Novel multifunctional iron chelators of the aroyl nicotinoyl hydrazone class that markedly enhance cellular NAD+/NADH ratios

Article abstract of DOI:10.1111/bph.14963

Background and Purpose: Alzheimer's disease (AD) is a multifactorial condition leading to cognitive decline and represents a major global health challenge in ageing populations. The lack of effective AD therapeutics led us to develop multifunctional nicot

Full text of DOI:10.1111/bph.14963

Novel Multi-Functional Iron Chelators of the Aroyl Nicotinoyl
Hydrazone Class that Markedly Enhance Cellular NAD/NADH
Ratios
Running Title: Novel Multi-Functional Chelators that Enhance Cellular NAD/NADH Ratios
†
†
§
Փ
Zhixuan Wu , Duraippandi Palanimuthu , Nady Braidy , Nor Hawani Salikin , Suhelen
Փ
†
†γ*
Egan , Michael L. H. Huang and Des R. Richardson
†Molecular Pharmacology and Pathology Program, Department of Pathology and Bosch
Institute, The University of Sydney, Sydney, New South Wales, 2006, Australia.
§Centre for Healthy Brain Ageing, School of Psychiatry, University of New South Wales,
Sydney, 2031, Australia.
ՓSchool of Biological, Earth and Environmental Sciences, Centre for Marine Science and
Innovation, University of New South Wales, Sydney, 2031, Australia.
γDepartment of Pathology and Biological Responses, Nagoya University Graduate School of
Medicine, 65 Tsurumai, Showa-ku, Nagoya 466-8550, Japan.
*
Corresponding authors: Dr. Des R. Richardson, Molecular Pharmacology and Pathology
Program, Discipline of Pathology, Medical Foundation Building (K25), The University of
Sydney, Sydney, New South Wales 2006, Australia. Tel.: + 61 2 9036-3026; Email:
d.richardson@med.usyd.edu.au.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/bph.14963
This article is protected by copyright. All rights reserved.

Word Counts
Abstract: 250 words
Total Word Count (excluding Abstract, Materials and Methods, References and Figure
Legends): 4032 words
Discussion and Conclusion: 723 words
References: 64 references
Figures and Tables: 10
Acknowledgments
This work was supported by a Project Grant from the National Health and Medical Research
Council (NHMRC) Australia [1021607] and a NHMRC Senior Principal Research
Fellowship to D.R.R. [1062607]. D.P. thanks the Sydney Medical School for an Early Career
Research Grant. N.B. is the recipient of the Australian Research Council Discovery Early
Career Research Award at the University of New South Wales [DE170100628]. D.R.R.
appreciate the award of a Judith Jane Mason and Harold Stannett Williams Memorial
Foundation National Medical Program Grant.
Bullet Point Summary
What is already known:

Alzheimer’s disease is a complex, multi-factorial condition leading to cognitive
decline.

No treatments are available to halt or reverse Alzheimer’s disease progression.
What this study adds:
Frontier therapeutics for Alzheimer’s possessing novel multi-targeted activity: NAD+-
donation, iron chelation, enhancement of lifespan, etc.
Clinical significance:
Development of novel therapeutics for the effective treatment of the multiple
pathologies that constitute Alzheimer’s.


Abstract
This article is protected by copyright. All rights reserved.

Background and Purpose: Alzheimer’s disease (AD) is a multi-factorial condition that leads
to cognitive decline and represents a major global health challenge in ageing populations. The
lack of effective AD therapeutics led us to develop multi-functional nicotinoyl hydrazones to
target multiple pathological AD hallmarks.
Experimental Approach: We synthesised 20 novel multi-functional agents based on the
nicotinoyl hydrazone scaffold, which acts as a metal chelator and a lipophilic delivery vehicle
+
to donate a NAD precursor to cells, in order to target metal-dyshomeostasis, oxidative stress,
+
β-amyloid (Aβ)-aggregation and a decrease in the NAD /NADH ratio.
Key Results: Our studies revealed 6-methoxysalicylaldehyde nicotinoyl hydrazone (SNH6)
as the most promising compound that demonstrated: (1) low cytotoxicity; (2) potent iron
(
Fe)-chelation efficacy; (3) significant inhibition of copper-mediated Aβ aggregation; (4)
+
oxidative stress alleviation; (5) effective donation of NAD to NAD-dependent metabolic
+
processes (PARP and sirtuin activity) and enhanced cellular NAD /NADH ratios, as well as
significantly increased median C. elegans lifespan (to 1.46-fold of the control); (6) partially
decreased BACE1 expression, resulting in significantly lower soluble amyloid precursor
protein-β (sAPPβ) and Aβ1-40 levels; and (7) favourable blood-brain barrier-permeation
properties. Structure-activity relationships demonstrated the ability of our nicotinoyl
+
hydrazones to increase NAD was dependent on the electron-withdrawing or -donating
substituents on the aldehyde- or ketone-derived moiety. Aldehyde-derived hydrazones
containing the ONO-donor set and electron-donating groups were required for NAD+-
donation and low cytotoxicity.
Conclusion and Implications: This study highlighted the potential of nicotinoyl hydrazones,
particularly SNH6, to act as multi-functional therapeutics and delivery vehicles for NAD+-
precursors for AD treatment.
This article is protected by copyright. All rights reserved.

+
Keywords: Alzheimer’s disease; Ageing; NAD ; Metals; Oxidative stress; Lifespan
1
. INTRODUCTION
Although Alzheimer’s disease (AD) pathogenesis remains complex, multiple distinctive AD
hallmarks are observed, including: (1) β-amyloid (Aβ) deposition, resulting in non-fibrillary
pre-amyloid deposits, fibrillary, dense amyloid plaques, and cerebral amyloid angiopathy
(
LaFerla & Oddo, 2005); (2) hyper-phosphorylated tau-aggregation, leading to neurofibrillary
tangles (LaFerla & Oddo, 2005); (3) metal dys-homeostasis (Greenough et al., 2013); (4)
oxidative stress (Eskici & Axelsen, 2012; Greenough et al., 2013); and (5) decreased
+
NAD /NADH ratios (Braidy et al., 2018).
Redox-active metals, like iron (Fe) and copper (Cu), are involved in reactive oxygen species
(
ROS) generation and are deposited in, and around, amyloid plaques (Eskici & Axelsen,
012; Greenough et al., 2013). Iron plays a role in amyloid precursor protein (APP)
2
expression (Greenough et al., 2013) and chelation therapy could be beneficial for correcting
metal dys-homeostasis. Encouragingly, clioquinol (CQ)- (Ritchie et al., 2003) or
desferrioxamine (DFO)-based (Crapper McLachlan et al., 1991) chelation therapy has
demonstrated some clinical trial success. The planned Deferiprone to Delay Dementia clinical
trial aims to give a further perspective of iron chelation therapy in AD (Adlard & Bush,
2
018).
+
Nicotinamide adenine dinucleotide (NAD ) is an essential co-factor and sole substrate for a
+
variety of enzymes. Oxidative stress during ageing leads to NAD -depletion and a decline in
+
NAD-dependent processes (Clement et al., 2018). Promotion of NAD anabolism represents
+
a new target for AD (Braidy et al., 2018). NAD regulates sirtuin deacetylase activity
This article is protected by copyright. All rights reserved.

associated with non-amyloidogenic APP-processing (Verdin, 2015). Increased nuclear NAD+
synthesis and SIRT1 activation, a nuclear sirtuin, protects against axonal degeneration (Khan
et al., 2007). Additionally, the NAD-dependent enzyme, poly(ADP-ribose) polymerase
(
PARP), is activated after oxidative DNA damage (Schreiber et al., 2006). Poly(ADP-ribose)
accumulates at higher levels in AD brains (Love et al., 1999) and PARP expression is also
+
elevated, suggesting increased NAD turnover (Khan et al., 2007).
There are no treatments to halt or reverse AD progression and current drugs target only one
AD hallmark (e.g., acetylcholinesterase or N-methyl-D-aspartate receptor overactivation) and
are palliative. Considering this, innovative multi-targeted agents are desirable for AD (Kaur
et al., 2019; Sang et al., 2019). We synthesised 20 novel nicotinoyl hydrazones (Fig.1A,B) as
multi-functional agents that act as a metal-chelator and a lipophilic delivery-vehicle to donate
+
a NAD -precursor to cells. In our rational design strategy, the hydrazone-moiety was chosen
for effective metal-ion chelation (Kalinowski & Richardson, 2005), whereas a nicotinoyl acid
+
fragment was incorporated to increase cellular NAD . We previously demonstrated that
aroylhydrazones donate isonicotinoyl acid hydrazide to cells (Ellis et al., 2014). Thus, we
hypothesised that nicotinoyl hydrazones may undergo hydrolysis, in lysosomes (Ellis et al.,
+
2
014), releasing NA which may fuel NAD production. These agents might act as lipophilic
“Trojan Horse” delivery-vehicles to gain efficient cellular access, which upon hydrolysis,
liberate their active “cargo”.
These analogues effectively targeted multiple AD hallmarks, including metal dys-
+
homeostasis, oxidative stress, Aβ aggregation and the decrease of the NAD /NADH ratio. 6-
Methoxysalicylaldehyde nicotinoyl hydrazone (SNH6; Fig.1A) was identified as a promising
agent, displaying multi-functional activity for targeting AD.
This article is protected by copyright. All rights reserved.

