role of cyclic amp receptor proteins in growth, differentiation, and

[CANCER RESEARCH 50, 7093-7100. November 15. 1990]

Perspectives in Cancer Research

Role of Cyclic AMP Receptor Proteins in Growth, Differentiation, and Suppressionof Malignancy: New Approaches to TherapyYoon S. Cho-Chung

Cellular Biochemistry Section, Laboratory of Tumor Immunology and Biology, \ational Cancer Institute, \IH, Bethesda, Maryland 20892

Abstract

Two isoforms of the regulatory subunits of cyclic AMP (cAMP)-dependent protein kinase that bind cAMP are inversely expressed duringontogeny and cell differentiation. These cAMP-binding receptor proteinsin harmony may regulate the growth of normal cells and their differentiation into nondividing states. Cancer cells can also be made to differentiate and stop growing when the functional balance of these cAMPreceptor proteins is restored by treatment with site-selective cAMPanalogues or by the use of an antisense oligodeoxynucleotide, suggestingnew approaches to cancer treatment.

Introduction

Cancer cells can be unlocked from their biological blockageand be forced to differentiate and cease dividing; once thishappens, the cells should eventually die. This idea is not new(1-5) and is now shared widely among the investigators incancer research. There exists great potential for clinical applications which are targeted toward noncytotoxic induction ofdifferentiation. There is a need for nontoxic biological agentsthat modulate intracellular regulatory molecules and therebyrestore normal control mechanisms in malignant cells.

1will discuss the role of cAMP' receptor proteins in ontogeny

and differentiation of normal cells and our experimental approaches which made the control and reversal of malignancypossible by the use of nontoxic biological agents. One suchbiological agent, 8-Cl-cAMP, is now in preclinical phase Istudies at the National Cancer Institute.

Two Types of cAMP Receptor Proteins

cAMP, discovered by Sutherland in 1957 as a mediator ofhormonal signals, is present in all cells and tissues, from bacteria to humans (6). The primary mediator of cAMP action ineukaryotic cells is cAMP-dependent protein kinase (7, 8). Unlike other protein kinases, cAMP-dependent protein kinase isuniquely composed of two genetically distinct catalytic (C) andregulatory (R) subunits. The activating ligand cAMP, whichbinds to the R subunit, induces conformational changes anddissociates holoenzyme R2C; into an R2-(cAMP)4 dimer andtwo free C subunits that are catalytically active (9, 10). Thereare two different classes of cAMP-dependent protein kinase,designated as type I and type II, which contain distinct Rsubunits, RI and RII, respectively, but share a common Csubunit (10). Four different regulatory subunits [RI,, (11), RI,,(12), RII,, (RII54) (13), and RII,, (RII,,) (14)] have been identifiedat the gene/mRNA level. Two distinct C subunits [C., (15) andCrf (16, 17)] have also been identified; however, preferentialcoexpression of either one of these C subunits with either the

Received 4/9/90: accepted 8/8/90.' The abbreviations used are: cAMP. cyclic AMP: 8-CI-cAMP, 8-chlorocyclic

AMP; CAP. catabolite gene activator protein; CRE. cAMP-reguIatory elements;dibutyryl cAMP. .V6.O2'-dibutyryl cyclic AMP: TGFÂ«. transforming growthfactor o; 8-Br-cAMP. 8-bromocyclic AMP.

type I or type II protein kinase R subunit has not been found(17).

The two general classes of R subunits, RI and RII, containtwo tandem cAMP-binding domains at the carboxyl terminusthe amino acid sequences of which are highly conserved, whilethe primary structure of the remaining one-third of the moleculeat the amino terminus is quite variable among the R subunits(18). RI and RII differ significantly in the amino terminus at aproteolytically sensitive hinge region that occupies the peptidesubstrate-binding site of the C subunit in the holoenzymecomplex. In this segment, RII contains the autophosphoryla-tion site (94Arg-Arg-Val-Ser"''-Val-''''Cys), whereas RI has a

high-affinity binding site for ATP at the pseudophosphorylationsite (94Arg-Arg-GIy-Ala-Ile-"Ser). Thus, the type II holoen

zyme undergoes autophosphorylation (phosphorylation of theR subunit by the C subunit) while the type I isozyme does not,and the type I holoenzyme binds ATP with high affinity whilethe type II enzyme has a low affinity for ATP (18).

A heat-stable protein inhibitor, protein kinase inhibitor (19),like RI, contains a psuedophosphorylation site, and the proteinkinase inhibitor-C subunit complex has a high-affinity bindingsite for ATP (20). This mechanism for inhibition is used byother protein kinases, e.g., myosin light chain kinase (21) andprotein kinase C (22). Therefore, it appears from an evolutionary point of view that type I protein kinase is similar to otherprotein kinases rather than to type II protein kinase. On theother hand, the RII shares extensive homology (23) with thecAMP-binding domain of bacterial CAP (24). The regulationof gene expression by cAMP in bacteria is achieved by modulating the DNA-binding potential of a single protein, CAP (24).The homologies between CAP and the RII subunit indicate thata large part of this structure has, with some modifications, beenconserved during evolution. Thus, RII subunits may have retained the bifunctionality of CAP (i.e., both cAMP and DNAbinding) (25). The cAMP-inducible eukaryotic genes have beenshown to contain CRE with a consensus palindromic sequence(26) and cAMP-dependent and sequence-selective binding ofRII to double helical DNA (CRE) has been shown (27). Accordingly, RII, either in its subunit form or in its holoenzymecomplex (type II protein kinase), may regulate gene transcription in eukaryotes as does bacterial CAP (25, 28).

Two isoforms of cAMP-dependent protein kinase in mammalian cells may thus serve distinct physiological functions.

Differential Expression of cAMP Receptor Isoforms in Ontogenyand Differentiation

The relative content of protein kinase type I and type II variesamong tissues in adult animals and also in the same tissueamong the different species of animals (29-31), although thetotal R subunit/C subunit molar ratio in all normal tissuesexamined was found to be 1:1 (31). The ratios of these isozymesin tissues also depend on physiological conditions and hormonal status (for review see Ref. 32). These findings suggest

7093

on April 13, 2018. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

TYPE 11 CAMP RECEPTOR INHIBITS TUMOR GROWTH

that protein kinase isozymes may have distinct roles in differentphysiological processes. In fact, the changes in the amounts oractivity of cAMP-dependent protein kinase isozymes have beencorrelated with highly regulated processes such as ontogenyand cell differentiation. Examples of such studies are describedhere.

In rat cerebrum the postnatal development of cAMP-depend-ent protein kinase was studied by measuring the amounts of RIand RII and C subunit activity (33). Neither RI nor RII nor Csubunit was found to show any significant change in amountduring postnatal development of rat cerebrum, indicating thepossibility that the relative distribution of protein kinases maybe established sooner, prenatally.

Two studies (34, 35) of the development of cAMP-dependentprotein kinase in mouse heart reached similar conclusions. Thetype I/type II isozyme activity ratio (or RI/RII ratio) decreasedfrom 3.0 in neonates (7- or 14-day-old) to 1.0 in adult hearts.It was noted (35) that in rodents after the third postnatal weekno more cardiac cell division occurs such that only cellularhypertrophy is responsible for further cardiac growth (36). Thisimplies that high levels of type I kinase (or RI) are associatedwith the developmental phase of cell growth.

Both total and cAMP-dependent protein kinase activity ofrat testes increased sevenfold between birth and 90 days of age(37). The increase in cAMP-dependent protein kinase was dueto type II enzyme because at birth it was mostly absent andtype I enzyme was already present at adult levels. The development of type II kinase in testes coincided with acquisition ofthe capacity for spermatogenesis and steroidogenesis.

One well-studied example of normal cellular differentiationis that of ovarian follicles shown to contain a follicle-stimulatinghormone-responsive adenylate cyclase which seems to be involved in mediation of the conversion of immature follicles toones able to ovulate, luteinize, and produce steroids (38). Astudy of the differentiation of ovarian follicle granulosa cellsfrom immature hypophysectomized rats treated with injectionsof estradiol and follicle-stimulating hormone showed a 10- to20-fold increase in the RII,, content of granulosa cells, withouta corresponding increase in the catalytic activity of cAMP-dependent protein kinase (39). On the other hand, an increasein RI or type I protein kinase with a decrease in RII or type IIkinase was shown in tissues for which steroids have a trophicrole. Three days postcastration, steroid-dependent tissues, e.g.,ventral prostate and levator ani muscle, showed a 50% decreasein type I protein kinase activity and little change in the type IIenzyme (40). An inverse relationship between estrogen receptorand RII cAMP receptor has been shown (41) in the growth andregression of hormone-dependent mammary tumors for whichestrogen has a trophic role. Thus, the growing tumors containedhigh ratios of estrogen receptor/RII, whereas in regressingtumors following ovariectomy or dibutyryl cAMP treatment,such estrogen receptor/RII ratios became inverted. These studies suggest an association among type I protein kinase levels,steroidogenesis, and growth, and an association between increased type II protein kinase and tissue differentiation.

