Chemical strategies to modify amyloidogenic peptides using iridium(iii) complexes: coordination and photo-induced oxidation† †Electronic supplementary information (ESI) available: Experimental section and Fig. S1–S19. See DOI: 10.1039/c9sc00931k

Effective chemical strategies, i.e., coordination and coordination-/photo-mediated oxidation, are rationally developed towards modification of amyloidogenic peptides and subsequent control of their aggregation and toxicity.


Introduction
A substantial amount of research effort has been dedicated towards identifying the association of amyloidogenic peptides with the pathologies of neurodegenerative diseases. Among these amyloidogenic peptides, amyloid-b (Ab), a proteolytic product of the amyloid precursor protein found in the ADaffected brain with a self-aggregation propensity, has been implicated as a pathological factor in Alzheimer's disease (AD). [1][2][3][4] As the main component of senile plaques, Ab accumulation is a major pathological feature of AD. [1][2][3]5 Recent developments in Ab research (e.g., clinical failures of Ab-directed therapeutics) have led to the re-evaluation of the amyloid cascade hypothesis. 6 Ab pathology, however, remains a pertinent facet of the disease with indications of Ab oligomers as toxic species responsible for disrupting neuronal homeostasis. [1][2][3]7 Furthering our elucidation of Ab pathology presents an investigative challenge arising from its heterogeneous nature and intrinsically disordered structure. 1,2 To overcome this obstacle and advance our understanding of the Abrelated contribution towards AD, in this study, we illustrate chemical approaches to modify Ab peptides at the molecular level using transition metal complexes.
Transition metal complexes have been reported to harness their ability to induce peptide modications (e.g., hydrolytic cleavage and oxidation), inhibit the activities of enzymes, and image cellular components.  In particular, the ability of transition metal complexes to alter peptides stems from their properties, such as the capacity for peptide coordination. [17][18][19][20][21][22][23][24][25][26][27][28][29]36,37 Herein, we report effective chemical strategies for modication of Ab peptides using a single Ir(III) complex in a photo-dependent manner (Fig. 1). Ab modications, achieved by our rationally engineered Ir(III) complexes, include two events: (i) complexation with Ab in the absence of light; (ii) Ab oxidation upon coordination and photoactivation, which can signicantly regulate their aggregation and toxicity. Through our multidisciplinary studies, presented in this work, we demonstrate the development of new chemical tactics for modication of amyloidogenic peptides using transition metal complexes, useful for identifying their properties, such as aggregation, at the molecular level.

Results and discussion
Rational strategies for peptide modication using Ir(III) complexes To chemically modify Ab peptides in a photoirradiationdependent manner (Fig. 1a), four Ir(III) complexes (Ir-Me, Ir-H, Ir-F, and Ir-F2; Fig. 1b) were rationally designed and prepared. Iridium is a third row transition metal exhibiting strong spinorbit coupling at the center of Ir(III) complexes with facile electronic transitions. 44,45 This spin-orbit coupling can be further strengthened by ne-tuning the ancillary ligands of Ir(III) complexes. As a result, Ir(III) complexes confer notable photophysical properties upon excitation by relatively low energy irradiation in the visible range, including their ability to generate reactive oxygen species [ROS; e.g., singlet oxygen ( 1 O 2 ) and the superoxide anion radical (O 2 c À )] via electron or energy transfer. [46][47][48] In addition, Ir(III) complexes with octahedral geometry are relatively stable upon light activation. 48 Incorporation of 2-phenylquinoline derivatives as ligands yielded high emission quantum yield (F) and robust ROS generation. 46 Therefore, ancillary ligands of four complexes were constructed based on the 2-phenylquinoline backbone by applying simple structural variations to provide appropriate structural and electronic environments to promote the photochemical activity of the corresponding Ir(III) complexes. 46 Moreover, uorine atoms were introduced into the ancillary ligand framework affording Ir-F and Ir-F2 to chemically impart the ability to interact with Ab through hydrogen bonding, alter photophysical properties of the complexes, and enhance the molecules' biocompatibility. [49][50][51] Two water (H 2 O) molecules were incorporated as ligands to enable covalent coordination to Ab via replacement with amino acid residues of the peptide, e.g., histidine (His). 20,52,53 The four Ir(III) complexes were synthesized following previously reported procedures with modications (Scheme 1 and Fig. S1-S3 †). 20,54-56 As depicted in Fig. S4 and S5, † these Ir(III) complexes were conrmed to coordinate to His or Ab in both H 2 O and an organic solvent [i.e., dimethyl sulfoxide (DMSO)] under our experimental conditions.

