Novel resveratrol derivatives have diverse effects on the survival, proliferation and senescence of primary human fibroblasts

Article abstract of DOI:10.1007/s10522-020-09896-6

Resveratrol alters the cytokinetics of mammalian cell populations in a dose dependent manner. Concentrations above 25–50 μM typically trigger growth arrest, senescence and/or apoptosis in multiple different cell types. In contrast, concentrations below 10 μM enhance the growth of log phase cell cultures and can rescue senescence in multiple strains of human fibroblasts. To better understand the structural features that regulate these effects, a panel of 24 structurally-related resveralogues were synthesised and evaluated for their capacity to activate SIRT1, as determined by an ex-vivo SIRT1 assay, their toxicity, as measured by lactate dehydrogenase release, and their effects on replicative senescence in MRC5 human fibroblasts as measured by their effects on Ki67 immunoreactivity and senescence-associated β galactosidase activity. Minor modifications to the parent stilbene, resveratrol, significantly alter the biological activities of the molecules. Replacement of the 3,5-dihydroxy substituents with 3,5-dimethoxy groups significantly enhances SIRT1 activity, and reduces toxicity. Minimising other strong conjugative effects also reduces toxicity, but negatively impacts SIRT1 activation. At 100 μM many of the compounds, including resveratrol, induce senescence in primary MRC5 cells in culture. Modifications that reduce or remove this effect match those that reduce toxicity leading to a correlation between reduction in labelling index and increase in LDH release. At 10 μM, the majority of our compounds significantly enhance the growth fraction of log phase cultures of MRC5 cells, consistent with the rescue of a subpopulation of cells within the culture from senescence. SIRT1 activation is not required for rescue to occur but enhances the size of the effect.

Full text of DOI:10.1007/s10522-020-09896-6

Biogerontology
https://doi.org/10.1007/s10522-020-09896-6(0123456789(). -, volV
(
)
0
1
2
3
4
5
8
9
(
)
.
-
v
o
l
V
)
RESEARCH ARTICLE
Novel resveratrol derivatives have diverse effects
on the survival, proliferation and senescence of primary
human fibroblasts
.
Vishal C. Birar Angela N. Sheerin Elizabeth L. Ostler
.
.
Richard G. A. Faragher
Received: 18 July 2020 / Accepted: 7 August 2020
Ó Springer Nature B.V. 2020
Abstract Resveratrol alters the cytokinetics of
mammalian cell populations in a dose dependent
manner. Concentrations above 25–50 lM typically
trigger growth arrest, senescence and/or apoptosis in
multiple different cell types. In contrast, concentra-
tions below 10 lM enhance the growth of log phase
cell cultures and can rescue senescence in multiple
strains of human fibroblasts. To better understand the
structural features that regulate these effects, a panel
of 24 structurally-related resveralogues were synthe-
sised and evaluated for their capacity to activate
SIRT1, as determined by an ex-vivo SIRT1 assay,
their toxicity, as measured by lactate dehydrogenase
release, and their effects on replicative senescence in
MRC5 human fibroblasts as measured by their effects
on Ki67 immunoreactivity and senescence-associated
b galactosidase activity. Minor modifications to the
parent stilbene, resveratrol, significantly alter the
biological activities of the molecules. Replacement
of the 3,5-dihydroxy substituents with 3,5-dimethoxy
groups significantly enhances SIRT1 activity, and
reduces toxicity. Minimising other strong conjugative
effects also reduces toxicity, but negatively impacts
SIRT1 activation. At 100 lM many of the com-
pounds, including resveratrol, induce senescence in
primary MRC5 cells in culture. Modifications that
reduce or remove this effect match those that reduce
toxicity leading to a correlation between reduction in
labelling index and increase in LDH release. At
10 lM, the majority of our compounds significantly
enhance the growth fraction of log phase cultures of
MRC5 cells, consistent with the rescue of a subpop-
ulation of cells within the culture from senescence.
SIRT1 activation is not required for rescue to occur but
enhances the size of the effect.
Keywords Resveratrol Á Senescence Á Toxicity Á
SIRT1
Introduction
Electronic supplementary material The online version of
this article (https://doi.org/10.1007/s10522-020-09896-6) con-
tains supplementary material, which is available to authorized
users.