Other researchers claimed (42) the RI subunit to be a bettermarker of differentiation because it increased steadily from fetalto adult ages and was higher in well-differentiated hepatomacells (Morris 9618A) than in those that were poorly differentiated (Yoshida AH 130). However, the RII subunit also increased with the degree of terminal differentiation and the RI/RII ratio was lower in the more differentiated cells. Differentiation of neuroblastoma x glioma hybrid cells (43) and neuroblastoma cells (44) after treatment with yV6,O2'-dibutyryl-

cAMP has also been correlated with the increase of RI freesubunit; there was no change in the amount of free C subunitor type II protein kinase holoenzyme. It may be that an elevatedlevel of RI is prerequisite to RII development, in which case anincrease of both RI and RII would be required during development or differentiation of organs in which RI is at or below itsthreshold level.

Differential expression of type I and type II protein kinase isalso cell cycle specific. In Chinese hamster ovary cells, type Ikinase activity is high in mitosis and relatively low during G,and S phases, while type II activity is low during mitosis andincreases at the G.-S phase border, again decreasing by mid- tolate S phase (45). Also, it was found that there is a selectiveactivation of type I isozyme during stimulation of mitogenesisin human lymphocytes (46). These data have implicated type Iprotein kinase as a positive effector of growth. On the otherhand, in instances in which cAMP stimulates growth, usuallyin G, of nonmalignant cells, the increase of type II proteinkinase during late G, phase was claimed to trigger DNA synthesis (47).

Several examples of changes in cAMP-dependent proteinkinase associated with differentiation induced in malignant cellsby means other than cAMP stimulus have been described.Friend erythroleukemia cells, virus-transformed murine eryth-roleukemia cells, can be stimulated by a wide variety of agentsto express several characteristics of erythroid cell differentiation. Induction of differentiation by dimethyl sulfoxide wasshown to result in a 3-fold increase in RII, a 3-fold decrease inRI, and an RII/RI ratio of 11, compared to 1.2 in control, intreated cells (48). Murine F-9 embryonal carcinoma cells, thepluripotent stem cells of malignant teratocarcinoma, respondedto retinoic acid followed by cAMP (but not the reverse order)by differentiating and becoming parietal endoderm cells (49).Because cAMP alone cannot produce this effect, it was reasonedthat the retinoic acid induced changes which altered the response to cAMP (50). The treatment of F-9 cells with retinoicacid resulted in an increase in RI, RII, and histone kinaseactivity in cytosol and a preferential increase in the amount ofRII associated with plasma membrane. Similarly, retinoic acidinhibited growth of mouse B16-F, melanoma cells and increased their total cAMP-dependent protein kinase activity,whereas it produced neither effect in MR-4 cells, a proteinkinase-deficient variant of F, (51). A selective activation of typeII isozyme was also correlated with the growth arrest inducedby calcitonin in a human breast cancer cell line (52). These dataindicate that cAMP receptors may act as intracellular effectormolecules of a common pathway for those different agents thatinduce differentiation in malignant cells.

These studies illustrate the distinct roles of cAMP receptorisoforms in the regulation of ontogeny and cell differentiationand support a hypothesis that protein kinase type II is primarilyinvolved in differentiation processes, whereas protein kinasetype I relates to cell proliferation (53, 54) (Table 1).

Disruption in the Normal Patterns of cAMP Receptor Proteinsin Malignancy

Since the changing ratio of type I versus type II protein kinaseappears to be involved in the ontogenic and differentiationprocesses as described above, it is logical to ask if there are infact any changes or abnormalities in the protein kinase ofneoplastic cells.

RI is the major or sole R subunit of protein kinase detectedin a variety of types of human cancer cell lines (55, 56). These

7094



TYPE II CAMP RECEPTOR INHIBITS TUMOR GROWTH

Table 1 Changes in cAMP receptor isoforms during ontogenic development, cell differentiation, and neoplastic transformation

cAMP receptors

Tissue, cell line Development, differentiation, dedifferentiation RI RII RI/RII Ref.

Rat cerebrum0-4 wkPostnatal

Synapses 33

Mouseheart7days â€”Â»adult14

days â€”Â»adultRat

testes0-90daysPostnatalRat

ovarian granulosacellsNeuroblastoma

cellsFrienderythroleukemiaRat

mammaryglandRatstomachBALB

3T3fibroblastsNIH/3T3fibroblastsNormal

rat kidney fibroblastsEnd

cellularproliferationOnsetcellularhypertrophySpermatogenesisSteroidogenesisDifferentiationMaturationDifferentiationDifferentiation

to hemoglobin-containingcellsDimethylbenz[iv]anthracenecarcinogenesisA'-Methyl-A/'-nitro-jV-nitrosoguanidinecarcinogenesisSV40

viraltransformationHarveymurine sarcoma virustransformationv-Ki-rasor TGFn transformationiÂ«-Â»Â«-Â»T1Ã®Ã®Ã®Ã®Ã®Ã®Ã®Ã®Â«-Â»Ã®1Â«->Â«-Â»11i

34,i

371

39Ã®43,|48Ã®69Ã®70Ã®74Ã®76Ã®

773544a f, increase; |. decrease; <-Â»,no change.

cell lines include hormone-dependent (MCF-7, T47D, and ZR-75-D) and -independent (MCF-7raâ€žMDA-MB-231, and BT-20) breast cancers; WiDr, LS-174T, and HT-29 colon carcinomas; A549 lung carcinoma; HT-1080 fibrosarcoma; A4573Ewing's sarcoma; FOG and U251 gliomas; and HL-60, K-562,

K-562myc, and Molt-4 leukemias. When these cancer cells weregrowth arrested and differentiated following treatment withsite-selective cAMP analogues (see following section), therewas a decrease in RI with an increase in RII, indicating acorrelation between the enhanced expression of RI with thegrowth of these cancer cells (55, 56).

cAMP-dependent protein kinases type I to type II have beenexamined in the neoplastically transformed BT5C glioma cellline as compared to fetal brain cells (57). The R and C subunitsof protein kinase were expressed to a similar degree in both celllines. However, the glioma cell line had a significantly higherratio (1.2) between type I and type II protein kinase than hadthe normal fetal cells (0.5). Normal and malignant osteoblastsdiffer also in their isozyme response to hormones, with arelative predominance of type I activation in malignant cellsand of type II activation in normal cells (58).

In primary breast tumors, as compared to normal breasttissues, significantly higher levels of type I/type II proteinkinase ratio (59), no change in RI/RII ratio (60), and increasein RI (61) have been reported. The ratio of type I to type IIprotein kinase in renal cell carcinomas was about twice that inrenal cortex, although the total soluble cAMP-dependent protein kinase activity was similar in both normal and malignanttissues (62). In surgical specimens of Wilms' tumor, the type I/

type II protein kinase ratio was 2 times greater than that innormal kidney and the RI/RII ratio was 0.79 for normal and2.95 for tumor (P < 0.001 ) (63). Specimens of colorectal cancer(50 patients) contained the RI, which comprise 80% of the totalcAMP-binding R subunits, with no detectable levels of RII,whereas both distant and adjacent mucosa contained detectableamounts of RII with decreased levels of RI (64).

Changes in cAMP-dependent protein kinase activity havebeen reported in human thyroid carcinomas (65), in humancerebral tumors (66), and in leukemia cells from patients (67).A comparison of RI and RII patterns in normal peripheralblood lymphocytes with those in lymphocytes from patientswith chronic lymphocytic leukemia showed that the poorlydifferentiated states of the chronic lymphocytic leukemia are

characterized by low absolute amounts of RI and RII and analtered molecular species of RII (68).