2
2
. METHODS
.1 Materials and Methods
All solvents and reagents were procured from Sigma-Aldrich (St. Louis, MO, USA). The β-
secretase inhibitor, C3, was purchased from Millipore (Burlington, MA, USA), while the α-
1
secretase inhibitor, batimastat, was purchased from Tocris Bioscience (Bristol, UK). The H
nuclear magnetic resonance (NMR; 400 MHz) and 13C NMR (100 MHz) spectra were
obtained using a Bruker Advance 400 NMR spectrometer using DMSO-d
6
as a solvent
(
unless stated otherwise) and tetramethylsilane was used as an internal standard. 3-(4,5-
Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and other biochemicals were
purchased from Sigma-Aldrich. Elemental analysis was performed on a Thermo Scientific
Flash 2000 CHNS/O analyser. Electrospray ionisation mass spectra (ESI-MS) were acquired
using a Bruker amaZon SL mass spectrometer in enhanced resolution mode. Partition
coefficients (Log Pcalc) and topological polar surface area (TPSA) for all ligands were
estimated using ChemBioDraw v.14.0.1. (Perkin-Elmer, Waltham, MA, USA). The purity of
all hydrazones was determined to be ≥95% using elemental analysis (C, H, N). X-ray
diffraction methods and ORTEP diagrams of the nicotinoyl hydrazones (Fig.S1) are
described in the Supporting Information.
2
.2 General Procedure for the Preparation of the Nicotinoyl Hydrazones
The nicotinoyl hydrazones were synthesised via a standard Schiff-base condensation reaction
(
(
Kalinowski et al., 2008). To a refluxing solution of the desired aldehyde (4 mmol) in ethanol
15 mL), nicotinic hydrazide (0.55 g, 4 mmol) followed by glacial acetic acid (5 drops) were
added and refluxed for 1 h. After cooling to room temperature, the precipitate formed was
collected by filtration and washed with ethanol and dried in vacuo.
This article is protected by copyright. All rights reserved.

The general synthetic procedure and characterisation of the Fe complexes of the nicotinoyl
hydrazones is described in the Supporting Information.
2
-Pyridinecarboxaldehyde Nicotinoyl Hydrazone (PCNH)
1
White crystals (0.80 g). Yield: 89%. H NMR: δ ppm (DMSO-d
6
) 7.44 (1H, m, CH), 7.59
(
1H, dd, J = 7.8, 4.9 Hz, CH), 7.90 (1H, t, J = 7.4 Hz, CH), 8.01 (1H, d, J = 8.1 Hz, CH),
8
.29 (1H, d, J = 7.8 Hz, CH), 8.48 (1H, s, CH=N), 8.63 (1H, d, J = 4.4 Hz, CH), 8.79 (1H, d,
1
3
J = 3.7 Hz, CH), 9.10 (1H, s, CH), 12.21 (1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 120.5,
1
24.1, 125.0, 129.4, 136.0, 137.4, 149.1, 149.2, 150.0, 152.9, 153.5, 162.5. ESI-MS in
+
+
CH
3
OH: found mass: 249.04 (100%), Calc. mass for C12
H
10
N
4
ONa: 249.08 [M+Na ] . Anal.
Calc. for C12
H
10
N
4
O·H O (%): C 59.00, H 4.95, N 22.94. Found (%): C 59.05, H 4.91, N
2
2
3.21.
2
-Quinolinecarboxaldehyde Nicotinoyl Hydrazone (QNH)
1
Pale yellow-brown solid (1.00 g). Yield: 90%. H NMR: δ ppm (DMSO-d
6
) 7.63 (2H, m,
2
×CH), 7.81 (1H, t, J = 7.2 Hz, CH), 8.05 (2H, m, 2×CH), 8.15 (1H, d, J = 8.5 Hz, CH), 8.31
(
1H, d, J = 7.6 Hz, CH), 8.45 (1H, d, J = 8.8 Hz, CH), 8.61 (1H, s, CH=N), 8.81 (1H, d, J =
1
3
3
1
1
2
.9 Hz, CH), 9.12 (1H, s, CH), 12.36 (1H, s, NHCO). C NMR: δ ppm (DMSO-d
24.2, 127.9, 128.4, 128.5, 129.4, 129.5, 130.6, 136.1, 137.3, 147.8, 149.1, 149.2, 153.0,
54.1, 162.6. ESI-MS in CH OH: found mass: 299.08 (100%), Calc. mass for C16 ONa:
99.09 [M+Na ] . Anal. Calc. for C16 O·1.5H O (%): C 63.36, H 4.99, N 18.47. Found
%): C 63.64, H 4.89, N 18.49.
6
) 118.0,
3
H
12
N
4
+
+
H
12
N
4
2
(
8
-Hydroxy-2-Quinolinecarboxaldehyde Nicotinoyl Hydrazone (8-OH-QNH)
This article is protected by copyright. All rights reserved.

1
Yellow solid (0.80 g). Yield: 69%. H NMR: δ ppm (DMSO-d
6
) 7.18 (1H, d, J = 7.1 Hz, CH),
.47 (2H, m, 2×CH), 7.62 (1H, m, CH), 8.16 (1H, d, J = 8.5 Hz, CH), 8.36 (2H, m, 2×CH),
.69 (1H, s, CH=N), 8.83 (1H, d, J = 4.2 Hz, CH), 9.16 (1H, s, CH), 9.88 (1H, s, OH), 12.41
7
8
1
3
(
1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 112.7, 118.2, 118.3, 124.1, 128.9, 129.4, 129.5,
OH: found
Na: 315.09 [M+Na ] . Anal. Calc. for
1
36.1, 137.1, 138.6, 149.0, 149.2, 152.0, 153.0, 153.9, 162.6. ESI-MS in CH
3
+
+
mass: 315.08 (100%), Calc. mass for C16
H
12
N O
4 2
C
16
H
12
N
4
2
O (%): C 65.75, H 4.14, N 19.17. Found (%): C 65.86, H 4.07, N 19.17.
Pyridoxal Nicotinoyl Hydrazone (PNH).HCl
1
Yellow solid (1.13 g). Yield: 88%. H NMR: δ ppm (DMSO-d
6
) 2.64 (3H, s, CH ), 4.79 (2H,
3
s, CH
2
), 7.67 (1H, dd, J = 7.9, 4.8 Hz, CH), 8.23 (1H, s, CH), 8.46 (1H, d, J = 8.1 Hz, CH),
8
1
1
3
.85 (1H, d, J = 4.2 Hz, CH), 9.10 (1H, s, CH), 9.22 (1H, s, CH=N), 13.04 (1H, bs, OH),
1
3
3.38 (1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 15.5, 58.7, 124.5, 126.9, 128.1, 130.7,
CN: Found mass:
36.8, 137.1, 144.3, 145.1, 149.0, 153.2, 153.3, 162.3. ESI-MS in CH
3
+
+
09.09 (90%), Calc. mass for C14
H
14
N
4
O
3
Na: 309.10 [M−HCl+Na ] . Anal. Calc. for
C
14
H
15
N
4
O
3
·(HCl)(H O) (%): C 49.35, H 5.03, N 16.44. Found (%): C 49.61, H 4.99, N
2
1
6.41.
2
-Hydroxy-1-Naphthaldehyde Nicotinoyl Hydrazone (NNH)
1
Yellow solid (1.05 g). Yield: 90%. H NMR: δ ppm (DMSO-d
6
) 7.26 (1H, d, J = 8.9 Hz, CH),
7
.42 (1H, t, J = 7.5 Hz, CH), 7.63 (2H, m, 2×CH), 7.91 (1H, d, J = 8.1 Hz, CH), 7.95 (1H, d,
J = 9.0 Hz, CH), 8.30 (1H, d, J = 8.6 Hz, CH), 8.33 (1H, dt, J = 7.9, 2.0 Hz, CH), 8.82 (1H,
dd, J = 4.8, 1.5 Hz, CH), 9.16 (1H, d, J = 1.6 Hz, CH), 9.47 (1H, s, CH=N), 12.36 (1H, s,
1
3
OH), 12.61 (1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 109.0, 119.3, 121.3, 124.1, 124.2,
1
28.3, 128.4, 129.0, 129.5, 132.1, 133.5, 135.9, 147.9, 149.1, 153.0, 158.6, 161.6. ESI-MS in
This article is protected by copyright. All rights reserved.

