Bioconjugates of Co(III) complexes with Schiff base ligands and cell penetrating peptides: Solid phase synthesis, characterization and antiproliferative activity

Article abstract of DOI:10.1016/j.jinorgbio.2020.111041

In this work we synthesized a chelating Schiff base by a single condensation of salicylaldehyde with 3,4-diamino benzoic acid (1). This ligand was used further for complexation to CoCl₂·6H₂O under nitrogen. In the next step, three six-coordinate Co(III) complexes were synthesized by coordinating this complex with imidazole (2), 2-methyimidazole (3) and N-Boc-L-histidine methyl ester (4) (Boc: tert.-butoxycarbonyl) in axial positions with simultaneous oxidation of Co(II) to Co(III) under ambient environment. All Co(III) complexes were characterized by multinuclear NMR spectroscopy (¹H, ¹³C and ⁵⁹Co NMR), FT-IR, mass spectrometry and HPLC. The Co(III) complexes were conjugated to three different cell penetrating peptides: FFFF (P1), RRRRRRRRRGAL (P2) and FFFFRRRRRRRRRGAL (P3). Standard solid-phase peptide chemistry was used for the synthesis of cell penetrating peptides. Coupling of N-terminal peptides with the cobalt complexes, possessing a carboxylic group on the tetradentate Schiff base ligand, afforded Co(III)-peptide bioconjugates, which were purified by semi-preparative HPLC and characterized by analytical HPLC and mass spectrometry. The antiproliferative activity of the synthesized compounds was studied against different human tumour cell lines: lung cancer A549, liver cancer HepG2 and normal human fibroblasts GM5657T, in comparison with the activity of cisplatin as a reference drug. The bioconjugate 21 containing the Co complex 4 and the combined phenylalanine and polyarginine cell penetrating sequence P3 shows better activity against the liver cancer line HepG2 than the parent Co(III) complex 4.

Full text of DOI:10.1016/j.jinorgbio.2020.111041

Journal Pre-proof
Bioconjugates of co(III) complexes with Schiff Base ligands and
cell penetrating peptides: Solid phase synthesis, characterization
and Antiproliferative activity
Dariusz Śmiłowicz, NilsMetzler-Nolte
PII:
S0162-0134(20)30069-6
DOI:
https://doi.org/10.1016/j.jinorgbio.2020.111041
JIB 111041
Reference:
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
20 October 2019
17 February 2020
17 February 2020
Please cite this article as: D. Śmiłowicz and NilsMetzler-Nolte, Bioconjugates of co(III)
complexes with Schiff Base ligands and cell penetrating peptides: Solid phase synthesis,
characterization and Antiproliferative activity, Journal of Inorganic Biochemistry (2020),
https://doi.org/10.1016/j.jinorgbio.2020.111041
This is a PDF file of an article that has undergone enhancements after acceptance, such
as the addition of a cover page and metadata, and formatting for readability, but it is
not yet the definitive version of record. This version will undergo additional copyediting,
typesetting and review before it is published in its final form, but we are providing this
version to give early visibility of the article. 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.
© 2020 Published by Elsevier.

Journal Pre-proof
Bioconjugates of Co(III) Complexes with Schiff Base Ligands and Cell
Penetrating Peptides: Solid Phase Synthesis, Characterization and
Antiproliferative Activity
Dariusz Śmiłowicz, NilsMetzler-Nolte*
InorganicChemistry I – BioinorganicChemistry
Ruhr-University Bochum, Universitätsstraße 150, 44801 Bochum, Germany
E-mail: nils.metzler-nolte@rub.de; Fax: +49 234 321 4378
Keywords: Bioconjugates, Cell Targeting Peptides, Cobalt(III), Medicinal Inorganic
Chemistry, Schiff base ligands, Solid Phase Peptide Synthesis
Abstract
In this work we synthesized a chelating Schiff base by a single condensation of
salicylaldehyde with 3,4-diamino benzoic acid (1). This ligand was used further for
complexation to CoCl2·6H2O under nitrogen. In the next step, three six-coordinate Co(III)
complexes were synthesized by coordinating this complex with imidazole (2), 2-
methyimidazole (3) and N-Boc-L-histidine methyl ester (4) (Boc: tert.-butoxycarbonyl) in
axial positions with simultaneous oxidation of Co(II) to Co(III) under ambient environment.
13
All Co(III) complexes were characterized by multinuclear NMR spectroscopy (1H, C and
59Co NMR), FT-IR, mass spectrometry and HPLC. The Co(III) complexes were conjugated to
three different cell penetrating peptides: FFFF (P1), RRRRRRRRRGAL (P2) and
FFFFRRRRRRRRRGAL (P3). Standard solid-phase peptide chemistry was used for the
synthesis of cell penetrating peptides. Coupling of N-terminal peptides with the cobalt
complexes, possessing a carboxylic group on the tetradentate Schiff base ligand, afforded
Co(III)-peptide bioconjugates, which were purified by semi-preparative HPLC and
characterized by analytical HPLC and mass spectrometry. The antiproliferative activity of the
synthesized compounds was studied against different human tumor cell lines: lung cancer
A549, liver cancer HepG2 and normal human fibroblasts GM5657T, in comparison with the
activity of cisplatin as a reference drug. The bioconjugate 21 containing the Co complex 4 and
the combined phenylalanine and polyarginine cell penetrating sequence P3 shows better
activity against the liver cancer line HepG2 than the parent Co(III) complex 4.
1

Journal Pre-proof
Introduction
Cobalt is an essential trace element in the human body where it exists exclusively in the form
of vitamin B12 (cobalamin), which is a cofactor for a number of enzymes, like isomerases,
methyl transferases or dehalogenases.[1] Further, cobalamin takes part in creating
neurotransmitters and stimulates the formation of erythrocytes in bone marrow.[2] Deficiency
of cobalt is strongly related to the vitamin B12 level and causes anaemia and thyroid
hypofunction.[3],[4]
Recently, other cobalt compounds have gained extended attention in the field of medicinal
inorganic chemistry, especially in the search for new anticancer drug candidates.[5] The
severe side effects of platinum-based anticancer drugs have forced researchers towards
finding metal-based anticancer drugs as alternatives.[6] From a chemical and physical point of
view cobalt, similar to platinum, adopts a wide variety of coordination numbers, geometries,
oxidation states, and ligand binding affinities, which makes cobalt chemistry thoroughly
investigated, not only but also in the field of new anticancer agents.[7], [8] Interestingly,
despite the fact that cobalt(III) complexes possess the same electron configuration (d6) as
platinum(IV) prodrugs, cobalt(III) complexes exhibit a different mode of action from
platinum-based anticancer drugs.[9] The very inert oxidised cobalt(III) state is reduced to
more labile cobalt(II), facilitating the exchange of axial ligands.[10],[11],[12]Binding of the
complexes to histidine residues in or nearby the active sites of proteins causes irreversible
inhibition of activity.[13],[14] Thus, Co(III) complexes inhibit histidine-containing proteins and
enzymes including zinc finger transcription factors (TFs) and metalloendopeptidases. Protein
inhibition occurs through a dissociative exchange of labile axial ligands for the imidazole
nitrogens of histidine residues, as drawn schematically in Figure 1.[15],[16]
Moreover, cobalt is generally less toxic to humans than platinum, which creates a basis for the
investigation of cobalt-containing compounds as less toxic alternatives to platinum-based
anticancer drugs.[17],[18] The simple Co3+ ion is unstable in water, but can be stabilized to
hamper reduction to Co2+ by coordination to chelating N,O or N,S donor ligands.[19]
Cobalt(III) complexes derived from these donor ligands find applications as antibacterial and
antiviral agents.[20] One of the most promising classes of Co(III) complexes containing N, O
donor ligands are those based on chelating Schiff bases.[21] So far, among all Co(III)
complexes, clinical trials have been reached only by a cobalt(III) Schiff base complex
containing bis(acetylacetone)ethylenediimine (acacen), with two axially coordinated 2-
2