During the latter decades of the twentieth century
organismal ageing was regularly conceptualised as the
potential result of the interplay of dozens, if not
hundreds, of biological mechanisms, each of which
exerted small effects on lifespan (Kirkwood and
Franceschi 1992; Medvedev 1990). However, more
recently the adoption of a criterion-based approach for
V. C. Birar Á A. N. Sheerin Á E. L. Ostler (&) Á
R. G. A. Faragher
School of Pharmacy and Biomolecular Sciences,
University of Brighton, Cockcroft Building,
Moulsecoomb, Brighton BN2 4GJ, UK
e-mail: e.ostler@brighton.ac.uk
123

Biogerontology
potential mechanisms (that they manifest during
normal ageing, that their experimental aggravation
accelerates ageing and that their amelioration retards
it) has substantially reduced the number of mecha-
nisms that can currently be considered plausible. The
best example of this ontological approach, the ‘Hall-
marks of ageing’ lists cell senescence, genomic
instability, loss of proteostasis, mitochondrial dys-
function, epigenetic alteration, stem cell exhaustion
and altered intercellular communication as candidate
reversing cell senescence after it has become estab-
lished (Latorre et al. 2017).
Understanding the structural motifs necessary for
interaction with different hallmark mechanisms would
greatly enhance the utility of resveratrol and its
analogues. To facilitate this we developed a rapid,
robust and convenient protocol for resveralogue
synthesis which we have used to create a structurally
diverse set of compounds, many of which are novel.
This paper reports the effects of these compounds on
one of the best characterised hallmark mechanisms,
the proliferation and senescence of primary human
fibroblasts.
´
´
major ageing mechanisms (Lopez-Otın et al. 2013).
On this model it is possible to envisage two distinct
ways in which a small molecule or similar intervention
might lengthen lifespan. Firstly, it might directly
interact with multiple hallmark mechanisms at once,
in a manner analogous to that in which phloroglucinols
such as hyperforin interact with the multiple pathways
that are involved in depression (Butterweck 2003).
Secondly, interconnectedness between the ageing
hallmarks might allow an intervention targeting one
hallmark to indirectly effect one or more of the others
(perhaps best described as ‘domino impingement’).
Distinguishing between these potential modes of
action is clearly desirable for many reasons, including
gaining a better understanding of the primary mode of
action of compounds currently being used clinically to
target hallmark mechanisms (such as quercetin, which
is both a senolytic and a proteostatic modulator see
Kirkland and Tchkonia (2017) and Wedel et al. (2018)
and aiding in the development of more potent and/or
selective next generation compounds.
Materials and methods
General synthetic procedures for resveralogues
Reaction schemes and V series identities are shown in
Schemes 1, 2 and 3 and Tables 1 and 2. Detailed
synthetic methods and characterisation data are pre-
sented in the electronic supplementary information.
Briefly, functionalised stilbenes were prepared
from commercially available substituted benzyl bro-
mides and benzaldehydes by using a modified Horner–
Wadsworth–Emmons reaction as previously described
(Birar et al. 2015). Four of these were subjected to
boron tribromide mediated deprotection to afford 3,5-
diphenol analogues to facilitate comparisons with the
parent 3,5,40-triphenol, resveratrol.
Resveratrol (trans-3,5,40-trihydroxystilbene) is an
excellent candidate for a molecule which modulates
multiple hallmark mechanisms simultaneously.
Although initially proposed to act as a dietary
restriction mimetic based in part on the observation
that resveratrol supplementation of the standard diet of
laboratory mice induces shifts in gene expression
similar to those seen when animals undergo calorie
restriction (Barger et al. 2008; Bhullar and Hubbard
2015). A wealth of data have subsequently become
available consistent with resveratrol affecting pro-
teostasis (Corpas et al. 2019), stem cell exhaustion
(Safaeinejad et al. 2018), mitochondrial dysfunction
(de Oliveira et al. 2016), genomic instability (Stopper
et al. 2005) and intercellular communication (Huang
et al. 2020). Recently we have also shown that
resveratrol has profound effects on the inflammatory
phenotype of senescent cells, and is capable of
A range of nitrogen-substituted resveralogues were
prepared from nitro- and cyano-substituted simple
resveralogues (see also Latorre et al. 2017). Anilines
were prepared by reduction of nitro-substituted stil-
benes, and then used to prepare a ranged of anilide and
sulphanilide derivatives by reaction with the appro-
priate carboxylic acid or sulphonyl chloride. Three of
these were also demethylated as above. Finally 2- and
4-tetrazole substituted analogues were afforded from
2- and 4-cyano stilbenes via Finnegan’s method.
