Also available as PDF (figures and tables appear in the PDF).
The serine-threonine kinase CK2, which targets over 300 cellular proteins, is over expressed in all cancers, presumably reflecting its ability to promote proliferation, spread, and survival through a wide range of complementary mechanisms. Via an activating phosphorylation of Cdc373, a co-chaperone which partners with Hsp90, CK2 prolongs the half-life of protein kinases that promote proliferation and survival in many cancers, including Akt, Src, EGFR, Raf-1, and several cyclin-dependent kinases. CK2 works in other ways to boost the activity of signaling pathways that promote cancer aggressiveness and chemoresistance, including those driven by Akt, NF-kappaB, hypoxia-inducible factor-1, beta-catenin, STAT3, hedgehog, Notch1, and the androgen receptor; it also promotes the epidermal-mesenchymal transition and aids efficiency of DNA repair. Several potent and relatively specific inhibitors of CK2 are now being evaluated as potential cancer drugs; CX-4945 has shown impressive activity in cell culture studies and xenograft models, and is now entering clinical trials. Moreover, it has long been recognized that the natural flavone apigenin can inhibit CK2, with a Ki near 1 micromolar; more recent work indicates that a range of flavones and flavonols, characterized by a planar structure and hydroxylations at the 7 and 4’ positions - including apigenin, luteolin, keampferol, fisetin, quercetin, and myricetin - can inhibit CK2 with Ki s in the sub-micromolar range. This finding is particularly intriguing in light of the numerous studies demonstrating that each of these agents can inhibit the growth of cancer cells lines in vitro and of human xenografts in nude mice. These studies attribute the cancer-retardant efficacy of flavones/flavonols to impacts on a bewildering array of cellular targets, including those whose activities are boosted by CK2; it is reasonable to suspect that, at least in physiologically achievable concentrations, these agents may be achieving these effects primarily via CK2 inhibition. Inefficient absorption and rapid conjugation limit the bioefficacy of orally administered flavonoids; however, the increased extracellular beta-glucuronidase of many tumors may give tumors privileged access to glucuronidated flavonoids, and nanopartical technology can improve the bioavailability of these agents. Enzymatically modified isoquercitrin has particular promise as a delivery vehicle for quercetin. Hence, it may be worthwhile to explore the clinical potential of flavones/flavonols as CK2 inhibitors for cancer therapy.
Keywords: Cancer; CK2; CX-4945; Hsp90; Flavones; Flavonols; Xenografts; Betaglucuronidase; HDA6; Sulforaphane; Beta- glucuronidase
Copyright ©2015 McCarty MF. This is an open access article distributed un- CK2 IS OVER-EXPRESSED AND UP-REGULATES PROLIFERATION, SPREAD, AND SURVIVAL der the Creative Commons Attribu- IN CANCER tion 4.0 International License (CC BY 4.0), which permits unrestricted CK2, a serine-threonine kinase once known as casein kinase 2, is a ubiquitously ex- use, distribution, and reproduction in any medium, provided the origi- pressed tetramer comprised of two catalytic subunits - α and/or α’ - and two regulatory β sub- nal work is properly cited. units that direct it to specific targets. CK2 is capable of phosphorylating a huge range of cellular proteins; over 300 physiological targets have been documented NF-kappaB to date (though ironically casein is not one of them!).1 Its level of expression and its sub-cellular localization determine its activ- Numerous studies show that CK2 inhibition suppresses ity, as post-translational modifications or allosteric interactions NF-kappaB activity in cancer cell lines, whereas overexpression 12-27 are thought to have little impact in that regard; moreover, no of this kinase boosts NF-kappaB activity. CK2 promotes gain-of-function mutants of this kinase are known. degradation of IkappaB; this can reflect an activating phos- phorylation of IKKbeta, as well as a direct phosphorylation of Virtually all cancer cell lines studied to date overex- IkappaB that renders it more sensitive to proteolytic cleavage 13,14,19,24 press CK2 protein, relative to its expression in normal tissues of by calpain. CK2 activity also has been reported to some- origin; moreover, they tend to route a higher proportion of this how boost the expression of IKK-i/IKKepsilon, an alternative protein to the cell nucleus.2 This is not likely to be accidental, IkappaB kinase complex capable of promoting IkappaB degra- as high CK2 activity works in a bewildering number of comple- dation.17 And the transcriptional activity of p65 is enhanced by a mentary ways to promote cellular proliferation and spread, while phosphorylation of Ser529 conferred by CK2.20 suppressing apoptosis. Hence, cancer cells which overexpress CK2 will tend to be selected for.
