1,5‐Diazacyclooctanes, as Exclusive Oxidative Polyamine Metabolites, Inhibit Amyloid‐β(1‐40) Fibrillization

Biologically relevant 1,5‐diazacyclooctanes derived from polyamines and acrolein, inhibit Aβ40 peptide fibrillization and significantly suppress cell cytotoxicity. Formal [4+4] cycloaddition reaction of imines is thus involved in modulating oxidative stress processes associated with neural diseases.


DOI: 10.1002/advs.201600082
of 3-formyl-3,4-dehydropiperidine (FDP), formed from two molecules of acrolein with the ε-amino group of lysine, is currently employed as an oxidative stress marker. [13][14][15] In a similar process, polyamines have also been shown to react with acrolein to produce the corresponding FDP derivatives. [ 16 ] With an increase of cellular polyamine or cytotoxic acrolein levels, research has shown a correlation with the progression of certain diseases, such as cancer or stroke. [ 17,18 ] In the brains of AD patients, observed levels of acrolein or spermine ( SPM ) are increased, [ 19,20 ] whereas spermidine ( SPD ) or putrescine levels are decreased. [ 19 ] Recent reports also indicate that polyamines can promote amyloid-β peptide 1-40 (Aβ40) fi brillization, which is implicated in the acceleration of the AD process. [ 21 ] Recently, we discovered that spermine and spermidine smoothly react with acrolein to produce 1,5-diazacyclooctanes (cyclic spermine and spermidine, cSPM and cSPD ) through a formal [4+4] cycloaddition of the intermediary unsaturated imines ( Scheme 1 ). [22][23][24][25] We demonstrated that these compounds are produced in much higher amounts and effi ciency than the oxidative stress marker, FDP, which thus far has only been detected under standard analytical conditions. This likely suggests that acrolein reacts with polyamines to exclusively produce the eight-membered heterocycles as initial acrolein-modifi ed intermediates. Given these results, there is considerable potential for these compounds to be implicated in biological processes that were previously unexplored or overlooked. This can be supported by our recent demonstration that diazaheterocycles produced from polyamines (e.g., cSPM ) can effi ciently neutralize the toxicity of acrolein, and that eight-membered polymers produced through sequential cycloaddition processes, both within and on the surface of oxidatively stressed cells, are responsible for damaging cellular function. [ 22 ] Our investigations into the biological signifi cance of diazaheterocycles led us to focus on Aβ fi brillization, largely due to the fact that acrolein is produced in the brain tissues of AD patients as a polyamine metabolite during oxidative stress processes. It was speculated that the eight-membered polyamineacrolein heterocycles (i.e., cSPM or cSPD ) may potentially control and/or modulate disease progression. Unlike previous reports suggesting that polyamines promote Aβ40 fi brillization, [ 21 ] this study clearly shows that the biologically relevant polyamine-acrolein conjugates inhibit fi brillization and hence cytotoxicity. Thus, the acrolein/polyamine-derived [4+4] cycloaddition process may effectively modulate the oxidative stress processes associated with neuronal diseases.
We initially investigated the effects of cSPM and cSPD on Aβ40 fi brillization. Samples were incubated with 25 × 10 −6 M of the Aβ40 peptide at 37 °C in phosphate buffered saline (PBS) for 5 d, and fi bril formation was evaluated based on the thiofl avin T (ThT) fl uorescence assay ( Figure 1 ). Although SPM and acrolein did not show any activity, fi brillization was effi ciently inhibited in the presence of cSPD and cSPM at concentrations exceeding 0.5 × 10 −6 M . Furthermore, one of the diazacyclooctanes, spermine-derived cSPM , effectively suppressed fi brillization for more than a month ( Figure 2 ), indicating that cSPM could steadily inhibit Aβ fi brillization for an extended period of time.
We next examined the activity of Aβ40 samples treated with cSPM using PC12 cells derived from transplantable rat pheochromocytoma ( Figure 3 ). In agreement with previous reports, [25][26][27] Aβ40 fi brils are expected to display cytotoxic activity. However, observations indicate that with cSPM treatment in a dose-dependent manner, Aβ40 cytotoxicity was   The mechanism by which polyamine-acrolein heterocycle cSPM inhibited Aβ40 fi brillization and cytotoxicity was examined by analyzing the molecular sizes of the cSPM-treated Aβ40 peptides using native PAGE/western blotting techniques and an anti-Aβ40 antibody (6E10) ( Figure 4 a, also see Figure S3a, Supporting Information). This method permitted detection of the soluble monomer at the bottom of the gel (lane 1), whereas the insoluble aggregates, such as the Aβ40 fi brils, remained at the top (lane 5). Intriguingly, a signifi cant quantity of the Aβ40 monomer remained in the Aβ40 samples after treatment with 0.5 × 10 −6 or 1.0 × 10 −6 M cSPM (lanes 3 and 4), although some insoluble species were also detected. The "soluble" oligomeric species, which are considered to be highly toxic to cells, [28][29][30] were not observed in these gels. The absence of notorious "soluble" aggregates, therefore, supported the signifi cant inhibitory effects of cSPM on cytotoxicity, as observed in Figure 3 .
