Interaction of amyloid beta with humanin and acetylcholinesterase is modulated by ATP

Both HN and AChE can bind Aβ in the absence of added ATP. Addition of ATP increases the binding affinity of Aβ to HN but not to AChE.

Humanin (HN) is known to bind amyloid beta (Ab)-inducing cytoprotective effects, while binding of acetylcholinesterase (AChE) to Ab increases its aggregation and cytotoxicity. Previously, we showed that binding of HN to Ab blocks aggregation induced by AChE and that HN decreases but does not abolish Ab-AChE interactions in A549 cell media. Here, we set out to shed light on factors that modulate the interactions of Ab with HN and AChE. We found that binding of either HN or AChE to Ab is not affected by heparan sulfate, while ATP, thought to reduce misfolding of Ab, weakened interactions between AChE and Ab but strengthened those between Ab and HN. Using media from either A549 or H1299 lung cancer cells, we observed that more HN was bound to Ab upon addition of ATP, while levels of AChE in a complex with Ab were decreased by ATP addition to A549 cell media. Exogenous addition of ATP to either A549 or H1299 cell media increased interactions of endogenous HN with Ab to a comparable extent despite differences in AChE expression in the two cell lines, and this was correlated with decreased binding of exogenously added HN to Ab. Treatment with exogenous ATP had no effect on cell viability under all conditions examined. Exogenously added ATP did not affect viability of cells treated with AChE-immunodepleted media, and there was no apparent protection against the cytotoxicity resulting from immunodepletion of HN. Moreover, exogenously added ATP had no effect on the relative abundance of oligomer versus total Ab in either cell line.
Besides its well-documented intracellular role as a molecular energy source, ATP is known to be a ubiquitous extracellular messenger that acts on purinergic receptors to activate a number of intracellular signaling cascades [1,2]. While the concentration of extracellular ATP in normal tissues has been found to be~1-5 lM, it is elevated in the tumor microenvironment to levels (> 100 lM), which might induce normal cells to undergo apoptosis [1,3]. Cancer cells that include non-small-cell lung carcinoma (NSCLC) A549 cells have been shown to release ATP and tolerate extracellular ATP concentrations that would otherwise lead to a cytotoxic response in normal cells [1].
Almost all types of cells produce amyloid beta (Ab), a peptide well recognized for its role in the development and progression of different stages of Alzheimer's disease (AD) [4][5][6][7][8]. The Ab peptide is~4 kDa and derived from the sequential processing of the higher molecular weight amyloid precursor protein by two membrane-bound endoproteases, band c-secretase [4,9]. Different C-terminal Ab heterogeneity results from processing by c-secretase, where Ab40 represents the most abundant isoform (~90%) as compared to Ab42 (~10%) [4,9]. Self-assembled Ab40/42 peptides into amyloid fibrils are thought to be implicated in the pathology of more than 20 devastating and serious human disorders, including AD and other neurodegenerative diseases [6][7][8][10][11][12][13]. The Ab40 peptide has a lower tendency to form oligomers, shows lower aggregation kinetics, and displays lower toxicity than Ab42 [7]. Of the two main forms of Ab in the brains of patients with AD, Ab42 has enhanced amyloidogenicity and is more toxic and fibrillogenic with faster aggregation kinetics [14]. The sequence of Ab is partitioned into a hydrophilic N-terminal region, while the C-terminal part is composed of nearly all hydrophobic amino acids, proposed to account for its propensity to aggregate at neutral pH [15].
The mechanisms by which Ab monomers are converted into functional entities and various types of dysfunctional assemblies are largely obscure [16]. Complementary approaches [17], employing molecular dynamics simulations and experimental methodology, have provided information about inhibitors that reduce aggregation and toxicity of different Ab species [18,19] and structural details of a broad range of interconverting Ab assemblies that range in size, conformation, and toxicity between monomers [20][21][22][23], oligomers [6,24], protofibrils [25], and fibrils [26].
A rapidly growing body of evidence has recently steadily emerged showing that AD patients might have a reduced risk and some protection against certain cancers. [27] Inverse associations between cancer and AD [28][29][30][31][32][33] have been reported showing that patients with AD generally had a significantly reduced risk of developing cancer with time, while individuals diagnosed with cancer have a reduced likelihood of living long enough to develop AD [27]. The incidence of AD was found to be reduced in glioblastoma and in other types of cancers including lung cancer [33]. Experimental evidence indicates that Ab is protective against certain types of cancer and could inhibit the growth of tumor cells [34,35]. Following treatment of cancer cell lines with conditioned media containing Ab, proliferation of human breast adenocarcinoma, melanoma and glioblastoma [35] was inhibited. Direct injection of Ab into human lung adenocarcinoma xenografts was also found to suppress tumor growth in mice [34]. Plasma levels of Ab40 and Ab42 were reported to be higher in all cancer patients compared with normal controls [36]. To gain further mechanistic insights into regulation of Ab in lung cancer cells, we used two human NSCLC cell lines [37], A549 (p53positive) and H1299 (p53-null) cells [38] in this study.
