Page 47 - Read Online

Girotti et al. J Cancer Metastasis Treat 2020;6:52 I http://dx.doi.org/10.20517/2394-4722.2020.107 Page 9 of 15

postulated that acetylation of specific lysine residues in p65 was necessary for stimulating transcription.
[63]
As supporting evidence, Fahey et al. showed that acetylation of lysine-310 (p65-acK310) increased
progressively during post-irradiation incubation of U87 cells, reaching > 3-times the control level after 24 h
[Figure 4A]. The rise in acK310 level was blocked by C646, an inhibitor of activated p300, confirming
[63]
that the latter had catalyzed this acetylation . The acetyltransferase p300 and its paralog CREB-binding
protein (CBP) act as transcriptional co-activators for several tumor-promoting transcription factors [64,65] .
p300 stimulates gene expression at promoter sites by catalyzing acetylation of specific lysine residues on
[64]
histones or transcription factors such as NF-κB . Therefore, we determined whether p300 is involved
in p65-K310 acetylation and, if so, how photodynamic stress might affect p300 expression/activity.
Immunoblot analysis revealed that photostress had no effect on overall p300 level relative to a dark control.
As with p65-acK310 build-up, however [Figure 3A], there was a progressive increase in activated p300,
i.e., Ser-1834-phosphorylated enzyme (p-p300), over at least a 6-h post-hν period [Figure 4B]. Moreover,
immunoprecipitation (pull-down) analysis revealed a striking photostress-enhanced interaction of activated
p300 with NF-κB-p65, thus favoring acetylation of the latter [Figure 4C] . Another striking finding of this
[59]
study is that Sirtuin-1 (Sirt1), a Class-III deacetylase that modulates gene expression by catalyzing acetyl
[66]
group removal , was strongly downregulated in photostressed U87 cells [Figure 4D], whereas a homolog,
[59]
Sirt2, was unaffected . Along with these effects, there was a striking post-hν upregulation of type-4
bromodomain and extra-terminal domain (BET) protein (Brd4), an epigenetic “reader” and transcriptional
[63]
co-activator for various stress-responding genes In contrast, Brd2 (a paralog of Brd4) was unaffected,
providing another example of signaling specificity in this system. Brd4-regulated expression of stress
[67]
proteins such as E-selectin and IL-8 was first demonstrated for lung cancer cells , but our studies were the
first to link Brd4 to iNOS expression in glioblastoma cells [59,63] . Looking at other events upstream of iNOS
[59]
transcription, Fahey et al. found that p65-acK310 formation in photostressed U87 cells was dependent
on phosphorylation-activation of PI3K. This stimulated phosphorylation-activation of protein kinase B
(Akt) which, in turn, depended on activation of phosphoinositide-dependent kinase-1 (PDK1). PI3K/Akt-
mediated signaling is known to play a central role in cancer cell survival and proliferation . A specific
[67]
PI3K inhibitor (LY294002) prevented p300 activation as well as iNOS upregulation after an ALA/light
challenge, thereby linking the iNOS response to upstream events set in motion by photodynamic stress.
Evidence for another key upstream event was also obtained, viz. oxidative inactivation of tumor suppressor
PTEN, which would have fostered PI3K/Akt activation via elevation of phosphatidylinositol triphosphate
[59]
(PIP ) level . Taken together, the above findings, which are depicted schematically in Figure 5, reveal a
3
well-coordinated stress signaling network leading ultimately to iNOS/NO induction and a pro-survival/
expansion outcome. Other pro-survival effectors, e.g., COX-2, Survivin, and S100A4, are upregulated by
[56]
photostress similar to iNOS/NO , but it is not yet clear whether this occurs independently of NO or
results from downstream signaling by NO [19,20] .

How ALA/light-induced NO can elicit photokilling resistance or greater aggressiveness of surviving cells
is a question of ongoing interest. Since NO does not scavenge O 2 [68] , this has been ruled out as a possible
1
cytoprotective mechanism, leaving open the possibility of downstream species scavenging. Studies by
Niziolek et al. [68,69] revealed that NO from the chemical donor spermine-NONOate (SPNO) could suppress
1
PpIX-sensitized ( O -initiated) lipid peroxidation in model membranes and also breast cancer cells. In
2
the latter case, ALA-induced PpIX was allowed to diffuse from mitochondria to plasma membrane before
cell irradiation in the absence vs. presence of SPNO. Irradiated cells died mainly by membrane-breaching
necrosis and NO protected against this by acting as a chain-breaking antioxidant, as was observed
[70]
previously by Rubbo et al. , using a non-photodynamic model system. Niziolek et al. [68,69] deduced that,
•
•
in their system, NO acted by intercepting chain-carrying lipid-derived radicals, i.e., LOO /LO , thereby
protecting cells against necrosis due to free radical-mediated membrane damage. There is no evidence
yet as to whether endogenous NO can act similarly on peroxidation of mitochondrial membrane lipids
sensitized by ALA-induced PpIX. Such peroxidation is highly likely, given that PpIX accumulates initially
in mitochondrial membranes [39,40] . Whereas chain breaking by NO occurs via an irreversible chemical