Hypoxia-inducible factor-1 (HIF-1) CK2 enhances the transcriptional activity of HIF-1, CK2 MODULATES A PLETHORA OF SIGNALING PATHWAYS even though it doesn’t increase the protein expression or nuclear 28-30 binding of this factor. Some evidence suggests that this re- One of CK2’s most intriguing and ramified effects is to 3,4 flects a reduction of p53 levels; nuclear p53 somehow antago- phosphorylate, and thereby activate, the co-chaperone Cdc37. nizes the transcriptional activity of HIF-1. CK2’s impact on Activated Cdc37 interacts with Hsp90 to provide chaperoning p53 level, in turn, may reflect phosphorylations of MDM2 that activity for a broad range of protein kinases, many of which play enhance its ability to promote proteasomal degradation of p53. a role in promoting cell proliferation and survival. These include Akt, Src, EGFR, PDGFR, Raf-1, IKK, RIP1, Cdc2, Cdk2, Cdk4, Beta-Catenin and Cdk6. This chaperoning activity tends to slow the prote- olytic degradation of these kinases, prolonging their effective Many studies show that CK2 inhibition decreases Wnt- half-lives; this activity is particularly crucial for the survival of 22,31-39 beta-catenin signaling. Down-regulation of Akt, which certain mutant constitutively active forms of these kinases of- stabilizes beta-catenin through inhibition of glycogen synthase ten found in cancers. To date, CK2 is the only upstream kinase kinase-3 and also via a direct phosphorylation on Ser552, evi- known to confer activation on Cdc37 - for which reason assess- 37,38 dently can contribute to this effect. However, CK2 also ment of Cdc37 phosphorylation at Ser13 has been proposed as a 5 phosphorylates beta-catenin directly on Thr393, an effect which strategy for determining CK2 activity in vivo.
likewise prolongs the half-life and promotes the transcriptional 32,33 activity of this factor.
But CK2 works in a number of additional ways to boost the activity of signaling pathways that make cancer more ag- STAT3 gressive and harder to kill: There are several reports that inhibition of CK2 sup- Akt presses STAT3 phosphorylation and activation in cancer cell 40-42 lines. The basis of this effect is not yet clear. In some cell While CK2 boosts Akt expression via Hsp90-cdc37- lines, suppression of IL-6 expression may contribute to this ef- mediated stabilization, it can also work in various complemen- fect.
tary ways to increase the phosphorylation and activation of this key kinase, which promotes cellular proliferation while acting Hedgehog in a number of ways to inhibit apoptosis. CK2 phosphorylates Akt directly at Ser129; this up-regulates the activation of Akt In human lung cancer cells, CK2 activity has been mediated by PDK1 and mTORC2, and facilitates its association shown to boost the mRNA and protein expression of Gli1, and 6,7 with Hsp90.6,7 And CK2 inhibits phosphatase activities that tar- to enhance the half-life and transcriptional activity of this key get Akt; it phosphorylates and thereby reduces the activity of the mediator of hedgehog signaling. CK2 can directly phosphory- crucial cancer suppressor PTEN, and also promotes proteasomal late Gli1, and it has been suggested that this may be responsible degradation of PML, a protein which is an obligate component for the positive impact of CK2 on hedgehog signaling. of a nuclear complex that dephosphorylates Akt within the 8-11 nucleus.8-11 CK2 also has the potential to work upstream from Notch1 Akt, enhancing its activation by up-regulating certain tyrosine kinase signaling pathways.
In human lung cancer cell lines expressing Notch1, in- hibition of CK2 activity suppresses Notch1-driven transcription, this agent was employed as a relatively specific inhibitor of CK2 44 64 whereas forced overexpression of CK2 has the opposite effect.44 in cell culture studies, with a Ki near 1 µM.64 There are indeed This may reflect the fact that CK2 activity increases the half-life a number of studies, both in cancer cell culture and in mouse of Notch1 protein.