Alteration in cAMP-dependent protein kinase isozymes hasbeen found during carcinogenesis. During dimethyl-benz(a)anthracene-induced mammary carcinogenesis in rats,the transient increase in type I protein kinase and RI subunitcoincided with the initial action of carcinogen in the mammarygland (69). The incidence of gastric adenocarcinoma by N-methyl-yV'-nitro-/V-nitrosoguanidine and the trophic action of

gastrin on gastric carcinoma production was correlated with anincrease in type I protein kinase activity (70). Changes in thecAMP-binding affinity to RII but not to RI have been shown(71, 72) to correlate with the progression of urethan-inducedmouse lung tumors. Thus, RII of normal adult lung containedboth high- and low-affinity cAMP-binding sites, whereas RIIin tumors exhibited only the low affinity site, and these changesin the cAMP-binding property correlated in degree with tumorsize and extent of anaplasticity. In contrast, a decreased expression of type I protein kinase and RI subunit was found in tumorcell lines of lung epithelial origin as compared to the immortalized nontumorigenic cell lines (73).

Changes in cAMP-dependent protein kinase isozyme patterns have been commonly observed in in vitro cell transformation. SV40 viral transformation of BALB 3T3 fibroblastswas accompanied by an increase in type I protein kinase activityand RI subunit; normal BALB 3T3 cells contain only type IIkinase isozyme (74). Transformation of rat 3Y1 cells by thehighly oncogenic human adenovirus type 12 correlated with 3-to 6-fold increase of type I kinase activity and RI subunit (75).A marked increase in RI with a decrease in RII subunit wasdetected in Harvey murine sarcoma virus-transformed NIH/3T3 clone 13-3B-4 cells (76), and in NRK cells tranformed withTGFa or v-Ki-ras oncogene (77). TGFa-induced transformation of mouse mammary epithelial cells brought about a markedincrease in the RI mRNA level along with a decrease in RIIÂ«and RUÃŸmRNA levels (78).

As these studies illustrate, there is abnormal expression oftype I and type II protein kinases and their regulatory subunitsassociated with malignant transformation (Tables 1 and 2).

Restoration of Normal cAMP Receptor Patterns in the Suppression of Malignancy by Site-selective cAMP Analogues

cAMP binds to its receptor proteins, the regulatory subunitsof protein kinase, at two different binding sites, termed Site A

7095



TYPE II cAMP RECEPTOR INHIBITS TUMOR GROWTH

Table 2 cAMP receptor isoform ratios between normal and neoplastic cells

cAMP receptorTissue, cell line ratio RI/RIIRef.BT5C

gliomacellsFetalbraincellsWilms'

tumorNormal

kidneyRenalcarcinomaColorectalcarcinomaBreastcarcinomaLung

epithelial tumor celllinesMCF-7.MDA-MB-231 breast cancercellsLS-1

74T colon cancercellsHL-60,K-562. Molt-4 leukemia cells1.20.52.950.79rTÃŽ/Â«Ã•ÃŽÃŽÃŽ5763626459-617355.55,56.565689

" ], increase; J, decrease; Â«->.no change compared with respective normal

tissues for primary tumors. For cell lines, the RI/RII ratios are expressed incomparison to that in normal counterpart cell lines or growth-inhibited cells.

(Site 2) and Site B (Site 1) (79, 80). These two different bindingsites have been identified by differences in their rate of cyclicnucleotide dissociation (79, 80) and by their specificities forcyclic nucleotide analogue binding (80). cAMP analogues modified at C-6 on the adenine ring (C-6 analogues) are generallyselective for Site A, whereas analogues modified at C-2 or C-8(C-2 and C-8 analogues, respectively) are selective for Site B.Importantly, Site B-selective analogues further exhibit specificity for either type I or type II protein kinase (81, 82); cAMPanalogues with nitrogen attached at C-8 of the adenine ringexhibit higher affinity toward type I kinase, and C-8 analogueswith sulfur or halogen attached possess higher affinity for typeII enzyme. Moreover, such selectivity toward type I and type IIkinase is synergistically enhanced when these C-8 analogues(Site B selective) are combined with C-6 analogues (Site Aselective). Thus, C-8 amino analogues are optimally synergisticwith C-6 analogues for type I kinase, whereas a sulfur- orhalogen-attached C-8 analogue is most effective toward type IIkinase when combined with the C-6 analogue (81, 82).

cAMP-dependent protein kinase is usually present in tissuesas a mixture of type I and type II isozymes (29-31). In orderto identify any specific function of these protein kinase isozymes, it is essential that these kinases can be selectively modulated in intact cells. With the use of site-selective cAMPanalogues it became possible to correlate the specific effect ofcellular protein kinase isozymes with cAMP-mediated responses in intact cells (83).

It was found that site-selective cAMP analogues demonstratea major regulatory effect on growth in a broad spectrum ofhuman cancer cell lines, including breast, colon, lung, andgastric carcinomas, fibrosarcomas, gliomas, and leukemias, aswell as on growth in athymic mice of human cancer xenograftsof various cell types, including breast, colon, and lung carcinomas (55, 56, 84, 85). The effect of the analogue on growthinhibition appeared to be selective toward transformed cancercells as opposed to nontransformed cells. The analogues produced little or no growth inhibition of NIH/3T3 cells, normalrat kidney fibroblasts, normal mammary epithelial cells, andnormal peripheral blood lymphocytes (84).

The growth-inhibitory effect of the site-selective cAMP analogues was not due to the cytotoxic effect of its adenosinemetabolites as experimentally documented using an adenosineanalogue (86). Thus, the growth inhibitory effect of site-selective cAMP analogues is different from that in previous reportsthat have shown a strong cytotoxicity using some of the amino-substituted C-8 analogues (87) and cyclic nucleotides of purineanalogues (88); the cytotoxicity was mainly due to the adenosinemetabolites of their nucleotides.

The growth inhibition induced by the analogues brought

about biochemical (oncogene and transforming growth factor asuppression) and morphological changes, differentiation, andreverse transformation (56, 84). Despite the appearance ofmarkers of mature phenotype and definitive growth arrest asshown in the analogue-treated leukemic cells, the cell cyclephase between the treated and untreated cells was unmodified;namely, the treated cells were not Gn-Gi arrested (89). Thus, itappears that site-selective cAMP analogues produce growthinhibition while allowing the cells to progress through theirnormal cell cycle, albeit at a slower rate, and this may lead toeventual restoration of a balance between cell proliferation anddifferentiation in cancer cells.

The site-selective cAMP analogues were able to inhibit thegrowth of cultured cancer cell lines that are resistant to thecAMP analogues previously studied, such as dibutyryl-cAMP.This important property of site-selective analogues was reproduced by the effect of 8-Cl-cAMP on in vivo growing tumors.The growth of transplantable hormone-independent DMBA-1and metastatic NMU-2 rat mammary carcinomas that arecompletely resistant to dibutyryl-cAMP was markedly inhibitedby 8-Cl-cAMP treatment, and in the treated NMU-2 tumor-bearing animals no metastatic lesions were detected at the timeof sacrifice, while the animals with untreated control tumordisplayed metastatic lesions (56).

In contrast to the previously studied cAMP analogues, thesite-selective analogues demonstrated their growth-inhibitoryeffect at micromolar concentrations. The analogues appearedto work directly through the cAMP receptor protein by substituting for endogenous cellular cAMP. Among the site-selectiveanalogues tested, 8-Cl-cAMP, which has a high Site B selectivity for type II protein kinase (90), exhibited the most potencyin all human cancer cell lines tested (55, 56, 84). The analogueeffects, in fact, correlated with a selective modulation of twotypes of cAMP receptor protein, i.e., a marked reduction in theRI receptor with an increase in the RII receptor. Thus, site-selective cAMP analogues restore normal cAMP receptor patterns in cancer cells.

This selective modulation of the RI and RII cAMP receptorprotein is not mimicked by the cAMP analogues previouslystudied, by cAMP itself, or by agents that increase cellularcAMP level. cAMP at high levels, having no site selectivity(80), activates both type I and type II isozymes maximally andequally without discrimination (91, 92). Selective modulationof type I versus type II protein kinase isozymes clearly correlateswith the potency of growth inhibition demonstrated among thesite-selective cAMP analogues. The analogues with a lower

concentration, inducing 50% inhibition of cell proliferation,bring about a greater RII/RI cAMP receptor ratio in cancercells. Thus, 8-Br-cAMP, despite its high Site B selectivitytoward the type II isozyme, when compared with 8-Cl-cAMP,exhibits greater ICM1(50% inhibitory concentration) and bringsabout less increase in the RII/RI ratio in cancer cells. Thesynergistic effect of the C-6- and C-8-substituted analoguecombinations on growth inhibition (55, 86) and differentiation(93) of cancer cells further supports the idea that the efficacyof an analogue is dependent on its ability to selectively bind totype II protein kinase.

