;
Wooley, K. Macromolecules 1992, 25, 2401-2406.
14. Ornelas, C.; Pennell, R.; Liebes, L.; Weck, M. Org. Lett. 2011, 13, 976-979.
15. Kenawy, E. R.; Worley, S. D.; Broughton, R. Biomacromolecules 2007, 8, 1359-
1384.
16. Nguyen, T.; Paluck, S.; McGahran, A.; Maynard, H. Biomacromolecules 2015, 16,
2684-2692.
17. Niki, Y.; Ogaea, M.; Makiura, R.; Mgata, Y.; Kojima, C. Nanomedicine 2015, 11,
2119-2127.
18. Jevprasesphant, R.; Penny, J.; Jalala, R.; Attwood, D.; McKeown, N. B.;
D’Emanuele, A. Int. J. Pharm. 2003, 252, 263–266.
19. Zhang, X.; Intra, J.; Salem, A. Bioconjugate Chem. 2007, 18, 2068-2076.
20. Kesharwani, P.; Tekade, R.; Jain, N. Biomaterials 2014, 35, 5539-5548.
21. Esfand, R.; Tomalia, D. A. Drug Discov. Today 2001, 6, 427-436.
22. Márquez-Miranda, V.; Peñaloza, J.; Araya-Durán, I.; Reyes, R.; Vidaurra, S.;
Romero, V.; Fuentes, J.; Céric, F.; Velásquez, L.; Gonzalez-Nilo, F.; Otero, C.
Nanoscale Res. Lett. 2016, 11.
23. Brazeau, G. A.; Attia, S.; Poxon, S.; Hughes, J. A. Pharm. Res. 1998, 680–684.
24. Berna, M.; Dalzoppo, D.; Pasut; G.; Manunta, M.; Izzo, L.; Jones, A.; Duncan, R.;
Veronese, F. Biomacromolecules 2006, 7, 146-153.
25. Kesharwani, P.; Jain, K.; Jain, N. Prog. Polym. Sci. 2014, 39, 268-307.
26. Malik, N.; Wiwattanapatapee, R.; Klopsch, R.; Lorenz, K.; Frey, H.; Weener, J.;
Meijer, E.; Paulus, W.; Duncan, R. J. Controlled Release 2000, 65, 133-148.
27. Galdiero, S.; Falanga, A.; Morelli, G.; Galdiero, M. Biochem. Biophys. Acta 2015,

ACCEPTED MANUSCRIPT
1848, 16-25.
28. Tarallo, R.; Accardo, A.; Falanga, A.; Guarnieri, D.; Vitiello, G.; Netti, P.; D'Errico,
G.; Morelli, G.; Galdiero, S. Chem. Eur. J. 2011, 17, 12659-12668.
29. Carberry, T.; Tarallo, R.; Falanga, A.; Finamore, E.; Galdiero, M.; Weck, M.;
Galdiero, S. Chem. Eur. J. 2012, 18, 13678-13685.
30. Tarallo, R.; Carberry, T.; Falanga, A.; Vitiello, M.; Galdiero, S.; Galdiero, M.; Weck,
M. Int. J. Nanomedicine 2013, 8, 521-534.
31. Newkome, G.; Shreiner, C. Chem. Rev. 2010, 110, 6338-6442.
32. Newkome, G. R.; Kotta, K. K.; Moorefield, C. N. J. Org. Chem. 2005, 70, 4893-
4896.
33. Kolb, H.; Finn, M.; Sharpless, B. Int. Ed. Angew. Chem. Int. Ed. 2001, 40, 2004-
2021.
34. Kolb, H.; Sharpless, B. Drug Discov. Today 2003, 8, 1128-1137.
35. Lutz, J.; Zarafshani, Z. Adv. Drug Deliv. Rev. 2008, 60, 958-970.
36. Galdiero, S.; Russo, F.; Falanga, A.; Cantisani, M. Biochemistry 2012, 51, 3121-
3128.

ACCEPTED MANUSCRIPT
•
•
Four generations of dendrimer peptide conjugates were synthesized.
Each peptidodendrimer is able to enter HeLa cells via a combination of active
and passive pathways.
•
The second generation (G2) peptidodendrimer is the optimal generation for
intracellular delivery.