0013-7227/92/1306-3669$03.00/O

Endocrinology Copyright 0 1992 by The Endocrine

Vol. Printed

Society

Retinoic Acid Regulates Insulin-Like Growth Expression in a Neuroblastoma Cell Line KAZUE CAROL

MATSUMOTO, J. THIELE

CARLO

GAETANO,

WILLIAM

H. DAUGHADAY,

Factor

130, No. 6 in U.S.A.

II

AND

Molecular Genetics Section, Pediatric Branch (K.M., C.G., C.J.T.), Division of Cancer Treatment, National Cancer Institute, Bethesda, Maryland; and Department of Internal Medicine (W.H.D.), University of Washington School of Medicine, St. Louis, Missouri 63110

ABSTRACT. Insulin-like growth factors (IGF-I and IGF-II) are mitogenic polypeptides that play an important role in normal growth and development. IGF-II has been shown to stimulate the growth of neuroblastoma tumors in an autocrine and paracrine fashion. Critical in determining the role of IGF-II in tumorigenesis is the necessity to delineate factors affecting the transcription of IGF-II in normal and tumor tissues. To date such factors are poorly characterized. In this study we find that retinoic acid (RA), a naturally occurring morphogen, that has been shown to be indispensable in the development of the chick limb bud, stimulates an increase in IGF-II messenger RNA

I

NSULIN-LIKE growth factors (IGF-I and IGF-II) are polypeptide hormones that are important paracrine and/or autocrine mediators of cell growth and differentiation. IGF-II is highly expressed in the fetus or neonate implicating it as a putative fetal growth and differentiation peptide (1). Accumulating evidence has focused on its role during fetal cell growth since a disrupted IGF-II gene (male allele) in transgenic mice results in substantial growth retardation (2). While in vitro studies of muscle development (3) indicate IGF-II may have a role in cell differentiation, transgenic mice lacking IGF-II, although small, develop normally indicating that, whatever its role may be, IGF-II is not an absolute requisite for normal development. IGF-II has been found to stimulate the growth of a number of tumors in an autocrine or paracrine manner. These tumors, which include Wilm’s tumor (4,5), neuroblastoma (6), and rhabdomyosarcoma (7), are derived from distinct cell lineages yet are all embryonal in nature. The role this fetal growth factor plays in the development of these embryonal tumors is not clear. The induction of differentiation in embryonal carcinoma cells from a tuReceived November 26.1991. Address all correspondence and requests for reprints to: Dr. Carol J. Thiele, Molecular Genetics Section, Pediatrics Branch, Division of Cancer Treatment, National Cancer Institute, Bethesda, Maryland 20892.

(mRNA) in the Lan-1-15N neuroblastoma cell line. This increase in IGF-II is coincident with RA mediated inhibition of DNA synthesis. An increase in the steady state levels of IGF-II mRNA is detectable within 2 h of RA treatment and maximal by 24 h. In RA-treated Lan-1-15N cells, IGF-II mRNA levels are regulated at the level of new gene transcription and result in an increase in IGF-II protein in the culture supernatant. These studies suggest one mechanism affecting the production of IGFII in uioo may be mediated by RA and detail a model system by which transcriptional regulation of IGF-II mRNA can be analyzed. (Endocrinology 130: 3669-3676, 1992)

morigenic to a nontumorigenic state is associated with an increase in the production of IGF-II as well as plateletderived growth factor-like growth factors and transforming growth factor (TGF)a (8). Key to deciphering the role of IGF-II in tumorigenesis is an understanding of the mechanisms capable of regulating IGF-II expression. IGF-II is expressed and regulated in a developmental and tissue-specific manner (1). To date, four promoters have been identified within the 40 kilobase (kb) of chromosomal DNA containing the nine exons encoding the IGF-II gene. Differential promoter utilization and RNA splicing yield a 5.3 kb messenger RNA (mRNA) species that is preferentially expressed in adult liver while four mRNA species of 6.0, 4.8, 2.2, and 1.8 kb predominate in fetal tissues (9). However the molecular mechanisms regulating IGF-II mRNA transcription and production are not well characterized. Circulating levels of IGF-I are dependent on GH and studies in GH-deficient animals indicate that levels of IGF-II are only partially affected (10). In fetal rat fibroblasts, placental lactogen but not GH stimulates IGF-II production (10). During the course of our studies evaluating the effects of retinoic acid (RA) on neuroblastoma (NB) tumor cell growth and differentiation, we noted that RA induced a dramatic increase in the steady state level of IGF-II in the Lan-1-15N (15N) NB cell line. The paucity of in vitro systems to study the regulation of IGF-II made this

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.

3670

RA REGULATES

a particularly intriguing model to characterize the mechanisms controlling IGF-II expression. Furthermore, RA is a naturally occurring morphogen that has been shown to be important in the growth and development of a variety of embryonic tissues (11). Thus, it is possible that expression of IGF-II may be regulated either directly or indirectly by RA in viva. In this report, we describe a cell culture system in which the steady state levels of IGF-II are transcriptionally increased in RA-treated NB tumor cells resulting in an increase in the production of IGF-II protein.