Dihydro analogues of both resveratrol and V14
were readily prepared by palladium-catalysed reduc-
tion as described previously (Faragher et al. 2010).
123

Biogerontology
Scheme 1 Synthesis of simple resveralogues
Scheme 2 Functional group modifications of nitrogen-substituted resveralogues
Scheme 3 Synthesis of dihydroresveralogues
123

Biogerontology
Determination of SIRT1 activation capacity
Table 1 Identities of simple resveralogues presented in V
series number order
of novel resveralogues by measurement of lysyl
deacetylase activity
Compound number
R1
R2
V1
4-NO2
2-F
–
A commercial kit (SIRT1 fluorometric drug discovery
kit, Cayman) based on the Fluor de Lys assay was used
to determine the potential of each resveralogues to
activate sirtuin 1. All reactions were carried in reaction
buffer (50 mM Tris/HCl, pH 8.0, 137 mM NaCl,
2.7 mM KCl, 1 mM MgCl2, 1 mg/cl BSA and 2% (v/
v) DMSO).
V2
–
V3
4-MeO
3,5-Dimethyl
2,6-Difluoro
2,4-Difluoro
3-Cl
–
V4 (V26: 3,5-dihydroxy)
–
V5
–
V6 (V40: 3,5-dihydroxy)
–
V7
4-MeO
V8
3-CF3
4-NMe2
Resveralogues were diluted from a DMSO stock
solution to a final concentrations of 25 lM in reaction
buffer. 5 lL of purified SIRT1 enzyme solution was
then added and 15 lL of substrate solution was added
to produce a final target peptide concentration of
125 lM (with 3 mM NAD? in deacetylase reaction).
The reaction mixture was incubated on a shaker for
45 min at room temperature. After this time, stop
solution was added and the mixture incubated for a
further 30 min at room temperature. Fluorescence
reading was measured using Perkin-Elmer VICTOR
1420 multi-label Plate Reader with excitation set at
360/40 nm and emission measured at 460/40 nm.
Activity was calculated according to the formula:
V9
4-COOMe
2-CN
–
V10 (V23: 3,5-dihydroxy)
–
V14 (V13: 3,5-dihydroxy)
4-Me
–
V16
V17
V18
V19
V20
V21
V22
V25
V27
V28
V29
V33
V42
2-CN
2-CN
3-Cl
4-NMe2
2-NO2
2-CN
4-MeO
4-MeO
3-NO2
2, 6-Dichloro
2-NO2
2-NO2
3-NO2
4-NO2
H
4-MeO
–
–
2-CN
2,4-Dimethoxy
2,4-Dimethoxy
H
–
Activity ¼ ½Fluorescence of sample
À Initial fluorescenceŠ  100:
4-CN
H
–
This was then normalised relative to enzyme only
controls and resveratrol to give activation relative to
resveratrol as a ratio, such that negative values
represent enzyme inhibition, positive values indicate
activation and compounds with values greater than
one are more effective activators than resveratrol.
Table 2 Identities of modified nitrogen-substituted resver-
alogues presented in V series number order
Compound number Ring 1 substitution Ring 2 substitution
V11
V12
V24
V30
V31
V32
V34
V36
V37
V38
V39
V41
4-NHCOMe
4-NH2
3,5-Dimethoxy
3,5-Dimethoxy
3,5-Dimethoxy
3,5-Dimethoxy
3,5-Dihydroxy
3,5-Dimethoxy
3,5-Dimethoxy
3,5-Dimethoxy
3,5-Dihydroxy
3,5-Dimethoxy
3,5-Dimethoxy
3,5-Dihydroxy
General tissue culture
2-Tetrazole
4-NHSO2Me
4-NHCOMe
4-NHCOiPr
4-Tetrazole
4-NHCOPh
4-NHCOPh
2-NH2
MRC5 human diploid foetal lung fibroblasts
(AG05965B) were obtained from the Coriell institute
for medical research and grown at a seeding density of
1.3 9 104 cells cm-2 in Modified Eagle’s Medium
with essential and non-essential amino acids, 15%
FCS, 2 mM glutamine, penicillin (50 U/L), strepto-
mycin (50 lg/L). Cells were sub-cultured every 5–10
days by trypsin-EDTA dispersion and viable cell
numbers determined by trypan blue staining. Replica-
tive lifespan was measured in Population Doublings
(PDs) calculated using the standard formula.