Journal Pre-proof
methylimidazole rings [Co(III)(acacen)(2-mimd)2] (A).[22] The drug formulation was
developed as Doxovir™ by the Redox Pharmaceutical Corporation. In 2013, Doxovir
successfully completed phase II clinical trials for the treatment of Herpes Simplex Virus Type
1 (HSV-1) labialis infections and phase I clinical trials for the treatment of two viral eye
infections, the major causes of blindness (ophthalmic herpetic keratitis and adenoviral
conjunctivitis).[23] The promising antiviral activity of Doxovir is attributed to the direct
interaction of the Co(III) Schiff base complex with its molecular target, the herpes virus
maturational protease, a serine protease containing large amounts of histidine.[24]
Another promising group of cobalt(III) complexes are those with nitrogen mustards ligands,
which are potent cytotoxins, utilizing the lone pair on the amine nitrogen to initiate a reaction
sequence, in which DNA is cross-linked by double alkylation.[25],[26],[27] The Co(III)
complex [Co(Meacac)2(DCE)]+ (Meacac = Methlyacetylacetonate, DCE=N,N-bis(2-
chloroethyl)ethylenediamine) (B) had 20-times greater activity against cancer cells under
hypoxic than oxic conditions, due to bioreduction of the starting Co(III) complex to Co(II) in
the hypoxic regions of solid tumours and subsequent release of the cytotoxic free nitrogen
mustard from the substitutionally more labile Co(II) ion.[28]
Along the same lines, the Hambley group has exploited bioreductive prodrugs of chaperone
cobalt(III) complexes for the delivery of cytotoxic ligands to hypoxic solid tumours.[29]
Firstly, Hambley et al. presented a bioreductively activated carrier system for the delivery and
release of curcumin. The dichlorido precursor complex, [CoCl2(tpa)]ClO4, (tpa = tris-(2-
pyridylmethyl)amine), exhibited no toxicity up to a concentration of 200 µM against the
colorectal cancer cell line DLD-1, whereas the Co(III) complex with coordinated curcumin
possesses an IC50 value of 39 ± 4 µM.[30] Secondly, Hambley et al. have prepared a series of
cobalt(III) complexes with the tripodal, ancillary tpa ligand as chaperones for delivery of
derivatives of hydroxamic acid, which inactivates enzymes by binding to catalytic zinc ions
through the hydroxamic acid moiety. The [Co(tpa)(c343ha)]ClO4 complex with a fluorescent
doubly deprotonated hydroxamic acid ligand (c343haH2: coumarin-343 hydroxamic acid)
showed enhanced antiproliferative activity against DLD-1 colon cancer cells (31 ± 2 µM) in
comparison to the free c343haH2 ligand (113 ± 8 µM).[31] Later, Hambley and co-workers
extended the concept of Co(III) prodrugs to selectively deliver inhibitors of matrix
metalloproteinases (MMP) enzymes, which are involved in the process of tumour metastasis.
The Co(III) carrier system consists of the MMP inhibitor marimastat with a tetradentate tpa
carrier ligand (C). In vivo antimetastatic activity tested against Balb/c mice with 4T1.2 tumour
3

Journal Pre-proof
implants of C showed a higher level of tumour-growth inhibition than free marimastat.[32]
Meade et al. have synthesized Co(III) Schiff base-DNA conjugates targeting C2H2
transcription factors, to inhibit the Hedgehog (Hh) pathway, which regulates the activity of
the Gli family of C2H2 zinc finger transcription factors in mammals. Such Co(III)-DNA
bioconjugates resulted in a targeted inhibitor of the single C2H2 zinc finger transcription
factor Cubitus Interruptus (Ci).[33],[34],[35] In the literature there are also examples of Co(III)
conjugates with bleomycin and pepleomycin, glycopeptide antitumor antibiotics, which are
used in the treatment of Hodgkin’s lymphoma, carcinomas of the skin, head and neck.[36],[37]
Such Co-glycopeptide conjugates bind to the DNA and under UV or visible light irradiation
result in DNA cleavage.[38],[39],[40],[41]
Despite the fact that cobalt(III) complexes with small molecular bioactive ligands as
described above were extensively studied as antiviral, antimicrobial and anticancer agents, the
development of cobalt(III) bioconjugates with larger biomolecules and suitable drug delivery
systems for them is still in its infancy. Our research is devoted to the design of peptide
delivery systems for various anticancer drugs.[42],[43],[44] In the past, we have synthesized a
dicobalt hexacarbonyl alkyne compound linked to the neurotensin peptide hormone. The
cobalt-peptide bioconjugates showed moderate cytotoxicity against HeLa cervical cancer cells
(26.4 ± 5.8 µM).[45] Thus, despite their demonstrated versatility cobalt derivatives, especially
cobalt-peptide bioconjugates have not been further studied as anticancer agents. This fact and
our previous work on bioconjugation techniques for inorganic pharmaceuticals,[46],[47],[48],[49]
prompted us to investigate cell penetrating peptides as a drug delivery system for promising
Co(III) anticancer candidates, as an alternative to platinum-based therapy.
Results and discussion
Synthesis and characterisation of cobalt(III) complexes
The key idea of this study is to synthesize a Schiff base ligand containing a functional group,
suitable for coupling with peptides. Such covalent linking to peptides via the ligand has not
been achieved before. The literature-known ligand H2salophen 1 was synthesized by a single
condensation of salicylaldehyde with 3,4-diamino benzoic acid (Scheme 1).[50],[51] The
ligand was fully characterized by FT-IR, ESI-MS, HPLC, 1H- and 13C-NMR (ESI†).
In the next step, the complexation reaction of Schiff base ligand 1 to cobalt(II) was carried out
under nitrogen atmosphere (Scheme 2). The ligand 1 was dissolved in absolute MeOH and
4