+
+
CH
3
OH: Found mass: 314.08 (100%), Calc. mass for C17
H
13
N
3
O Na: 314.09 [M+Na ] . Anal.
2
Calc. for C17
H
13
N
3
O
2
(%): C 70.09, H 4.50, N 14.42. Found (%): C 70.29, H 4.46, N 14.31.
Salicylaldehyde Nicotinoyl Hydrazone (SNH1)
1
White solid (0.72 g). Yield: 75%. H NMR: δ ppm (DMSO-d
6
) 6.94 (2H, m, 2×CH), 7.32
(
1H, m, CH), 7.58 (2H, m, 2×CH), 8.29 (1H, dt, J = 7.9, 1.9 Hz, CH), 8.66 (1H, s, CH=N),
8
.79 (1H, dd, J = 4.9, 1.5 Hz, CH), 9.10 (1H, d, J = 2.0 Hz, CH), 11.15 (1H, s, OH), 12.24
1
3
(
1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 116.9, 119.1, 119.9, 124.1, 129.2, 129.8, 132.1,
35.9, 149.0, 149.1, 152.9, 157.9, 161.9. ESI-MS in CH OH: found mass: 264.05 (100%),
Na: 264.08 [M+Na ] . Anal. Calc. for C13 (%): C 64.72,
1
3
+
+
Calc. mass for C13
H
11
N
3
O
2
H
11
N
3
O
2
H 4.60, N 17.42. Found (%): C 65.00, H 4.54, N 17.46.
3
-Chlorosalicylaldehyde Nicotinoyl Hydrazone (SNH2)
1
White solid (0.85 g). Yield: 78%. H NMR: δ ppm (DMSO-d
6
) 6.97 (1H, m, CH), 7.51 (2H,
m, 2×CH), 7.60 (1H, m, CH), 8.30 (1H, d, J = 8.1 Hz, CH), 8.62 (1H, s, CH=N), 8.80 (1H, m,
1
3
CH), 9.11 (1H, d, J = 1.7 Hz, CH), 12.28 (1H, s, OH), 12.51 (1H, s, NHCO). C NMR: δ
ppm (DMSO-d ) 120.0, 120.5, 120.9, 124.2, 128.7, 130.0, 132.0, 136.0, 149.1, 149.6, 153.1,
53.7, 162.0. ESI-MS in CH
6
1
3
CN: found mass: 298.04 (100%), Calc. mass for
+
+
C
13
H
10
N
3
O
2
ClNa: 298.04 [M+Na ] . Anal. Calc. for C13
H
10
N
3
O
2
Cl·H O (%): C 53.16, H
2
4
.12, N 14.31. Found (%): C 53.34, H 4.06, N 14.31.
3
-Bromosalicylaldehyde Nicotinoyl Hydrazone (SNH3)
1
White solid (0.95 g). Yield: 75%. H NMR: δ ppm (DMSO-d
6
) 6.91 (1H, t, J = 7.8 Hz, CH),
7
.53 (1H, dd, J = 7.7, 1.3 Hz, CH), 7.61 (2H, m, 2×CH), 8.29 (1H, d, J = 7.8 Hz, CH), 8.58
(
1H, s, CH=N), 8.80 (1H, bs, CH), 9.12 (1H, s, CH), 12.46 (1H, s, OH), 12.53 (1H, s,
1
3
NHCO). C NMR: δ ppm (DMSO-d ) 110.5, 119.8, 121.1, 124.2, 128.7, 130.9, 135.0, 136.0,
6
This article is protected by copyright. All rights reserved.

1
49.1, 149.7, 153.1, 154.7, 162.0. ESI-MS in CH
3
CN: found mass: 341.97 (100%), Calc.
+
+
mass for C13
H
10
N
3
O
2
BrNa: 341.99 [M+Na ] . Anal. Calc. for C13 Br (%): C 46.17, H
H
10
N
3
O
2
3
.58, N 12.43. Found (%): C 46.41, H 3.62, N 12.33.
3
-Hydroxysalicylaldehyde Nicotinoyl Hydrazone (SNH4)
1
Pale yellow-brown solid (0.86 g). Yield: 84%. H NMR: δ ppm (DMSO-d
6
) 6.76 (1H, m,
CH), 6.89 (1H, dd, J = 7.8, 1.5 Hz, CH), 7.01 (1H, dd, J = 7.8, 1.5 Hz, CH), 7.59 (1H, dd, J =
7
4
.8, 4.9 Hz, CH), 8.29 (1H, dt, J = 8.0, 1.9 Hz, CH), 8.62 (1H, s, CH=N), 8.78 (1H, dd, J =
.9, 1.5 Hz, CH), 9.11 (1H, s, CH), 9.27 (1H, s, OH), 10.98 (1H, s, NHCO), 12.26 (1H, s,
OH). 13C NMR: δ ppm (DMSO-d
) 118.0, 119.2, 119.7, 120.4, 124.1, 129.1, 135.9, 146.1,
46.6, 149.1, 149.9, 152.9, 161.9. ESI-MS in CH OH: found mass: 280.06 (100%), Calc.
Na: 280.07 [M+Na ] . Anal. Calc. for C13 (%): C 60.70, H 4.31,
6
1
3
+
+
mass for C13
H
11
N
3
O
3
H
11
N
3
O
3
N 16.31. Found (%): C 61.04, H 4.20, N 16.23.
5
-Trifluoromethoxysalicylaldehyde Nicotinoyl Hydrazone (SNH5)
1
White crystals (0.78 g). Yield: 60%. H NMR: δ ppm (DMSO-d
6
) 7.03 (1H, d, J = 9.0 Hz,
CH), 7.31 (1H, dd, J = 8.9, 2.6 Hz, CH), 7.59 (1H, dd, J = 7.9, 4.8 Hz, CH), 7.66 (1H, d, J =
2
.4 Hz, CH), 8.29 (1H, dt, J = 8.0, 1.7 Hz, CH), 8.69 (1H, s, CH=N), 8.79 (1H, d, J = 3.9 Hz,
CH), 9.10 (1H, s, CH), 11.21 (1H, bs, OH), 12.31 (1H, bs, NHCO). 13C NMR: δ ppm
(
DMSO-d
6
) 118.2, 119.4, 120.7 (JC-F = 10.0 Hz), 122.0, 124.1, 124.9, 129.1, 136.0, 141.2,
CN: found mass: 348.05 (100%), Calc.
Na: 348.06 [M+Na ] . Anal. Calc. for C14 (%): C 51.70, H
.10, N 12.92. Found (%): C 51.92, H 3.24, N 12.85.
1
46.3, 149.1, 153.0, 156.6, 162.1. ESI-MS in CH
3
+
+
mass for C14
H
10
N
3
O
3
F
3
H
10
N
3
O F
2 3
3
This article is protected by copyright. All rights reserved.