xenograft models, showing that apigenin can exert cancer-retar- dant and chemo-potentiating effects. In xenograft models, api- Androgen Receptor genin has shown activity whether administered parenterally or 65-81 orally, alone or as an adjuvant to chemotherapy. Intriguingly, CK2 inhibitors suppress androgen receptor-mediated many of the effects of apigenin on signaling pathways reported transcription in prostate cancer cell lines, at least in part by block- in cell culture or xenograft studies are parallel to those of CK2 45-47 4,74,82-86 ing androgen-induced nuclear translocation of the receptor. inhibition, including down-regulated activity of Akt, HIF- 65,67,87-91 4,17,21,28,92-95 4,94,96 97,98 The direct target of CK2 in this effect has not been identified. 1, NF-kappaB, STAT3, beta-catenin, 99 100,101 4 94,102-109 Gli1, AR, and Cdc37, and up-regulated p53. In- DNA Repair deed, Zhao and colleagues have recently proposed that inhibi- tion of CK2 is a key mediator of apigenin’s anti-cancer activity CK2-mediated phosphorylations of XRCC1 and in multiple myeloma cells.4 A survey of the burgeoning cancer MDC1, nuclear proteins which play a key role in the repair of research literature involving apigenin - 476 citations on Pubmed DNA single-strand and double-strand breaks, respectively, are at present - reveals apigenin can influence a truly dizzying array 48-53 required for their proper activity. Hence, inhibition of CK2 of molecular targets in cancer cells; it is reasonable to suspect can boost the killing activity of DNA-damaging cytotoxins not that, rather than directly inhibiting dozens of separate targets, only by up-regulating mechanisms of apoptosis, but also by im- it must be influencing one or more signaling factors that have peding the efficiency of DNA repair.
a remarkably broad impact on the molecular biology of cancer cells. CK2 may be the crucial target in this regard. However, Epidermal-Mesenchymal Transition none of the studies in which apigenin has been administered in cancer-retardant doses to xenograft-bearing mice have assessed Studies with CK2 inhibitors demonstrate that CK2 ac- the impact of apigenin on tumor CK2 activity. A study assess- tivity can promote the epidermal-mesenchymal transition neces- ing this - perhaps by measuring Ser13 phosphorylation of Cdc37 sary for invasive behavior by boosting expression of vimentin, in tumors - would be worthwhile; and it would also be intrigu- 54-58 snail, and smad2/3, while suppressing that of E-cadherin. For ing to see whether apigenin administration has any significant some reason this effect is most prominent in cancer cells which additional impact on cancer growth in animals that are already overexpress CK2 alpha catalytic subunits, relative to CK2 beta receiving potent doses of CX-4945; if CK2 is apigenin’s key 59,60 regulatory subunits.
target, little additional benefit might be seen. NEW DRUGS FOR INHIBITION OF CK2 - CX-4945 Although apigenin is considered the prototype flavone inhibitor of CK2, recent studies show that other naturally-occur- These considerations make it abundantly clear that well ing flavones and flavonols have similar or slightly more potent tolerated and effective pharmaceutical inhibitors of CK2 may inhibitory activity. Working in vitro with human recombinant have a bright future in oncology - both as agents for slowing can- CK2, Lolli and colleagues have recently reported that apigenin, cer growth and spread, and as adjuvants to chemo- or radiother- luteolin, kaempferol, fisetin, quercetin, and myricetin can inhibit apy. Some pharmaceutical companies are moving aggressively CK2 with Kis of 0.8, 0.5, 0.4, 0.35, 0.55, and 0.92 µM, respec- to evaluate the potential of this approach, and the highly potent tively.110 This inhibition is competitive with respect to the phos- and orally active CK2 inhibitor CX-4945 has shown impressive phodonor substrate ATP. All effective compounds are planar and anti-cancer activity in mouse xenograft models, in doses which are hydroxylated at the7 and 4’ positions. Hydroxylations at 5, 3, 61,62 the animals appear to tolerate well. Moreover, in doses that and 3’ positions do not greatly add to or detract from activity. don’t greatly retard tumor growth, CX-4945 considerably ampli- fies response of an ovarian cancer xenograft to gemcitabine and These findings may help to explain the curious fact cisplatin - though the somewhat greater weight loss in the mice that every one of these flavones or flavonols has been report- receiving combination therapy suggests that toxicity might also ed to exert anti-cancer effects, both in cancer cell cultures, be increased to a degree.63 This agent is now entering clinical and in xenografted mice. Here are citations for the xenograft 65-81 111-122 123 124-126 trials, and its progress should be followed with the greatest inter- studies: apigenin, luteolin, kaempferol, fisetin, 127-145 146 est.
quercetin, myrcetin. There are at least 53 published stud- ies in which flavones or flavonols have decreased the growth of FLAVONES/FLAVONOLS AS NATURAL INHIBITORS OF CK2 human xenografts in nude mice.