Journal Pre-proof
subsequently CoCl2·6H2O was added. The formation of the expected cobalt(II) complex of
ligand 1 in situ was confirmed by electrospray mass spectrometry (see ESI†). In the full-scan
mass spectra, the molecular ion peak [M+H]+ (m/z = 417) was found and assigned to
cobalt(II) complexes with ligand 1. 3 hours were obligatory to obtain the cobalt(II) complex
in situ. To the in situ formed cobalt(II) complex bearing the Schiff base ligand 1, three
different types of N-donors were added as the axial ligands (2 eq.), namely imidazol, 2-
methyimidazol and N-Boc-L-histidine methyl ester (Boc: tert.-butoxycarbonyl). The
imidazole-based ligands were chosen because of the previously noted remarkable activity of
cobalt(III) Schiff base complexes with coordinated 2-methylimidazole rings.[22] In the case
of N-Boc-L-histidine methyl ester, both functional groups (amino and carboxylic group) were
protected to avoid any side reactions during future coupling with peptides by solid phase
peptide synthesis. After adding the N-donors, the reaction mixture was opened to air to allow
oxidation of cobalt(II) to cobalt(III) (Scheme 2).
All cobalt(III) complexes were characterized by FT-IR, mass spectrometry, HPLC and
multinuclear NMR spectroscopy (1H, 13C and 59Co NMR).The IR spectra of the cobalt(III)
complexes 2 – 4 were compared to those of the free ligands. The IR spectral bands with their
assignment observed in the region 4000-400 cm-1 are listed in Table 1. The medium and
broad absorption peaks in the range 3102-3356 cm-1 were assigned to ν(O–H) stretching
frequencies showing the existence of a carboxylic acid. The sharp bands indicative of NH
vibrations are located at 2900-2999 cm-1. IR spectra show that the vibration bands of the
ν(C=N) imino groups were shifted to 1600–1675 cm-1. These shifts by 15 cm-1 compared to
the free ligand 1indicate that the nitrogen atoms of the imino groups coordinate to the cobalt
centre.[52] The sharps bands of the νas(COO) stretching frequencies at 1502–1633 cm-1 and
νsym(COO) at 1289–1370 cm-1 indicate the existence of carboxylate salts. All IR spectra are
reproduced in the ESI†.
Electrospray ionization mass spectrometry (ESI-MS) was applied to confirm the mass of the
desired compounds (Figure 3a). In the full-scan mass spectra, the molecular ion peaks
[M+H]+ were found and assigned to all complexes. The m/z values are consistent with the
proposed constitution of the compounds. The mass spectra of cobalt(III) complexes displayed
two additional peaks with moderate intensity corresponding to the [M-L]+ and [M-2L]+
species. This feature confirms the expected primary fragmentation pathway, which would be
5

Journal Pre-proof
characterized by the gradual loss of one and the second ligand in the axial positions. The full-
scan mass spectra of complexes 2 - 4 are available in the ESI†.
To fully accomplish characterization of the Co(III) complexes also their 59Co NMR spectra
were measured. Representative spectra are shown in Figure 3, and chemical shifts (δ) and line
59
widths (ν1/2) are summarized in Table 1. The remaining Co NMR spectra are available in
the ESI†. All spectra showed only one peak, which confirms the oxidation of the cobalt(II)
59
precursor to only one cobalt(III) complex. The Co NMR data show that the nature of the
different axial ligands affect the amount of electron density at the Co nucleus.[53] The signal
is shifted downfield, when the axial ligand is altered from imidazole (2) to 2-methylimidazole
(3) and to N-Boc-L-histidine methyl ester (4).
Synthesis and characterisation of peptides
To enhance the cellular uptake and hence potentially the cytotoxicity of cobalt(III) Schiff base
complexes we decided to utilize cell penetrating peptides (CPPs). CPPs are usually rich in
positively charged amino acids, such as arginine (Arg, R) and lysine (Lys, K).[54]
Oligoarginine / arginine-rich cell-penetrating peptides have been successfully used as vectors
for the intracellular delivery of small anticancer drugs without specific receptors.[55]It has
been also reported that the attachment of a short peptide segment, namely hydrophobic
sequences of phenylalanine residues (Phe, F) results in enhanced translocation through cell
membranes of arginine-rich CPPs, due to stronger interaction with membranes.[56],[57]
Therefore, we prepared two peptides containing an arginine-rich sequence, one with an
additional FFFF segment at the N-terminus. A third peptide, consisting only of phenylalanine
residues, was design as a control peptide for cytotoxicity experiments. Solid-phase peptide
synthesis (SPPS) with Fmoc strategy (Fmoc: Fluorenylmethoxy-carbonyl) was employed to
synthesize those CPPs (Scheme 3).[58] Peptide P1, which consists of only a
tetraphenylalanine sequence (FFFF), was synthesized manually. The longer peptides P2 and
P3, were assembled on an automated, microwave-assisted synthesizer. Deprotection of
temporary Fmoc protecting groups was performed with 20% piperidine in DMF. The coupling
was performed with an excess of the Fmoc-amino acid (4 eq.), activated with 2-(1H-
benzotriazole-1-yl)-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU, 4 eq.) in the
presence of N-hydroxybenzotriazole (HOBt, 4 eq.) and an excess of diisopropylethylamine
(DIEA, 6 eq.). SPPS consists of repeated cycles of N-terminal protecting group removal on
6

Journal Pre-proof
the last amino acid, followed by coupling of the next incoming amino acid. Through
repetitions of a reaction cycle of peptide coupling, washing, deprotection and washing the
desired peptides were obtained.
After completion of peptide synthesis and before the bioconjugation reaction with Co(III)
complexes 2 – 4 the successful synthesis of peptides P1, P2 and P3 was established by HPLC
and ESI-MS. For this purpose, 10 mg of the resin from each peptide was treated with 20% of
piperidine in DMF in order to remove the Fmoc group from the N-terminus. Then, after
washing and drying the peptides were cleaved from the resin by treatment with a mixture of
TFA/phenol/H2O (95%/2.5%/2.5%). After cleavage from the resin the peptides were
lyophilized. The crude peptides were purified by reverse-phase semi-preparative HPLC.
Finally, the peptides were characterized by HPLC and ESI-MS measurements. The purity of
the peptides was confirmed by analytical HPLC (ESI†). ESI-MS spectra confirm the
molecular mass [M+H]+ for peptide P1 at m/z = 606, for peptide P2 at m/z = 1664 and for
peptide P3 at m/z = 2253. Besides the signals of molecular masses, multiply charged species
were formed, in the case of peptide P2 (M+2H)2+ at m/z = 832, (M+3H)3+ at m/z = 555,
(M+4H)4+ at m/z = 416, and for peptide P3 (M+2H)2+ at m/z = 1127 and (M+3H)3+ at
m/z = 752 (ESI†).
Synthesis and characterisation of bioconjugates
In the last synthesis step, the peptides were coupled to our Co(III) complexes which contain a
free carboxylic acid group for bioconjugation. The coupling of Co(III) complexes 2, 3 and 4
with the Fmoc-deprotected peptides P1, P2, P3 was performed with the fully side-chain
protected peptides on the resin in order to avoid any undesired side reactions (Scheme 4). The
carboxylic acid group on the Schiff base ligand provided a suitable functionality for coupling
by SPPS. The metal complex was dissolved in DMF and coupling was performed with an
excess of the cobalt complexes (4 eq.), activated with 2-(1H-benzotriazole-1-yl)-1,1,3,3-
tetramethyluronium tetrafluoroborate (TBTU,
4
eq.) in the presence of N-
hydroxybenzotriazole (HOBt, 4 eq.) and an excess of diisopropylethylamine (DIEA, 8 eq.).
After 24 h, an excess of reagents was removed by filtration, the resin was washed with DMF,
CH2Cl2, Et2O, and then dried under vacuum. Cleavage from the resin and deprotection of the
pbf side-chain protecting groups of arginine (pdf: 2,2,4,6,7-pentamethylhydrobenzofuran-5-
7