Soluble oligomers of proteins implicated in different diseases are primarily thought to be the main toxic form as opposed to the larger fibrillar assemblies [13,39,40]. Ab regions containing Tyr and Ser (H 6 DSGY 10 and G 25 SNKG 29 ) (Fig. 1) along with post-translational modifications of these regions, have been implicated in misfolding, oligomerization, or fibril formation of Ab [5,10,11,16,41]. ATP is known to be protective against Ab-mediated cytotoxicity [42], and lower extracellular ATP levels were found to correlate with increased misfolded extracellular Ab in AD [43,44]. Ab proteins are known to bind DNA and RNA [42,45] with residues 25-35 comprising the DNA binding region. This region is within the GxxxG motif on Ab (Fig. 1), involved in both Ab oligomerization and nucleotide binding [42,46]. Computational and biochemical studies demonstrated that ATP strongly interacts with both Tyr10 and Ser26 of Ab fibrils ( Fig. 1) and that both ATP and ADP reduced misfolding of Ab at physiological intracellular concentrations, an effect that was enhanced by magnesium, the levels of which are known to be lowered in AD [42,47]. In aqueous solution, monomeric Ab is known to be intrinsically disordered but upon conversion into fibrils, amino acid residues Val12-Val24 and Ala30-Val40 each form a b-strand with Gly25-Gly29 forming a bend that results in parallel b-sheets [41,48]. Formation of this bent structure composed of Gly25-Gly29 is thought to be an early event in self-association of Ab into fibrils [41,48] of Ab [49], suggesting that phosphorylation of Ser26 may impact Ab oligomerization and assembly. A~10 6 -fold difference has been previously reported between the extracellular concentrations of ATP (nMlow lM range) and the intracellular concentrations of ATP (mM range) [2,42], a difference that was suggested to affect intracellular and extracellular folding of Ab. Phosphorylation of Ab was detected in primary cultures of mouse cortical neurons at low nanomolar concentrations of ATP, suggesting that Ab can be phosphorylated in vivo at physiological concentrations of extracellular ATP [50]. In addition to ATP, the glycosaminoglycan, heparan sulfate (HS), has been previously reported to interact with Ab peptides, increasing their aggregation [51,52]. In both Ab40/42, amino acid residues 12-18 (VHHQKLV) are reported to be important for interaction of Ab with HS ( Fig. 1).
Humanin (HN) is a secreted 21-to 24-amino acid mitochondrial-derived peptide [53,54]. Certain amino acid residues in HN have been identified to be involved in different functions, including binding to Ab [55,56]. Growing evidence suggests that HN is a peptide with broad spectrum cyto-and neuroprotective actions that prevent different types of stress [55,57,58]. HN was identified previously as a binding partner of Ab, likely modulating its aggregation pathways and counteracting its deleterious effects [56,57,59]. The morphology of Ab40 was altered by HN from fibrillary to amorphous [60], likely protecting against Abinduced cytotoxicity. Using circular dichroism and NMR [61], HN was found to be unstructured and flexible in aqueous solutions. HN was shown to take up a helical structure (Gly5 to Leu18) in a less polar environment, however, which might enable it to pass through membranes in its helical conformation forming specific interactions, while conformational changes leading to an unstructured form might allow the peptide to interact with different receptors [61].
Numerous attempts that employ a broad range of small molecule and peptide inhibitors are currently underway to delay the self-assembly of monomeric Ab into oligomeric forms [5,13,62]. While similar structures are adopted by Ab40 and Ab42 when part of the fibril, minimal information of the three-dimensional structures of monomers and oligomers of either Ab40 or Ab42 in aqueous solution is currently available [6]. HN has been shown earlier to directly interact with Ab oligomers [63]. Therefore, and due to its known function as a natural broad spectrum cytoprotective peptide, direct binding of HN to Ab may enable it to block formation and/or toxicity of aggregated Ab assemblies.
Acetylcholinesterase (AChE) is an enzyme known for its role in terminating acetylcholine-mediated neurotransmission at the synaptic cleft [67]. The majority of the cortical AChE in the Alzheimer's brain is mainly associated with the amyloid core of senile plaques [68][69][70][71][72]. AChE forms a stable complex with Ab during its assembly into filaments, increasing the aggregation and neurotoxicity of Ab fibrils to levels higher than those of the Ab aggregates alone [72,73]. AChE increases Ab42 oligomeric formation [74] and is known to be associated with amyloid plaque accumulation of abnormally folded Ab40, considered as a main component of the amyloid plaques found in the brains of AD patients [67][68][69][70][71][72][73]. Addition of AChE significantly accelerated the aggregation of Ab40 and assembly into Alzheimer's fibrils via decreasing the lag phase of the aggregation of the peptide, likely by a mechanism affecting the nucleation step and/or fibril elongation [68,[70][71][72][73]. Noncatalytic functions of AChE were suggested earlier since the catalytic active center of the enzyme was not required for Ab40 amyloid fibril formation [75]. The AChE peripheral anionic site was identified as the site where Ab interacts, accelerating formation of amyloid fibrils and leading to a highly toxic complex [74]. Higher toxicity was associated with the AChE-amyloid complexes as compared to the toxicity of the Ab aggregates alone [72]. Binding assays indicated [71] that AChE binds to Ab (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28), as well as to the Ab (1-16) peptide ( Fig. 1), directly promoting aggregation of Ab40 and assembly into amyloid fibrils.
Access to the central domain of Ab (residues 17-24) flanked by Lys-16 and Lys-28, known to be a critical structural element in fibrillar Ab aggregates [76][77][78], might be regulated by binding of AChE and HN to their overlapping binding sites on Ab. Despite the known opposing effects of HN and AChE on the oligomerization of Ab [53,54,60,[63][64][65]68,[70][71][72]74], we recently found that while the binding of AChE to Ab is decreased in the presence of HN, it is not abolished, and that Ab aggregation is greatly diminished in the presence of both HN and AChE to levels close to those induced by HN alone [79]. Moreover, we showed that the relative amount of Ab oligomer versus total Ab was increased upon immunodepletion (ID) of HN from A549 and H1299 lung cancer cell-conditioned media, decreasing cell viability and increasing apoptosis [79].