Materials

and Methods

Cell culture

The neuronal subclone of the human NB cell line Lan-l15N (15N) (12), was cultured in RPMI-1640 containing 10% fetal calf serum (FCS) (Biofluid, Rockville, MD) at 37 C with 5% CO,. Cells were plated at a density of 2 x lo6 in 150-mm tissue culture disheswith indicated amounts of all-trans-RA (SigmaChemicalCo., St. Louis, MO) dissolvedin 95% ethanol or control solvent. Cells were fed every 2 days with fresh medium and harvested for isolation of nucleic acids at the indicated time points. For short incubation studiesof 2 days or lessthe medium was removed and added back to cells after adjusting RA or solvent concentration. Actinomycin D (5 pg/ ml) or cycloheximide (CHX, 5 Kg/ml) (Sigma) were added to cultures for the indicated period of time. Cell growth

assay

15N cells were plated at a density of 2.5 X lo3 cells per 96well flat bottom plates with varying amountsof RA or solvent control. After 2-8 days of RA treatment, cellswere labeledwith 1 j&i of [3H]thymidine per well for 18 h and harvested with PHD cell harvester (CambridgeTech., Inc., Watertown, MA) and radioactivity wasmeasuredin a LS 1801liquid scintillation counter (Beckman Instruments Inc., Fullerton, CA). Isolation

and analysis of RNA

Poly(A)+ RNA was isolated from cultured cells using “Fast Track” procedure (Invitrogen, San Diego, CA). The poly(A)+ RNA (l-3 pg) was electrophoresedin a 1% agaroseformaldehyde gel and transferred to Nytran membrane(Schliecher and Schuell, Keene, NH) by capillary transfer in 10x standard salinecitrate buffer (IX SSC = 0.15 M sodiumchloride, 15 mM trisodium citrate pH 7.2). The membranewasbaked 2 h at 80 C and usedfor Northern analysis.DNA probesfor IGF-II [0.8 kb complementaryDNA (cDNA) of IGF-II cloned into the PstI site of pKT218) (10) or pB4 [l.l kb PstI fragment of IGF-II gene correspondingto exon 7 nucleotides 2932-4060 of the IGF-II geneclonedin pGEM (13)] werelabeledwith [32P]dCTP (New England Nuclear, Boston, MA) by nick translation according to manufacture’s specification (Amersham, Arlington Heights, IL) to a specific activity of 3 x 10’ cpm/pg. Blots were incubatedwith 2 x lo6 cpm/ml of hybridization solution for 16 h at 42 C with moderateshakingand subsequentlywashedtwo times for 20 min in 2~ SSC containing 0.1% sodium dodecyl

IGF-II

Endo * 1992 Vol130. No 6

sulfate (SDS) at room temperature (RT) then twice for 30 min at 65 C in 0.15~ SSC and 0.5% SDS. Autoradiograms were madewith X-Omat-AR film (Kodak, Rochester, NY) using a Lightening Plus intensifying screen (DuPont Cronex, Clifton, NJ) and exposedfor various times. Rehybridization of filters wasperformedafter treating Nytran membranesfor 1 h in 50% formamide, 1X SCC at 75 C. The IGF-II cDNA recognizesa 6.0,5.3,4.8, and 2.2 kb mRNA dependingon the tissueanalyzed while pB4 recognizes6.0, 4.8, and 1.8 kb mRNA species. Nuclei preparation

and in vitro nuclear

transcription

assay

Nuclei wereisolatedfrom cultured cellsthat had beentreated with RA or solvent control for 27 h aspreviously described(14) and stored at -70 C. The nuclei were thawed and washedin 1 ml Transcription Buffer (TB) (TB = 20 mM HEPES, 20% glycerol, 175 mM KCl, 10 mM MgCl, 1 mM dithiothreitol, and 1 mM each of ATP, CTP, and GTP (Boehringer-Mannheim, Indianapolis, IN). Transcription was performed in TB buffer with 1mCi [32P]UTP (800 Ci/mmol) (New England Nuclear) for 60 min at 26 C. The reaction was stoppedby adding CaClz to 4 mM, subsequentlytreated with 30 U/p1 deoxyribonuclease (Promega,Madison, WI) for 15 min at 30 C, and incubated 90 min at 42 C with 100pg/ml ProteaseK (Sigma). After phenol extraction, RNA wasprecipitated with 3 M ammoniumacetate in isopropanolon dry ice. The RNA was dissolvedin 1 x TE buffer (10 mM Tris, pH 7.4, 1 mM EDTA) and centrifuged through SephadexG-50 spin column (Boehringer-Mannheim) and reprecipitated with 300 mM sodium acetate and 2.5 vol of ethanol overnight at -20 C. To evaluate the relative gene transcription, DNA fragments correspondingto the indicated geneswere treated with 0.3 M NaOH for 60 min at 60 C, adjustedto 1 M ammonium acetate, and 1 Kg was transferred to Nytran membraneusing a Slot Blot apparatus (Minifold II, Schliecher and Schnell, Keene, NH) and vacuum aspiration. The membrane was baked for 2 h at 80 C. Equal amounts (cpm) of 32P-labeledRNA were hybridized to filters for 3 days at 42 C in 50% formamide,5~ SSPE, 0.5% SDS, 5~ Denharts, and 250pg/ml of salmonspermDNA with gentle shaking. The filters were washedtwice in 2~ SSC, 0.1% SDS for 30 min at RT, then washed in 2~ SSC, 0.1% SDS for 15 min at 50 C. Autoradiogramswere prepared as describedabove. In situ hybridization