Journal Pre-proof
sulfonyl), for peptides P2 and P3, was achieved with 95% TFA. The obtained metal-peptide
bioconjugates were purified by preparative HPLC to obtain the desired products. All metal-
peptide bioconjugates were characterized by analytical HPLC, ESI-MS and MALDI-MS.
Analysis by RP-HPLC showed main peaks (ESI†), which were isolated and characterized by
ESI mass spectrometry. After purification by semi-preparative HPLC and lyophilization, all
bioconjugates were obtained as pale brown solids. ESI-MS analysis of the conjugates showed
m/z values that were consistent with that of the charged species [M+H]+, [M+Na]+ and doubly
charged species [M + 2H]2+, also showing the expected isotopic mass distribution patterns
(ESI†). The observed m/z values are listed in Table 2.
Cytotoxicity
Our next objective was to evaluate the anti-proliferative activity of Co(III) complexes and
cobalt-peptide bioconjugates against human liver cancer cells (HepG2), human lung cancer
cells (A549) and normal human fibroblast cells (GM5657T) by the MTT assay (MTT assay:
A cytotoxicity assays based upon formation of a purple-coloured precipitated from the yellow
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide dye after reduction in
metabolically active cells). The choice of cell lines was guided by our previous studies on
platinum, gold and metallocene bioconjugates.[44],[45],[46],[59] Cytotoxicity results are
expressed as IC50 values (half-maximal inhibitory concentration) and are summarized in
Table 3. Dose-response curves are shown in the ESI†. For comparison, the IC50 value of
cisplatin was also determined using the same assay, and identical conditions. As often in the
field, stock solutions of our compounds were prepared in DMSO for solubility reasons,
however final DMSO concentrations were in all cases well below 1 %. Even though not all
compounds were insoluble in water (or buffer) at the required concentrations, and certainly
cisplatin is soluble enough, we still utilized DMSO as the solvent for coherence, so that all
compounds are treated in an absolutely identical fashion. While we are aware of the effect of
DMSO on cisplatin activity as reported by Hall et al.,[60] we did not observe a similar
influence of DMSO on ligand exchange in our Co complexes by 59Co NMR.
The effect of solvent and, more generally, assay conditions deserves further discussion. It is
well known that DMSO exhibits a significant, diminishing effect on the cytotoxicity of
cisplatin.[60] In our studies we used DMSO stock solutions in order to compare the activity of
bioconjugates and cobalt(III) complexes with cisplatin in a range of different cell lines.
According to the literature,[61] cisplatin dissolved in H2O alone has an IC50 value of 3.6 ± 0.4
8

Journal Pre-proof
μM against A549 human lung cancer cells, whereas in our experiments its IC50 value was 25.3
± 2.3 μM (dissolved in DMSO). Against HepG2 human liver cancer cells, the values for
cisplatin were (2.09 ± 0.15 μM) in water,[62] compared to our experiments (13.3 ± 1.4 μM)
with DMSO present. However, other differences must also be taken into account, such as
assay type, incubation time, and probably general conditions of the cell line (e.g. no. of
passages for a given cell line). In this case, we note that the MTT assay was used in all studies
discussed above. However, while we incubate our compounds with cells routinely for 48 h,
Yao et al. incubated the A549 cells for 24 h only with cisplatin, then the media with
compounds were replaced by fresh media, and cells were cultured for another 48 h. As
normally no additional information is available on the age or “fitness” of the cell line (and
also it can only be assumed that identity checks of cell lines are regularly performed), such
data can be compared only with appropriate caution, and there is clearly not one “true” IC50
value for a given cell line. For the same reason, we feel most comfortable with performing all
cytotoxicity assays at the same time, on the same cell lines with identical treatment of all
samples including negative and positive controls – knowing that especially the conditions for
cisplatin may be considered “sub-optimal”.
Firstly, we investigated the cytotoxic activity of all cobalt complexes. As indicated in Table 3
all Co(III) complexes display moderate in vitro activity. HepG2 liver cancer cells appeared to
be the most sensitive to cobalt complexes among all testes cell lines. Complex 4 was the most
active towards HepG2 human liver cancer cells (13.2±0.8 µM) comparable with cisplatin
(13.3±1.4 µM). The GM5657T non-cancerous human fibroblast cells were the most sensitive
(22.0±0.9 µM) to complex 2.
Secondly, we investigated the influence of cell penetrating peptides attached to the cobalt(III)
complexes on in vitro cytotoxicity (Table 3). Since complex 4 appeared to be the most potent
cytotoxic agent among all cobalt complexes, only bioconjugates containing this complex (15,
18, 21) were investigated by the MTT assay. Bioconjugate 15 with the short phenylalanine
sequence showed only moderate activity towards all cancer cells, lower than complex 4.
Bioconjugate 18 consisting of an polyarginine domain showed activity similar to the parent
Co(III) complex. Finally, bioconjugate 21 possessing the combined polyarginine sequence
and a short uptake-accelerating tetra-phenylalanine sequence on the N-terminus shows
slightly improved activity in comparison to complex 4 against all tested cells. Similar
observations were made by Takayama et al., who observed that the attachment of a small
9

Journal Pre-proof
hydrophobic peptide segment to arginine rich CPPs improves cellular uptake.[63] Also Sadler
et al. described the improved cytotoxic activity of osmium(II) complexes after conjugation to
polyarginine peptides.[64] Since cell penetrating peptides without any particular selectivity
were used in this work cancer cells as well as normal human fibroblast cells were about
equally affected.
In a control experiment we investigated the anti-proliferative impact of the cell penetrating
peptides alone, i.e. without cobalt(III) Schiff base complexes attached. Together, the metal-
free peptides P1, P2 and P3 were examined by MTT assays using the same cell lines and
conditionsas for the parent bioconjugates (Table 3). These peptides exhibit no in vitro activity
at concentrations of up to 250 μM. These results are in agreement with observations made by
Gross et al.,[44] where similar peptides with only a slight C-terminal modification appeared to
be inactive as well.
Conclusion
Cobalt complexes emerge as a promising alternative to classical platinum-based anticancer
drugs. Here we present the synthesis of Co(III) complexes with salen-type Schiff base ligand
as chelating donor and three different imidazole derivatives containing N-donors as axial
ligands. Notably, a free carboxylic group on the salen-type ligand provides a handle for the
synthesis of covalent bioconjugates. By reaction with this carboxylate group, cobalt(III)-
bioconjugates with cell penetrating peptides have been synthesized and characterized for the
first time. First, the cell penetrating peptides were synthesized by solid phase peptide
synthesis (SPPS). Then, an SPPS reactions scheme of resin-bound peptides and the Co(III)
complexes with one free carboxylate group afforded monoconjugated cobalt-peptide species,
which were purified by RP-HPLC and characterized by ESI-MS. All Co complexes and
selected Co-peptide bioconjugates were screened for their cytotoxicity against selected cancer
cell lines. The Co(III) complexes alone show promising cytotoxicity in the mid-to-low
micromolar range. Among all bioconjugates, the bioconjugate which consists of a CPP-
polyarginine peptide sequence with four addition phenylalanine residues (21), was found to be
most active against HepG2 liver cancer cells. It is noteworthy that the peptide sequences
chosen here are not know for any particular intra-cellular targeting other than uptake through
lysosomes. On the other hand, adding a peptide amplifies the molecular weight of the (small)
10