The affinity and activation potency of some cAMP analoguesfor purified preparations of protein kinase are shown in Table3. Among C-8-substituted (Site B-selective) analogues (79, 80),the halogen derivatives, 8-C1- and 8-Br-cAMP, which demonstrate a higher affinity binding for Site B of RII than RIreceptor, are the potent growth inhibitors (55, 90), while theamino derivatives, 8-amino- and 8-aminohexylamino-cAMP,

7096




Table 3 Affinity and activation potency of cAMP analogues for protein kinaseSite A and Site B are two distinct cAMP binding sites (79. 80) on receptor

proteins (RI and RII); the affinities of analogues for Site A and Site B areexpressed as the relative affinity, K,' = KÂ¡cAMP/K, analogue (82). The relativeactivation constant. A',' = K. cAMP/K, analogue, where A",is the concentration

of eAMP or analogue sufficient for half-maximum activation of protein kinasetype I (I) and type II (II) (82).

Relative affinity(A'RIAnalogue8-CI-cAMP8-Br-cAMP8-NH;-cAMP8-Aminohexvlamino-cAMP/V'-Benzoyl-cAMPiV'-Monobutyrvl-cAMPSite

A2.71.30.150.113.53.6SiteB2.01.03.91.60.190.093i')Relative

activationconstantRII(A.')Site

A0.0510.110.0880.0216.90.74SiteB4.66.83.00.290.0280.041I2.31.42.51.10.930.37II0.701.70.620.120.450.39I/II3.30.834.09.22.10.95

which show higher affinity Site B binding for RI than RII, arethe weak growth inhibitors (55). Among C-6-substituted (SiteA-selective) analogues (79, 80), yV6-benzoyl-cAMP, which

shows higher affinity binding for Site A of RII than RI, is morepotent than yV6-monobutyryl-cAMP, which exhibits higher af

finity Site A binding for RI than RII receptor (55).In the activation of protein kinase, 8-Cl-cAMP exhibits 2.3-

fold greater potency than cAMP toward type I protein kinasewhile showing 30% less potency than cAMP for type II proteinkinase (90). In contrast, 8-Br-cAMP, which not only exhibits ahigh-affinity Site B binding but also shows potent activationfor type II protein kinase (1.7-fold of cAMP), is less potent ingrowth inhibition than 8-Cl-cAMP (55). Thus, the efficacy ofcAMP analogues in growth inhibition is dependent on twoimportant properties: high-affinity binding (either Site A orSite B) toward RII but not RI; and low activation capacity fortype II protein kinase. This latter property implicates type IIprotein kinase holoenzyme in the growth inhibition by cAMP.

The changing levels of RI and RII cAMP receptor proteinswere reflected in changes in the mRNA levels and the rates oftranscription of these receptor genes. In LS-174T human coloncancer cells, the 8-Cl-cAMP-induced growth inhibition was

preceded by an increase in the transcription of RII,, gene witha decreased transcription of RI,, gene and with correspondingchanging levels of RII,, and RI,, proteins (90). The inhibition ofgrowth of LX-1 human lung carcinoma in athymic mice by 8-Cl-cAMP (continuous infusion, s.c.) correlated with an increasein RII,, mRNA and protein levels and with a decrease in RI,,mRNA and protein levels (85). A marked increase in RII,,mRNA along with a decrease in RI,, mRNA also preceded themegakaryocytic differentiation induced in K-562 human leukemia cells by site-selective cAMP analogues (93). The antagonistic effect of 8-Cl-cAMP toward TGFÂ«also brought aboutchanging levels of cAMP receptor mRNA levels (78).

It was an important observation that these changes in therates of transcription of RI,, and RII,, were preceded by nucleartranslocation of RII., protein, which occurred within 10 min of8-Cl-cAMP treatment, while the changes in the transcriptionof the cAMP receptor genes were not yet detected (90). Inearlier studies, nuclear translocation of RII or type II proteinkinase has been observed during normal development (94),hormone-dependent tumor regression (95), and reverse transformation (96). Since cAMP-dependent protein kinase is acytoplasmic enzyme its translocation into nucleus is an essentialprerequisite for any role it may play at the cell nucleus.

The selective nuclear translocation of RII,, but not RI,, (90)

is interpreted in light of the primary structure of RII,,, whichcontains four positively charged amino acids, ^K K R K247

(14). Based on the studies (97) on SV40 large T-antigen, thecontrolled exposure of this sequence might provide a signal forregulating the transport of the protein into the nucleus. Interestingly, despite the large structural homology between RI andRII receptor proteins, the K K R K sequence is a uniquesequence of RII that is not present in RI. It was found that thenuclear translocation of RII,, induced by 8-Cl-cAMP is 5- to10-fold greater than that induced by 8-Br- or N6,O2 -dibutyryl-

cAMP (98), indicating that the magnitude of RII,, translocationparallels the growth-inhibitory potency of cAMP analogues.

Involvement of cAMP in the control of transcription inmammalian cells is demonstrated by the discovery of a specificDNA sequence, CRE (26), which is common to promoterregions of genes for proteins the synthesis of which is regulatedby changes in cellular levels of cAMP. The possibility wasexamined that in cancer cells, site-selective cAMP analoguesincrease transcription factors that bind specifically to CRE andwhether this effect can be correlated with the ability of theanalogues to promote nuclear translocation of RII,,. It wasfound that such factors indeed increased in nuclear extractsfrom colon (LS-174T) and gastric (TMK-1) cancer and pheo-chromocytoma (PC 12) cell lines after treatment with the analogues (99). Importantly, the ability of the analogues to increasethe nuclear factors closely paralleled their ability to induce bothnuclear translocation of RII,, and growth inhibition (98, 100).These results suggest that the fundamental basis for the anti-neoplastic activity of site-selective cAMP analogues may residein the restoration of normal gene transcription in cancer cellsin which the RII,, cAMP receptor protein plays an importantrole.

Use of Antisense Oligodeoxynucleotide in the Suppression ofMalignancy

Antisense RNA sequences have been described as naturallyoccurring biological inhibitors of gene expression in both pro-karyotes (101) and eukaryotes (102), and these sequences presumably function by hybridizing to complementary mRNAsequences, resulting in hybridization arrest of translation (103).Antisense oligodeoxynucleotides are short synthetic nucleotidesequences formulated to be complementary to a specific geneor RNA message. Through the binding of these oligomers to atarget DNA or mRNA sequence, transcription or translation ofthe gene can be selectively blocked and the disease processgenerated by that gene can be halted. The cytoplasmic locationof mRNA provides a target considered to be readily accessibleto antisense oligodeoxynucleotides entering the cell; hencemuch of the work in the field has focused on RNA as a target.Currently, the use of antisense oligodeoxynucleotides providesa useful tool for exploring regulation of gene expression in vitroand in tissue culture (104).

As described in the preceding sections, the cAMP-dependentprotein kinase type I and type II ratio varies among tissues andan increased expression of protein kinase type I or RI correlateswith active cell growth, cell transformation, or early stages ofdifferentiation, suggesting that RI may be involved in ontogenicgrowth. Constitutive expression of RI may disrupt normalontogenic processes, resulting in a pathogenic outgrowth, suchas malignancy.

Antisense strategy was used to determine whether the presence of RI,, is essential for neoplastic cell growth. Exposure ofhuman HL-60 promyelocytic leukemia cells to 21-mer RIâ€ž

7097




antisense oligodeoxynucleotide brought about growth inhibition and monocytic differentiation, bypassing the effects of anexogenous cAMP analogue, with no sign of cytotoxicity (105).In cells unexposed to RI,, antisense oligomer, 8-CI-cAMP plusyV6-benzyl-cAMP induce a monocytic morphological change

characterized by a decrease in nuclear/cytoplasm ratio, abundant ruffled and vacuolated cytoplasm, and loss of nucleoli (89).Strikingly, the same morphological change is induced whencells are exposed to RI,, antisense oligodeoxynucleotide, andthe morphological change induced by the oligomer is indistinguishable from that induced by 12-O-tetradecanoylphorbol-13-acetate. Moreover, exposure of these cells to RI., antisenseoligodeoxynucleotide induces a marked increase (10- to 50-fold) in the expression of monocytic surface antigens (Leu-15and Leu-M3) along with a decrease in markers related to theimmature myelogenous cells (My9) (105). Treatment of thesecells exposed to RI,, antisense oligomer with 8-Cl-cAMP plus/Vfi-benzyl-cAMP does not further change the antigen expres

sion.The effect of RI,, antisense oligodeoxynucleotide correlates

with a decrease in RI,, receptor and an increase in RII,, receptorlevel. In control cells, treatment with 8-Cl-cAMP plus N6-benzyl-cAMP brought about a 70% reduction of RI,, with a 3-fold increase in RII,f, resulting in a 10-fold increase in the RII.(/RI,, ratio. Exposure of these cells to RI,, antisense oligodeoxynucleotide brought about marked changes in RI,, and RII,,levels; an 80% reduction in RI,, with a 5-fold increase in RII,,resulted in a 25-fold increase in the RII,,/RIâ€žcompared withthat in control cells. Thus, suppression of RI,, by its antisenseoligodeoxynucleotide results in a compensatory increase in RII,,level. Such coordinated expression of RI and RII has also beenshown in other cells (106).