Here, we set out to understand the effect of HS and ATP on the ability of Ab to interact with either HN or AChE. We show that binding of HS does not alter the interaction of Ab with either HN or AChE. However, upon addition of ATP, there was increased interaction between HN and Ab but decreased affinity of Ab for AChE. Moreover, exogenous ATP, added at concentrations that abolished AChE binding to Ab in A549 cell media, did not protect against the cytotoxicity resulting from ID of HN in either A549 cells that express AChE or H1299 cells with minimal expression of the enzyme.

ELISA
ELISAs were carried out as we previously reported [65,80,82]. Nunc MaxiSorp 96-well flat bottom plate (Thermo Fisher) wells were coated with samples as indicated. The plates were incubated overnight at 4°C on a shaker to allow binding of the samples to the plate wells. After the incubation, the wells were washed 49 with TBST, filled with 400 µL blocking buffer (110 mM KCl, 5 mM NaHCO 3 , 5 mM MgCl 2 , 1 mM EGTA, 0.1 mM CaCl 2 , 20 mM HEPES, 1% BSA, pH 7.4), and incubated with shaking overnight at 4°C. The wells were then washed 49 with TBST, and 100 µL of sample at the desired concentration was added to each well and incubated with shaking overnight at 4°C. TBST was then used to wash the wells 49 before proceeding in one of two ways: (a) biotinylated samples were analyzed by adding 100 µL streptavidin-HRP conjugate in TBST (1 : 2500 dilution) to the samples followed by incubation for 3 h at RT on a shaker, or (b) samples without biotin were analyzed by adding 100 µL TBST containing the primary antibody as per the manufacturer's recommendation, incubating for 3 h at RT on a shaker, followed by washing the wells 49 with TBST. The secondary antibody in 100 µL TBST was then added to the samples following the manufacturer's recommendation and incubated for 1 h at RT on a shaker. Plates containing either biotinylated or nonbiotinylated samples were then washed 59 with TBST followed by the addition of 100 µL TMB, which resulted in a blue color change. The reaction was stopped with 100 µL 2 M H 2 SO 4 after incubating at RT for 0.5-15 min, resulting in a yellow color change, measured by absorbance at 450 nm. To monitor nonspecific binding, negative control wells on the plates included, for example, bound Ab peptide then adding all components, streptavidin-HRP and TMB, but without addition of biotin-HN. Some wells were coated with 2.5, 10, 50, 100, 500, and 5000 nM biotin-HN or Ab to allow conversion of the OD measurements to concentrations of bound material. Before analysis, the OD from the data was corrected for nonspecific binding by subtracting the mean background absorbance for the negative controls. Typically, in control wells incubated on each plate, the background binding is about 10-15% of the maximum binding seen with addition of biotin peptides or antibodies. Statistical analysis was determined by the GRAPHPAD PRISM 8.4.3 software (San Diego, CA, USA). Data were expressed as the mean AE SD. Three independent experiments were carried out in triplicate for each assay condition.

Quantitation of Ab
Ab ELISAs were carried out according to previous protocols [83,84] for determining the oligomeric and monomeric concentrations of Ab and as we recently reported [79]. Briefly, total Ab (monomers + oligomers) was measured by two-site binding ELISAs using the capture 6E10 monoclonal antibody and 4G8-conjugated biotin as the detection antibody, which recognizes a distinct epitope, then quantitated using streptavidin-HRP.
Using the same samples, oligomerized Ab was measured by a single-site ELISA in which antibodies targeting the same primary sequence epitope were used for both capture (4G8) and detection (4G8-biotin). Only oligomers are detected with this approach since the 4G8-biotin antibody cannot bind to the captured monomer because the epitope is blocked by the 4G8 capture antibody. Therefore, only oligomeric or multimeric Ab containing additional exposed 4G8 epitopes, not engaged by the capture antibody, are reported by the streptavidin-HRP. The amount of the monomer was then estimated as the difference between the concentration of total Ab and the concentration of the oligomer.

MTT assay
The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) reduction assay (Sigma-Aldrich), used to measure cell viability, was carried out as we reported earlier [80,82,85]. Cells were seeded in 96-well plates as indicated in 200 lL 10% FBS-supplemented media per well and maintained overnight at 95% humidity and 5% CO 2 . After an overnight incubation, the media were replaced with 200 lL serum-free media, and the cells were further incubated, without or with different treatments, for 24, 48, or 72 h. The final concentration of DMSO in each well never exceeded 0.1%. The cells were then incubated for 4 h with MTT (0.5 mgÁmL À1 ) in the dark. The media were carefully removed, and DMSO (100 lL) was added to dissolve the formazan crystals. The absorbance was measured at 570 nm in a plate reader. Untreated cells or wells containing only DMSO and media were used as a positive and negative control, respectively. Statistical analysis was conducted using GRAPHPAD PRISM version 8.4.3 for Windows. Significant values were considered at P < 0.05 and more significant values at P < 0.01, compared with the control.