Journal Pre-proof
Co complex significantly, and this might potentially alter its propensity to interact with the
intracellular target. On the other hand, the fact that we do see acceptable bioactivity of the Co-
peptide conjugates in our work indicates that at least a fraction of the Co complex will reach
the intracellular target and exert its activity. Further investigations as to what this target would
be shall be carried out and the results reported subsequently.
Experimental section
Materials and reagents
All reagents and chemicals were purchased from commercial sources and used without further
purification. Salicylaldehyde, 3,4-diamino benzoic acid and CoCl2·6H2O were purchased
from Merck. All Fmoc-protected amino acids were purchased from Iris Biotech GmbH. All
the materials and organic solvents including absolute methanol, ethanol, dimethyl sulfoxide
and dimethylformamide were commercially obtained from Merck, Aldrich or Fluka. Unless
otherwise noted, all manipulations were performed using standard Schlenk techniques under
nitrogen atmosphere. For the biological experiments, Dulbecco’s Modified Eagle’s Medium
(DMEM), fetal bovine serum, penicillin/streptomycin mixture, trypsin/EDTA, and phosphate-
buffered saline (PBS) were purchased from Sigma-Aldrich. Compounds were dissolved in
DMSO and diluted with the tissue culture medium before use.
General Methods: Solvents were dried according to standard procedures and stored over
molecular sieves (4 Å). Solid-supported reactions were performed in 5 mL plastic syringes
with a porous polypropylene disc as filter. HPLC was performed by using two buffer systems
(buffer A: H2O/MeCN/TFA, 95:5:0.1, v/v/v; buffer B: MeCN/H2O/TFA, 95:5:0.1, v/v/v) as
the mobile phase. Preparative HPLC runs were performed by using a Dr. Maisch reprosil C18
reversed-phase column (250 × 20 mm) at a flow rate of 10 mL/min with a linear gradient of
buffer B (100 % in 40 min) from 100 % buffer A with a total run time of 60 min. Analytical
HPLC runs were performed by using a Knauer Eurospher-II C18 reversed-phase column (250
× 4.6 mm) at a flow rate of 1.0 mL/min with a linear gradient of buffer B (100 % in 20 min)
from 100 % buffer A with a total run time of 40 min. FT-IR spectra were recorded with ATR
technique on a Spectrum 65 PerkinElmer instrument. NMR spectra were recorded at room
1
13
temperature with a Bruker Avance 300 Digital (1H at 300 MHz) spectrometer. H and C
NMR spectra were referenced using the residual solvent chemical shift (DMSO-d6). Mass
spectra were recorded with a Bruker Esquire 6000 (ESI-MS) spectrometer. UV-Vis
absorption spectra were recorded on a Spectrophotometer Microplate Reader (Berthold
Detection System).
11

Journal Pre-proof
Synthesis of the ligand
(1): Salicylaldehyde (1.60 g, 0.13 mmol) was mixed with ethanol (10.0 mL) and
simultaneously 3,4-diamino benzoic acid (1.00 g, 0,01 mmol) with the same solvent (10.0
mL). This solution of 3,4-diamino benzoic acid was added dropwise to the salicylaldehyde
solution and the reaction mixture was stirred overnight at room temperature. After 24h, the
orange precipitation was collected and washed three times with Et2O. The product was
dissolved and purified by preparative HPLC using 0.01% TFA as the aqueous mobile phase
and acetonitrile as the organic mobile phase with a linear gradient from 10% to 90% of the
organic mobile phase over 20 min. The final product was dried by rotary evaporation,
followed by lyophilization, to give a pale yellow oil. Yield: (2.09 g, 52%). Anal. Calc. for
C21H16N2O4: C, 69.99; H, 4.48; N, 7.77. Found: C, 70.11; H, 5.03; N, 7.51. MS (ESI-MS, pos.
mode, m/z): obs.: 361.00 [M+H]+; calc.: 361.11 [M+H]+, obs.: 383.00 [M+Na]+; calc.: 383.10
[M+Na]+; obs.: 255.40 [M-L]+; calc.: 254.07 [M-L]+. 1H NMR (300 MHz, DMSO-d6) δ: 8.94
(s, 2H), 8.13 (d, 2H), 7.80 (ddd, 2H), 7.71 (m, 4H), 7.65 (d, 2H), 7.35 (d, 2H), 7.32 (s, 2H),
6.93 (t, 4H). 13C NMR (126 MHz, DMSO-d6) δ: 163.6, 161.1, 160.0, 144.6, 132.4, 132.1,
130.1, 127.2, 123.6, 123.5, 121.4, 120.5, 117.8.
Synthesis of Co(III) complexes
(2): To a methanolic suspension (5 mL) of ligand 1 (40 mg, 0.0989 mmol), CoCl2·6H2O
complex (51.7 μL, 0.297 mmol) was added directly under nitrogen atmosphere at room
temperature. After 2 hours, 2 eq. of imidazol (12.54 mg, 0.0989 mmol) was added to the
reaction. The solution was opened to air and the reaction was stirred for 12 h to produce a
brown solution. The solution was concentrated. Addition of diethyl ether precipitated a brown
solid. The product was filtered and washed with cold diethyl ether. Yield: (0.657 g, 49%).
Anal. Calc. for C27H22ClCoN6O4: C, 55.07; H, 3.77; N, 14.27. Found: C, 55.42; H, 3.83; N,
14.49. MS (ESI-MS, pos. mode, m/z): obs.: 564.80 [M+Na]+; calc.: 574.09 [M+Na]+, obs.:
485.90 [M-L]+; calc.: 484.06 [M-L]+, obs.: 485.90 [M-2L]+; calc.: 417.03 [M-2L]+. 1H NMR
(300 MHz, DMSO-d6) δ: 8.94 (s, 2H), 8.13 (d, 2H), 7.80 (ddd, 2H), 7.71 (m, 4H), 7.65 (d,
2H), 7.35 (d, 2H), 7.32 (s, 2H), 6.93 (t, 4H). 13C NMR (126 MHz, DMSO-d6) δ: 163.6,
161.1, 160.0, 144.6, 132.4, 132.1, 130.1, 127.2, 123.6, 123.5, 121.4, 120.5, 117.8.
12