This increase in RII,, may be responsible for the self-differentiation in these cells exposed to RI,, antisense oligodeoxynucleotide. HL-60 cells that were exposed to RII,, antisense oligodeoxynucleotide have been shown (107) to become refractoryto treatment with cAMP analogues and to continue to grow,indicating the essential role of RII,, in cAMP-induced differentiation. The increase in RII, mRNA or RII,, protein level hasbeen correlated with cAMP analogue-induced differentiation inK-562 chronic myelocytic leukemic cells (93) and in erythroiddifferentiation of Friend erythrocytic leukemic cells (108).

The essential role of RII,, in differentiation of HI-60 cells wasfurther demonstrated when these cells were exposed to bothRI,, and RII,, antisense oligodeoxynucleotides simultaneously(105). Cells exposed to both RI,, and RII,, antisense oligomerswere neither growth inhibited or self-differentiated nor differentiated upon cAMP analogue treatment. These results mayreflect an escape from the cAMP-dependent growth regulatorypathway. Cells are no longer cAMP dependent but survive andproliferate probably through an alternate pathway. Thus,suppression of both RI,, and RII,, gene expression led to anabnormal cellular growth regulation similar to that in mutantcell lines (109), which contain either deficient or defectiveregulatory subunits of cAMP-dependent protein kinase and areno longer sensitive to cAMP stimulus.

These results suggest that two isoforms of cAMP-bindingreceptor proteins, RI,, and RII,,, the regulatory subunits ofprotein kinase, in harmony, regulate cell growth and differentiation in HL-60 cells. The RI,, antisense oligodeoxynucleotidethat appears to restore the functional balance of these cAMPreceptor proteins led to terminal differentiation in HL-60 leu

kemia with no sign of cytotoxicity.These effects of RI,, antisense oligodeoxynucleotide are not

limited to leukemic cells. The RI,. antisense oligomer is similarly effective toward the growth control of epithelial humancancer cell lines including colon, breast, and gastric carcinomaand neuroblastoma (110). These studies with the use of anti-sense strategy provided direct evidence for the distinct roles ofcAMP receptor isoforms in cell proliferation and differentiation.

Conclusions

The following conclusions can be drawn from studies on therole of cAMP receptor proteins in the molecular control ofgrowth and differentiation in normal development, malignanttransformation, and suppression of malignancy: (a) malignancycan be suppressed by inducing differentiation with site-selectivecAMP analogues which demonstrate a high affinity for the RIIcAMP receptor or with an antisense oligodeoxynucleotide targeted against the RI cAMP receptor mRNA; (h) this suppression of malignancy is due to restoration of the normal functionalbalance of cAMP receptor isoforms, RI and RII; (e) this functional balance of these cAMP receptor proteins underlies thecontrolled expression of growth factors and cellular oncogenes;and (d) since cAMP receptor proteins are expressed in everytype of cell and tissue, this suppression of malignancy is effective toward a broad spectrum of neoplasia. This suppression ofmalignancy by restoration of normal function of intracellularregulatory molecules that induce differentiation can be of valuein cancer therapy (84).

References

1. Osgood, E. E. A unifying concept of (he etiology of the leukemius. lympho-mas, and cancers. J. Nail. Cancer Inst., 18: 155-166, 1957.

2. Pierce. G. B.. Nakanc. P. K.. and Mazurkiewicz, J. E. Natural history ofmalignant stem cells. In: \V. Nakahara. T. Ono, T. Sugimura, and H.Sugano (eds.). Differentiation and Control of Malignancy of Tumor Cells,pp. 453-472. Tokyo: University of Tokyo Press. 1974.

3. Sachs. L. Control of normal cell differentiation and the phenotypic reversionof malignancy in myeloid leukemia. Nature (Lond.). 274: 5.35-539, 1978.

4. Potter. V. R. Phenotypic diversity in experimental hcpatomas: the conceptof partially blocked ontogeny. Br. J. Cancer. 38: 1-23. 1978.

5. Cho-Chung. Y. S. On the mechanism of cyclic AMP-mediated growth arrestof solid tumors. Adv. Cyclic Nucleotide Res.. 12: 111-121. 1980.

6. Robison, G. A. The biological role of cycilc AMP: an updated reivcw. In:R. H. Kahn and \V. E. M. Lands (eds.). Proslaglandins and Cyclic AMP:Biological Actions and Clinical Applications, pp. 229-247. New York:Academic Press. 1973.

7. Kuo. J. F.. and Grcengard. P. Cyclic nuclcotide-dcpendent protein kinases.IV. Widespread occurrence of adenosine 3':5'-monophosphate-depcndent

protein kinase in various tissues and phyla of the animal kingdom. Proc.Nati. Acad. Sci. USA, 64: 1349-1355, 1969.

8. Krebs, E. G., and Beavo, J. A. Phosphorylation-dcphosphorylation of enzymes. Annu. Rev. Biochem., 48: 923-939, 1979.

9. Bramson. H. N.. Kaiser. E. T., and Mildvan. A. S. Mechanistic studies ofcAMP-dependent protein kinase actions. CRC Crii. Rev. Biochem.. /5: 93-124, 1983.

10. Beebe, S. J.. and Corbin. J. D. Cyclic nucleotide-dependem protein kinases.In: P. D. Boyer and E. G. Kerbes (eds.). The Enzymes: Control by Phos-phorylation. Part A, Vol. 17. pp. 43-111. New York: Academic Press. 1986.

11. Lee, D. G. Carmichael. D. F.. Krebs. E. G., and McKnight, G. S. Isolationof a cDN A clone for the type I regulatory subunit of bovine cAMP-dependentprotein kinase. Proc. Nati. Acad. Sci. USA, Â«ft-3608-3612. 1983.

12. Clegg, C. H., Cadd. G. G., and McKnight, G. S. Genetic characterizationof a brain-specific form of the type I regulatory subunil of cAMP-dependentprotein kinase. Proc. Nati. Acad. Sci. USA, 85: 3703-3707. 1988.

13. Scott, J. D., Glaccum, M. B., Zollcr, M. J., Uhler, M. D.. Holfman. D. M.,McKnight. G. S., and Krebs, E. G. The molecular cloning of a type IIregulatory subunit of the cAMP-dependent protein kinase from rat skeletalmuscle and mouse brain. Proc. Nati. Acad. Sci. USA, 84: 5192-5196. 1987.

14. Jahnsen, T., Hedin, L., Kidd, V. J.. Beattie. W. Gâ€žl.ohmann. S. M.,Walter, U.. Durica. J., Schulz. T. Z., SchÃ¼tz,E., Browner, M., Lawrence,C. B., Goldman. D.. Ratoosh, S. L., and Richards. J. S. Molecular cloning.cDNA structure, and regulation of the regulatory subunit of type II cAMP-dependent protein kinase from rat ovarian granulosa cells. J. Biol. Chem.,261: 12352-12361. 1986.

15. Uhler. M. D.. Carmichael. D. F.. Lee, D. C.. Chrivia, J. C.. Krebs, E. G..

7098




and McKnight, G. S. Isolation of cDNA clones coding for the catalyticsubunit of mouse cAMP-dependent protein kinase. Proc. Nati. Acad. Sci.USA, 83: 1300-1304, 1986.

16. Uhler, M. D., Chrivia, J. O. and McKnight, G. S. Evidence for a secondisoform of the catalytic subunit of cAMP-dependent protein kinase. J. Biol.Chem., 261: 15360-15363, 1986.

17. Showers, M. O., and Maurer. R. A. A cloned bovine cDNA encodes analternate form of the cataltyic subunit of cAMP-dependent protein kinase.J. Biol. Chem., 267: 16288-16291, 1986.

18. Taylor, S. S., Bubis, J.. Toner-Webb, J., Saraswat, L. D., First, E. A.,Buechler, J. A., Knighton, D. R., and Sowadski. J. cAMP-dependent proteinkinase: prototype for a family of enzymes. FASEB J., 2: 2677-2685. 1988.