2-NHCOPh
2-NHCOPh
123

Biogerontology
Determination of growth fraction by Ki67
immunostaining
Fibroblast populations were dosed with resveralogues
over a range of 10–100 lM for 24 h after which the
levels of cell death, senescence and proliferation were
assessed.
MRC5 cells on cover slips were washed with PBS, and
fixed in a 1:1 mixture of methanol and acetone for
5–10 min at room temperature. The fixed cells were
incubated for one hour at room temperature with a
primary antibody raised against Ki67 (Dako M0722,
mouse anti-Ki67, diluted in 1% (v/v) FCS in PBS 1:20
dilution). The cells were then washed three times in
PBS and incubated for a further hour in the dark at
room temperature with a secondary antibody conju-
gated with FITC (Dako F0261, rabbit anti-mouse
FITC, diluted using 1% (v/v) FCS in PBS, 1:20
dilution). Coverslips were washed three times in PBS
and once in water to remove excess salt. They were
then mounted in a DAPI-containing mountant (Vec-
tashield) and viewed on a confocal microscope. Cells
traversing the cell cycle showed a distinctive pattern of
nuclear staining. To determine the proliferating frac-
tion, either 400 Ki67 positive or 1000 total nuclei were
counted in random fields on each of three coverslips.
Determination of toxicity by lactate
dehydrogenase release
A commercial LDH cytotoxicity assay (Pierce LDH
Cytotoxicity Assay Kit) was used to determine cell
viability. MRC5 fibroblasts were seeded in 24 well
plates with 1.3 9 105 cells cm-2 and allowed to
recover from trypsinisation for 24 h under standard
tissue culture conditions. The cultures were then
exposed the 10–100 lM of the resveralogue for a
further 24 h. 50 lL of media from each well was then
mixed with an equal volume of LDH assay reaction
mixture and incubated at room temperature in the dark
for 30 min. 50 lL of LDH assay stop solution was
added to each well and the absorbance of the solution
was measured by spectrophotometry at 490 nm.
Complete cell lysis and vehicle only controls were
included.
Statistical analysis
Determination of senescent fraction by catalytic
histochemistry
All statistical analysis was carried out using Graph
Pad-Prism.
MRC5 fibroblasts were grown on 13 mm diameter
glass coverslips at identical seeding densities to the
parent culture. After exposure to resveralogues they
were washed and fixed in 3% formaldehyde for 10 min
at room temperature. The cells were washed and then
incubated overnight at 37 °C in a solution of 4-chloro-
5-bromo-3-indolyl-b-D-galactoside in citric acid
phosphate buffer containing potassium ferrocyanide
(5 mM) and potassium ferricyanide (5 mM). As a
positive control for lysosomal b-galactosidase activ-
ity, one set of coverslips was incubated in buffer at pH
4.0. For the determination of senescence-associated b-
galactosidase, the incubation buffer pH was raised to
pH 6.0. The fixed cells were counterstained using
hematoxylin QS and Nuclear Fast Red and viewed
under a light microscope. To determine the senescent
fraction, 1000 cells were counted in random fields on
two separate coverslips and the positive fraction
calculated.
Results
The structures of the V series of resveralogues are
shown in Schemes 1, 2 and 3 together with Tables 1
and 2.
SIRT activation
The resveralogues’ capacity to activate or inhibit
SIRT1 in the well-known ‘Fluor de Lys’ fluorescence-
based histone deacetylase activity assay is shown in
Fig. 1a. Relative to resveratrol, the resveralogues
displayed a spectrum of activities from increased
activation (maximum of approximately two fold)
through to inhibition.
The effect of single substituent modifications can
be evaluated by comparison of the 3,5-dimethoxy
series, and their deprotected 3,5-dihydroxy counter-
parts (see Table 3). In this series the highest values for
SIRT 1 activation are observed both for compounds
123

Biogerontology
a
b
c
d
Fig. 1 Selected cell physiological effects of the V series
resveralogues relative to resveratrol and vehicle only controls.
a SIRT1 activation capacities at 25 lm relative to resveratrol
(shown in red, activation = 1) and the SIRT1 inhibitor sirtinol.
Error bars SD. b Capacity of the V series resveralogues to
induce necrosis in MRC5 human embryonic lung fibroblasts as
measured by the release of lactate dehydrogenase following
exposure to each compound at 100 lM for 24 h. Error
bars SD. c Stacking bar graph of Ki67 (grey) and senescence
associated beta galactosidase indices (white) after treatment
with 100 lM V series resveralogues for 24 h. d Ki67 indices
after treatment with 10 lM V series resveralogues for 24 h.