Immunodepletion
Conditioned media were immunodepleted (ID) according to the methods previously described [86] and our recently published report [79]. Briefly, specific antibodies were bound to ELISA wells overnight (1 : 1000 dilution). The wells were then blocked and washed, then 300 lL of the conditioned medium (0.5 lgÁlL À1 ) treated as indicated, 72 h postserum starvation, was incubated with the antibodies bound to ELISA wells for 24 h. The ID media were then carefully removed and analyzed for the presence of the target protein or peptide by ELISA. The same amount of protein (3 µL of 600 µgÁmL À1 total protein) of each sample was analyzed in the experiments. Significant depletion (95-100%) was observed upon using each of the antibodies employed in this study.

Statistical analysis
The analysis was carried out as we previously reported [79,81,82]. Each experiment in this study was performed in triplicate and repeated a minimum of three times. Statistical values are expressed as the mean AE SD. To evaluate the statistical differences, the Mann-Whitney or Kruskal-Wallis (ANOVA) tests were used. All the statistical tests were two-sided and a P value of < 0.05 was considered statistically significant in all cases. GRAPHPAD PRISM (GraphPad Software, 8.4.3) was used for the statistical analysis.
Our previously published reports show that biotinylated-Ab interacts with the same affinity as the nonbiotinylated peptide with HN, and similarly, the binding of either HN or biotinylated-HN to Ab is indistinguishable [65,79]. To examine the binding of Ab40 and Ab42 to HS, ELISA plate wells were coated with HS (100 nM). Increasing concentrations of biotinylated-Ab were then added ( Fig. 2A) to the wells and processed as described in the Materials and methods. Optical densities (450 nm) were normalized for both curves by expressing each point relative to the best-fitted E max value (set to 100%). The data were plotted as a function of increasing biotinylated-Ab concentrations and fit to a single binding site model with a nonlinear regression curve fitting approach, using the GRAPHPAD PRISM 8.4.3 software. Both Ab40/42 were found to bind HS with comparable affinities (Fig. 2A). We next tested whether increasing concentrations of HS can compete with binding of Ab to either HN (Fig. 2B) or AChE (Fig. 2C). Ab (100 nM) was bound to ELISA plate wells. Biotinylated-HN (300 nM) was then added to the wells in the absence or presence of increasing concentrations of HS (Fig. 2B). Similarly, AChE (10 nM) was bound to ELISA wells followed by the addition of biotinylated-Ab (1 lM) in the absence or presence of increasing concentrations of HS (Fig. 2C). The negative controls had the same HN or Ab and AChE concentrations, but water was substituted in place of biotinylated-Ab or biotinylated-HN. Data were processed using the GRAPH-PAD PRISM 8.4.3 software and presented as the mean AE SD of three independent assays. No effects of HS on the binding of Ab40 or Ab42 to either HN (Fig. 2B) or AChE (Fig. 2C) were observed at any of the HS concentrations used.

ATP weakens interactions between AChE and Ab but strengthens those between Ab and HN
Ab peptides are known to bind DNA and RNA [42,45] with amino acid residues 25-35 comprising the DNA binding region found within the Ab GxxxG motif (Fig. 1), involved in both Ab oligomerization and nucleotide binding [42,46]. ATP was shown to strongly interact with both Tyr10 and Ser26 of Ab fibrils ( Fig. 1) and reduce misfolding and fibrillation of Ab at physiological intracellular concentrations, an effect that was enhanced by magnesium [42,47]. Moreover, the aggregation of Ab16-22 was reported, by an in silico study, to be highly unfavorable in the presence of ATP that formed hydrogen bonding, p-p stacking, and NHÀp interactions with the Ab16-22 peptide, preventing its aggregation [88].
To examine the effect of added ATP, if any, on the binding of either AChE or HN to Ab, AChE (10 nM) was bound to ELISA plate wells (Fig. 3). Increasing concentrations of biotinylated-Ab, preincubated for 30 min at RT in 50 mM Tris/HCl, pH 7.5, 0.1 mM EDTA, 10 mM MgCl 2 , 2 mM DTT buffer without or with 200 lM ATP, were added to the wells and processed as described in the Materials and methods. Similarly, Ab (100 nM) preincubated for 30 min at RT in 50 mM Tris/HCl, pH 7.5, 0.1 mM EDTA, 10 mM MgCl 2 , 2 mM DTT buffer without or with 200 lM ATP was bound to the wells (Fig. 4), and then, increasing concentrations of biotinylated-HN were added to the wells and processed. Optical density measurements (450 nm) were normalized by expressing each point in relation to the best-fitted E max value (set to 100%) and then plotted as a function of increasing biotinylated-Ab or biotinylated-HN concentrations. The data were fit to a single binding site model with a nonlinear regression curve fitting approach using the GRAPHPAD PRISM 8.4.3 software.
Addition of ATP reduced the affinity of AChE to both Ab40 and Ab42 (Fig. 3). Conversely, the affinity of either Ab40 or Ab42 for HN was increased (Fig. 4). The concentration of ATP used (200 lM) was chosen to be within the extracellular physiological concentrations previously found (> 100 lM) in the lung cancer cell lines used in this study [1][2][3]. These results might suggest that ATP binding to Ab promotes optimal alignment of the amino acids needed to interact with HN increasing binding affinity, while conversely, it renders the amino acid residues on Ab important for binding AChE, less accessible.