Cells(1 x 105)were cytocentrifuged onto poly-1-lysinecoated sialized glass slidesand fixed with paraformaldehyde-lysineperiodate-gluteraldehydefixative for 20 min at RT and stored at -70 C asdescribedpreviously (15). A 30-nucleotideantisense oligo correspondingto nucleotide 535-565 of IGF-II (gift from Dr. El-Badry, NIH, Bethesda,MD) waslabeledwith [35S]dATP (1000 Ci/mmol) (New England Nuclear) with terminal deoxynucleotide transferase method (Bethesda Research Lab, Gaithersburg, MD) to a specific activity of 6 x lO’cpm/pg. 35SAntisense oligo nucleotide (5 X lO?pm) was hybridized with cells for 16 h at 37 C and washedoncewith 2~ SSC for 10 min at RT, twice with IX SSC, 0.1% SDS for 15 min at RT, and twice with 1X SSC,0.1% SDS at 42 C for 15min and dehydrated with ascendingconcentrations of ethanol (16). Autoradiography wasperformed by dipping slidesin photographic emulsion (Kodak, Rochester,NY) diluted 1:l with water, exposedat -70

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.

RA REGULATES C for 7 days, then developed in D19 developer for 4 min at 15 C, stopped by acetic acid, and fixed with 30% sodium thiosulphate for 2.5 min. Cells were stained with hematoxylin and eosin and examined using a Nikon microscope.

IGF-II protein analysis Cells were incubated in control or RA-treated RPMl-1640 supplemented with 10% FCS for indicated times. Twenty hours before harvest, cells were shifted to medium lacking serum. Serum-free culture supernatants were passed through Sep-Pak C-18 cartridges as previously described (17) except that IGF-II was eluted with 2 ml of 70% ethanol, 0.1 M acetic acid. IGF-II was measured by RIA with a monoclonal antibody raised against rat IGF-II (Amano Pharmaceutical Co., Troy, VA). IGF-II (Eli Lilly Co., Indianapolis, IN) was used for iodination and standards. Separation of bound label was performed by the double antibody method.

Results RA treatment of 15N

Northern analysis of 15N cell line during RA treatment

15N cells were treated, as described in Materials and 1 x 10m6 M RA for 2-8 days. Northern blot analysis of RNA isolated from control and treated cells revealed a 50-fold increase in the 6.0-kb, a 25-fold increase in the 4.8-kb, and a 29-fold increase in the 1.8kb IGF-II mRNAs by 2 days with the relative levels declining at later time points (Fig. 2, panel A). The induction of the 6.0-kb, 4.8-kb, 1.8-kb mRNA by RA is reproducible in seven different experiments. While these species of IGF-II mRNA do not appear to be coordinately regulated, the differences in their relative abundance in Northern analysis may be due to the fact that the 1.8-kb mRNA specie does not include, and is 3’ to, the coding

Methods, with

of the IGF-II

gene. However,

that the differences reflect differential

we cannot

blotting

rule out

efficiency

3671

A s A ;

600

-

400

-

ii 2

300

-

200

-

1, 2 500 8

.OOSuMRA 5

.OSuMRA .5uMRA

2 (3

2

0

4

6

5.0uMRA 1 10

a

DAYS

B F z

600

-

g

500

-

400

-

: 2

300

-

5

200

-

&

Previous studies indicated that 15N was among the most sensitive of the NB cell lines tested to the growthinhibiting effects of RA (18). A RA dose-response curve ranging from 5 X lo-’ M to 5 X 10m6M revealed that 15N showed a reduction in [3H]thymidine uptake greater than 50% at 8 days of culture in 5 X lo-’ M RA while the concentrations from 1 X 10-s M to 5 X 10m6M RA resulted in more than 90% inhibition of cell growth (Fig. 1A). However, after 8 days in culture we noted a small, albeit, reproducible increase in DNA synthesis in cell cultures (Fig. 1B). Prolonged culture (~30 days) of cells in 5 X 10e7 M and 5 X lo-’ M RA resulted in populations of RA resistant cells (data not shown). Since a previous study noted an association between resistance to the growthinhibiting effects of RA and expression of IGF-II we analyzed 15N treated with RA for the expression of IGFII mRNA (18).

portion

IGF-II

0

CarrROL

2

4

6

8

10

12

DAYS

FIG. 1. A, Growth of 15N cells was evaluated after varying periods of time in different concentrations of RA by incubating quadruplicate cultures of cells with [3H]thymidine as described in Materials and Methods. Upperpanel cells were incubated in control condition or, 5 x 10m6M RA, (5 X lo-’ M) RA, 5 X 1OmRM RA, and 5 X 1O-9 M RA and treated for 8 days. B, Cells were treated with control solvent or 5 x lo-’ M RA for 10 days.