Figure produced in GraphPad-Prism
123

Biogerontology
Table 3 SIRT1 activation ratios for dimethoxy resveralogues and their dihydroxy equivalents
R
R0 = 3,5-dimethoxy SIRT activation ratio (SD) R0 = 3,5-dihydroxy SIRT activation ratio (SD)
p
3,5-Dimethyl V4
1.8 (0.16)
0.33 (0.05)
1.14 (0.09)
0.2 (0.04)
V26
V13
V31
V37
V41
V23
V40
0.53 (0.10)
- 1.33 (0.19)
0.76 (0.19)
0.00030
0.00012
0.03516
0.00030
0.00751
0.00001
0.81942
4-Methyl
V14
V11
4-Acetamide
4-Benzamide V36
2-Benzamide V39
- 1.12 (0.19)
- 0.24 (0.07)
- 0.11 (0.02)
0.48 (0.09)
- 0.03 (0.02)
0.68 (0.05)
0.5 (0.11)
2-Cyano
V10
V6
2,4-Difluoro
bearing electron donating (V3, V9) or electron with-
drawing (V1) substituents on the second ring. Activa-
tion is further enhanced by replacement of the 3,5-
dimethoxy substituents with 4-methoxy equivalents
(V22 vs. V18, V10 vs. V19), but is reduced for
analogous 2,4-dimethoxy compounds (V20 vs. V27,
V1 vs. V28). However, no other consistent patterns in
activity emerge in relation to variations in substituent
positioning.
significant. The highest toxicities in the V series are
associated with 3,5,40-trihydroxy (resveratrol),
4-amino (V8, V12, V17), dihalo (V5, V6, V21),
2-cyano (V10, V19, V23) and benzamide (V36, V39)
substitution. In contrast, conversion of amines to
amides (with the exception of 2-benzamide), and
cyano to tetrazole derivatives, consistently reduces
toxicity (p = 0.049), suggesting that substituents that
exert strong conjugative effects are causal for
increased toxicity. This is in contrast with the effects
on SIRT1 activity where removal of the hydroxyl
group increases activity but loss of the amino group
has the opposite effect.
Deprotection of any one analogue to its 3,5-
dihydroxy counterpart significantly (with the sole
exception of the difluoro analogues) reduces SIRT1
activity, and in many cases converts a SIRT1 activator
to an inhibitor (negative values) as seen in Table 3.
In contrast, conversion of 2- and 4-amino-substi-
tuted resveralogues to amide equivalents (V12 and
V38 to V11, V30, V32, V36 and V39) reduces SIRT1
activity significantly (p \ 0.01) for all compounds
except the 4-acetamide (V11).
Reductive removal of the central double bond (RSV
to DHRSV, and V14 to V15) reduces or maintains low
toxicity, possibly by disrupting conjugation between
the rings.
Hydrogenation of the central double bond (RSV to
DHRSV, and V14 to V15) maintains or enhances
SIRT activation activity.
Toxicity
The capacity of the novel resveralogues to induce
necrotic cell death at 100 lM, as measured by the
release of lactate dehydrogenase into the culture
medium is shown in Fig. 1b. As with SIRT1 activation
small modifications to one ring of resveratrol gives
rise to a wide range of toxicities ranging from
indistinguishable from vehicle only control values to
over 30% of positive control.
Fig. 2 Comparison of matched 3,5-dihydroxy and 3,5-
dimethoxy V series resveralogues’ capacity to induce necrosis
in MRC5 human embryonic lung fibroblasts as measured by the
release of lactate dehydrogenase following exposure to each
compound at 100 lM for 24 h. Error bars SD. Figure pro-
duced in GraphPad-Prism
Dihydroxy analogues are generally slightly more
toxic than their dimethoxy counterparts, as can be seen
in Fig. 2, but the differences observed are not
123

Biogerontology
Induction of senescence
did not. However, several compounds with no signif-
icant SIRT activity (V30, V32, V36, V39) also
provoked an increased labelling index. This dichot-
omy explains the weak but significant (p = 0.011,
r2 = 0.4361) correlation between SIRT1 activity and
labelling index after treatment at 10 lM (Fig. 4).