6
-Methoxysalicylaldehyde Nicotinoyl Hydrazone (SNH6)
1
White crystals (0.83 g). Yield: 77%. H NMR: δ ppm (DMSO-d
6
) 3.87 (3H, s, OCH ), 6.57
3
(
2H, t, J = 7.4 Hz, 2×CH), 7.29 (1H, t, J = 8.3 Hz, CH), 7.59 (1H, dd, J =7.8, 4.8 Hz, CH),
8
.30 (1H, dt, J =7.9, 1.9 Hz, CH), 8.79 (1H, d, J = 4.2 Hz, CH), 8.96 (1H, s, CH=N), 9.11
1
3
(
1H, s, CH), 12.12 (1H, s, OH), 12.33 (1H, bs, NHCO). C NMR: δ ppm (DMSO-d
6
) 56.4,
1
02.2, 107.2, 109.9, 124.2, 128.8, 133.1, 135.9, 146.8, 149.1, 153.0, 159.1, 159.8, 161.6. ESI-
+
+
MS in CH
3
CN: found mass: 294.08 (100%), Calc. mass for C14
H
13
N
3
O
3
Na: 294.09 [M+Na ] .
Anal. Calc. for C14
N 14.37.
H
13
N
3
O
3
·H
2
O (%): C 58.13, H 5.23, N 14.53. Found (%): C 58.39, H 5.18,
5
-Fluoro-3-methylsalicylaldehyde Nicotinoyl Hydrazone (SNH7)
1
Yellow crystals (0.70 g). Yield: 64%. H NMR: δ ppm (DMSO-d
1H, dd, J = 9.0, 2.9 Hz, CH), 7.25 (1H, dd, J = 9.0, 2.9 Hz, CH), 7.60 (1H, dd, J = 7.9, 4.8
Hz, CH), 8.29 (1H, d, J = 8.1 Hz, CH), 8.55 (1H, s, CH=N), 8.80 (1H, d, J = 3.7, CH), 9.11
6
) 2.23 (3H, s, CH ), 7.13
3
(
13
(
1H, s, CH), 11.61 (1H, s, OH), 12.47 (1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 16.0,
1
7
2
C
13.8 (JC-F = 118.0 Hz), 118.0 (JC-F = 8.5 Hz), 119.6 (JC-F = 119.6 Hz), 124.1, 127.7 (JC-F
=
=
.7 Hz), 128.8, 136.0, 149.1, 149.6 (JC-F = 2.7 Hz), 152.7 (JC-F = 1.5 Hz), 153.1, 155.2 (JC-F
33.0 Hz), 162.0. ESI-MS in CH
3
CN: found mass: 296.06 (100%), Calc. mass for
+
+
14
H
12
N
3
O
2
FNa: 296.08 [M+Na ] . Anal. Calc. for C14 F·H O (%): C 57.73, H 4.85,
H
12
N
3
O
2
2
N 14.43. Found (%): C 57.96, H 4.91, N 14.48.
5
-Bromo-3-fluorosalicylaldehyde Nicotinoyl Hydrazone (SNH8)
1
Pale yellow-brown solid (1.20 g). Yield: 89%. H NMR: δ ppm (DMSO-d
6
) 7.60 (2H, m,
2
×CH), 7.69 (1H, s, CH), 8.29 (1H, dt, J = 8.0, 1.7 Hz, CH), 8.64 (1H, s, CH=N), 8.79 (1H,
dd, J = 8.0, 1.3 Hz, CH), 9.10 (1H, d, J = 1.7 Hz, CH), 11.58 (1H, bs, OH), 12.44 (1H, s,
This article is protected by copyright. All rights reserved.

NHCO). 13C NMR: δ ppm (DMSO-d
C-F = 4.0 Hz), 124.1, 126.8 (JC-F = 3.0 Hz), 128.9, 136.0, 145.1 (JC-F = 14.0 Hz), 146.1 (JC-F
3.0 Hz), 149.2, 151.7 (JC-F = 240.0 Hz), 153.1, 162.2. ESI-MS in CH CN: found mass:
) 109.8 (JC-F = 9.0 Hz), 120.9 (JC-F = 21.0 Hz), 123.6
6
(
J
=
3
+
+
3
59.98 (100%), Calc. mass for C13
H
9
N
3
O
2
BrFNa: 359.98 [M+Na ] . Anal. Calc. for
C
13
H
9
N
3
O
2
BrF·H
2
O (%): C 43.84, H 3.11, N 11.80. Found (%): C 44.01, H 3.10, N 11.69.
3
,4-Dihydroxysalicylaldehyde Nicotinoyl Hydrazone (SNH9)
1
Pale yellow-brown solid (0.65 g). Yield: 60%. H NMR: δ ppm (DMSO-d
6
) 6.41 (1H, d, J =
7
.8 Hz, CH), 6.82 (1H, d, J = 8.5 Hz, CH), 7.58 (1H, m, CH), 8.26 (1H, dt, J = 8.2, 1.9 Hz,
CH), 8.47 (1H, s, OH), 8.50 (1H, s, CH=N), 8.77 (1H, dd, J = 4.9, 1.7 Hz, CH), 9.08 (1H, s,
1
3
CH), 9.51 (1H, s, OH), 11.35 (1H, s, OH), 12.10 (1H, s, NHCO). C NMR: δ ppm (DMSO-
) 108.2, 111.2, 121.7, 124.1, 129.2, 133.2, 135.8, 148.0, 149.0, 149.4, 151.0, 152.8, 161.5.
ESI-MS in CH OH: found mass: 296.05 (100%), Calc. mass for C13 Na: 296.07
M+Na ] . Anal. Calc. for C13 (%): C 57.14, H 4.06, N 15.38. Found (%): C 57.18,
H 4.05, N 15.36.
d
6
3
H
11
N
3
O
4
+
+
[
H
11
N
3
O
4
4
'-Fluoro-2'-hydroxyacetophenone Nicotinoyl Hydrazone (AcNH1)
1
White solid (0.83 g). Yield: 75%. H NMR: δ ppm (MeOH-d
4
) 2.52 (3H, s, CH ), 6.90 (1H, d,
3
J = 8.8 Hz, CH), 7.42 (1H, m, CH), 7.62 (1H, dd, J = 7.8, 4.9 Hz, CH), 7.76 (1H, d, J = 2.2
1
Hz, CH), 8.38 (1H, d, J = 7.6, CH), 8.76 (1H, dd, J = 4.9, 1.5 Hz, CH), 9.11 (1H, s, CH). H
NMR: δ ppm (DMSO-d
6
) 6.75 (2H, m, 2×CH), 7.59 (1H, ddd, J = 7.9, 4.8, 0.7 Hz, CH), 7.71
(
1H, m, CH), 8.29 (1H, dt, J = 8.0, 1.9 Hz, CH), 8.79 (1H, dd, J = 4.8, 1.6, CH), 9.09 (1H, s,
1
3
CH), 11.55 (1H, s, OH), 13.77 (1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 14.8, 104.4 (JC-
F
= 24.0 Hz), 106.3 (JC-F = 22.0 Hz), 116.8 (JC-F = 3.0 Hz), 124.0, 129.2, 131.1 (JC-F = 10.0
Hz), 136.4, 149.4, 152.9, 158.5, 161.2 (JC-F = 13.0 Hz), 163.2 (JC-F = 247.0 Hz), 163.5. ESI-
This article is protected by copyright. All rights reserved.

+
+
MS in CH
found mass: 569.10 (100%), Calc. mass for C28
F (%): C 61.53, H 4.43, N 15.38. Found (%): C 61.73, H 4.47, N 15.34.
3
CN: found mass: 296.08 (80%), Calc. mass for C14
H
12
N
3
O FNa: 296.08 [M+Na ] ,
2
+
+
H N
24 6
O F
4 2
Na: 569.17 [2M+Na ] . Anal. Calc.
for C14
H
12
N
3
O
2
5
'-Bromo-2'-hydroxyacetophenone Nicotinoyl Hydrazone (AcNH2)
1
Pale yellow-brown solid (0.90 g). Yield: 68%. H NMR: δ ppm (MeOH-d
4
) 2.52 (3H, s, CH
.66 (1H, s, CH), 6.69 (1H, m, CH), 7.62 (1H, dd, J = 7.3, 4.9 Hz, CH), 7.69 (1H, dd, J = 9.3,
.6 Hz, CH), 8.37 (1H, d, J = 8.1, CH), 8.77 (1H, dd, J = 4.9, 1.5 Hz, CH), 9.09 (1H, s, CH).
) 6.90 (1H, d, J = 9.0, 2.9 Hz, CH), 7.45 (1H, dd, J = 9.0, 2.9 Hz,
3
),
6
6
1H NMR: δ ppm (DMSO-d
6
CH), 7.58 (1H, dd, J = 7.9, 4.8 Hz, CH), 7.77 (1H, d, J = 8.1 Hz, CH), 8.28 (1H, s, CH), 8.80
1H, d, J = 3.7 Hz, CH), 9.09 (1H, s, CH), 11.61 (1H, s, OH), 13.35 (1H, s, NHCO). 13C
(
NMR: δ ppm (DMSO-d
6
) 14.8, 110.1, 120.1, 121.9, 124.0, 129.2, 131.1, 134.2, 136.5, 149.5,
53.0, 157.6, 158.3, 163.6. ESI-MS in CH CN: found mass: 356.01 (100%), Calc. mass for
BrNa: 356.00 [M+Na ] . Anal. Calc. for C14 Br (%): C 50.32, H 3.62, N
1
3
+
+
C
14
H
12
N
3
O
2
H
12
N
3
O
2
1
2.57. Found (%): C 50.52, H 3.58, N 12.52.
5
'-Bromo-2'-hydroxypropionylphenone Nicotinoyl Hydrazone (PrNH1)
1
White solid (0.90 g). Yield: 65%. H NMR: δ ppm (DMSO-d
6
) 1.15 (3H, t, J = 7.4 Hz, CH ),
3
3
.03 (2H, q, CH
2
), 6.92 (1H, d, J = 8.8 Hz, CH), 7.46 (1H, dd, J = 8.8, 2.2 Hz, CH), 7.59 (1H,
dd, J = 7.8, 5.0 Hz, CH), 7.75 (1H, d, J = 2.2 Hz, CH), 8.25 (1H, d, J = 8.1 Hz, CH), 8.80 (1H,
1
3
d, J = 3.7 Hz, CH), 9.05 (1H, s, CH), 11.65 (1H, s, OH), 13.40 (1H, s, NHCO). C NMR: δ
ppm (DMSO-d ) 11.7, 19.9, 110.2, 120.4, 123.9, 129.3, 130.7, 134.2, 136.6, 149.6, 152.9,
58.8, 160.7, 163.9. ESI-MS in CH CN: found mass: 370.01 (100%), Calc. mass for
BrNa: 370.02 [M+Na ] . Anal. Calc. for C15 Br (%): C 51.74, H 4.05, N
6
1
3
+
+
C
15
H
14
N
3
O
2
H
14
N
3
O
2
1
2.07. Found (%): C 51.80, H 4.03, N 12.04.
This article is protected by copyright. All rights reserved.