for large us. smaller sized mRNA species. Viability of cells in 1 X 10m6 M RA was 80% of control values at 2 days and declined thereafter. To evaluate the earliest detectable increase in IGF-II mRNA in RA-treated cells, RNA was isolated from cells after 2, 5, 24, and 48 h exposure to RA. As depicted in Fig. 2C, a detectable increase in IGF-II mRNA expression is apparent within 2 h of RA treatment and increased to 4-fold greater than basal levels after 5 h of RA treatment. Maximal levels of IGF-II mRNA are detected by 24 h of treatment (Fig. 2C). A RA dose response curve indicated an increase in the 6.0,4.8, and 1.8 kb IGF-II mRNA levels occurs within 24 h in 15N cells treated with doses of RA as low as 5 x lo-’ M (Fig. 3, lane 2). IGF-I mRNA was not detected in control or RA-treated 15N cells (data not shown). Although 15N is a clonal cell line, NB cell lines have the propensity to transdifferentiate in culture (19) and the possibility existed that the increase in the steady state level of IGF-II mRNA reflected a preexisting sub-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.

RA REGULATES

3672

1

2

3456

IGF-II

Endo. 1992 I’01130 - No 6

123456

78

28s -

28S-

18S-

18S-

B

28S-

28S-

12345678

C

FIG. 3. Expression of IGF-II mRNA in 15N cells was evaluated in cultures treated for 24 h with varying concentrations of RA. Poly(A)+ mRNA (1 pg) from cultures treated with solvent control (lane l), 5 x 10m9M RA (lane 2), 5 X lo-* M RA (lane 3), 5 X 10e7 M RA (lane 4), 1 X 10V6 M RA (lane 5), 5 X 10m6M RA (lane 6) was used for Northern blot hybridization. After washing, blots were exposed to X-AR-5 film with an intensifying screen at -70 C for 6 h. Ethidium bromide staining of the gel indicates that equivalent amounts of RNA were loaded into each lane.

28s -

18S-

D 28s

-

FIG. 2. Expression of IGF-II mRNA in 15N cells at varying times after RA treatment. A, The relative levels of IGF-II expression were evaluated by hybridization of 32P-labeled pB4 DNA probe to Northern blot containing 3 pg poly(A)+ mRNA extracted from cells at varying times after RA (1 X 10e6 M) treatment: cultures were treated with control solvent for 2 days (lane l), 4 days (lane 3), 6 days (lane 5), 8 days (lane 7), or 1 PM RA for 2 days (lane 2), 4 days (lane 4), 6 days (lane 6), and 8 days (lane 8). Quantitative scanning of appropriately exposed autoradiograms using a Bio-Rad (Richmond, CA) densitometer were performed to estimate relative changes in mRNA levels. B, Ethidium bromide staining of 1% agarose-formaldehyde gel indicates equivalent amounts of samples in each lane. C, Evaluation of the relative levels of IGF-II expression at the varying times after RA treatment: cultures were treated with control solvent for 2 h (lane l), 5 h (lane 3), 24 h (lane 5), 48 h (lane 7), or with 1 x 10m6M RA for 2 h (lane 2), 5 h (lane 4), 24 h (lane 6), 48 h (lane 8). D, Ethidium bromide staining of 1% agarose formaldehyde gel indicates equivalent amounts of samples in each lane.

population of cells expressing high levels of IGF-II. To evaluate on a per cell level the expression of IGF-II, control, and RA-treated cells were analyzed by in situ hybridization for the expression of IGF-II mRNA. An antisense oligo corresponding to nucleotides 535-565 of IGF-II mRNA was 35S-labeled and hybridized to 15N cells treated for 3 days with 1 X 10e6 M RA. When over 5000 control cells were examined we were unable to

FIG. 4. In situ hybridization of RA-treated 15N cells with ‘S-IGF-II antisense oligo. 1B and 2B, Dark field images (silver grain appears as bright spots) of the same views shown in 1A and 2A using bright field optics (silver grain appears as dark spot) with ~500 magnification. 1A and lB, 15N cells treated with solvent control for 3 days and 2A and 2B were treated with 1 X 10e6 M RA for 3 days. Slides were incubated with NTB-2, Kodak emulsion for 7 days at -20 C and developed.

detect a subpopulation of cells expressing significant grains over background in control cells (Fig. 4, panels lA, 1B). A statistically significant increase in grains corresponding to IGF-II mRNA was detected in all cells in the RA treated cultures (Table 1) (Fig. 4, panels 2A, 2B). The uniform increase in grain count was also detected in cells treated with RA for 7 days (data not

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.