Like resveratrol, at 100 lM, many of the novel
resveralogues also induce cellular senescence as
demonstrated by the increased absence of staining
for the proliferation marker pKi67, although as above
there is a very wide range of activity (from no effect, to
a reduction to less than half the labelling index
observed with resveratrol, Fig. 1c). For a subset of the
V series compounds this was confirmed by observa-
tion of a concomitant increased senescence-associated
beta galactosidase activity (Fig. 1c). A clear correla-
tion is present between the capacity of resveralogues
to induce necrosis and their capacity to induce
senescence (Fig. 3, p [ 0.0006, r2 = 0.53). This is
reinforced by the observation that the functional group
interconversions that reduce toxicity (described
above) also result in reduced induction of senescence
at 100 lM.
Discussion
We have determined the effects on growth, death and
senescence of a panel of novel resveralogues com-
pared to the parent compound, resveratrol, and com-
mon metabolites such as dihydroresveratrol.
Accurately measuring the toxicity of resveralogues is
more complex than generally assumed. Many
researchers seeking to determine the toxicity of
polyphenols use viability assays based on the reduc-
tion of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide) which is soluble in
aqueous solution but reduced to an insoluble formazan
as a result of mitochondrial activity. The accuracy of
this assay is predicated on a linear relationship
between viable cell number, mitochondrial mass and
mitochondrial activity. However, alterations in either
mitochondrial content or metabolism influence the
apparent viability of cell populations and resveratrol-
like molecules are established activators of mitochon-
drial function (Bernhard et al. 2003). Other simple
options for measuring cytotoxicity (e.g. neutral red
Increase in growth fraction
Conversely, at 10 lM, treatment with a significant
number of the V series compounds enhances the
growth fraction of primary human fibroblasts as
measured by staining for pKi67 (Fig. 1d). Of all the
compounds examined, those with SIRT1 activation
activity similar to or greater than resveratrol (V2, 5,
11, V31) displayed significant increases in labelling
index, whilst SIRT inhibitors (V13, V23, V37, V41)
Fig. 3 Capacity of the V series resveralogues to induce
senescence and necrosis in log phase cultures of MRC5 human
embryonic lung fibroblasts as measured by a combination of loss
of immunoreactivity for Ki67 (Y axis) and release of lactate
dehydrogenase (X axis). Vehicle only control and resveratrol
shown as open and solid squares respectively. Figure produced
in GraphPad-Prism
Fig. 4 Relationship between the capacities of the V series
resveralogues to activate SIRT1 and to alter the growth fraction
of MRC5 human embryonic lung fibroblasts. Vehicle only
control shown as an open square. Resveratrol and dihy-
droresveratrol shown as solid squares. V series compounds
previously shown to rescue senescence (Latorre et al. 2017)
shown as solid triangles. Figure produced in GraphPad-Prism
123

Biogerontology
uptake assays) can also be compromised because
many polyphenols absorb strongly in the relevant
wavelengths used to measure death.
always immunologically detectable across all cell
cycle phases except G0 in log phase cell cultures
(Kalashnik et al. 2000) and modulation of Ki67 levels
does not affect cell cycle kinetics (Sobecki et al.
2016). Thus, the most plausible explanation for this
increase is the ‘rejuvenation’ of a fraction of the
senescent cell subpopulation within these cultures
(rather than, for example, altered total cell cycle time
or cell cycle phase duration). It is noteworthy that all
the V series compounds which we have previously
shown to allow senescent cells to divide again also
induce a significant pKi67 spike (Latorre et al. 2017).
Accordingly, we hypothesise that the compounds
which increase the fraction of cells within log phase
cultures staining positive for Ki67 will also rescue cell
cultures composed entirely of senescent cells.
Although activation of SIRT1 is not necessary for
these effects to occur there is a clear trend for
compounds which also activate SIRT1 to show a
larger increase in the culture growth fraction. This is
consistent with both SIRT1-dependent and indepen-
dent rescue mechanisms which can potentially be
elucidated using the appropriate combination of
compounds. It must be stressed that the ‘rejuvenation’
effect we have observed to date are transient and are
dependent on the continued presence of the
compound.
Cognizant of these issues but above all because we
wished to obtain an accurate measure of the induction
of necrotic cell death we applied the lactate dehydro-
genase release assay to our panel of resveralogues. In
general, substituents such as halogens, hydroxyl and
amine groups that exert strong conjugative effects
increase the toxicity of these compounds, which is
abrogated by conversion to ‘‘protected’’ methoxy or
anilide equivalents. In contrast whilst methoxy ana-
logues have better SIRT1 activation capacity than
hydroxyl equivalents, anilide analogues display
reduced SIRT1 activation when compared to the
parent amine. Reductive removal of the central double
bond also reduces toxicity and maintains or enhances
SIRT1 activation. To our knowledge this is the first
study of the structure activity relationships of resver-
alogues with regard to induction of necrosis.