More HN is found in a complex with Ab upon addition of ATP to the conditioned media of either A549 or H1299 cells, while levels of AChE found in a complex with Ab are decreased by ATP addition to A549 cell media We next set out to examine whether addition of ATP modulates the binding of either HN or AChE to Ab using the conditioned media from A549 cells and H1299 cells that we previously used to examine these interactions [79]. Cells (0.2 9 10 5 cells per well) were seeded in 96-well plates in 10% FBS-supplemented media. The next day, the cells were incubated in serum-free medium for 72 h. Specific antibodies were added (1 : 1000 dilution) to ELISA wells (Fig. 5). After blocking the wells, 300 lL of the A549 or H1299 cell-conditioned medium (0.5 lgÁlL À1 ), 72 h postserum starvation, was added in the absence or presence of ATP (100 lM, 1 mM). The proteins/peptides were detected using their corresponding primary antibodies and then processed as described in Materials and methods section. It was previously shown that the antibody, 6E10, is highly specific for nonphosphorylated Ab, while the antibody, 82E1, detects both phosphorylated and nonphosphorylated peptides [50]. Since the objective of this experiment was to test the binding of HN and AChE to Ab in the conditioned media without and with added ATP, both antibodies that recognize all species of Ab without regard to conformation were used in the case of any possible complications that might occur due to phosphorylated Ab, if any (Fig. 5). The 6E10 antibodies are known to react with monomers, oligomers, and fibrils of Ab [89,90] and recognize the Nterminal hydrophilic sequence, amino acids 1-16 of Ab. This epitope, previously reported to be exposed in Ab aggregates [89], has been shown by a high-resolution mapping approach, to be residues 4-10 [91]. The 82E1 monoclonal antibodies are known to be specific to the N terminus of Ab and recognize residues 1-16 [50].
The effects on the binding of HN to Ab were comparable using either 6E10 or 82E1 antibodies bound to ELISA wells (Fig. 5) incubated with the conditioned media of either A549 or H1299 cells. Upon addition of 100 lM ATP to the conditioned media of A549 cells, there was an~1.85-fold increase in the amount of HN bound to Ab (Fig. 5A), a level that was further increased to~2.5-fold upon addition of 1 mM ATP, relative to samples without added ATP. Similar results were obtained upon binding of ELISA wells with anti-HN antibodies and using either 6E10 or 82E1 to detect Ab from A549 cell-conditioned media (Fig. 5B). Increased levels of HN bound to Ab in the presence of ATP were also comparable using H1299 cell-conditioned media (Fig. 5C,D). Recently, using human lung carcinoma NSCLC cell lines [82], we found that treatment of A549 cells (p53-positive) with p53 siRNA blocked AChE expression while no change in the levels of the enzyme was found with this siRNA treatment using H1299 cells with a p53-null genotype due to a biallelic deletion of the TP53 gene [38]. Binding of AChE from the conditioned media of A549 cells to Ab was comparable when either 6E10 or 82E1 was bound to the ELISA plate wells (Fig. 5A). The levels of AChE found in a complex with Ab were decreased bỹ 2.3-fold upon addition of 100 lM ATP, then decreased to almost blank levels upon addition of 1 mM ATP (Fig. 5A). Binding anti-AChE antibodies to the plate wells also showed a comparable decrease in the levels of Ab detected by either 6E10 or 82E1 antibodies, upon addition of ATP (Fig. 5B). No AChE was detected above background using H1299 cell- conditioned media (Fig. 5C,D), consistent with our previous report showing negligible levels of AChE in H1299 cells relative to those in A549 cells [82]. Despite differences in AChE expression in the two cell lines, however, interaction of HN from the conditioned media of both cell lines with Ab appears to be augmented by the addition of ATP to a comparable extent (Fig. 5). These results might suggest that increased interactions of HN with Ab by ATP may not be modulated by AChE.  Exogenously added ATP increased interactions of HN from A549 and H1299 cell-conditioned media with Ab and correlated with decreased binding of exogenously added HN, and less AChE in a complex with Ab from A549 cell-conditioned media We previously reported that while HN weakens the interactions of AChE with Ab, it does not abolish the enzyme's ability to bind Ab [79]. Since our results showed (Fig. 5) that addition of ATP increases the interaction of HN with Ab using the conditioned media from either A549 or H1299 cells but decreases the binding of AChE to Ab in A549 cell media, we set out to understand how addition of ATP might affect binding of exogenously added HN to Ab. Anti-Ab-specific antibodies (82E1) were added (1 : 1000 dilution) to ELISA plate wells (Fig. 6). The wells were blocked, and then, 300 µL of the conditioned media (0.5 µgÁµL À1 ) of A549 or H1299 cells, 72 h postserum starvation, was added in the absence or presence of increasing HN concentrations, and the HN and AChE bound were detected using the corresponding specific primary antibodies. Fold change relative to controls that included all components but without the primary antibodies was calculated and fit with a nonlinear regression curve using the GRAPHPAD PRISM 8.4.3 software. While binding of AChE to Ab immobilized to its antibody, 82E1, was detected using the A549 cell-conditioned media, barely detectable levels of AChE from the conditioned media of H1299 cells were found to bind Ab (Fig. 6). This finding is not surprising since we recently showed [82] that there are minimal levels of AChE in the conditioned media of the p53-null cell line, H1299, as compared to the media from the p53positive cell line, A549. Binding of Ab to AChE from the A549 cell-conditioned media was decreased upon addition of ATP (Fig. 6A). In the conditioned media of both cell lines, HN was found bound to Ab (Fig. 6). To examine the ability of exogenously added HN to bind Ab upon addition of ATP, we incubated the wells with increasing concentrations of the HN peptide followed by washing the unbound material as described in Materials and methods. Addition of exogenous HN resulted in its increased binding to Ab from both A549 and H1299 media bound to its immobilized antibody, 82E1. In both cases, increasing ATP concentrations resulted in increased binding of endogenous HN from the conditioned media and corresponded to decreased binding of exogenously added HN, an effect that was more pronounced using 1 mM as opposed to 0.1 mM ATP. These results likely indicate that the affinity of endogenous HN to Ab is increased upon incubation with higher ATP concentrations in the media of both cell lines, which is expected to correlate with reduced binding of exogenously added HN. This interpretation is supported by the observed increased affinity of HN to either Ab40 (Fig. 4A) or Ab42 (Fig. 4B) in the presence of ATP. Moreover, the curves from each cell line treatment leveled off at comparable higher concentrations of exogenously added HN. The signal obtained upon adding HN increased, then began to level off around 50 nM (Fig. 6), concentration of HN was~50 nM when measured previously [80] in the A549 media. Since the effect of added ATP on HN interactions with Ab is comparable using either A549 or H1299 cell-conditioned media and as AChE levels in H1299  media are negligible relative to those found in A549 cell media, one possibility is that AChE is inefficient at regulating the effects of ATP on the interaction of HN with Ab. Our data show (Fig. 6) that the affinity of HN, added exogenously at lower concentrations, to Ab is increased upon addition of ATP. The lack of an increase in the binding of HN, added exogenously at concentrations that exceed 50 nM in the absence or presence of ATP, suggests that binding of HN to Ab is closer to reaching saturating levels under these conditions.