RA REGULATES TABLE

1. Effect of RA on IGF-II mRNA as determined by in situ

analysis Grain count per cell

IGF-II

of IGF-II

3

4

5

6

7

8

9 101112

c-

IGF-II

-

IGF-II

a--

pB4

+

GAPDH

-

BG

18S-

28S-

1234

B

5676

28s -

mRNA during 15N RA treatment

To investigate the mechanism by which RA regulates IGF-II mRNA levels in 15N cells, we analyzed IGF-II mRNA stability. 15N cells were treated for 48 h with 1 x 10m6 M RA or control solvent and cultures were incubated for up to 2 h with the transcriptional blocker actinomycin D (5 pg/ml). A Northern blot analysis showed that the stability of the 6.0 and 4.8 kb of IGF-II mRNA species in RA-treated cells was not significantly different than the stability of the IGF-II detected in control cells (Fig. 5A). To determine if protein synthesis is necessary for RA to stimulate IGF-II mRNA levels, 15N cells were treated with RA or the control solvent in the presence or absence of CHX (5 pg/ml) for 5 h, a time in which there is a significant RA stimulated increase in IGF-II mRNA compared to control cultures (Fig. 5B, lanes 1 and 3). Northern analysis of mRNA isolated from these cultures indicated that inhibition of protein synthesis prevented a RA-stimulated increase in IGF-II mRNA (Fig. 5B, lanes 3 and 4). The basal level of IGFII mRNA synthesis detected in control cultures was also depressed in the presence of CHX (Fig. 5B, lanes 1 and 2). These studies indicate that protein synthesis is required to sustain a portion of the basal level of IGF-II mRNA synthesis and the RA-stimulated increase in IGFII mRNA is dependent on new protein synthesis. While the decrease in the 6.0-kb species is most apparent, regulation of the 4.8 and 1.8 kb IGF-II mRNA species was similar. The levels of IGF-II mRNA do not significantly change if cultures pretreated with RA for 48 h are incubated for 4 h in CHX (Fig. 5B, panels 1 and 2, lanes 5-8). However, the stability of IGF-II mRNA (>2 h) may obscure any change in IGF-II mRNA synthesis under these conditions (Fig. 5A). To evaluate the relative transcription of IGF-II mRNA in control and RA-treated cells, we performed nuclear

2

28s -

Control 7.6 + 2.4 0.0001” 3 dav RA 23.4 + 6.6 Per cell level of silver grain count of control and 3 day RA (10M6 M) treated cells. a Statistics were performed using Student’s t test two tail analysis on Statview SE+ Graphics (Abacus Concept, Inc.).

Regulation

1

A

P value

shown). The NB cell line SMS-KCNR which doesn’t make detactable IGF-II mRNA on Northern analysis (16) was used as a negative control and contained 3.4 + 2 grains per cell (data not shown). These results, taken together, clearly indicate that RA stimulates IGF-II mRNA levels uniformly in the 15N cell line.

3673

18S-

28s -

C

FIG. 5. Regulation of IGF-II mRNA. Poly(A)+ mRNA (1 rg) was used for Northern blot analysis and after hybridization the blots were exposed to X-AR-5 film with an intensifying screen at -70 C for overnight. Ethidium bromide staining of 1% agarose formaldehyde gel shows an equivalent amount of RNA was loaded into each well. A, Stability of IGF-II mRNA was evaluated by analyzing Northern blots of 1 fig poly(A)+ mRNA. RNA was isolated from 15N NB cells treated with control solvent (lanes l-6) or 1 X 10e6 M RA (lanes 7-12) for 48 h and subsequently incubated for 0 min (lanes 1, I), 10 min (lanes 2, 8), 30 min (lanes 3, 9), 60 min (lanes 4, lo), 90 min (lanes 5, ll), and 120 min (lanes 6,12) with 5 pg/ml actinomycin D. B, IGF-II expression in CHX-treated cultures of 15N during RA treatment. Cells were treated with control solvent (lane 1) or 1 x 1Om6M RA (lane 3) for 5 h, control solvent + 5 rg/ml CHX (lane 2), or 1 x 1Om6M RA + 5 pg/ml CHX for 5 h (lane 4). After 2 days with solvent control (lanes 5 and 6) or 1 X lo-” M RA (lanes 7 and 8) cells were incubated with 5 pg/ml CHX (lanes 6 and 8) or solvent control (lanes 5 and 7) for an additional 4 h with the presence or absence of RA. Poly(A)+ mRNA (1 pg) was evaluated by Northern hybridization analysis and blots were exposed to Kodak X-AR-5 film with an intensifying screen at -70 C for 2 days (lanes l-4) or exposed for 6 h (lanes 5-8). C, Nuclear transcription of IGF-II and pB4 genes by 15N nuclei after 27 h of RA treatment. One microgram of indicated DNAs were immobilized on Nytran filters according to manufacture’s recommendation. ‘*P-Labeled nuclear RNA transcripts were prepared and isolated from 10’ nuclei harvested from solvent-treated cells (lane l), lo7 nuclei from 1 x 10e6 M RA-treated cells (lane 2), and 10’ nuclei from 5 X 10m6M-treated cells (lane 3). BG indicates background.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.