Journal Pre-proof
(3): To a solution of the ligand 1 (0.50 g, 2.45 mmol) in MeOH (5.0 mL) was added directly
solid CoCl2·6H2O complex (0.24 g, 1.025 mmol) under nitrogen atmosphere at room
temperature with continuous stirring. The mixture was stirred at room temperature for 2 h.
After that, 2 eq. of 2-methylimidazol (12.54 mg, 0.0989 mmol) was added to the reaction. The
reaction was subjected to stirring for 18 h in an open-air environment to produce a bright
brown precipitation. Finally, the precipitate was filtered, washed with Et2O (5.0 mL) and
dried. Yield: (0.657 g, 49%). Anal. Calc. for C29H26ClCoN6O4: C, 56.46; H, 4.25; N, 13.62.
Found: C, 56.86; H, 4.33; N, 14.01. MS (ESI-MS, pos. mode, m/z): obs.: 582.00 [M+H]+;
calc.: 581.48 [M+H]+, obs.: 499.80 [M-L]+; calc.: 498.70.10 [M-L]+, obs.: 416.90 [M-2L]+;
calc.: 417.03 [M-2L]+. 1H NMR (300 MHz, DMSO-d6) δ: 8.94 (s, 2H), 8.13 (d, 2H), 7.80
(ddd, 2H), 7.71 (m, 4H), 7.65 (d, 2H), 7.35 (d, 2H), 7.32 (s, 2H), 6.93 (t, 4H). 13C NMR (126
MHz, DMSO-d6) δ: 163.6, 161.1, 160.0, 144.6, 132.4, 132.1, 130.1, 127.2, 123.6, 123.5,
121.4, 120.5, 117.8.
(4): The ligand 1 (0.31 g, 1.5 mmol) was suspended in MeOH (10 mL) and to the suspension
was added immediately solid CoCl2·6H2O complex (0.24 g, 1.025 mmol) under nitrogen
atmosphere at room temperature with continuous stirring. The reaction was carried out for 3
h, and then 2 eq. of N-Boc-L-histidine methyl ester (12.54 mg, 0.0989 mmol) was added to
the reaction. The solution was opened to air and an immediate colour change to deep brown
was observed. After stirring at room temperature for 24 h, the resulting solution was allowed
to evaporate resulting in the precipitation of a deep brown solid, which was collected by
filtration and washed with Et2O. Yield: (0.657 g, 49%). Anal. Calc. for C45H52ClCoN8O12: C,
54.52; H, 5.29; N, 11.30. Found: C, 54.92; H, 5.37; N, 11.46. MS (ESI-MS, pos. mode, m/z):
obs.: 458.88 [M+H]+; calc.: 458.11 [M+H]+; obs.: 480.82 [M+Na]+; calc.: 481.10 [M+Na]+;
(ESI-MS, neg. mode, m/z): obs.: 456.85 [M-H]-; calc.: 457.10 [M-H]-; obs.: 478.73 [M+Na-
H]-; calc.: 480.09 [M+Na-H]-. 1H NMR (300 MHz, DMSO-d6) δ: 8.94 (s, 2H), 8.13 (d, 2H),
7.80 (ddd, 2H), 7.71 (m, 4H), 7.65 (d, 2H), 7.35 (d, 2H), 7.32 (s, 2H), 6.93 (t, 4H). 13C NMR
(126 MHz, DMSO-d6) δ: 163.6, 161.1, 160.0, 144.6, 132.4, 132.1, 130.1, 127.2, 123.6, 123.5,
121.4, 120.5, 117.8.
Peptide Synthesis: The FFFF peptide (P1) was synthesized manually by means of the Fmoc
protocol on an Fmoc-Phe-Wang resin (0.64 mmol/g). This peptide was synthesized on a 0.25
mmol scale. Each synthetic cycle consisted of the following steps: Fmoc removal, washing,
coupling, washing. Fmoc removal: The resin (357 mg, 0.250 mmol) was treated with a 20 %
solution of piperidine in DMF (3.0 mL; 2 × 10 min). The solution was removed and the resin
13

Journal Pre-proof
washed with DMF (3.0 mL; 3 × 1 min) and CH2Cl2 (3.0 mL; 3 × 1 min). The presence of free
α-amino functionalities was checked by the bromophenol blue (BPB) test (blue-green beads).
Coupling of standard amino acids: A mixture of Fmoc-Xxx-OH (1.00 mmol, 4 equiv.), 2-(1H-
benzotriazole-1-yl)-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU, 1.00 mmol, 4 eq.),
N-hydroxybenzotriazole (HOBt, 1.00 mmol, 4 eq.) and DIEA (1.5 mmol, 6 equiv.) in DMF
(3.0 mL) was added to the resin and the suspension mixed by shaking at room temperature for
45 min. Reagents and solvents were removed by filtration, and the resin was subsequently
washed with DMF (3.0 mL; 3 × 1 min) and CH2Cl2 (3.0 mL; 3 × 1 min). Completion of the
coupling (as monitored by the absence of free α-amino functionalities) was verified either by
the Kaiser test or the BPB test (BPB: bromophenol blue, colorless beads in both cases). Then,
the resin was washed with CH2Cl2 (3.0 mL; 3 × 1 min), Et2O (3.0 mL; 3 × 1 min), and finally
was dried under vacuum. The resin bound polyarginine peptides RRRRRRRRRGAL (P2) and
FFFFRRRRRRRRRGAL (P3) were synthesized on a CEMLibertyBlueTM automated
microwave peptide synthesizer using standard protocols (amino acid coupling: TBTU in DMF
(0.5 M), HOBt in DMF (0.5 M), DIEA in NMP (2 M) and amino acids in DMF (0.2 M);
arginine coupling: 25 min, 75 1C, 0 W followed by 5 min, 75 1C, 25 W; standard amino acid
coupling: 5 min, 75 1C, 24 W. For deprotection: 20% piperidine in DMF: initial deprotection:
0.5 min, 75 1C, 30 W followed by deprotection: 3 min, 75 1C, 50 W). Afterwards, aliquots of
100 mg of peptide-containing resin were transferred into a filter-containing syringe for further
derivatization.
Deprotection and Cleavage. The Fmoc deprotection was performed twice by treating the
Fmoc protected peptide with 2 mL of 20% piperidine in DMF (each time 10 min). For the
synthesis of all compounds, the N-terminal Fmoc group was deprotected, and after each step
the resin was washed 5 times using 2 mL of DMF. TFA cleavage: The resin was swirled in a
mixture of TFA/phenol/H2O (2.0 mL; 95:2.5:2.5, v/v/v) at room temperature for 1 h. Then the
resin was removed and the residual TFA solution transferred to a 50 mL falcon tube and an
ice-cold mixture of Et2O/hexane (20 mL; 1:1 v/v) was added to precipitate the peptide. The
supernatant was removed after centrifugation, and the peptide pellets were washed twice with
Et2O/hexane. The crude peptides were dissolved in MeCN/H2O (1:1, v/v) and lyophilized.
Crude peptides were purified by preparative HPLC, and the pure fractions were pooled and
lyophilized. Finally, the peptides were analyzed by analytical HPLC and characterized by
ESI-MS.
14

Journal Pre-proof
Cell culture. General procedure. Cells were grown in RPMI 1640 cell culture medium with
1% sodium pyruvate, 1% L-glutamine, 100 units per mL Pen Strep, 10% fetal bovine serum.
The cells were maintained at 37°C in a humidified incubator under an atmosphere containing
5% CO2.
Cytotoxicity experiments: Dulbecco’s Modified Eagle’s Medium (DMEM), containing 10%
fetal calf serum, 1% penicillin and streptomycin, was used as growth medium. A549, HepG2
and GM5657T cells were detached from the wells with trypsin and EDTA, harvested by
centrifugation and re-suspended again in cell culture medium. The assays were carried out on
96 well plates with 6000 cells per well for all cell lines: human lung cancer (A549), human
liver cancer (HepG2) and normal human fibroblast (GM5657T). After 24 h of incubation at
37.8 °C and 10 % CO2, the cells were treated with the compounds (with final DMSO
concentrations of 0.5 %) with a final volume of 200 mL per well. For a negative control, one
series of cells was left untreated. The cells were incubated for 48 h followed by adding 50 mL
MTT (2.5 mg mL-1). After an incubation time of 2 h, the medium was removed and 200 mL
of DMSO were added. The formazan crystals were dissolved and the absorption was
measured at 550 nm, using a reference wavelength of 620 nm. Each test was repeated in
quadruplicates in three independent experiments for each cell line.
Conflicts of interest
The authors declare no conflict of interest.
Author Contributions (CRediT)
D.S.: Methodology, Investigation, Visualization, Writing – Original Draft. N.M.N.:
Conceptualization, Visualization, Writing – Review & Editing, Supervision, Project
Administration, Funding Acquisition.
15