The quantities of the monomeric Aβ40 species present in the cSPM-treated samples in Figure 4 a were determined by separating the "insoluble" Aβ40 aggregates from the monomeric peptide by centrifugation (see details in Figure S3a, Supporting Information). The percentages of Aβ40 monomers that remained in the cSPM-treated Aβ40 samples were calculated to be 70%-80% of the mixtures (for samples treated with 0.5 or 1.0 × 10 −6 M cSPM , Figure S3b, Supporting Information). Therefore, it can be concluded that cSPM effi ciently suppresses insoluble aggregate formation. As a side note, the insoluble species isolated from the cSPM-treated Aβ40 mixture as the minor product was found to match the cytotoxicity of Aβ40 fi brils (prepared as a control) to PC12 cells (Figure 4 b). Together with the TEM images of Figure 5 , which show that cSPM could noticeably reduce fi bril formation, the collective data suggest that cSPM inhibits cytotoxicity by a mechanism that involves both blocking the formation of highly toxic "soluble" oligomer   species and minimizing the formation of toxic "insoluble" Aβ40 fi brils. As a consequence, Aβ40 peptides are expected to be maintained in a monomeric state, thereby reducing cytotoxicity.
Observations also show that cSPM pre-incubated in PBS solution for various time intervals did not induce decomposition or affect the inhibitory activity (see Figure S4, Supporting Information). These eight-membered heterocycles also did not produce any conjugate products involving the lysine groups of model peptides. With these points in mind, it can be strongly speculated that cSPM was therefore the active structure that inhibited fi bril formation.
In addition, NMR studies of the Aβ40 peptide titrated using various concentrations of cSPM in PBS revealed chemical shifts at several residues were changed, which includes positions Arg5, His6, Ala21, Ser26, and Asn27 ( Figure 6 ; Figure S5, Supporting Information). [ 31 ] Therefore, it could be suggested that these residues may be responsible for suppressing peptide aggregation.
Finally, Aβ40 fi brillization was found to be directly suppressed by in situ generated cSPM (i.e., by simultaneously treating with polyamine and acrolein) by virtue of the facile [4+4] cycloaddition process that occurs in aqueous media ( Figure 7 ). Note that neither polyamine nor acrolein alone inhibited fi brillization, as shown in Figure 1 . Given that these heterocycles were produced in vivo as the oxidative metabolites of the polyamines, a unique strategy for neuronal disease treatment may be envisioned.
In summary, we found that 1,5-diazacyclooctanes, the exclusive and biologically relevant products between polyamines and acrolein, inhibited Aβ40 peptide fi brillization and significantly suppressed cytotoxicity. These compounds may inhibit the formation of the highly toxic "soluble" oligomer species while minimizing the toxic "insoluble" Aβ40 fi brillization process. There was no signifi cant difference in inhibitory activity between cSPM and cSPD in Figure 1 . The results therefore show that the cyclic 1,5-diazacyclooctane structure is critical to show the activity. The polyamines, SPM and SPD, have different expression level in AD process, but once smoothly reacted with acrolein in/out of the cells under the oxidatively stressed conditions, the corresponding cSPM and cSPD products might similarly inhibit the fi brillization. The results described in this Communication corroborate our discovery that the formal [4+4] cycloaddition reaction is involved in modulating oxidative stress processes associated with neural diseases.   Aβ40 fi brils (control); Aβ40 was incubated without compounds. Aβ40 + SPM + Acrolein; Aβ40 was incubated with mixture of equimolar SPM and acrolein at 0.1 × 10 −6 , 0.5 × 10 −6 , 1.0 × 10 −6 , 5.0 × 10 −6 , or 10 × 10 −6 M . Aβ40 + cSPM; Aβ40 was incubated with cSPM (0.5 × 10 −6 M , control). Aβ40 monomer (control); Aβ40 was not incubated. Thiofl avin T (ThT) Fluorescence Assay : The amyloid solution prepared above was diluted to 12.5 × 10 −6 M in PBS, and 40 µL of the solution was mixed with 160 µL of 25 × 10 −6 M ThT solution in 50 × 10 −3 M Glyine-NaOH buffer (pH 8.0) (fi nal concentration, 2.5 × 10 −6 M Aβ and 20 × 10 −6 M ThT). ThT fl uorescence was monitored using a spectrofl uorometer (Ex = 440 nm, Em = 495 nm). The average values of three-wells are shown. Experimental results for optimization of the ThT assay conditions are shown in Figure S1a-c (Supporting Information). It was also confi rmed that the presence of cSPM does not affect ThT intensity of Aβ fi brils ( Figure S1d, Supporting Information). In addition, the effect of DMSO on Aβ40 aggregation was examined and it was shown that although at higher DMSO concentration (10%) Aβ aggregation was almost completely inhibited, signifi cant amount of Aβ aggregation was still observed in the presence of 2.5% DMSO, supporting that cSPM could inhibit Aβ aggregation even in the presence of a small amount of DMSO ( Figure S2, Supporting Information).

Inhibition of Amyloid Peptide Aggregation
Cytotoxicity Assay : Cell viability was determined using cell proliferation kit (Roche, Basel, Switzerland). PC12 cells (a clonal line of rat pheochromocytoma) were plated in PDL-coated 96-well plates at a density of 40 000 cells per well and grown overnight. The cSPMtreated Aβ samples prepared above was diluted with PBS to the various concentrations (20 µL), and were added to the PC12 cells (in 80 µL of medium). The fl uorescence intensity of formazan product at 550 nm was measured by microplate reader (see details in the Supporting Information).

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.