Exogenously added ATP has no effect on A549 or H1299 cell viability Since addition of ATP appears to increase binding of HN to Ab, and conversely, decreases the interaction of Ab with AChE (Figs 3-6), we tested the effect of added ATP on A549 and H1299 cell viability (Fig. 7). Exogenously added ATP at either 100 µM or 1 mM had no effect on viability of either A549 (Fig. 7A) or H1299 cells (Fig. 7B). These results are consistent with previous reports showing that there is no effect upon addition of either 100 lM or 1 mM ATP on A549 cell viability [3] suggesting that, compared with normal cells, lung cancer cells exhibit reduced cytotoxicity upon treatment with these extracellular ATP concentrations. Earlier, we found that ID of HN from A549 or H1299 cell-conditioned media led to diminished cell viability [79]. To determine the effect of ID of AChE or HN in the absence or presence of added ATP on cell viability, ID media were prepared by first seeding 0.2 9 10 5 cells per well in 96-well plates in 10% FBSsupplemented media. The next day, the cells were incubated in serum-free medium for 72 h, then ID of AChE or HN as described in Materials and methods. For cell viability assays, cells were seeded in 96-well plates at 0.2 9 10 5 cells per well in 200 µL 10% FBSsupplemented media followed by incubation in serumfree medium for 12 h, then treated with the ID media for 48 h. The medium containing the specific components in the different treatments was replaced every 12 h. HN ID resulted in an approximate 0.45-and 0.35-fold decrease in A549 and H1299 cell viability, respectively (Fig. 7). No change in cell viability was observed, however, upon treatment of A549 or H1299 cells with media ID of HN, with added ATP at either 100 µM or 1 mM concentrations as compared to that measured in the absence of added ATP (Fig. 7). Similarly, addition of ATP did not affect cell viability (Fig. 7) upon cell incubation with AChE-depleted media. ID of AChE had no effect on H1299 cell viability (Fig. 7B) since they have minimal expression of the enzyme [82] while A549 cell viability increased (Fig. 7A)~1.45-fold upon AChE ID, effects that were not further modulated upon addition of ATP. This increase in cell viability upon ID of AChE from A549 cell media might reflect the recognized role of AChE as a tumor suppressor and a pro-apoptotic gene in NSCLC cells that attenuates cell growth when its expression is upregulated [92,93]. These tumor suppressor functions are known to be in part due to the catalytic hydrolysis of acetylcholine [92][93][94][95]. Therefore, this increase in cell viability might reflect attenuating the general adverse effects of AChE on A549 cell viability that include its ability to induce Ab aggregation.
These results indicate that differences in the binding of HN and AChE to Ab upon addition of exogenous ATP have no impact on viability of either A549 or H1299 cells. Therefore, factors affecting cell viability appear to depend not on additionally added ATP or its consequent effects on modulating the binding of either HN or AChE to Ab, but rather on depleting HN or AChE from the media. ATP is known to be protective against Ab-mediated cytotoxicity [42] and leads to reduced misfolding of Ab at physiological intracellular concentrations [42,47]. Reduction of extracellular ATP levels has been shown to correlate with increased misfolded extracellular Ab in AD [43,44]. In normal tissues, the concentration of extracellular ATP has been shown to be~1-5 lM; however, in the tumor microenvironment, ATP levels rise to concentrations greater than 100 lM, amounts that might lead normal cells to undergo apoptosis [1,3]. Cancer cells including NSCLC A549 cells have been found to release ATP and tolerate extracellular ATP concentrations that would otherwise lead to a cytotoxic response in normal cells [1]. Our results show that at the highest concentrations of ATP (1 mM) added extracellularly to A549 cell-conditioned media, known to express higher levels of AChE as compared to H1299 cells [82], the levels of the enzyme bound to Ab were close to blank values suggesting that the interaction of AChE with Ab was blocked by addition of 1 mM ATP (Figs 5 and 6). Under these conditions, more HN was found bound to Ab in both A549 and H1299 cells (Figs 5 and 6). However, addition of higher ATP concentrations does not seem to protect against the cytotoxicity resulting from ID of HN in either A549 or H1299 cells (Fig. 7) suggesting the need for further investigation into this observation.