Journal Pre-proof
References
[1] M. Kobayashi, S. Shimizu, Eur. J. Biochem., 261 (1999) 1-9.
[2] L.O. Simonsen, H. Harbak, P. Bennekou, Sci. Total. Environm., 432 (2012) 210-215.
[3] K. Czarnek, S. Terpiłowska, A.K. Siwicki, Central-Eur. J. Immunol., 40 (2015) 236.
[4] K. Yamada, Cobalt: its role in health and disease, in: Interrelations between Essential Metal Ions
and Human Diseases, Springer, 2013, pp. 295-320.
[5] B.Y.K. Law, Y.Q. Qu, S.W.F. Mok, H. Liu, W. Zeng, Y. Han, F. Gordillo-Martinez, W.-K. Chan, K.M.-C.
Wong, V.K.W. Wong, Oncotarget, 8 (2017) 55003.
[6] K.J. Franz, N. Metzler-Nolte, Chem. Rev., 119 (2019) 1-3.
[7] M.C. Heffern, N. Yamamoto, R.J. Holbrook, A.L. Eckermann, T.J. Meade, Curr. Opinion Chem. Biol.,
17 (2013) 189-196.
[8] X. Wang, X. Wang, S. Jin, N. Muhammad, Z. Guo, Chem. Rev., 119 (2018) 1138-1192.
[9] P. Zhang, P.J. Sadler, Eur. J. Inorg. Chem., 2017 (2017) 1541-1548.
[10] N. Graf, S.J. Lippard, Adv. Drug Deliv. Rev., 64 (2012) 993-1004.
[11] A.K. Renfrew, Metallomics, 6 (2014) 1324-1335.
[12] I.C.A. de Souza, L.V. Faro, C.B. Pinheiro, D.T.G. Gonzaga, F.d.C. da Silva, V.F. Ferreira, F. da Silva
Miranda, M. Scarpellini, M. Lanznaster, Dalton Trans., 45 (2016) 13671-13674.
[13] M. Sonnay, T. Fox, O. Blacque, F. Zelder, Chem. Sci., 7 (2016) 3836-3842.
[14] S.V. Wegner, J.P. Spatz, Angew. Chem. Int. Ed., 52 (2013) 7593-7596.
[15] M.C. Heffern, V. Reichova, J.L. Coomes, A.S. Harney, E.A. Bajema, T.J. Meade, Inorg. Chem., 54
(2015) 9066-9074.
[16] J.-Y. Lee, Bull. Korean Chem. Soc., 33 (2012) 2762-2764.
[17] K.S. Egorova, V.P. Ananikov, Organometallics, 36 (2017) 4071-4090.
[18] A.K. Renfrew, E.S. O'Neill, T.W. Hambley, E.J. New, Coord. Chem. Rev., 375 (2018) 221-233.
[19] C.R. Munteanu, K. Suntharalingam, Dalton Trans., 44 (2015) 13796-13808.
[20] A.K. Ghosh, M. Mitra, A. Fathima, H. Yadav, A.R. Choudhury, B.U. Nair, R. Ghosh, Polyhedron,
107 (2016) 1-8.
[21] A. Böttcher, T. Takeuchi, K.I. Hardcastle, T.J. Meade, H.B. Gray, D. Cwikel, M. Kapon, Z. Dori,
Inorg. Chem., 36 (1997) 2498-2504.
[22] M.D. Hall, T.W. Failes, N. Yamamoto, T.W. Hambley, Dalton Trans., (2007) 3983-3990.
[23] E.L. Chang, C. Simmers, D.A. Knight, Pharmaceuticals, 3 (2010) 1711-1728.
[24] E.A. Bajema, K.F. Roberts, T.J. Meade, Cobalt-Schiff Base Complexes: Preclinical Research and
Potential Therapeutic Uses, in: P.L. Carver (Ed.) Essential Metals in Medicine: Therapeutic Use and
Toxicity of Metal Ions in the Clinic, Vol. 19 of Metal Ions in Life Sciences, de Gruyter, Berlin, Germany,
2019, pp. 267-295.
[25] D.C. Ware, B.D. Palmer, W.R. Wilson, W.A. Denny, J. Med. Chem., 36 (1993) 1839-1846.
[26] A.M. Downward, M.I. Polson, W.R. Kerr, J. Kariyawasam, R.M. Hartshorn, Polyhedron, 52 (2013)
617-622.
[27] P.R. Craig, P.J. Brothers, G.R. Clark, W.R. Wilson, W.A. Denny, D.C. Ware, Dalton Trans., (2004)
611-618.
[28] D.C. Ware, P.J. Brothers, G.R. Clark, W.A. Denny, B.D. Palmer, W.R. Wilson, J. Chem. Soc. Dalton
Trans., (2000) 925-932.
[29] N. Yamamoto, A.K. Renfrew, B.J. Kim, N.S. Bryce, T.W. Hambley, J. Med. Chem., 55 (2012) 11013-
11021.
[30] A.K. Renfrew, N.S. Bryce, T.W. Hambley, Chem. Sci., 4 (2013) 3731-3739.
[31] B.P. Green, A.K. Renfrew, A. Glenister, P. Turner, T.W. Hambley, Dalton Trans., 46 (2017) 15897-
15907.
[32] T.W. Failes, C. Cullinane, C.I. Diakos, N. Yamamoto, J.G. Lyons, T.W. Hambley, Chem. Eur. J., 13
(2007) 2974-2982.
16

Journal Pre-proof
[33] R.R. Hurtado, A.S. Harney, M.C. Heffern, R.J. Holbrook, R.A. Holmgren, T.J. Meade, Mol. Pharm.,
9 (2012) 325-333.
[34] A. Louie, T. Meade, Proc. Nat. Acad. Sci. USA, 95 (1998) 6663-6668.
[35] M.C. Heffern, J.W. Kurutz, T.J. Meade, Chem. Eur. J., 19 (2013) 17043-17053.
[36] K.D. Goodwin, M.A. Lewis, E.C. Long, M.M. Georgiadis, Proc. Nat. Acad. Sci. USA, 105 (2008)
5052-5056.
[37] V. Murray, J. Chen, L. Chung, Int. J. Mol. Sci., 19 (2018) 1372.
[38] J. Dabrowiak, M. Tsukayama, J. Am. Chem. Soc., 103 (1981) 7543-7550.
[39] J. Caceres-Cortes, H. Sugiyama, K. Ikudome, I. Saito, A.H.-J. Wang, Biochemistry, 36 (1997) 9995-
10005.
[40] F. Fedeles, M. Zimmer, Inorg. Chem., 40 (2001) 1557-1561.
[41] R. Subramanian, C.F. Meares, J. Am. Chem. Soc., 108 (1986) 6427-6429.
[42] F. Battistin, D. Siegmund, G. Balducci, E. Alessio, N. Metzler-Nolte, Dalton Trans., 48 (2019) 400-
414.
[43] B. Albada, N. Metzler-Nolte, Chem. Rev., 116 (2016) 11797-11839.
[44] A. Gross, H. Alborzinia, S. Piantavigna, L.L. Martin, S. Wölfl, N. Metzler-Nolte, Metallomics, 7
(2015) 371-384.
[45] A. Gross, M. Neukamm, N. Metzler-Nolte, Dalton Trans., 40 (2011) 1382-1386.
[46] D. Śmiłowicz, N. Metzler-Nolte, Dalton Trans., 47 (2018) 15465-15476.
[47] B. Albada, N. Metzler-Nolte, Acc. Chem. Res., 50 (2017) 2510-2518.
[48] J.C. Slootweg, P. Prochnow, S. Bobersky, J.E. Bandow, N. Metzler‐Nolte, Eur. J. Inorg. Chem.,
2017 (2017) 360-367.
[49] J.C. Slootweg, H.B. Albada, D. Siegmund, N. Metzler-Nolte, Organometallics, 35 (2016) 3192-
3196.
[50] K. Mohammadi, M. Niad, A. Irandoost, Spectrochim. Acta A: Mol. Biomol. Spectrosc., 107 (2013)
145-150.
[51] V. Oliveri, G. Vecchio, Eur. J. Med. Chem., 46 (2011) 961-965.
[52] C. Jing, C. Wang, K. Yan, K. Zhao, G. Sheng, D. Qu, F. Niu, H. Zhu, Z. You, Bioorg. Med. Chem., 24
(2016) 270-276.
[53] A.P. King, H.A. Gellineau, J.-E. Ahn, S.N. MacMillan, J.J. Wilson, Inorg. Chem., 56 (2017) 6609-
6623.
[54] Z. Guo, H. Peng, J. Kang, D. Sun, Biomed. Rep., 4 (2016) 528-534.
[55] X. He, M. Lin, J. Guo, Z. Qu, F. Xu, RSC Adv., 6 (2016) 30454-30459.
[56] E.J. Sayers, K. Cleal, N.G. Eissa, P. Watson, A.T. Jones, J. Contr. Rel., 195 (2014) 55-62.
[57] D. Kalafatovic, E. Giralt, Molecules, 22 (2017) 1929.
[58] R. Malishev, S. Nandi, D. Smilowicz, S. Bakavayev, S. Engel, N. Bujanover, R. Gazit, N. Metzler-
Nolte, R. Jelinek, ACS Chem. Neurosci., (2019).
[5 ] D. Śmiłowicz, J.C. Slootweg, N. Metzler-Nolte, Mol. Pharm., (2019).
[60] M.D. Hall, K.A. Telma, K.-E. Chang, T.D. Lee, J.P. Madigan, J.R. Lloyd, I.S. Goldlust, J.D. Hoeschele,
M.M. Gottesman, Cancer Res., 74 (2014) 3913-3922.
[61] H. Yao, Z. Xu, C. Li, M.-K. Tse, Z. Tong, G. Zhu, Inorg. Chem., 58 (2019) 11076-11084.
[62] S. Hadizadeh, N. Najafzadeh, M. Mazani, M. Amani, H. Mansouri-Torshizi, A. Niapour, Biochem.
Res. Intl., 2014 (2014).
[63] K. Takayama, H. Hirose, G. Tanaka, S. Pujals, S. Katayama, I. Nakase, S. Futaki, Mol. Pharm., 9
(2012) 1222-1230.
[64] S.H.v. Rijt, H. Kostrhunova, V. Brabec, P.J. Sadler, Bioconjugate Chem., 22 (2011) 218-226.
17