RA REGULATES

3674

transcription assays. Nuclei were isolated from cells treated with 1 X 10m6M, 5 X lop6 M RA or control solvent for 27 h. [32P]RNA was synthesized as described in Materials and Methods and hybridized to identical filter strips containing 1 pg IGF-II cDNA, pB4 DNA, and the housekeeping gene GAPDH (glyceraldehyde-3-phosphate dehydrogenase). Relative densitometric analysis revealed an increase in IGF-II mRNA transcription of approximately 3- to 6-fold depending on the concentration of RA utilized (Fig. 5C and Table 2). Essentially similar results were obtained when analysis was performed at 17 h of RA treatment (data not shown). Transcription of GAPDH mRNA is not significantly altered by RA. Since all the previous studies indicate the 6.0-, 4.8-, and 1.8-kb mRNA species are coordinantly regulated, we infer that the increase in the relative transcription of the IGF-II mRNA reflects an increase in transcription of these three mRNA species as well. IGF-II peptide analysis

To determine if the RA stimulated increase in IGF-II mRNA transcription resulted in an increase in IGF-II protein synthesis, culture supernatants were analyzed for the presence of IGF-II. Cells were treated for the indicated periods of time in RA and prior to harvest of culture supernatant the cells were washed twice in medium lacking FCS and cultured for 20 h in medium lacking FCS yet containing RA or control solvent. IGFII protein was determined by RIA. The results indicate a 2-fold increase in secreted IGF-II protein in cultures incubated with RA for 1 or 2 days and a 4-fold increase in cultures incubated for 4 days in RA (Fig. 6). Repeat experiments confirm an approximately 4-fold increase in secreted IGF-II in cultures treated for 4 days with RA. Total protein secreted by control and RA-treated cultures was not significantly different in these experiments (data not shown). Discussion Stimulation of embryonal tumor cell growth by the autocrine or paracrine production of IGF-II has suggested the possibility the IGF-II contributes to the neoTABLE

tion

2. Densitometric

analysis

of effect

of RA on IGF-II

transcrip-

of 15N 27 h Treatment

GAPDH IGF-II pB4

Control

1 PM

5fiM

0.55” 0.13 0.11

0.47 0.34 0.30

0.59 0.66 0.66

Densimetric analysis of Fig. densimeter to determine relative u Values expressed in relative

5C was performed using increase of transcription. densitometric units.

Bio-Rad

620

IGF-II

Endo. 1992 Vol130. No 6

3

y b ii is c

2

'

0 1

2

4

DAYS 6. Evaluation of the relative levels of IGF-II secreted in medium at varying times after RA treatment. Cells were treated with solvent control or 1 x 10e6 M RA for 1 day, 2 days, and 4 days in 10% FCS. The cells were washed twice with serum-free medium and incubated for an additional 20 h with the presence or absence of RA in medium lacking serum. IGF-II was extracted as previously described (17). Bars represent average of duplicate samples of extracts of supernatants from control (solid) and 1 X 1O-6 M RA (hatched) treated cells.

FIG.

plastic evolution of tumor cells such as NB (6), rhabdomyosarcoma (7), and Wilm’s tumor (4, 5). Although the molecular mechanisms regulating IGF-II production in vivo are obscure, in this study we demonstrate that in vitro RA is capable of increasing the steady state levels of IGF-II mRNA and IGF-II protein in the 15N NB cell line. This is of particular interest because RA has been shown to be a naturally occurring morphogen that regulates growth and differentiation in uiuo (11). The RA-induced increase in IGF-II mRNA levels in the 15N NB cell line is demonstrable within 2 h, maximal by 24 h of RA treatment and results in increases in the 6.0, 4.8, and 1.8 kb IGF-II mRNA species. While all the IGF-II mRNA species encode the same polypeptide, the 6.0 and 2.2 kb mRNA species utilize the P2 promoter and arise from differential usage of polyadenylation signals in the 3’ untranslated region of the gene. The 4.8-kb mRNA species uses the P4 promotor and is exclusively detected in the polysomes of a rhabdomyosarcoma cell line, suggesting that it alone is involved in the synthesis of prepro-IGF-II (20). Whether this can be extrapolated to other cell types expressing IGF-II remains to be tested. Differential promoter utilization may provide for translational discrimination among IGF-II transcripts and afford an additional level of gene regulation. In the 15N NB cells the RA-induced increase in the steady state level of the 6.0-kb IGF-II mRNA is the greatest. Nevertheless, the relative kinetics of expression of all three mRNA transcripts are similar in their temporal expression, stability, and inducibility in the presence of CHX.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.