19. Cheng, H-C.. van Patten, S. M., Smith. A. J., and Walsh. D. A. An activetwenty-amino-acid residue peptide derived from the inhibitor protein of thecAMP-dependent protein kinase. Biochem. J., 231: 655-661, 1985.

20. Cheng. H-C., Kemp. B. E.. Pearson, R. B., Smith, A. J., Misconi, L., vanPatten, S. M., and Walsh, D. A. A potent synthetic peptide inhibitor of thecAMP-dependent protein kinase. J. Biol. Chem., 261: 989-992, 1986.

21. Kemp. B. E., Pearson. R. B.. Guerriero, V.. Bagchi. I. C., and Means, A.R. The calmodulin binding domain of chicken smooth muscle myosin lightchain kinase contains a pseudosubstrate sequence. J. Biol. Chem., 262:2542-2548. 1987.

22. House, C.. and Kemp. B. E. Protein kinase C contains a pseudosubstrateprototype in its regulatory domain. Science (Washington DC), 238: 1726-1728. ÃŒ987.

23. Weber, I. T., Takio, K., Titani, K., and Steitz, T. A. The cAMP-bindingdomains of the regulatory subunit of cAMP-dependent protein kinase andthe catabolite gene activator protein are homologous. Proc. Nati. Acad. Sci.USA. 79:7679-7683. 1982.

24. Crombrugghe, B. D.. Busby. S.. and Bue. H. Activation of transcription bythe cyclic AMP receptor protein. In: R. F. Goldberger and K. R. Vamamoto(eds.). Biological Regulation and Development. Vol. 3B. pp. 129-167. NewYork: Plenum Publishing Corp.. 1984.

25. Nagamine, Y., and Reich. E. Gene expression and cAMP. Proc. Nati. Acad.Sci. USA, 82:4606-4610, 1985.

26. Roesler. W. J.. Vandenbark. G. R.. and Hanson. R. W. Cyclic AMP andthe induction of eukaryotic gene transcription. J. Biol. Chem., 263: 9063-9066, 1988.

27. Wu, J. C.. and Wang, J. H. Sequence-selective DNA binding to the regulatory subunit of cAMP-dependent protein kinase. J. Biol. Chem., 264:9989-9993, 1989.

28. Cho-Chung, Y. S., Clair. T., and Huffman, P. Loss of nuclear cyclic AMPbinding in cyclic AMP-unresponsive Walker 256 mammary carcinoma. J.Biol. Chem.. 252: 6349-6355. 1977.

29. Corbin. J. D., Keely, S. L.. and Park. C. R. The distribution and dissociationof cyclic adenosine 3':5'-monophosphate-dependent protein kinases in adipose, cardiac, and other tissues. J. Biol. Chem., 250: 218-225, 1975.

30. Sugden. P. H.. and Corbin. J. D. Adenosine 3':5'-cyclic monophosphate-binding proteins in bovine and rat tissues. Biochem. J., 759:423-437. 1976.

31. Hofmann. F., Bechtel, P. J., and Krebs. E. G. Concentrations of cyclicAMP-dependent protein kinase subunits in various tissues. J. Biol. Chem..252: 1441-1447, 1977.

32. Lohmann, S. M.. and Walter, U. Regulation of the cellular and subcellularconcentrations and distribution of cyclic nucleotide-dependent protein kinases. Adv. Cyclic Nucleotide Protein Phosphorylation Res.. 18: 63-117,1984.

33. Lohmann, S. M.. Walter, U., and Greengard. P. Protein kinases in developing rat brain. J. Cyclic Nucleotide Res.. 4: 445-452. 1978.

34. Haddox. M. K.. Roeske. W. R.. and Russell, D. H. Independent expressionof cardiac type I and II cyclic AMP-dependent protein kinase during murineembryogenesis and postnatal development. Biochim. Biophys. Acta, 585:527-534, 1979.

35. Malkinson, A. M.. Hogy, L.. Gharrett, A. J., and Gunderson, T. J. Onto-genetic studies of cyclic AMP-dependent protein kinase enzymes frommouse heart and other tissues. J. Exp. Zool., 205: 423-432, 1978.

36. Claycomb, W. C. Biochemical aspects of cardiac muscle differentiation. J.Bioi. Chem., 250: 3229-3235, 1975.

37. Lee, P. C.. Radloff, D., Schweppe, J. S.. and Jungmann, R. A. Testicularprotein kinases. J. Biol. Chem., 257: 914-921, 1976.

38. Wittmaack, F. M., Weber, W., and Hilz, H. Isoenzymes of cAMP-depend-ent protein kinase in developing rat liver and in malignant hepatic tissues.Eur. J. Biochem., 729: 669-674, 1983.

39. Jonassen. J. A.. Bose. K.. and Richards, J. S. Enhancement and desensiti-zation of hormone-responsive adenylate cyclase in granulosa cells of prean-tral and antral ovarian follicles: effects of estradiol and follicle-stimulatinghormone. Endocrinology, 777: 74-79, 1982.

40. Richards, J. S.. and Rolfes. A. I. Hormonal regulation of cyclic AMPbinding to specific receptor proteins in rat ovarian follicles. J. Biol. Chem..255:5481-5489, 1980.

41. Fuller, D. J. M., Byus, C. V., and Russell, D. H. Specific regulation bysteroid hormones of the amount of type I cyclic AMP-dependent proteinkinase holoenzyme. Proc. Nati. Acad. Sci. USA, 75: 223-227, 1978.

42. Cho-Chung, Y. S. Minireview: on the interaction of cyclic AMP-bindingprotein and estrogen receptor in growth control. Life Sci.. 24: 1231-1240.

43. Walter, U., Costa. M. R. C., Breakefield. X. O.. and Greengard. P. Presenceof free cyclic AMP receptor protein and regulation of its level by cyclicAMP in neuroblastoma-glioma hybrid cells. Proc. Nati. Acad. Sci. USA.76:3251-3255. 1979.

44. Prashad, N.. Rosenberg, R. N., Wischmeyer, B., Ulrich. C., and Sparkman,D. Induction of adenosine 3',5'-monophosphate binding proteins by A",O2 -dibutyryl-adenosine 3',5'-monophosphate in mouse neuroblastoma cells:analysis by two-dimensional gel electrophoresis. Biochemistry, 18: 2717-2725, 1979.

45. Costa, M., Gerner, E. W., and Russell. D. H. Cell cycle-specific activity oftype I and type II cyclic adenosine 3':5'-monophosphate-dependent proteinkinases in Chinese hamster ovary cells. J. Biol. Chem., 257: 3313-3319,1976.

46. Byus, C. V., Kumpel, G. R.. Lucas. D. O., and Russell. D. H. Type I andtype II cyclic AMP-dependent protein kinase as opposite effectors of lymphocyte mitogenesis. Nature (Lond.), 26Ã„:63-64. 1977.

47. Boynton, A. L., and Whitfield, J. F. The role of cyclic AMP in cellproliferation: a critical assessment of the evidence. Adv. Cyclic NucleotideRes., 75: 193-294. 1983.

48. Schwartz, D. A., and Rubin, C. S. Regulation of cAMP-dependent proteinkinase subunit levels in Friend erythroleukemic cells. J. Biol. Chem., 258:777-784. 1983.

49. Strickland. S.. Smith. K. K.. and Marotti, K. R. Hormonal induction ofdifferentiation in teratocarcinoma stem cells: generation of parietal endo-derm by retinoic acid and dibutyryl cAMP. Cell 27: 347-355, 1980.

50. Plet, A.. Evain, D., and Anderson. W. B. Effect of retinoic acid treatmentof F9 embryonal carcinoma cells on the activity and distribution of cyclicAMP-dependent protein kinase. J. Biol. Chem., 257: 889-893. 1982.

51. Ludwig, K. W., Lowey, B., and Niles, R. M. Retinoic acid increases cyclicAMP-dependent protein kinase activity in murine melanoma cells. J. Biol.Chem., 255: 5999-6002. 1980.

52. Ng, K. W., Livesey, S. A.. Larkins, R. G., and Martin, T. J. Calcitonineffects on growth and on selective activation of type II isoenzyme of cyclicadenosine 3':5'-monophosphate-dependent protein kinase in T47D humanbreast cancer cells. Cancer Res.. 43: 794-800. 1983.

53. Cho-Chung, Y. S. Hypothesis: cyclic AMP and its receptor protein in tumorgrowth regulation in vivo. J. Cyclic Nucleotide Res.. 6: 163-177, 1980.

54. RÃ¼ssel,D. H. Type I cyclic AMP-dependent protein kinase as a positiveeffector of growth. Adv. Cyclic Nucleotide Res., 9: 493-506, 1978.