Conversely, necrotic toxicity and the induction of
replicative senescence by resveralogues at high doses
correlate well. Given that resveratrol and its deriva-
tives are regularly viewed as potential anti-cancer
agents this could be seen as a useful feature - a single
compound which potentially either kills tumour cells
or renders them senescent. However, evidence is
growing that the release of cytoplasmic proteins (e.g.
HMGB1) into the extracellular space acts as a
proinflammatory and angiogenic stimulus (Lee et al.
2018; Jain et al. 2013). Given that development of a
necrotic core in tumours is associated with poor
patient prognosis the potential induction of necrosis by
resveralogues is another reason, alongside the high
doses required to induce senescence, that many of
these compounds may not represent especially promis-
ing anti-tumour agents.
Finally, in contrast, to the focus on SIRT1 activa-
tors such as resveratrol as anti-cancer drugs it has been
proposed that SIRT1 inhibition may represent a utile
strategy for the prevention of tumour growth. The
canonical SIRT1 inhibitor, sirtinol, has been shown to
induce senescence and apoptosis in MCF-7 cells
(Wang et al. 2012) with an IC50 of * 44-–48 lM
depending on exposure time. At least one of our
compounds (V13) may be especially well suited for an
approach of this type since it shows higher SIRT1
inhibition than sirtinol, lower toxicity than resveratrol
at 100 lM (p = 0.039) and significantly reduces the
pKi67 index of MRC5 fibroblasts at 25 lM. This is
perhaps surprising given that the only structural
difference between resveratrol and V13 is the replace-
ment of the 4-phenol with a 4-methyl substituent, but
highlights the importance of resonance effects in
determining the activities of these molecules.
More promisingly, treatment of mid-passage cul-
tures of MRC5 human fibroblasts with low concen-
trations of many, but not all, of the V series
compounds increased the pKi67 index and decreased
the fraction of cells in the cultures positive for
senescence-associated beta galactosidase. Previously,
we have shown that dosing senescent fibroblasts from
multiple different tissues with resveratrol, dihy-
droresveratrol and four of our V series compounds
results in a SIRT1-independent, SASP-independent
and senolysis-independent exit from cellular senes-
cence which we have termed ‘rejuvenation’. Ki67 is
Author contributions All authors made substantial
contributions to the design of the work, analysis of data and
123

Biogerontology
production of the manuscript. All authors read and approved the
final manuscript.
metabolism and related dysbiosis of gut programed by
prenatal high-fat diet and postnatal high-fat diet exposure.
J Nutr Biochem 75:108260
Jain MV, Paczulla AM, Klonisch T, Dimgba FN, Rao SB,
Roberg K, Schweizer F, Lengerke C, Davoodpour P,
Palicharla VR, Maddika S, Łos M (2013) Interconnections
between apoptotic, autophagic and necrotic pathways:
implications for cancer therapy development. J Cell Mol
Med 17(1):12–29
Funding We thank the University of Brighton for
studentship for VB.
a
Data availability ESI includes details of procedures and
characterisation. Novel compounds available in limited
quantities.
Kalashnik L, Bridgeman CJ, King AR, Francis SE, Mikhalovsky
S, Wallis C, Denyer SP, Crossman F, Faragher RG (2000)
A cell kinetic analysis of human umbilical vein endothelial
cells. Mech Ageing Dev 120(1–3):23–32
Kirkland JL, Tchkonia T (2017) Cellular senescence: a trans-
lational perspective. EBioMedicine 21:21–28
Kirkwood TB, Franceschi C (1992) Is aging as complex as it
would appear? New perspectives in aging research. Ann N
Y Acad Sci 663:412–417
Latorre E, Birar VC, Sheerin AN, Jeynes JCC, Hooper A, Dawe
HR, Melzer D, Cox LS, Faragher RGA, Ostler EL, Harries
LW (2017) Small molecule moderation of splicing factor
expression is associated with rescue from cellular senes-
cence. BMC Cell Biol 18:31. https://doi.org/10.1186/
s12860-017-0147-7
Compliance with Ethical Standards
Conflict of interest None.