However, there are other known inhibitors of CK2, one It seems likely that, ultimately, a drug such as CX-4945 being the dietary flavone apigenin. Indeed, long before the de- will offer the most convenient and effective way to address the drugs will not be available for several years, and when available restoring its activity. Hence, type II histone deacetylase inhibi- will initially only be approved for use in a limited number of tors have the potential to complement the impact of CK2 inhibi- cancers - and will doubtless be staggeringly expensive to use for tors on the chaperoning of many pro-oncogenic kinases. More- off-label purposes. For this reason, it would be prudent to give over, it has recently emerged that sulforaphane can function as serious attention to the possibility that apigenin or related fla- an inhibitor of HDAC6 within cells;171 this phenomenon may be vones/flavonols might be clinically useful for suppressing CK2 clinically relevant, as acute ingestion of 68 g of broccoli sprouts activity in some sufficiently high dosage schedule. This might be has been reported to suppress global histone deacetylase activity assessed by pharmacokinetic studies in which a marker for CK2 in peripheral blood mononuclear cells.166 This finding may be of activity, such as phosphorylation of Cdc37 - or Thr145 phospho- particular interest, in light of the fact that HDAC6, rather like rylation of p21, employed as a marker in studies with CX-454961 CK2, works in multifarious ways to sustain malignant cellular 172,173 - is determined in leukocytes or some other accessible cell type. behavior, and is emerging as a key target for cancer therapy. The efficacy of a given agent will presumably reflect it absorb- It would be of interest to determine whether flavones/flavonols ability, the rapidity with which it is conjugated once absorbed and sulforaphane might complement each other’s efficacy in in- (glucuronidation or sulfation), and its capacity to pass through tegrative cancer therapy.
cell walls. Pharmaceutical innovations which optimize absorb- ability might make this approach more feasible.141 With respect A POTENTIAL COUNTERVAILING EFFECT - NRF2 ACTIVATION to quercetin, the approved food additive enzymatically-modified isoquercitrin (EMIQ), unlike quercetin, is highly soluble, but is Although the great majority of reports examining quer- metabolized to yield free quercetin at the intestinal brush bor- cetin’s impact on cancer, in vitro or in vivo, with or without con- der; a human pharmacokinetic study found that when equimolar current chemotherapy, conclude that quercetin has cancer sup- amounts of quercetin and EMIQ were administered orally, the pressive activity, one recent study found that, in low micromolar plasma levels of quercetin achieved were twenty-fold higher concentrations, quercetin protected a human ovarian cancer cell 147-151 with EMIQ. This agent apparently has not yet been tested in line from a range of cytotoxic drugs; concurrent quercetin ad- rodent tumor models.
ministration decreased the cancer-retardant efficacy of cispla- tin in a xenograft model. This effect was traced to quercetin’s Rapid conjugation of absorbed flavonoids limits their ability to activate nrf2 and thereby increase the expression of capacity to exert intracellular effects.152 It is therefore fortu- antioxidant enzymes, glutathione, and glutathione-dependent nate that some tumors may have privileged access to flavone/ detoxicant enzymes. The ability of phase 2 induction via nfr2 flavonol glucuronide conjugates, owing to the fact that extracel- activation to promote chemoresistance in some cancers has lular beta-glucuronidase activity tends to be elevated in tumors, been demonstrated. Hence, while a number of studies describe 153,154 135,136, particularly in their hypoxic/necrotic regions. Infiltrating a chemosensitizing effect for quercetin in cancer models 174-176 immune cells may be the chief source of this activity. Moreover, including a report that low concentrations of quercetin sen- the tendency of extracellular pH to be acidic in such regions can sitize some ovarian cancer cell lines to cisplatin - the possibility 155-157 be expected to amplify their beta-glucuronidase activity. remains that quercetin (and presumably other phase 2-inductive Many investigators have proposed or presented evidence that flavonols) may promote chemoresistance in some cancers. (The glucuronide-masked anti-cancer agents - including flavonoids - “flip side” of this observation is that quercetin has potential for 154,157-165 can be selectively activated within tumors. Hence, the protecting healthy tissues from chemotherapy drugs, as demon- 177-181 rapid glucuronidation of flavones and flavonols may not be an strated in mice. These considerations, in any case, do not insuperable obstacle to the capacity of these compounds to in- speak to quercetin’s potential utility as an adjuvant for slowing hibit CK2 in vivo. Perhaps this mechanism contributes to the cancer growth.