7
-Hydroxy-1-indanone Nicotinoyl Hydrazone (InNH1)
1
Pale yellow-brown solid (0.66 g). Yield: 62%. H NMR: δ ppm (DMSO-d
6
) 3.09 (4H, m,
2
×CH
2
), 6.77 (1H, d, J = 8.1 Hz, CH), 6.91 (1H, d, J = 7.3 Hz, CH), 7.31 (1H, t, J = 7.7 Hz,
CH), 7.57 (1H, dd, J = 7.6, 4.9 Hz, CH), 8.24 (1H, d, J = 7.8 Hz, CH), 8.77 (1H, d, J = 3.7 Hz,
13
CH), 9.05 (1H, s, CH), 10.19 (1H, s, OH), 11.26 (1H, s, NHCO). C NMR: δ ppm (DMSO-
d
1
2
6
6
) 28.3, 29.0, 113.4, 116.9, 122.9, 124.0, 129.8, 133.2, 136.3, 149.3, 150.2, 152.6, 155.7,
62.8, 167.4. ESI-MS in CH CN: found mass: 290.08 (100%), Calc. mass for C15 Na:
(%): C 67.40, H 4.90, N 15.72. Found (%): C
3
H
13
N
3
O
2
+
+
90.09 [M+Na ] . Anal. Calc. for C15
7.63, H 4.82, N 15.60.
H
13
N
3
O
2
7
-Hydroxy-6-methyl-1-indanone Nicotinoyl Hydrazone (InNH2)
1
Yellow solid (0.70 g). Yield: 63%. H NMR: δ ppm (DMSO-d
6
) 2.19 (3H, s, CH ), 3.07 (4H,
3
broad singlet, 2×CH
2
), 6.81 (1H, d, J = 7.6 Hz, CH), 7.19 (1H, d, J = 7.6 Hz, CH), 7.57 (1H,
dd, J = 7.8, 4.8 Hz, CH), 8.24 (1H, d, J = 7.8 Hz, CH), 8.77 (1H, dd, J = 3.5, 1.5 Hz, CH),
9
1
1
3
6
.05 (1H, s, CH), 10.34 (1H, s, OH), 11.24 (1H, s, NHCO). 13C NMR: δ ppm (DMSO-d
4.8, 28.5, 116.4, 121.8, 122.4, 124.0, 129.8, 134.5, 136.3, 147.5, 149.3, 152.6, 153.7, 162.8,
67.6. ESI-MS in CH CN: found mass: 304.10 (100%), Calc. mass for C16 Na:
(%): C 68.31, H 5.37, N 14.94. Found (%): C
6
)
3
H
15
N
3
O
2
+
+
04.11 [M+Na ] . Anal. Calc. for C16
8.24, H 5.42, N 14.90.
H
15
N
3
O
2
4
-Bromo-7-hydroxy-1-indanone Nicotinoyl Hydrazone (InNH3)
1
Yellow solid (0.82 g). Yield: 60%. H NMR: δ ppm (DMSO-d
6
) 3.05 (4H, m, 2×CH ), 6.77
2
(
1H, d, J = 8.8 Hz, CH), 7.46 (1H, d, J = 8.8 Hz, CH), 7.57 (1H, dd, J = 7.7, 5.0 Hz, CH),
8
.23 (1H, d, J = 7.8 Hz, CH), 8.77 (1H, d, J = 3.7 Hz, CH), 9.04 (1H, s, CH), 10.31 (1H, s,
This article is protected by copyright. All rights reserved.

1
3
OH), 11.36 (1H, s, NHCO). C NMR: δ ppm (DMSO-d
6
) 27.9, 30.6, 108.9, 116.2, 123.9,
24.7, 129.6, 135.1, 136.3, 149.2, 149.4, 152.7, 155.0, 163.0, 166.3. ESI-MS in CH CN:
BrNa: 368.00 [M+Na ] . Anal. Calc.
1
3
+
+
Found mass: 367.98 (100%), Calc. mass for C15
H
12
N
3
O
2
for C15
H
12
N
3
O
2
Br·H O (%): C 49.47, H 3.88, N 11.54. Found (%): C 49.56, H 3.87, N 11.46.
2
2
.3 Cell culture
Human SK-N-MC neuroepithelioma cells (Cat # HTB-10; RRID: CVCL_0530) were
obtained from and authenticated by the American Type Culture Collection (Manassas, VA).
The cells were cultured by standard procedures (Richardson et al., 1995; Yuan et al., 2004) in
MEM supplemented with 10% foetal calf serum (Sigma-Aldrich), penicillin (100 U/mL),
streptomycin (100 mg/mL), L-glutamine (2 mM), non-essential amino acids, sodium pyruvate
®
(
1 mM) and Fungizone (0.25 μg/mL) and incubated at 37°C in a humidified atmosphere
containing 5% CO /95% air (Forma Scientific, Marietta, OH).
2
2
.4 MTT assay
The cytotoxic potential of nicotinoyl hydrazones was determined by the 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay against SK-N-MC cells
using a standard procedure (Richardson et al., 1995). The cytotoxicity of the nicotinoyl
hydrazones was examined between 0.2 - 100 µM. Using the SK-N-MC cell-type, we showed
that MTT results were shown to be directly proportional to cell numbers derived from viable
cell counts using Trypan blue (Richardson et al., 1995).
Based on data from a previous study assessing the effect of chelators on the proliferation of
SK-N-MC cells (Richardson et al., 1995), a sample size of >4 biological replicates was
determined using a priori power analysis (G*Power 3.1; RRID: SCR_013726). This resulted
This article is protected by copyright. All rights reserved.

in >80% statistical power to detect at least 20% difference in mean with an effect size of 2.22
and standard deviation of 0.09.
2
.5 Alleviation of H
2
2
O -Mediated Cytotoxicity
A previously reported protocol using the MTT assay (Lim et al., 2008) was used to assess the
ability of the nicotinoyl hydrazones to protect SK-N-MC neuroepithelioma cells against
4
H
2
O
2
-medated cytotoxicity. SK-N-MC cells were seeded at a density of 1.5 x 10 cells/well
in 96-well microtiter plates and incubated at 37°C/24 h. Following this incubation, cells were
pre-incubated with serum-free medium alone or serum-free medium containing the nicotinoyl
hydrazones (10 µM) or the control agents, 2-pyridinecarboxaldehyde 2-thiophenecarboxyl
hydrazone (PCTH), clioquinol (CQ), nicotinic hydrazide (NH), salicylaldehyde benzoyl
hydrazone (SBH), or salicylaldehyde isonicotinoyl hydrazone (SIH), at 10 µM for 2 h/37°C.
After removing the medium, the cells were then treated with serum-free medium alone or
serum-free medium containing H
2
2
O (150 µM) for 24 h/37°C. Cell viability was measured
using the MTT assay, as described above.
5
9
2
.6 Labelling of Transferrin with Fe
The Fe-transport protein, transferrin (Tf), was labelled with 59Fe (PerkinElmer Life and
5
9
Analytical Sciences, Boston, MA) to form Fe -Tf using standard methods (Richardson et
2
al., 1995).
5
9
2
.7 Cellular Fe Mobilisation
The effect of chelators on 59Fe mobilisation from SK-N-MC cells was determined by 59Fe
efflux experiments using a standard protocol (Richardson et al., 1995; Yuan et al., 2004).
Cell viability examination using Trypan blue staining and phase contrast microscopy
This article is protected by copyright. All rights reserved.