RA REGULATES

Therefore, it is reasonable that the transcriptional increase detected in the nuclear-run on experiment reflects an increase in the relative transcription of all the detectable IGF-II mRNA species. While a transcriptional increase in IGF-II mRNA occurs in 15N NB cells treated with RA, a demonstrable decrease in IGF-II mRNA occurs in cells treated with either RA and CHX or control solvent and CHX culture. This indicates de nouo protein synthesis is required to induce IGF-II mRNA. This does not rule out that RA receptors (RAR) directly mediate the increase in IGF-II mRNA levels, as newly synthesized RAR or its coregulator proteins (21), which increase RAR-dependent gene transcription, may be required to affect the increase in IGF-II mRNA. However, RA may stimulate the synthesis of other transcription factors which, in turn, modulate IGF-II mRNA synthesis. While several transcriptional binding factors may be involved in regulating IGF-II mRNA levels, recent studies indicating that the transcriptional factor AP-2 is developmentally expressed in neural crest cells (22) and can be induced by RA (23) indicate AP-2 may be important in this system. Definitive evaluation of the proteins involved in mediating the RA-induced response require transient transfection and DNA binding analyses of the IGF-II promoter regions and such studies utilizing the Pq promoter are currently in progress in our laboratory. The RA-induced increase in IGF-II mRNA levels results in the accumulation of IGF-II peptide in the supernatant of 15N NB cells. Increases in IGF-II, a peptide that has been shown to be a potent growth factor for some NB tumor cell lines (24) and can stimulate neurite extension in primary neuroblasts (2), again raises the question as to the role of IGF-II in stimulating growth and/or differentiation associated properties in this tumor cell line. In the 15N NB cells treated for 4 days with RA there is a more than 90% decrease in DNA synthesis despite the production of IGF-II during this time. Furthermore, RA-treated 15N NB cells also show increases in neurofilament mRNA, although cells do not show evidence of morphologic differentiation (our unpublished observation). This is similar to results recently reported in rat hepatoma cells in which heparin together with insulin, glucagon, and GH induced transcriptional increases in IGF-II and TGFP mRNA levels as well as inhibition of cell growth (25). Also, in embryonal carcinoma cells induced to differentiate there are increases in a number of growth factors including IGF-II (8). However, in 15N NB cells treated with RA, there is a relatively rapid development (within 10 days) of cells which proliferate despite the presence of RA suggesting that during prolonged culture the accumulation of IGF-II may stimulate the proliferation of these cells (18). Two recent studies present interesting aspects of the interaction of

IGF-II

3675

RA and growth modulating factors which offer a possible context in which to view the apparent contradictory processes occurring during the response of 15N NB cells to RA. RA can stimulate an autocrine growth suppressing loop in HL60 cells that is mediated by TGFBl (26). Furthermore, in uitro studies indicate that AP-1, a transcription factor induced by some growth factors, can block transcription of some genes regulated by RA receptors (27). Thus, it is possible that although a growth inhibitory process is initially stimulated in RA-treated 15N cells, the accumulation of IGF-II over time may induce factors which partially inhibit or compete with growth inhibitory effects of RA. Current studies are aimed at evaluating signals transduced by IGF-II using antisense oligomers directed to the IGF-II gene or transfection of a plasmid expressing an antisense IGF-II mRNA. Such studies are needed to determine the role of IGF-II in this system. The question arises whether the regulation of IGF-II in these NB cells reflects an abnormal response of these cells to RA or is a component of the response of the normal cells from which these tumor cells are derived, i.e. neural crest cells. Southern analysis and karyotypic analysis indicates a normal gross structure and number of IGF-II genes in 15N NB cells (our unpublished data). If alterations in the promoter regions of the IGF-II gene were invoked to account for this response, one would have to predict alterations in both the P3 and P4 promoters since increases in mRNA from both these promoters are detected. This seems unlikely. It is possible that alterations in regulation or expression of the putative short-lived protein that mediates this response are responsible for this effect. It should therefore be considered that IGF-II may have a role in stimulating the growth of neural crest-derived chromaffin precursors during the formation of the adrenal medulla. In fact, in situ analysis of IGF-II mRNA in the developing human adrenal reveled high levels of IGF-II mRNA expression in 7-8 week fetal adrenocortical cells, suggesting a potential role for IGF-II in the paracrine stimulation of medullary cell proliferation (16). A recent study indicates that IGF-I stimulates cell growth or potentiates cell differentiation depending on the presence of differentiation signals (28). Thus, the production of IGF-II by the 15N NB cell line which exhibits characteristics similar to normal developing adrenal chromaffin cells may thus reflect a physiological rather than an abnormal response to RA. Acknowledgments We would like to thank Ms. Judi Jourabchi for preparing the manuscript, Dr. Pamela Cohen for excellent advice on in situ hybridization work, and Drs. Peter Nissley, Derek Leroith, and Caterina Minniti for their critical review of our manuscript.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.