Coordination-dependent photophysical properties and ROS production of Ir(III) complexes
Photophysical properties of the prepared Ir(III) complexes were investigated by UV-vis and uorescence spectroscopy. As shown in Table 1 and Fig. S6, † in the absence of His or Ab, low F values of the four Ir(III) complexes were observed, along with relatively poor 1 O 2 generation with photoactivation. Note that a solar simulator (Newport IQE-200) was used to irradiate the samples at a constant intensity (1 sun light; 100 mW cm À2 ). Upon addition of His, the F values of the four Ir(III) complexes drastically increased (e.g., F Ir-F ¼ 0.0071 versus F Ir-F+His ¼ 0.26; Table 1), indicating His coordination of the complexes, which was further conrmed by electrospray ionization-mass spectrometry (ESI-MS) (Fig. S5a †). The F values and 1 O 2 formation of the four Ir(III) complexes with His binding exhibited trends similar to their binding affinity with His (Ir-F > Ir-H > Ir-Me > Ir-F2). Ir-F, indicating the strongest binding affinity with His ( Fig. S5b †), among the four Ir(III) complexes, showed notable binding affinities towards different Ab species (for monomers, K d ¼ 1.6 Â 10 À4 M; for oligomers, K d ¼ 2. photoactivation ( Fig. S6 and S8 †). Based on these properties, we selected Ir-F as a representative candidate of our Ir(III) complexes and illustrated its ability to modify Ab peptides in detail (vide infra).     (Fig. 2a,  bottom). Ab 40 oxidation manifested a conformational change as probed by IM-MS (Fig. 2d). The most dominant arrival time indicated a peak at 9.92 ms. These results suggest that Ab 40 oxidation induced by Ir-F can alter the structural distribution of Ab 40 . Similar observations were observed with Ir-Me, Ir-H, and Ir-F2, where the complexes were able to oxidize Ab 40 and consequently vary its structural distribution (Fig. S9 and S10 †). In order to determine the location of peptide oxidation, the Ab fragment ions, generated by selectively applying collisional energy to singly oxidized Ab, were analyzed by ESI-MS 2 (Fig. 2e). All b fragments smaller than b 13 were detected in their nonoxidized forms, while those larger than b 34 were only monitored in their oxidized forms. The b fragments between b 13 and b 34 were indicated in both their oxidized and nonoxidized forms. Such observations, along with previous reports regarding Ab oxidation, 19,57 suggest His13, His14, and Met35 of Ab as plausible oxidation sites. Collectively, our studies demonstrate that Ab peptides can be modied upon treatment with Ir-F [(i) coordination to Ab by replacing two H 2 O molecules with the peptide in the absence of light; (ii) coordination-mediated oxidation of Ab at three possible amino acid residues (e.g., His13, His14, and Met35) upon photoactivation (Fig. 1a)]. Note that the Ab samples produced by treatment of photoactivated Ir-F showed high uorescence intensity and were relatively stable in both H 2 O and cell growth media (Fig. S11 †).