demonstrable efficacy of flavones/flavonols in mouse xenograft studies; co-administration of a beta-glucuronidase inhibitor EVALUATING THE HYPOTHESIS might clarify this. A corollary of this consideration, however, is that measurement of CK2 activity in healthy tissues following As noted above, EMIQ may be the most appropriate oral administration of flavones/flavonols may underestimate the agent to study in pre-clinical and clinical trials, owing to its abil- capacity of these agents to inhibit CK2 within tumors. ity to promote absorption of quercetin. In cancer xenograft mod- els, the impact of EMIQ administration on Ser13 phosphoryla- JOINT INHIBITION OF HDAC6 AND CK2 TO TARGET HSP90 tion of Cdc37 in the tumor could be determined to assess this FUNCTION agent’s ability to suppress CK2 activity in vivo. Positive results in such studies could then encourage clinical cancer trials with In light of the fact that inhibition of the chaperoning EMIQ. Rather than expecting objective response, it would be function of Hsp90-Cdc37 plays a key role in the cancer-retardant more realistic to hope that flavonol administration will slow the efficacy of CK2 inhibitors, it is pertinent to note that acetylation growth and spread of cancer, as it does in rodent models. A pla- 166-170 of Hsp90 notably reduces its chaperoning activity. The cy- cebo-controlled design might thus be required to establish clini- tosolic deacetylase HDAC6 targets these acetylations of Hsp90, cal efficacy. The extent of Ser13 phosphorylation of Cdc37 in leukocytes could be measured as a surrogate for CK2 inhibition 7. Siddiqui-Jain A, Drygin D, Streiner N, et al. CX-4945, an in the cancer - bearing in mind, however, that quercetin metabo- orally bioavailable selective inhibitor of protein kinase CK2, lites might have greater activity within inflamed tumor tissue. inhibits prosurvival and angiogenic signaling and exhibits an- titumor efficacy. Cancer Res. 2010; 70(24): 10288-10298. doi: In regard to toxicity considerations, it should be noted 10.1158/0008-5472.CAN-10-1893 that knockout of the alpha subunits of CK2 results in embryonic 182,183 lethality. However, flavonols in vivo would achieve at best 8. Shehata M, Schnabl S, Demirtas D, et al. Reconstitution only partial inhibition of CK2. In rodent studies with CX-4945, of PTEN activity by CK2 inhibitors and interference with the cancer control is noted with doses that are not overtly toxic to PI3-K/Akt cascade counteract the antiapoptotic effect of human the animals. Phase I clinical trials with this agent have not yet stromal cells in chronic lymphocytic leukemia. Blood. 2010; been reported, so it is not clear what the dose-limiting toxicities 116(14): 2513-2521. doi: 10.1182/blood-2009-10-248054 of CK2 inhibitors will be. Flavonols are of course prominent phytochemicals in natural diets. The toxicological evaluation 9. Barata JT. The impact of PTEN regulation by CK2 on PI3K-de- of EMIQ in rodents has been described by Valentova and col- pendent signaling and leukemia cell survival. Adv Enzyme Regul. leagues; 147 when fed at up to 2.5% of diet to rats for 13 weeks, 2011; 51(1): 37-49. doi: 10.1016/j.advenzreg.2010.09.012 yellowish discoloration of bones and urine was noted, and weight gain was slightly decreased at the highest doses. At 5% of diet, 10. Kang NI, Yoon HY, Kim HA, et al. Protein kinase CK2/ isoquercetin feeding to male rats was associated with significant PTEN pathway plays a key role in platelet-activating factor-me- declines in body weight, hemoglobin, triglycerides, bilirubin, diated murine anaphylactic shock. J Immunol. 2011 1; 186(11): and phosphorus, with small increases in the relative weights of 6625-6632. doi: 10.4049/jimmunol.1100007 the lungs and testes.184 EMIQ has been accorded GRAS status for use as a food additive. These considerations suggest that it 11. Chatterjee A, Chatterjee U, Ghosh MK. Activation of protein would be reasonably safe to test EMIQ in doses of several grams kinase CK2 attenuates FOXO3a functioning in a PML-depen- daily in Phase I cancer trials.
dent manner: implications in human prostate cancer. Cell Death Dis. 2013; 4: e543. doi: 10.1038/cddis.2013.63