3676

RA REGULATES

References 1. Daughaday WH, Rotwein P 1989 Insulin-like growth factors I and II. Peptide, messenger ribonucleic acid and gene structures, serum and tissue concentrations. Endoc Rev 10:68-91 2. DeChiara TM, Efstratiadis A, Robertson EJ 1990 A growth-deticiency phenotype in heterozygous mice carring an insulin like growth factor-II gene disruptedby targeting. Nature 345:78-80 3. Tollefsen S. Sadow JL. Rotwein P 1989 Coordinate exuression of insulin-like growth factor II and its receptor during muscle differentiation. Proc Nat1 Acad Sci USA 861543-1547 4. Reeve AE. Eccles MR. Wilkins I&J. Bell GI. Millow LJ 1985 Expression of insulin-like growth factor II transcripts in Wilms’ tumour. Nature 317:258-260 5. Scott J, Cowell J, Robertson ME, Priestly LM, Wadey R, Hopkins B, Pritchard J, Bell GI, Rall LB, Graham CF, Knott JT 1985 Insulin-like growth factor-II gene expression in Wilms’ tumour and embryonic tissues. Nature 317:260-262 6. El-Badry OM, Romanus JA, Helman LJ, Cooper MJ, Rechler MM, Israel MA 1989 Autonomous growth of a human neuroblastoma cell line is mediated by insulin-like growth factor II. J Clin Invest 84:829-839 7. El-Badry OM, Minniti C, Kohn EC, Houghton PJ, Daughaday WH, Helman LJ 1990 Insulin-like growth factor II acts as an autocrine growth and motility factor in human rhabdomyosarcoma tumors. Cell Growth Differ 1:325-331 8. vanzolen EJ, Koornneef I, Holthuis JCM, Ward-Van Oostwood THJ, Feijeni A, de Poorter TL, Mummery CL, Buul-Offers SC 1989 Production of insulin-like growth factors, platelet-derived growth factor, and transforming growth factors and their in the density-dependent growth regulation of a differentiated embryonal carcinoma cell line. Endocrinology 124:2029-2041 9. Gloudemans T, Prinsen I, Van Unnik JAM, Lips CJM, Den Otter W, Sussenbach JS 1990 Insulin-like growth factor gene expression in human smooth muscle tumors. Cancer Res 50:6689-6695 10. Adams SO, Nissley SP, Handwerger S, Rechler MM 1983 Developmental patterns of insulin-like growth factor-I and -II synthesis and regulation in rat fibroblasts. Nature 302:150-153 11. Thaller C, Eichele G 1987 Identification and spatial distribution of retinoids in the developing chick limb. Nature 327:625 12. Ciccarone V, Spengler BA, Meyers MB, Biedler JL, Ross R 1989 Phenotypic diversification in human neuroblastoma cells: Expression of neural crest lineages. Cancer Res 49:219-225 13. Helman LJ. Thiele CJ. Linehan WM. Nelkin BD. Bavlin SB. Israel MA 1987 Molecular markers of neuroendocrine’development and evidence of environmental regulation. Proc Nat1 Acad Sci USA 84:2336-2339 14. Thiele CJ, Deutsch LA, Israel MA 1988 The expression of multiple

15.

16.

17. 18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

IGF-II

Endo. Voll30.

1992 No 6

proto-oncogenes is differentially regulated during retinoic acid induced maturation of human neuroblastoma cell lines. Oncogene 3:281-288 Cohen PS, Seeger RC, Triche TJ, Israel MA 1988 Detection of Nmyc gene expression in neuroblastoma tumors by in situ hybridization. Am J Path01 131:391-397 El-Badry OM, Helman LJ, Chatten J, Steinberg SM, Evans AE, Israel MA 1991 Insulin-like growth factor II-mediated nroliferation of human neuroblastoma. JClin Invest 87:648-657 Daughaday WH 1987 Radioligand assays for insulin-like growth factor II. Methods Enzymol 146:248 Gaetano C, Matsumoto K, Thiele CJ 1991 Retinoic acid resistant neuroblastoma cells and the expression of insulin-like growth factor-II. Adv Neuroblastoma Res 3:165-172 Ross RA, Spengler BA, Beidler JL 1983 Coordinate morphological and biochemical interconversion of human neuroblastoma cells. J Nat1 Cancer Inst 71:741-747 Nielsen FC, Gammeltoft S, Christiansen J 1990 Translational discrimination of mRNAs coding for human insulin-like growth factor II. J Biol Chem 265:13431-13434 Glass CK. Devarv OV. Rosenfeld MG 1990 Multinle cell twespecific proteins differentially regulate target sequence recognition by the (Y retinoic acid receptor. Cell 63:729-738 Mitchell PJ, Timmons PM, Hebert JM, Rigby WJ, Tjian R 1991 Transcription factor AP-2 is expressed in neural crest cell lineages during mouse embryogenesis. Genes Dev 5:105-119 Luscher B, Mitchell PJ, Williams T, Tjian R 1989 Regulation of transcription factor AP-2 by the morphogen retinoic acid and by second messages. Genes Dev 3:1507-1517 Bothwell M 1983 Insulin and somatomedin MSA promote nerve growth factor-independent neurite formation by cultured chick dorsal root ganglionic sensory neurons. Prog Clin Biol Res 118:225231 Zvibel I, Halay E, Reid LM 1991 Heparin and hormonal regulation of mRNA synthesis and abundance of autocrine growth factors: relevance to clonal growth of tumors. Mol Cell Biol 11:108-116 Falk LA, De Benedetti F, Lohres N. Birchenall-Roberts MC. Ellingsworth LW, Faltynek CR, Ruscetti FW 1991 Induction of TGF-B, recentor exnression and TGF-B, nrotein nroduction in . retinoic acid:treated-HL-60 cells: possible TGF-&-mediated autocrine inhibition. Blood 77:1246-1255 Schule R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM 1990 Jun-Fos and receptors for Vitamins A and D recognize a common response element in the human osteocalcin gene. Cell 61:497-504 Pahlman S, Meyerson G, Lindgren E, Schalling M, Johansson I 1991 Insulin-like growth factor I shifts from promoting cell division to potentiating maturation during neuronal differentiation. Proc Nat1 Acad Sci USA 889994-9998

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 06:19 For personal use only. No other uses without permission. . All rights reserved.