Page 49 - Read Online

Girotti et al. J Cancer Metastasis Treat 2020;6:52 I http://dx.doi.org/10.20517/2394-4722.2020.107 Page 11 of 15

i.e., bystanders. Such a phenomenon is well documented for cancer cells exposed to ionizing radiation
(e.g., X-rays and γ-rays), and various signaling mediators have been described, including NO [77,78] . To
determine whether bystander effects might also apply to PDT, Bazak et al. [79,80] developed a novel approach
involving impermeable silicone rings to initially separate targeted cells (ALA/light-treated, outside rings)
from non-targeted bystanders (light-only, inside rings) on a large culture dish. At some interval (e.g., 2 h)
2
after a given light fluence (e.g., 1 J/cm ) from an LED source, rings are removed and responses in both cell
compartments are monitored during subsequent dark incubation, e.g., iNOS/NO levels and proliferation/
migration rates. Initial experiments with human prostate carcinoma PC3 cells revealed not only an
expected boost in iNOS/NO level and growth/migration rate of targeted cells, but similar responses in
[79]
non-stressed bystander cells . Although the latter responses were more moderate, they were inhibited by
1400W, cPTIO, or knockdown of targeted cell iNOS, implying that NO produced by targeted cell iNOS was
responsible for the bystander effects. Use of a NO fluorescence probe (DAF-FM-DA) provided more direct
[79]
evidence for this . Conditioned medium from targeted cells did not induce bystander effects, suggesting
that short-lived, continuously generated NO was solely responsible. In addition to iNOS, several other pro-
[79]
tumor effectors were upregulated in PC3 bystanders, including Akt, ERK1/2, and COX-2 . In more recent
studies, similar NO-mediated bystander effects were observed using glioblastoma U87 cells, and they were
compared with those obtained with prostate PC3, breast MDA-MB-231, and melanoma BLM cells. After
ALA treatment, irradiation conditions were adjusted to produce the same cell kill for all four types (~25%),
thus allowing clear conclusions to be made about NO-elicited resistance. Under these conditions, bystander
proliferation and migration rates increased with extent of iNOS upregulation in surviving targeted cells
[80]
in the following order: BLM < U87 < MDA-MB-231 < PC3 . Thus, targeted cells with the greatest iNOS/
NO induction after an ALA/hν challenge elicited the greatest increases in bystander aggressiveness. These
findings suggest that a NO-based “relay” process is set in motion by photodynamic stress. In this process,
NO overproduced by targeted cells (e.g., U87 or U251) diffuses to non-stressed bystanders and induces
iNOS/NO there, thus beginning a NO “feed-forward” process that propagates through the bystander
population. Whereas photodynamic stress activates NF-κB and thence iNOS transcription in targeted
cells, the transcription factor responsible for NO-initiated iNOS induction in bystander cells has not yet
been defined. If occurring in an actual tumor, e.g., GBM, after a PDT challenge, NO-mediated bystander
effects might stimulate tumor growth and metastatic expansion. While this unfortunate possibility is well
[78]
recognized in connection with therapeutic ionizing radiation , it is still not so with regard to PDT for any
solid malignancies, including glioblastomas. As discussed in the next section, these negative effects of NO
from targeted cells could be attenuated by pharmacologic interventions aimed at either inhibiting iNOS
enzymatic activity or iNOS transcription. This would be expected to increase the overall anti-tumor efficacy
of PDT at the clinical level.

PHARMACOLOGIC MITIGATION OF NITRIC OXIDE’S ANTI-PDT EFFECTS
Although not yet tested in the clinic, it is likely, based on evidence presented above, that inhibiting iNOS
activity or expression would significantly improve PDT outcomes against glioblastoma and other solid
tumors. At least two iNOS activity inhibitors, L-NIL and GW274150, have already been tested in clinical
trials, but these were unrelated to cancer or PDT [81,82] . Instead, both agents were tested for relieving
asthmatic inflammation and, importantly, neither one had any negative side effects. As indicated above,
[58]
GW274140 significantly improved PDT efficacy in a human breast tumor xenograft model , suggesting
that this inhibitor would be a good test adjuvant for clinical PDT against gliomas and other solid tumors.
As already discussed, iNOS transcription in glioblastoma cells is regulated by NF-κB subunit p65, which is
[59]
activated by p300-catalyzed acetylation of lysine-310 . Knowing this and that (1) Brd4 is a necessary co-
activator of iNOS transcription; (2) Brd4 is increasingly upregulated by photostress; (3) p65 is increasingly
K310-acetylated by photostress; and (4) that the latter promotes Brd4 interaction with p65 [59,63] , we asked
how the latter response might be suppressed in order to reduce iNOS upregulation in ALA/light-challenged
glioblastoma cells. Bromo- and extra-terminal domain (BET) proteins act as epigenetic “readers” of
acetylated lysine residues on histones and transcription factors, thereby co-regulating gene transcription at