demonstrated there was no decrease in viability over the short incubation period with the
chelators (i.e., 3 h/37°C).
Based on data from previous studies assessing the effect chelators on cellular 59Fe
mobilisation from SK-N-MC cells (Richardson et al., 1995; Yuan et al., 2004), a sample size
of >7 biological replicates was determined using a priori power analysis (G*Power 3.1). This
resulted in >80% statistical power to detect at least 20% difference in mean with an effect
size of 1.67 and standard deviation of 0.12.
5
9
2
.8 Cellular Fe Uptake
5
9
In order to estimate the ability of chelators to prevent the cellular uptake of Fe from the Fe-
transport protein, 59Fe -Tf, 59Fe uptake experiments were performed using standard
2
procedures (Richardson et al., 1995; Yuan et al., 2004). As in Section 2.7 above, there was
no decrease in cell viability during the short incubation with the chelators (i.e., 3 h/37°C).
5
9
Based on data from previous studies assessing the effect of chelators on cellular Fe uptake
by SK-N-MC cells (Richardson et al., 1995; Yuan et al., 2004), a sample size of >3
biological replicates was determined using a priori power analysis (G*Power 3.1). This
resulted in >90% statistical power to detect at least 20% difference in mean with an effect
size of 7.14 and standard deviation of 0.03.
2
.9 Ascorbate Oxidation Assay
The ability of chelator Fe complexes to catalyse ascorbate oxidation was examined as
described previously (Richardson et al., 2006). These studies utilised an iron binding
equivalent (IBE) of 1 due to the different ligand denticity (i.e., hexadentate ligands,
ethylenediaminetetraacetic acid (EDTA) and desferrioxamine (DFO), form 1:1 ligand:Fe
This article is protected by copyright. All rights reserved.

complexes,
thiosemicarbazone (Dp44mT) and the hydrazones, form 2:1 ligand:Fe-complexes
Kalinowski & Richardson, 2005; Richardson et al., 2006) and indicates complete filling of
while the tridentate ligands,
di-2-pyridylketone 4,4-dimethyl-3-
(
the Fe coordination shell.
Based on data from a previous study assessing the effect of Fe chelators on ascorbate
oxidation (Richardson et al., 2006), a sample size of >3 biological replicates was determined
using a priori power analysis (G*Power 3.1). This led to >90% statistical power to detect at
least 20% difference in mean with an effect size of 6.89 and standard deviation of 0.03.
2
.10 Inhibition of Cu(II)-Mediated Aggregation of Aβ
A well characterised turbidimetric assay was utilised to examine the ability of the nicotinoyl
hydrazones to inhibit Cu(II)-mediated aggregation of Aβ1–40 (Gomes et al., 2014; Jones et al.,
2
012). Based on data from previous studies assessing Cu(II)-mediated Aβ aggregation
(
Gomes et al., 2014; Jones et al., 2012), a sample size of 12 biological replicates was
determined using a priori power analysis (G*Power 3.1). This generated >80% statistical
power to detect at least 20% difference in mean with an effect size of 1.25 and standard
deviation of 0.16.
2
.11 Blood-Brain Barrier (BBB)-Penetration
The ability of nicotinoyl hydrazones to permeate the BBB was examined using the parallel
artificial membrane-permeation assay (PAMPA), following a well-established protocol (Di et
2
al., 2003). Our studies demonstrated a linear correlation (R = 0.9112) between experimental
and reported permeability of 5 commercial drugs, namely Theophylline, Verapamil,
This article is protected by copyright. All rights reserved.

Progesterone, Chlorpromazine, and Donepezil, obtained from the PAMPA-BBB assay
Fig.S2).
(
Reported in Table S1 are the predictions of CNS penetration based on ranges of permeability
values from the PAMPA BBB assay determined in Fig.S2, and as indicated by Di and
associates (Di et al., 2003) (see Supporting Information).
2
.12 Human Primary Astrocyte Cell Culture
Human primary astrocytes were cultured from resected normal adult brain tissue with
informed consent at the Minimally Invasive Cancer Centre, Prince of Wales Hospital, Sydney,
Australia (Reference number X12-0314 and HREC/12/RPAH/481). Astrocytes were prepared
using a previously described protocol (Guillemin et al., 2001). The mixed brain cell cultures
were shaken for 2 h at 37℃ at 220 rpm, and the floating cells were centrifuged and
5
transferred to cell culture flask at a density of 1× 10 cells/mL. Cells were grown in RPMI
with the following supplements: 10% fetal calf serum (FCS), 1% L-glutamax, and 1%
antibiotic/antifungal. Approval was obtained from the Human Research Ethics Committee of
the University of New South Wales (human brain tissue reference number: HC12563).
+
2
.13 NAD /NADH Assay
+
Intracellular NAD and NADH were extracted from treated and non-treated astrocyte cell
homogenates, as previously described (Trammell & Brenner, 2013). Briefly, 100ꢀμL of cell
homogenate was added to 20ꢀμL of [D ]-NA in 5% formic acid (v/v). Samples were then
4
mixed with 300ꢀμL of acetonitrile, vortexed for 15ꢀs, and then centrifuged, and dried via
speed vacuum overnight at 37°C. Samples were then reconstituted in 100ꢀμL of 10ꢀmM
This article is protected by copyright. All rights reserved.

ammonium acetate with 0.1% formic acid. Standard curves were prepared in water, as
described above.
+
Separation and quantitation of NAD and NADH were performed with a Sciex QTRAP
5
500 mass spectrometer operated in positive ion multiple reaction monitoring mode, as
+
described previously (Trammell & Brenner, 2013). NAD and NADH were separated on a
Phenomenex NH₂ column (150 mm x 2 mm x 3 µm; Phenomenex, Torrance, CA) using the
same gradient and mobile phase as described for the acid separation (Trammell & Brenner,
2
013).
The analytes were separated using a binary solvent gradient consisting of 5 mM NH
4
OAc
(
pH 9.5) adjusted with ammonia (mobile phase A) and acetonitrile (mobile phase B) with a
flow rate of 250 µL/min. The initial solvent composition at injection was 25% A, followed by
a 2 min gradient to 45% A and a fast gradient ramp to 80% A (0.1 min), which was
maintained for 5.9 min. Then, A was increased again to 95% (2 min), held for 13 min, and
then reverted to initial conditions (0.1 min) for equilibration, with a total run time of 30 min.
+
LC-MS standard curves, representative chromatograms and retention times for NAD and
NADH are shown in the Supporting Information (Fig.S3 and Table.S2).
+
Based on data from a previous study assessing NAD /NADH levels of human primary
astrocytes (Braidy et al., 2011a), a sample size of >4 biological replicates was determined
using a priori power analysis (G*Power 3.1). This resulted in >80% statistical power to
detect at least 20% difference in mean with an effect size of 2.50 and standard deviation of
0
.08.
This article is protected by copyright. All rights reserved.

+
2
.14 NAD -Consumption Activity Assay
+
NAD -consumption activity of nuclear extracts (Braidy et al., 2014) was measured as
previously described (Putt et al., 2005). The final buffer solution contained MgCl (10 mM),
2
+
Triton X-100 (1%), and NAD (20 M) in Tris buffer (50 mM; pH 8.1). The plate was
+
incubated for 1 h/37℃ and the amount of NAD consumed was measured using the NAD
assay, as described above.
+
Based on data from a previous study assessing NAD -consumption (Putt et al., 2005), a
sample size of 5 biological replicates was determined using a priori power analysis
(
G*Power 3.1). This generated >80% statistical power to detect at least 20% difference in
mean with an effect size of 2.22 and standard deviation of 0.09.
2
.15 SIRT1 Deacetylase Assay
Sirtuin 1 (SIRT1) deacetylase activity was evaluated using the Cyclex SIRT1/Sir2
Deacetylase Flourometric Assay Kit (CycLex, Nagano, Japan), as previously described
(
Braidy et al., 2011b). The final reaction mixture (100 L) contained 50 mM Tris-HCl (pH
8
2
.8), 4 mM MgCl , 0.5 mM DTT, 0.25 mA/mL lysyl endopeptidase, 1 M Trichostatin A,
2
+
00 M NAD , and 5 L of nuclear sample. The samples were mixed and incubated for 10
min at room temperature and the fluorescence intensity (ex. 340 nm, em. 460 nm) measured
every 30 s for a total of 60 min immediately after the addition of fluoro-substrate peptide (20

M final concentration). These measurements implemented a Fluostar Optima Fluorometer
NY, USA) and were normalised by the protein content, with the results being reported as
relative fluorescence/g of protein (AU).
(
This article is protected by copyright. All rights reserved.