Effects of peptide modications triggered by Ir(III) complexes on Ab aggregation
Based on the photoirradiation-dependent Ab modications by Ir(III) complexes, the impact of such variations on the aggregation of Ab was determined employing Ab 40 and Ab 42 , two main Ab isoforms found in the AD-affected brain. [2][3][4][58][59][60][61][62] For these experiments, freshly prepared Ab solutions were treated with Ir(III) complexes with and without light under both aerobic and anaerobic conditions. The molecular weight (MW) distribution and the morphology of resultant Ab species were analyzed by gel electrophoresis with Western blotting (gel/Western blot) using an anti-Ab antibody (6E10) and transmission electron microscopy (TEM), respectively (Fig. 3a).
Under aerobic conditions (Fig. 3b, le), the aggregation of Ab 40 was affected by treatment with Ir-F prompting a shi in the MW distributions in the absence of light. Photoactivation of the Ir-F-treated Ab 40 sample resulted in a more diverse MW distribution compared to that of the corresponding sample without light (light, MW # 100 kDa; no light, MW < 15 kDa). The distinct modulation of Ab 40 aggregation upon addition of Ir-F with photoirradiation is likely a consequence of the complex's ability to generate 1 O 2 and oxidize Ab through photoactivation as observed in our spectrometric studies (vide supra; Fig. 2). Therefore, the same experiments were performed under anaerobic conditions to directly monitor the role of O 2 in Ir-F 0 s modulative reactivity against Ab 40 aggregation. In the absence of O 2 (Fig. 3b, right), Ab 40 aggregation was also altered by Ir-F regardless of light treatment. Our results suggest that both light and O 2 are important in the regulation of Ab 40 aggregation through coordination-/photo-mediated peptide oxidation triggered by Ir-F. In addition, in the absence of light and O 2 , Ab 40 aggregation is directed by the covalent interactions between Ir-F and the peptide. Similar modulation of Ab 42 aggregation was observed upon incubation with Ir-F exhibiting different MW distributions compared to the Ab 42 samples without Ir-F in the absence and presence of light and O 2 (Fig. 3c). Moreover, smaller amorphous aggregates of both Ab 40 and Ab 42 , reported to be less toxic, 63,64 were visualized by TEM from the samples containing Ir-F regardless of irradiation ( Fig. 3d and S12c †).
Furthermore, preformed Ab aggregates, generated at various preincubation time points (i.e., 2, 4, and 24 h), were disassembled and their aggregation pathways were altered when Ir-F was introduced (Fig. S13 †). Such Ir-F-induced effects on preformed Ab aggregates were observed to be dependent on photoirradiation. Moreover, the aggregation of both Ab 40 and Ab 42 was also changed with addition of the other Ir(III) complexes (i.e., Ir-Me, Ir-H, and Ir-F2) with and without light (Fig. 3e, S12 and S13 †). In addition to Ab, Ir-F was able to interact with and modify other amyloidogenic peptides [i.e., a-synuclein (a-Syn) and human islet amyloid polypeptide (hIAPP)] affecting their aggregation pathways (Fig. S14 †).

Cytotoxicity of Ab species generated upon incubation with Ir(III) complexes
Prior to cytotoxicity measurements, the resultant species upon 24 h treatment of Ab 40 with Ir-F with light exposure were incubated with murine Neuro-2a (N2a) neuroblastoma cells in order to determine their cellular uptake. As depicted in Fig. S15, † the lysates of the cells added with the resultant species for 24 h, analyzed by inductively coupled plasma-mass spectrometry (ICP-MS), indicated an Ir concentration of 39 mg L À1 , demonstrating the cellular uptake of the species containing Ir(III). Note that the Ir concentration (0.17 and 34 mg L À1 ) was measured from the lysates of the cells treated only with either Ab 40 or Ir-F, respectively. Moving forward, the toxicity of Ab species produced by treatment with our Ir(III) complexes was monitored by the MTT assay [MTT ¼ 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide] (Fig. 4). The cytotoxicity of Ab 40 species incubated with our Ir(III) complexes was noticeably reduced in a photoirradiation-dependent manner. In the absence of light, the Ab 40 samples incubated with our Ir(III) complexes exhibited a decrease in cytotoxicity (ca. 20%) compared to the sample of the complex-free Ab 40 . As for the photoirradiated samples, Ab 40induced toxicity was lowered by ca. 35% by treatment with our Ir(III) complexes. This result suggests that modication of Ab,  such as oxidation, could attenuate Ab-triggered toxicity in living cells. 65 Furthermore, the cytotoxicity of Ab 42 species formed with Ir(III) complexes was also diminished by ca. 20% regardless of photoactivation. Note that the survival ($80%) of cells treated with our Ir(III) complexes at the concentration used for cell studies with Ab peptides was observed with and without light exposure (Fig. S16 †).