55. Katsaros, D., Tortora. G.. Tagliaferri. P.. Clair. T.. Ally. S., Neckers. L.,Robins. R. K., and Cho-Chung, Y. S. Site-selective cyclic AMP analogsprovide a new approach in the control of cancer cell growth. FEBS Lett-,225:97-103, 1987.

56. Cho-Chung, Y. S.. Clair. T.. Tagliaferri, P., Ally. S.. Katsaros. D., Tortora,G., Neckers, L.. Avery, T. L., Crabtree, G. W., and Robins. R. K. Basicscience review: site-selective cyclic AMP analogs as new biological tools ingrowth control, differentiation and proto-oncogene regulation. Cancer Invest., 7: 161-177. 1989.

57. Ekanger, R., 0greid. D., Evjen, O., Vintermyr. O., Laerum, O. D., andD0skeland, S. O. Characterization of cyclic adenosine 3':5'-monophos-phate-dependent protein kinase isozymes in normal and neoplastic fetal ratbrain cells. Cancer Res., 45: 2578-2583, 1985.

58. Livesey, S. A.. Kemp, B. E., Re, C. A., Partridge, N. C., and Martin, T. J.Selective hormonal activation of cyclic AMP-dependent protein kinaseisoenzymes in normal and malignant osteoblasts. J. Biol. Chem., 257:14983-14987, 1982.

59. Eppenberger, U.. Biedermann, K., Handschin, J. C., Fabbro, D., Kung. W.,Huber. P. R.. and Roos, W. Cyclic AMP-dependent protein kinase type Iand type II and cyclic AMP binding in human mammary tumours. Adv.Cyclic Nucleotide Res.. 72: 123-128, 1980.

60. Handschin. J. C., Handloser, K., Takahashi, A., and Eppenberger, U. Cyclicadenosine 3':5'-monophosphate receptor proteins in dysplastic and neo

plastic human breast tissue cytosol and their inverse relationship withestrogen receptors. Cancer Res., 43: 2947-2954. 1983.

61. Weber, W.. Schwoch, G., SchrÃ¶der,H., and Hilz, H. Analysis of cAMP-dependent protein kinases by immunotitration: multiple formsâ€”multiplefunctions? Cold Spring Harbor Conf. Cell Proliferation, 8: 125-140, 1981.

62. Fossberg. T. M.. D0skeland. S. O.. and Ueland, P. M. Protein kinases inhuman renal cell carcinoma and renal cortex. A comparison of isozymedistribution and of responsiveness to adenosine 3':5'-cyclic monophosphate.Arch. Biochem. Biophys.. 189: 372-381, 1978.

63. Nakajima, F., Imashuku, S., Wilimas, J., Champion, J. E.. and Green, A.A. Distribution and properties of type 1 and type II binding proteins in thecyclic adenosine 3:5'-monophosphate-dependent protein kinase system inWilms' tumor. Cancer Res., 44: 5182-5187, 1984.

64. Bradbury, A. W., Miller. W. R., Clair, T.. Yokozaki, H.. and Cho-Chung,Y. S. Overexpressed type I regulatory subunit (RI) of cAMP-dependentprotein kinase (PKA) as tumor marker in colorectal cancer. Proc. Am.Assoc. Cancer Res., 31: 172, 1990.

65. Sand, G., Jortay, A., Pochet. R., and Dumont, J. E. Adenylate cyclase andprotein phosphokinase activities in human thyroid. Comparison of normalglands, hyperfunctional nodules and carcinomas. Eur. J. Cancer, 72: 447-453, 1976.

66. Trabucchi, M.. Canal, N.. and Frattola, L. Cyclic nucleotides in humancerebral tumors: role of the protein kinase system. Adv. Cyclic NucleotideProtein Phosphorylation Res., 77:671-676, 1984.

67. Pena, J. M.. Itarte, E., Domingo, A., and Cusso. R. Cyclic adenosine 3':5'-monophosphate-dependent and -independent protein kinases in human leu-kemic cells. Cancer Res.. 43: 1172-1175, 1983.

68. Weber. W., Schwoch, G., Wielckens, K., Gartemann, A., and Hilz, H.cAMP receptor proteins and protein kinases in human lymphocytes: fun-

7099




damental alterations in chronic lymphocytic leukemia cells. Eur. J.Biochem., 120: 585-592, 1981.

69. Cho-Chung, Y. S., Clair, T.. and Shepheard. C. Anticarcinogenic effect ofA",O2 -dibutyryl cyclic adenosine 3':5'-monophosphate on 7,12-dimethyl-

benz(a)anthracene mammary tumor induction in the rat and its relationshipto cyclic adenosine 3':5'-monophosphate metabolism and protein kinase.Cancer Res.. 43: 2736-2740, 1983.

70. Yasui, W., and Tahara, E. Effect of gastrin on gastric mucosal cyclicadenosine 3':5'-monophosphate-dependent protein kinase activity in ratstomach carcinogenesis induced by A'-methyl-A"-nitro-iV-nitrosoguanidine.Cancer Res., 45:4763-4767, 1985.

/I. Malkinson, A. M., and Butley, M. S. Alterations in cyclic adenosine 3':5'-monophosphate-dependent protein kinases during normal and neoplasticlung development. Cancer Res., 41: 1334-1341, 1981.

72. Butley, M. S., Stoner, G. D., Beer, D. G., Beer, D. S., Mason. R. J.. andMalkinson, A. M. Changes in cyclic adenosine 3':5'-monophosphate-de-pendent protein kinases during the progression of urethan-induced mouselung tumors. Cancer Res., 45: 3677-3685, 1985.

73. Lange-Carter, C. A., Fossli, T., Jahnsen, T., and Malkinson, A. M. Decreased expression of the type 1 isoenzyme of cAMP-dependent proteinkinase in tumor cell lines of lung epithelial origin. J. Biol. Chem.. 74:7814-7818. 1990.

74. Wehner, J. M., Malkinson, A. M., Wiser, M. F., and Sheppard, J. R. CyclicAMP-dependent protein kinases from Balb 3T3 cells and other 3T3 derivedlines. J. Cell. Physio!.. 108: 175-184, 1981.

75. Ledinko. N., and Chan, 1-J. D. Increase in type 1 cyclic adenosine 3':5'-monophosphate-dependent protein kinase activity and specific accumulationof type I regulatory subunits in adenovirus type 12-transformed cells. CancerRes.. 44: 2622-2627, 1984.

76. Tagliaferri, P.. Katsaros. D.. Clair, T., Neckers, L.. Robins. R. K., and Cho-Chung, Y. S. Reverse transformation of Harvey murine sarcoma virus-transformed NIH/3T3 cells by site-selective cyclic AMP analogs. J. Biol.Chem.. 265:409-416, 1988.

77. Tortora, G., Ciardiello, F., Ally, S., Clair, T., Salomon, D. S., and Cho-Chung, Y. S. Site-selective 8-chloroadenosine 3'.5'-cyclic monophosphateinhibits transformation and transforming growth factor a production in Ki-rai-transformed rat fibroblasts. FEBS Lett., 242: 363-367, 1989.

78. Ciardiello. F., Tortora, G., Kim, N., Clair. T., Ally, S., Salomon, D. S., andCho-Chung, Y. S. 8-Chloro-cAMP inhibits transforming growth factor ntransformation of mammary epithelial cells by restoration of the normalmRNA patterns for cAMP-dependent protein kinase regulatory subunitisoforms which show disruption upon transformation. J. Biol. Chem., 265:1016-1020, 1990.

79. D0skeland, S. O. Evidence that rabbit muscle protein kinase has twokinetically distinct binding sites for adenosine 3',5'-cyclic monophosphate.Biochem. Biophys. Res. Commun., SÃ¬:542-549. 1978.

80. Rannels, S. R., and Corbin, J. D. Two different intrachain cAMP bindingsites of cAMP-dependent protein kinases. J. Biol. Chem., 255: 7085-7088.1980.

81. Robinson-Steiner. A. M.. and Corbin. J. D. Probable involvement of bothintrachain cAMP binding sites in activation of protein kinase. J. Biol.Chem.. 25Â«:1032-1040, 1983.

82. 0greid, D.. Ekanger. R., Suva, R. H., Miller. J. P.. Sturm. P.. Corbin, J.D., and Daskeland. S. O. Activation of protein kinase isozymes by cyclicnucleotide analogs used singly or in combination. Eur. J. Biochem.. 750:219-227. 1985.