Based on data from a previous study assessing SIRT1 deacetylase activity in human primary
astrocytes (Braidy et al., 2011a), a sample size of 5 biological replicates was determined
using a priori power analysis (G*Power 3.1). This produced >80% statistical power to detect
at least 20% difference in mean with an effect size of 2.32 and standard deviation of 0.08.
2
.16 Western Blotting
Western blotting was conducted and described in accordance to (Alexander et al., 2018). The
following primary antibodies were used in this study: β-site APP cleaving enzyme 1
(
BACE1) rabbit mAb (catalog #5606T; rabbit IgG; Epitope: residues surrounding H490 of
human BACE1) and APP/β-amyloid (NAB228) mouse mAb (catalog #2450; mouse IgG2A;
Epitope: N-terminus of β-amyloid) from Cell Signaling Technology; anti-amyloid precursor
protein C-terminal rabbit pAb (catalog #A8717; rabbit IgG; Epitope: C-terminal of human
APP) and anti-β-actin mouse mAb (catalog #A5316; mouse IgG2a; Epitope: N-terminal of β-
actin; RRID: AB_476743) from Sigma-Aldrich; purified anti-sAPPβ rabbit pAb (catalog
#
1
813401; rabbit IgG; Epitope: N-terminus of sAPPβ) from Biolegend and transferrin receptor
(TfR1) mouse mAb (catalog #13-6890; mouse IgG1; Epitope: N-terminal region of TfR1)
from Thermo Fisher Scientific. All primary antibodies were used at a dilution of 1:1000,
except for the anti-β-actin antibody, which was used at a 1:10,000 dilution and were prepared
®
in 5% w/v BSA (A7906; ≥98%; Sigma-Aldrich) in 1x TBS with 0.1% Tween 20 for single
use. The secondary antibodies, goat anti-rabbit IgG pAb (A0545; Epitope: rabbit IgG) and
goat anti-mouse pAb (A9917; Epitope: mouse IgG Fab fragment) conjugated with
horseradish peroxidase, were used at a 1:5,000 dilution and were purchased from Sigma-
Aldrich. Proteins were electro-blotted onto PVDF membranes (0.45 μm pore size; GE
Healthcare, WI, USA) at 30 V for 16 h at 4˚C, which were blocked using 5% w/v BSA
®
(
A7906; ≥98%; Sigma-Aldrich) in 1x TBS with 0.1% Tween 20. The ChemiDoc MP
Imaging System (Bio-Rad, CA, USA) was used for protein visualisation and the analysis of
This article is protected by copyright. All rights reserved.

blots was performed using Image Lab Software (Bio-Rad). All densitometry was normalised
to the relative β-actin loading control.
Based on our previous western blotting analysis (Palanimuthu et al., 2017), a sample size of
>3 biological replicates was determined using a priori power analysis (G*Power 3.1). This
generated >90% statistical power to detect at least 20% difference in mean with an effect size
of 20.00 and standard deviation of 0.01.
2
.17 Enzyme-linked Immunosorbent Assay
The human Aβ1-40 ELISA kit (catalog #: KHB3482) was purchased from Thermo Fisher
Scientific. The cell culture medium was harvested in the presence of the protease inhibitor,
AEBSF, after a 24 h/37℃ incubation of primary human cultures of astrocytes with 5 µM
batimastat, 5 µM C3, 25 µM SNH2, or 25 µM SNH6. Cell debris was removed by
centrifugation at 1400 rpm/1 min/4°C. The secreted Aβ1-40 was detected using the ELISA kit,
following the manufacturer’s instructions.
Based on data provided by the manufacturer (ThermoFisher Scientific), a sample size of >3
biological replicates was determined using a priori power analysis (G*Power 3.1). This
resulted in >90% statistical power to detect at least 20% difference in mean with an effect
size of 4.88 and standard deviation of 0.04.
2
.18 C. elegans Strain and Culture Conditions
The wild-type Bristol N2 strain of Caenorhabditis elegans (RRID: SCR_007341) was
obtained from the Caenorhabditis Genetics Center and were maintained on solid nematode
growth medium (NGM) with a lawn of E.coli OP50 that was incubated at 20°C (Brenner,
This article is protected by copyright. All rights reserved.

1
974).
2
.19 C. elegans Lifespan Assay
All test agents were dissolved in 800 µL methanol and then added to liquid NGM to reach a
final concentration of 100 µM, the exception was for FeCl and Fe(SNH6) , which were
3
2
maintained at 50 µM. Methanol alone was used for the control. Assays were performed in 35
mm petri dishes with E.coli OP50 as the food source for C. elegans. The C. elegans were
synchronised to obtain the L4 larval stage (Amrit et al., 2014). Thereafter, 25 healthy L4
worms were transferred onto fresh NGM plates containing the relevant test agent (Day 0) and
shifted to 25°C for the subsequent longevity experiments (Amrit et al., 2014).
The parental nematodes were transferred to new NGM plates every day to avoid counting of
the second-generation individuals. The survival (measured as response to touch) of C.
elegans was scored each day. At least 100 worms were used for each treatment per
experiment and experiments were repeated at least 5 times (N = 5). Kaplan-Meier survival
curves were plotted using GraphPad Prism (GraphPad Software, San Diego, CA, USA;
RRID: SCR_002798). Sex as a biological variable is not applicable as C. elegans are
hermaphrodites.
Based on data from a previous study (Wood et al., 2004), a sample size of >197 nematodes
was determined using a priori power analysis (G*Power 3.1). This produced >80% statistical
power to detect a 10% increase in survival with an effect size of 0.2.
This article is protected by copyright. All rights reserved.

2
.21 Data and Analysis
The Shapiro-Wilk test was utilised to assess data normality. Homogeneity of variation was
assessed using the Breusch-Pagan test. With the exception of C. elegans lifespan studies, all
datasets were found to be homoscedastic with a Gaussian distribution, thus no data
normalisation was necessary. Gaussian distributed data was subjected to the parametric
Student’s t-test (two-tailed, independent groups), or one-way and two-way ANOVA with
Tukey’s post-hoc test. Student’s t-test was performed using Microsoft Excel (RRID:
SCR_016137), while one-way and two-way ANOVAs were performed using GraphPad
Prism 7. Significance between groups for lifespan studies was determined log-rank (Mantel-
Cox) test using GraphPad Prism 7. The results were considered significant when p<0.05. For
data expressed as “fold mean of the controls”, the average Control group values were
calculated from all biological replicates of the Control, and all data were then universally
divided by this average Control group value. This transformation maintains the coefficient of
variation of the data.
The table and figure legends state the number of independent biological replicates (N) utilised
and statistical analyses were conducted only using these independent biological replicates
where each group was at least N = 5. The statistical analysis of data comply with the
recommendations on experimental analysis and design in pharmacology (Curtis et al., 2018).
Studies were designed to generate groups of equal size using randomisation and blinded
analysis during experimentation and data analysis. There was no outlier exclusion.
2
.22 Nomenclature
Key protein targets and ligands in this article are hyperlinked to corresponding entries in
http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS
This article is protected by copyright. All rights reserved.

Guide to PHARMACOLOGY (Harding et al., 2018), and are permanently archived in the
Concise Guide to Pharmacology 2017/18 (Alexander et al., 2017; Alexander et al., 2019).
3
. Results
.1 Synthesis
The nicotinoyl hydrazones (Fig.1A,B) were synthesised by Schiff-base condensations
Kalinowski et al., 2008). Various aldehydes and ketones were chosen to form 20 tridentate
3
(
ligands (NNO- or ONO-donors) upon reaction with nicotinic acid hydrazide. The structure of
SNH2 and SNH6 were determined by X-ray crystallography (Supporting Information Fig.S1).
As aroylhydrazones effectively chelate Fe, the 1:2 Fe:ligand complexes of selected
hydrazones (PCNH, SNH6, SNH8 and PrNH1) were synthesised and characterised
(
Supporting Information).
3
.2 Cytotoxicity of the Nicotinoyl Hydrazones
The cytotoxicity of the hydrazones and their selected Fe-complexes was examined using SK-
N-MC neuroepithelioma cells (Table 1) that represent a neural model implemented to assess
therapeutics for AD (Palanimuthu et al., 2017; Plaschke & Kopitz, 2015). The well-
characterised Fe chelators, DFO, (Fig.1C) and Dp44mT (Fig.1C), were used as controls
because of their low and potent cytotoxicity, respectively, is well-described in SK-N-MC
cells (Richardson et al., 2006). As additional controls, we also assessed the cytotoxicity of
CQ, a chelator previously investigated clinically for AD (Ritchie et al., 2003) (Fig.1C), and
the structurally-related hydrazones, salicylaldehyde benzoyl hydrazone (SBH, Fig.1C) and
salicylaldehyde isonicotinoyl hydrazone (SIH, Fig.1C). In these studies, DFO and Dp44mT
displayed IC50 values of 20.91 and 0.009 µM, respectively (Table.1). CQ, SBH and SIH had
IC50 values of 14.11, 9.43 and 19.38 µM, respectively.
This article is protected by copyright. All rights reserved.