Ternary complexation with Ab and intramolecular and intermolecular Ab oxidation
Premised on Ir-F 0 s covalent bond formation with Ab and oxidation of Ab (vide supra), additional studies regarding ternary complexation and promotion of intermolecular oxidation of Ab were carried out employing Ir-F (Fig. 5). Ab 28 , a fragment of Ab equipped with the metal binding and selfrecognition sites of the peptide with a relatively low propensity to aggregate than the full-length peptides, Ab 40 and Ab 42 , 1,66-68 was used to form a complex with Ir-F 0 (Fig. 2b) as evidenced by ESI-MS (1301 m/z; Fig. 5b) and increased uorescence (Fig. 5b, inset). As shown in Fig. 5a, following incubation, the sample of the Ab 28 -Ir-F 0 complex was treated with freshly prepared Ab 42 to monitor its effect on Ab 42 aggregation. Based on the gel/Western blot and TEM analyses, the aggregation of Ab 42 was modulated by the Ab 28 -Ir-F 0 complex ( Fig. 5c and d). Such modulative reactivity of the Ab 28 -Ir-F 0 complex was also observed against Ab 40 aggregation (Fig. S17 †). Our mass spectrometric studies conrmed that such control of Ab 42 aggregation by the Ab 28 -Ir-F 0 complex was a result of ternary complex formation with Ab 42 , i.e., (Ab 28 -Ir-F 0 )-Ab 42 , and (ii) oxidation of Ab, both intramolecular and intermolecular, upon photoactivation (Fig. 6). Based on previous reports detailing intermolecular interactions between Ab peptides, hydrophobic interactions between the self-recognition sites (LVFFA; Fig. 3a and 5a) of Ab are likely responsible for ternary complexation, 1,2,69 consequentially altering the aggregation pathways of Ab in the absence of photoirradiation. Furthermore, these studies indicate that intermolecular oxidation of Ab can be promoted by Ir-F upon photoactivation (Fig. 6, S18, and S19 †). This observation may explain the distinct difference between the modulation of Ab aggregation with and without light as the intermolecular oxidation of Ab by Ir(III) complexes could modify Ab at sub-stoichiometric levels.

Conclusions
Effective chemical strategies (i.e., coordination to Ab and coordination-/photo-mediated oxidation of Ab) for modication of Ab peptides using a single Ir(III) complex were rationally developed. Such dual mechanisms (i.e., coordination and oxidation) exhibiting photo-dependency for altering Ab peptides are novel and effective in controlling peptide aggregation and cytotoxicity. Our Ir(III) complexes can covalently bind to Ab by replacing two H 2 O molecules bound to the Ir(III) center with Ab regardless of light and O 2 [coordination to Ab; Fig. 1a(i)]. In the presence of light and O 2 , Ir(III) complexes bound to Ab are capable of inducing the intramolecular and intermolecular oxidation of Ab at His13, His14, and/or Met35 [oxidation of Ab; Fig. 1a(ii)]. Taken together, our multidisciplinary studies demonstrate the feasibility of establishing new chemical approaches towards modication of amyloidogenic peptides (e.g., Ab) using transition metal complexes designed based on their coordination and photophysical properties. In general, chemical modications in peptides of interest can assist in furthering our understanding of principles of their properties, such as peptide assembly. Furthermore, peptide aggregation and cytotoxicity can be affected by biomolecules, including lipid membranes; 70-73 thus, the regulatory reactivity of Ir(III) complexes towards amyloidogenic peptides in the presence of lipid membranes will be investigated in the future.

Conflicts of interest
There are no conicts to declare.