Journal Pre-proof
FIGURES AND SCHEMES
Figure 1. Cobalt binding to histidine residues through dissociative axial ligand exchange.
Figure 2. Examples of cobalt(III) inorganic drug candidates.
Scheme 1. Synthesis of Schiff base ligand1.
18

Journal Pre-proof
Scheme 2. Synthesis of cobalt(III) complexes (Imd: imidazole; 2-mimd: 2-methyl-imidazole.
RT: room temp.).
Figure 3. a) ESI-MS spectrum of complex 4; b) 59Co NMR spectra of complex 4 in dmso-d6.
19

Journal Pre-proof
Scheme 3. Synthesis of peptides P1, P2 and P3 by SPPS.
20

Journal Pre-proof
Scheme 4. Synthesis of Co(III)-peptide bioconjugates.
21

Journal Pre-proof
Figure 4. Structures of Co(III)-peptide bioconjugates.
22

Journal Pre-proof
TABLES
Table 1. Principal IR bands, assignments, and 59Co NMR chemical shifts.
Compound
IR assignment
ν(N-H) ν(C=N) νas(COO-) νs(COO-) δ (ppm) ν1/2 (Hz)
59Co NMR
ν(O-H)
3344
3230
-
1670
1680
1633
1537
1340
1324
-
-
1
2
2988
4550
5200
3223
3220
2999
2987
1688
1695
1566
1588
1354
1365
4670
4812
2900
2400
3
4
Table 2. Characterization of peptides and bioconjugates after purification by semi-preparative
HPLC.
Compound
Yield [%]
44
m/zexptl
606.40
m/zcalcd[M+H]+
606.28
tR [min]
6.12
P1
P2
P3
13
14
15
16
17
18
19
20
21
38
1664.03
2252.87
1140.02
1168.08
1342.55
2198.42
2226.00
2400.09
2785.40
2813.80
2969.90
1664.06
2252.34
1139.09
1167.12
1341.46
2197.09
2225.12
2399.24
2785.43
2813.46
2969.52
9.30
41
9.80
12
7.19
25
10.30
11.50
7.30
23
18
32
10.72
12.10
8.20
17
36
33
12.60
13.25
22
23

Journal Pre-proof
Table 3. Cytotoxicity data of cisplatin, Co complexes, metal-free peptides and Co-peptide
conjugates determined by the MTT assays after 48 h of incubation in our lab, all under
identical conditions.
IC50 [µM]a
Cell lines
A549
HepG2
GM5657T
25.3±2.3
43.8±1.6
65.3±0.9
32.0±2.2
>250
13.3±1.4
28.7±0.9
43.0±1.5
13.2±0.8
>250
9.0±2.2
22.0±0.9
46.5±1.3
51.2±1.1
>250
Cisplatin
2
3
4
P1
P2
P3
15
18
21
>250
>250
>250
>250
>250
>250
64.3±2.8
35.8±1.1
25.3±1.9
49.3±1.9
20.0±0.9
9.0±1.2
72.4±3.3
61.0±3.1
36.4±1.6
a
The IC50 values are averages of three independent determinations. All compounds were
dissolved in DMSO for stock solutions (final concentration 0.5 %).
24

Journal Pre-proof
Graphical Abstract
Bioconjugates of Co(III) Complexes with Schiff Base Ligands and Cell
Penetrating Peptides: Solid Phase Synthesis, Characterization and
Antiproliferative Activity
Dariusz Śmiłowicz, Nils Metzler-Nolte*
Inorganic Chemistry I – Bioinorganic Chemistry
Ruhr-University Bochum, Universitätsstraße 150, 44801 Bochum, Germany
E-mail: nils.metzler-nolte@rub.de
GA Text: Bioconjugates of octahedral Co(III) complexes with various cell targeting peptides
were prepared by solid phase peptide synthesis (SPPS), and a first structure-activity
relationship with regard to the peptide sequence was derived from cytotoxicity assays.
R
R
O
C
O
C
O
C
OH
PEPTIDE
OH
SPPS
OXIDATION
N
O
N
O
N
O
N
O
N
O
N
O
Co
Co
R
Co
R
25

Journal Pre-proof
Highlights
Bioconjugates of Co(III) Complexes with Schiff Base Ligands and Cell
Penetrating Peptides: Solid Phase Synthesis, Characterization and
Antiproliferative Activity
Dariusz Śmiłowicz, Nils Metzler-Nolte*
Inorganic Chemistry I – Bioinorganic Chemistry
Ruhr-University Bochum, Universitätsstraße 150, 44801 Bochum, Germany
E-mail: nils.metzler-nolte@rub.de
• Selective oxidation of Co(II) to Co(III) complexes with biocompatible ligands
• Synthesis of octahedral Co(III) complexes with conjugated cell targeting peptides
• Co-peptide conjugates prepared by solid phase peptide synthesis (SPPS)
• Comprehensive characterization and stability studies of Co-peptide conjugates
• Structure-activity relationship with regard to peptide sequence from cytotoxicity assay
26