The novel hydrazones demonstrated an IC50 between 0.56 to >100 µM. The ketone-derived
analogues generally showed greater cytotoxicity than aldehyde-derived analogues (Fig.1A,B).
Many aldehyde-derived hydrazones, including PCNH, 8-OH-QNH, PNH, SNH2, SNH3,
SNH5, SNH6, SNH8 and SNH9, displayed low cytotoxicity (i.e., IC50>10 µM; Table.1).
The four Fe-complexes assessed were selected as representatives for each subset of
hydrazones (i.e., ketone vs. aldehyde; and ONO vs. NNO-donor sets). The cytotoxicity of the
PCNH- and SNH8-Fe complexes was increased versus their free ligands (Table.1). In
contrast, SNH6 and PrNH1 demonstrated low cytotoxicity as Fe-complexes (Table.1). Hence,
several hydrazones and their Fe-complexes demonstrated low cytotoxicity, in particular
SNH6, suggesting its suitability for long-term treatment.
5
9
3
.3 Effect of Nicotinoyl Hydrazones on Cellular Fe Release
Brain Fe dys-homeostasis and Fe-loading is an AD hallmark, leading to oxidative stress
Greenough et al., 2013). Therefore, we examined the efficacy of the nicotinoyl hydrazones
(
to mobilise intracellular Fe versus the controls: (1) nicotinic acid hydrazide (NH), a precursor
of all nicotinoyl hydrazones herein; (2) the Fe chelators, DFO and Dp44mT (Yuan et al.,
2
004); (3) CQ; and (4) the structurally-related hydrazones, SBH and SIH, that show high Fe-
chelation efficacy (Richardson et al., 1995).
5
9
Control medium alone resulted in minimal Fe release from SK-N-MC cells (Fig.2A). NH
did not induce significant 59Fe release. DFO mobilised low amounts of intracellular 59Fe,
whereas Dp44mT, CQ, SBH and SIH markedly increased 59Fe release (Fig.2A). All
analogues, except SNH9 and agents with NNO-donor atoms (i.e., PCNH, QNH, 8-OH-QNH)
5
9
demonstrated high Fe chelation efficacy and were comparable to Dp44mT, CQ, SBH and
This article is protected by copyright. All rights reserved.

SIH (Fig.2A). PCNH and SNH9 displayed moderate 59Fe-chelation efficacy, while the
quinoline analogues, QNH and 8-OH-QNH, were comparable to DFO. The high activity of
several agents, including SNH6, demonstrated their potential to target cellular Fe-
accumulation in AD.
3
.4 Effect of Nicotinoyl Hydrazones in Inhibiting 59Fe Uptake from 59Fe-Transferrin
(
Tf)
5
9
Considering the marked activity of the hydrazones on cellular Fe release, we examined their
ability to inhibit 59Fe uptake by SK-N-MC cells from 59Fe
2
-Tf (Fig.2B). As for the efflux
studies, NH, DFO, Dp44mT, CQ, SBH and SIH were utilised as controls. The precursor, NH,
5
9
59
slightly reduced Fe uptake (Fig.2B). In contrast, Dp44mT markedly inhibited Fe uptake,
while CQ was less effective (Fig.2B). SBH and SIH showed activity greater than DFO and
CQ. Many novel analogues, including SNH6, showed potent activity at decreasing 59Fe
uptake (Fig.2B), and thus, their potential for targeting cellular Fe-loading in AD.
3
.5 Effect of the Fe-Complexes of Nicotinoyl Hydrazones on Ascorbate Oxidation
Plaque Fe accumulation, Fe dys-homeostasis and the resultant associated ROS generation are
a feature of AD (Greenough et al., 2013). An important property of chelators for the
treatment of Fe-loading is that they should remove Fe without forming redox-active
complexes. Hence, the ability of the Fe-nicotinoyl hydrazone complexes to catalyse ascorbate
oxidation via Fe-mediated Fenton chemistry was examined (Richardson et al., 2006). The
well-known, redox-active Fe-complexes of EDTA and Dp44mT were positive controls, while
the redox-inactive DFO-Fe(III) complex was a negative control (Richardson et al., 2006).
This article is protected by copyright. All rights reserved.

Dp44mT-Fe and EDTA-Fe markedly accelerated ascorbate oxidation, while the redox-
inactive DFO-Fe complex decreased ascorbate oxidation (Fig.3A). All hydrazone-derived Fe-
complexes, except that of 8-OH-QNH, did not show pro-oxidative activity (Fig.3A). Fe-
complexes of QNH, PNH, SNH1, SNH4, SNH7, SNH9, AcNH1 and InNH1 inhibited
ascorbate oxidation, while the remaining Fe-hydrazone complexes had no effect. Thus, the
Fe-hydrazone complexes do not induce ROS and have appropriate chelation properties for
AD treatment.
3
.6 Effect of Nicotinoyl Hydrazones on Hydrogen Peroxide-Induced Cytotoxicity
As metal-induced oxidative stress is a factor in AD (Greenough et al., 2013), the ability of
the novel agents to protect SK-N-MC cells from H -mediated cytotoxicity was examined
Lim et al., 2008). Representative nicotinoyl hydrazones from each subset (i.e., ketone vs.
2
O
2
(
aldehyde; and ONO vs. NNO donor atom set) that showed low cytotoxicity (Table.1),
namely PCNH, 8-OH-QNH, SNH2, SNH3, SNH5, SNH6, SNH8 and PrNH1, were
selectively investigated as promising agents to examine structure-activity relationships in all
further studies.
Control cells incubated with H
control hydrazones, PCTH, SBH and SIH, inhibited H
cytotoxic effects alone and increased H -mediated cytotoxicity, while NH could not rescue
cells from H -mediated cytotoxicity (Fig.3B).
2
O
2
showed a distinct decrease in viability (Fig.3B). The
2
O
2
-mediated cytotoxicity. CQ showed
2
O
2
2
O
2
The nicotinoyl hydrazones alone showed minimal effects on viability (Fig.3B). However, in
the presence of H , agents utilising the NNO-donor set, which showed poor iron release
Fig.2A), could not rescue cells from H -mediated cytotoxicity, while hydrazones utilising
2
O
2
(
2
O
2
This article is protected by copyright. All rights reserved.

the ONO-donor set (e.g., SNH6), that mediated high iron mobilisation (Fig.2A), protected
cells from H
2
O
2
-induced cytotoxicity, highlighting their potential for AD treatment.
+
3
.7 Effect of Nicotinoyl Hydrazones on Intracellular NAD /NADH Ratios
+
The NAD /NADH ratio is an important measure of redox state, with higher ratios suggestive
of improved metabolic function (Verdin, 2015). Primary human astrocytes were selected as
an ideal brain cell model to investigate the potential of the nicotinoyl hydrazones to increase
+
intracellular NAD (Bustamante et al., 2018). Astrocytes were utilised considering their
important role in neuronal and synaptic support and evidence of their involvement in AD
pathogenesis, suggesting their contribution to: (1) plaque formation by secreting considerable
levels of Aβ1-42; (2) oxidative stress via increased ROS production; and (3) mediating altered
Ca2+ signalling in neurons (Oksanen et al., 2017; Oyabu et al., 2019). The de-novo NAD+
synthesis pathway and the NAD+ salvage pathway, which resynthesises NAD from
nicotinamide (NAM), are present in these cells (Harlan et al., 2016).
+
+
Astrocytes were incubated with the NAD precursors, nicotinic acid (NA) or NAM as
controls, DFO, or selected hydrazones, PCNH, 8-OH-QNH, SNH2, SNH6 and PrNH1 (0-10
+
μM) for 24 h/37°C. Intracellular NAD (Fig.4A) and NADH (Fig.4B) were measured and the
+
+
NAD /NADH ratio calculated (Fig.4C). NAM increased intracellular NAD to a greater
+
extent than NA (Fig.4A). DFO did not markedly alter NAD , indicating that Fe chelation did
+
+
not affect NAD (Fig.4A). PCNH and 8-OH-QNH slightly increased NAD , while SNH2,
+
SNH6 and PrNH1 mediated a marked, dose-dependent increase in NAD (Fig.4A). SNH6
+
was the most effective at increasing NAD , being more effective than NA or NAM at ≥500
nM (Fig.4A).
This article is protected by copyright. All rights reserved.