83. Beebe, S. J., Holloway, R.. Rannels, S. R., and Corbin, J. D. Two classesof cAMP analogs which are selective for the two different cAMP-bindingsites of type II protein kinase demonstrate synergism when added togetherto intact adipocytes. J. Biol. Chem., 259: 3539-3547, 1984.

84. Cho-Chung, Y. S. Commentary: site-selective 8-chloro-cyclic adenosine3',5'-monophosphate as a biologic modulator of cancer: restoration ofnormal control mechanisms. J. Nati. Cancer Inst., 81: 982-987, 1989.

85. Ally, S., Clair. T.. Katsaros. D.. Tortora, G.. Yokozaki, H.. Finch, R. A.,Avery, T. L.. and Cho-Chung, Y. S. Inhibition of growth and modulationof gene expression in human lung carcinoma in athymic mice by site-selective 8-Cl-cyclic adenosine monophosphate. Cancer Res., 49: 5650-5655, 1989.

86. Tagliaferri. P.. Katsaros. D., Clair, T., Ally, S., Tortora, G., Neckers, L.,Rubalcava. B.. Parandoosh, Z.. Chang. Y-A.. Revankar, G. R., Crabtree. G.W., Robins. R. K., and Cho-Chung. Y. S. Synergistic inhibition of growthof breast and colon human cancer cell lines by site-selective cyclic AMPanalogues. Cancer Res., 48: 1642-1650, 1988.

87. Koontz, J. W., and Wicks. W. D. Cytotoxic effects of two novel 8-substitutedcyclic nucleotide derivatives in cultured rat hepatoma cells. Mol. Pharma-col., 18: 65-71. 1980.

88. Koontz. J. W., and Wicks. W. D. Comparison of the effects of 6-thio- and6-methylthio-purine ribonucleoside cyclic monophosphates with their cor

responding nucleosides on the growth of rat hepatoma cells. Cancer Res.,37: 651-657, 1977.

89. Tortora, G., Tagliaferri, P., Clair, T., Colamonici, O., Neckers, L. M.,Robins, R. K., and Cho-Chung. Y. S. Site-selective cAMP analogs atmicromolar concentrations induce growth arrest and differentiation of acutepromyelocytic, chronic myelocytic, and acute lymphocytic human leukemiacell lines. Blood, 71: 230-233. 1988.

90. Ally. S.. Tortora. G., Clair, T., Cricco, D., Merlo, G., Katsaros, D., 0greid.D., D0skeland, S. O., Jahnsen, T., and Cho-Chung, Y. S. SÃ©lectivemodulation of protein kinase isozymes by the site-selective analog 8-chloroadenosine 3'.5'-cyclic monophosphate provides a biological means for controlof human colon cancer cell growth. Proc. Nati. Acad. Sci. USA, 85: 6319-6322, 1988.

91. Krebs. E. G. Protein kinase. Curr. Top. Cell. Regul., 5: 99-133, 1972.92. Hofmann. F.. Beavo. J. A., Bechtel, P. J.. and Krebs. E. G. Comparison of

adenosine 3'.5'-monophosphate-dependent protein kinase from rabbit skeletal and bovine heart muscle. J. Biol. Chem., 250: 7795-7801, 1975.

93. Tortora. G.. Clair, T., Katsaros, D.. Ally, S.. Colamonici. O., Neckers, L.M., Tagliaferri, P. Jahnsen, T., Robins. R. K.. and Cho-Chung, Y. S.Induction of megakaryocytic differentiation and modulation of proteinkinase gene expression by site-selective cAMP analogs in K-562 humanleukemic cells. Proc. Nati. Acad. Sci. USA, 86: 2849-2852, 1989.

94. Jungmann, R. A., and Kranias. E. G. Nuclear phosphoprotein kinases andthe regulation of gene transcription. Int. J. Biochem.. 8: 819-830, 1977.

95. Cho-Chung, Y. S., Archibald, D., and Clair, T. Cyclic AMP receptor triggersnuclear protein phosphorylation in a hormone-dependent mammary tumorcell-free system. Science (Washington DC), 205: 1390-1392. 1979.

96. Nesterova, M. V., and Severin. E. S. Role of cAMP-binding proteins in theregulation of cell activity. Life Chem. Rep.. 4: 391-465. 1987.

97. Kalderon. D.. Roberts. B. L.. Richardson. W. D.. and Smith, A. E. A shortamino acid sequence able to specify nuclear location. Cell, 39: 499-509,1984.

98. Yokozaki. H., Clair, T., Mednieks, M.. Tortora, G., Ally, S., Merlo, G.,and Cho-Chung. Y. S. Nuclear translocation of the regulatory subunit oftype II protein kinase correlates with increased DNA (CRE)-protein bindingactivity in human cancer cell lines treated with site-selective analogs ofcAMP. Proc. Am. Assoc. Cancer Res.. 30: 446, 1989.

99. Mednieks. M. I., Yokozaki. H., Merlo. G. R., Tortora, G., Clair, T., Ally,S., Tahara, E., and Cho-Chung, Y. S. Site-selective 8-Cl-cAMP which causesgrowth inhibition and differentiation increases DNA (CRE)-binding activityin cancer cells. FEBS Lett., 254: 83-88, 1989.

100. Cho-Chung, Y. S., Tortora, G., Clair, T., Ally, S., and Yokozaki, H. Nuclearlocation signal and nuclear translocation of type H protein kinase regulatorysubunit in site-selective cAMP analog-induced growth control and differentiation. J. Cell Biol., 107: 492a, 1988.

101. Mizuno. T., Chou, M-Y., and Inouye, M. A unique mechanism regulatinggene expression: translational inhibition by a complementary RNA transcript (micRNA). Proc. Nati. Acad. Sci. USA. 81: 1966-1970. 1984.

102. Heywood, S. M. tcRNA as a naturally occurring antisense RNA in eukar-yotes. Nucleic Acids Res., 14: 6771-6772, 1986.

103. Paterson, B. M., Roberts, B. E., and Kuff, E. L. Structural gene identification and mapping by DNA-mRNA hybrid-arrested cell-free translation.Proc. Nat). Acad. Sci. USA. 74:4370-4374, 1977.

104. Rothenberg, M., Johnson, G.. Laughlin.C., Green, I.. Craddock, J., Sarver.N., and Cohen, J. S. Commentary: oligodeoxynucleotides as anti-senseinhibitors of gene expression: therapeutic implications. J. Nati. Cancer Inst..81: 1539-1544. 1989.

105. Tortora, G., Pepe, S., Yokozaki, H., Clair, T.. Rohlff, C., and Cho-Chung,Y. S. A RI., subunit antisense oligodeoxynucleotide of cAMP-dependentprotein kinase inhibits proliferation of human HL-60 promyelocytic leukemia. Proc. Am. Assoc. Cancer Res., 31: 38. 1990.

106. Otten. A. D.. and McKnight, G. S. Overexpression of the type II regulatorysubunit of the cAMP-dependent protein kinase eliminates the type I holoen-zyme Â¡nmouse cells. J. Biol. Chem.. 264: 20255-20260. 1989.

107. Tortora, G.. Clair, T., and Cho-Chung, Y. S. An antisense oligodeoxynucleotide targeted against the type IIâ€žregulatory subunit mRNA of proteinkinase inhibits cAMP-induced differentiation Â¡nHL-60 leukemia cells without affecting phorbol ester effects. Proc. Nati. Acad. Sci. USA. 87: 705-708. 1990.

108. Schwartz. D. A., and Rubin, C. Identification and differential expression oftwo forms of regulatory subunits (RII) of cAMP-dependent protein kinaseII in Friend erythroleukemic cells. J. Biol. Chem., 260: 6296-6303, 1985.

109. Gottesman, M. M. Genetic approaches in cyclic AMP effects in culturedmammalian cells. Cell. 22: 329-330, 1980.

110. Cho-Chung, Y. S., Tortora. G.. Yokozaki. H.. Meissner, S.. Urakami. K.I., and Miki. K. A RI,, subunit antisense oligodeoxynucleotide of cAMP-dependent protein kinase blocks proliferation in human and rodent cancercell lines by-passing exogenous cAMP effect. Proc. Am. Assoc. Cancer Res.,.J7.-29. 1990.

7100



1990;50:7093-7100. Cancer Res Yoon S. Cho-Chung and Suppression of Malignancy: New Approaches to TherapyRole of Cyclic AMP Receptor Proteins in Growth, Differentiation,

Updated version

http://cancerres.aacrjournals.org/content/50/22/7093

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/50/22/7093To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



role of cyclic amp receptor proteins in growth, differentiation, and

Documents