The relative amount of oligomer versus total Ab upon immunodepletion of either AChE or HN from the cell-conditioned media is unaffected by the addition of ATP Both epitopes recognized by the 6E10 and 4G8 antibodies have been previously shown to be exposed in Ab aggregates [89,96]. The 4G8 : 6E10 ratio was suggested to be a marker for the relative amount of aggregated versus monomeric Ab [96]. To determine the effect of ID of AChE or HN on this ratio in the absence or presence of added ATP, 0.2 9 10 5 cells per well were seeded in 96-well plates in 10% FBS-supplemented media. The next day, the cells were incubated in serum-free medium for 72 h, then ID of AChE or HN as described in Materials and methods. The antibodies 6E10 or 4G8 were bound (1 : 1000 dilution) to ELISA wells (Fig. 8). The wells were blocked, and then incubated with 300 lL of the ID medium (0.5 lgÁlL À1 ). Biotin-4G8 was then added and the signal processed as described in Materials and methods section. Fold change relative to controls using anti-6E10 or anti-4G8 antibodies incubated with 300 lL of the medium not incubated with cells was calculated.
Relative to total Ab, there was more oligomer in A549 cell-conditioned media (~63%) as compared to that found in media of H1299 cells (~48%) (Fig. 8). ID of HN reduced the total Ab in A549 cell media tõ 63% and to~58% in H1299 cell media. Compared to the total Ab that remained after HN ID, there was relatively higher oligomer (~86% of total) in A549 cell media as compared to H1299 media (~69%). While those results might suggest that depletion of HN promotes the ability of AChE to increase Ab oligomer formation in A549 cell media, this suggestion is unlikely to be correct since no change was found upon exogenously added ATP to A549 cell media (Fig. 8A), conditions under which the binding of the enzyme to Ab was largely abolished (Figs 5 and 6). ID of AChE from A549 cell media decreased the total Ab in the media to~57% (Fig. 8A). The amount of the oligomer was~47% of the total Ab remaining after the ID of AChE suggesting that removing AChE from the A549 cell media reduces the ratio of oligomer to total Ab relative to that in undepleted media. No change upon ID of AChE was observed in H1299 cell media (Fig. 8B), which is not surprising since expression of AChE is minimal in this cell line [82].
Our data show that the relative amount of oligomer versus total Ab is increased upon ID of HN from A549 or H1299 cell-conditioned media (Fig. 8) and correlates with diminished cell viability (Fig. 7). The mechanisms employed by HN in protecting against Ab oligomerization or toxicity under these conditions are unclear. One can imagine, however, that depletion of HN renders Ab susceptible to modifications increasing its oligomerization despite the higher concentrations of ATP. While binding of ATP might reduce misfolding of Ab, the lack of effect found with exogenously added ATP on the relative amount of oligomer versus total Ab (Fig. 8) using nondepleted media or media ID of either HN or AChE might indicate active processes occurring that use ATP to cause Ab aggregation, leading to an overall negligible change in the oligomer versus total Ab ratio. For example, various types of cancer cells, including lung adenocarcinoma, were reported to excrete extracellular cAMP-dependent protein kinase [97], the activity of which is elevated, and in serum samples from patients with different types of cancers, constitutive kinase activity compared with normal, was reported. Extracellular Ab was previously shown to be phosphorylated on Ser8 by protein kinase A that is either secreted or localized on the cell surface [98]. This phosphorylation, detected in transgenic mice and brains from human AD patients, promoted formation of toxic oligomeric and fibrillar Ab assemblies both in vitro and in vivo [50,98], and increased the stability of Ab aggregates against dissociation into monomers by SDS [99]. In monomeric Ab, Ser8 was found in a region of high conformational flexibility that upon phosphorylation undergoes structural changes favoring a less compact conformation in the N-terminal region of Ab present in insoluble aggregates [100]. In addition, unphosphorylated or Ser8 phosphorylated monomeric Ab remained largely unstructured and disordered [100]. Besides Ser8, Ab can also be phosphorylated on Ser26 by cdc2 or CK1 or nitrated on Tyr10 [50,98,101,102]. Cdk1/cdc2 expression was shown to be upregulated in lung adenocarcinoma and correlated directly with the pathological clinical features and poor prognosis of the disease [103]. Depletion of cdk1 was found to slow G2-M progression in the H1299 cell line [104]. Ab Ser26 phosphorylation was reported to result in the formation and stability of the soluble oligomeric assembly of the peptide without further formation of larger prefibrillar or fibrillar aggregates, increasing neurotoxicity [39]. While nonphosphorylated Ab and pSer8 Ab were both detected using the antioligomer A11 and the antiamyloid fibril LOC antibodies, very little detection was observed with these antibodies using pSer26 Ab [39]. Phosphorylation of Ser26 was found to rigidify the turn region around this modified residue leading to prevention of formation of fibrillar Ab aggregates while stabilizing the monomeric and nontoxic soluble nonfibrillar assemblies [105].  Fig. 8. Addition of ATP does not affect the relative amount of oligomer versus total Ab upon ID of HN from A549 (A) or H1299 (B) cell-conditioned media. Cells (0.2 9 10 5 ) were grown in 10% FBS-supplemented media for 24 h. The cells were then incubated in serum-free medium for 72 h without or with ATP and the media collected and ID from either HN or AChE as described in Materials and methods section. The antibodies 6E10 or 4G8 were bound (1 : 1000 dilution) to ELISA wells. The wells were blocked, and then incubated with 300 lL of the control and ID medium (0.5 lgÁlL À1 ). Biotin-4G8 was then added and the signal processed as described in Materials and methods section. Fold change relative to controls using anti-6E10 and anti-4G8 antibodies and 300 lL of the medium not incubated with cells was calculated. Data were processed using the GRAPHPAD