Consent to participate Not applicable.
Consent for publication All authors consent.
Ethics approval Not applicable.
References
Lee SY, Ju MK, Jeon HM, Jeong EK, Lee YJ, Kim CH, Park
HG, Han SI, Kang HS (2018) Regulation of tumor pro-
gression by programmed necrosis. Oxid Med Cell Longev
2018:3537471
Barger JL, Kayo T, Vann JM, Arias EB, Wang J, Hacker TA,
Wang Y, Raederstorff D, Morrow JD, Leeuwenburgh C,
Allison DB, Saupe KW, Cartee GD, Weindruch R, Prolla
TA (2008) A low dose of dietary resveratrol partially
mimics caloric restriction and retards aging parameters in
mice. PLoS ONE 3(6):e226
´
´
Lopez-Otın C, Blasco MA, Partridge L, Serrano M, Kroemer G
(2013) The hallmarks of aging. Cell 153:1194–1217
Medvedev ZA (1990) An attempt at a rational classification of
theories of ageing. Biol Rev 65:375–398
Bernhard D, Schwaiger W, Crazzolara R, Tinhofer I, Kofler R,
Csordas A (2003) Enhanced MTT-reducing activity under
growth inhibition by resveratrol in CEM-C7H2 lympho-
cytic leukemia cells. Cancer Lett 195(2):193–199
Bhullar KS, Hubbard BP (2015) Lifespan and healthspan
extension by resveratrol. Biochim Biophys Acta (BBA)
1852:1209–1218
Birar V et al (2015) A facile, stereoselective, one-pot synthesis
of resveratrol derivatives. Chem Cent J 9:26
Butterweck V (2003) Mechanism of action of St John’s Wort in
depression: what is known? CNS Drugs 17(8):539–562
Safaeinejad Z, Kazeminasab F, Kiani-Esfahani A, Ghaedi K,
Nasr-Esfahani MH (2018) Multi-effects of resveratrol on
stem cell characteristics: effective dose, time, cell culture
conditions and cell type-specific responses of stem cells to
resveratrol. Eur J Med Chem 155:651–657
`
Sobecki M, Mrouj K, Camasses A, Parisis N, Nicolas E, Lleres
D, Gerbe F, Prieto S, Krasinska L, David A, Eguren M,
´
Birling M-C, Urbach S, Hem S, Dejardin J, Malumbres M,
Jay P, Dulic V, Lafontaine DLJ, Feil R, Fisher D (2016)
The cell proliferation antigen Ki-67 organises heterochro-
matin. eLife 5:e13722
´
´
´
´
`
Corpas R, Grin˜an-Ferre C, Rodrıguez-Farre E, Pallas M, San-
feliu C (2019) Resveratrol induces brain resilience against
Alzheimer neurodegeneration through proteostasis
enhancement. Mol Neurobiol 56(2):1502–1516
Stopper H, Schmitt E, Kobras K (2005) Genotoxicity of phy-
toestrogens. Mutat Res 574(1–2):139–155
Wang J, Kim TH, Ahn MY, Lee J, Jung JH, Choi WS, Lee BM,
Yoon KS, Yoon S, Kim HS (2012) Sirtinol, a class III
HDAC inhibitor, induces apoptotic and autophagic cell
death in MCF-7 human breast cancer cells. Int J Oncol
41(3):1101–1109
Wedel S, Manola M, Cavinato M, Trougakos IP, Jansen-Du¨rr P
(2018) Targeting protein quality control mechanisms by
natural products to promote healthy ageing. Molecules
23(5):1219
de Oliveira MR, Nabavi SF, Manayi A, Daglia M, Hajheydari Z,
Nabavi SM (2016) Resveratrol and the mitochondria: from
triggering the intrinsic apoptotic pathway to inducing
mitochondrial biogenesis, a mechanistic view. Biochim
Biophys Acta 1860(4):727–745
Faragher RGA et al (2010) Resveratrol, but not dihydroresver-
atrol, induces premature senescence in primary human
fibroblasts. Age. https://doi.org/10.1007/s11357-010-
9201-5
Publisher’s Note Springer Nature remains neutral with
regard to jurisdictional claims in published maps and
institutional affiliations.
Huang YC, Huang LT, Sheen JM, Hou CY, Yeh YT, Chiang CP,
Lin IC, Tiao MM, Tsai CC, Lin YJ, Chen CC, Tain YL, Yu
HR (2020) Resveratrol treatment improves the altered
123