Conclusion
HN is known to bind Ab protecting against its cytotoxic effects, while AChE binding to Ab increases its aggregation and cytotoxicity [53,54,60,[63][64][65]68,[70][71][72]74]. Previously, we found that HN and AChE can simultaneously bind Ab in the A549 cell-conditioned media and that HN abolishes aggregation of Ab induced by addition of AChE [79]. We also showed that ID of HN from the media of A549 and H1299 cells increased the relative abundance of Ab oligomer versus total Ab, the A11-positive prefibrillar oligomers, and to a lesser extent, the LOC-positive fibrillar oligomers, results that correlated with diminished cell viability and increased apoptosis [79]. In this study, we set out to further understand factors affecting the interaction of Ab with HN and AChE. We show that the glycosaminoglycan, HS, reported earlier to interact with amino acid residues 12-18 (VHHQKLV) of Ab40/42 peptides (Fig. 1) [51], has no effect on the binding of Ab to either HN or AChE (Fig. 2). ATP is known to reduce misfolding and fibrillation of Ab [42,47]. ATP, at concentrations (200 lM) chosen to be within the extracellular physiological concentrations previously found (> 100 lM) in the lung cancer cell lines used in this study [1][2][3], was found to weaken interactions between AChE and Ab ( Fig. 3) but strengthens those between Ab and HN (Fig. 4). These findings might suggest that ATP binding to Ab promotes optimal alignment of the amino acids needed to interact with HN increasing binding affinity, while, conversely, renders the amino acid residues on Ab important for binding AChE, less accessible (Fig. 9). Using the conditioned media of either A549 or H1299 cells, more HN was found in a complex with Ab upon addition of ATP, while levels of AChE found in a complex with Ab were decreased by ATP addition to A549 cell media (Fig. 5). Interaction of HN from the conditioned media of both cell lines with Ab appears to be increased by the addition of ATP to a comparable extent, despite differences in AChE expression in the two cell lines [79,82] likely suggesting that increased interaction of HN with Ab by ATP is not regulated by AChE under these conditions. We also found that addition of exogenous ATP to A549 and H1299 cell-conditioned media increased interaction of endogenous HN with Ab and correlated with decreased binding of exogenously added HN (Fig. 6). Moreover, reduced levels of AChE were found in a complex with Ab using A549 cell-conditioned media with exogenously added ATP (Fig. 6). AChE might not be efficient at regulating the effects of ATP on the interaction of HN with Ab since addition of ATP had comparable effects on the interaction of HN with Ab using conditioned media of either A549 cells that express AChE or H1299 cells with minimal expression of the enzyme. We also show that exogenously added ATP, as high as 1 mM, had no effect on viability of either A549 or H1299 cells despite increased interactions between HN and Ab, and, conversely, reduced binding of Ab with AChE (Fig. 7). The lack of effect on cell viability upon addition of ATP is consistent with previous publications [3] reporting that compared to normal cells, lung cancer cells exhibit reduced cytotoxicity upon treatment with these extracellular ATP concentrations. Treatment of A549 or H1299 cells with HN-ID media with added ATP at either 100 µM or 1 mM concentrations had no effect on cell viability as compared to that measured in the absence of added ATP (Fig. 7). Similarly, while A549 cell viability increased~1.45-fold upon AChE ID (Fig. 7), no further effects were observed upon addition of ATP. Whether using nondepleted media, or one ID of HN or AChE, no change in the relative levels of oligomer versus total Ab was found (Fig. 8). Factors affecting cell viability appear to depend not on additionally added ATP or its consequent effects on modulating the binding of either HN or AChE to Ab, but rather on depleting HN or AChE from the media. Reduction of extracellular ATP levels has been previously found to correlate with increased misfolded extracellular Ab in AD [43,44]. While the concentration of extracellular ATP has been shown in normal tissue to be in the 1-5 lM range, in the tumor microenvironment, the concentration rises to greater than 100 lM which can induce normal cells to undergo apoptosis [1,3]. Cancer cells including NSCLC A549 cells have been reported to release ATP and tolerate extracellular ATP concentrations that would otherwise lead to a cytotoxic response in normal cells [1]. Here, we found that treatment with exogenous ATP had no effect on cell viability under all conditions tested. Addition of higher ATP concentrations did not affect viability of cells treated with AChE-ID media, and there was no apparent protection against the cytotoxicity resulting from ID of HN by added ATP (Fig. 7). One possibility among many is that ATP may serve to bind Ab decreasing its oligomerization while simultaneously serving as a substrate for extracellular kinases that might phosphorylate Ab promoting its aggregation, resulting in the observed comparable balance of oligomer to total Ab ratios (Fig. 8). In both cell lines, ATP might promote HN binding to Ab enabling it to regulate access of kinases to their sites on Ab modulating its cytotoxic effects, while AChE might provide an additional layer of regulation of Ab phosphorylation and/or aggregation in A549 cells. Mass spectrometry analysis of extracellular fluids from cancer patients found a substantial amount of phosphorylated proteins as compared to fluids from healthy patients [106] with 84 and 32 phosphorylated sites found in the proteins from breast cancers and lung cancer samples, respectively. Whether HN or AChE plays a role in regulating Ab phosphorylation and/or aggregation in lung cancer cells is currently a focus of research investigation in our laboratory.