(CANCER RESEARCH 58. 3163-3172.
`
`July 15. 1W8|
`
`Retinoic Acid Induced Mitogen-activated Protein (MAP)/Extracellular Signal-
`regulated Kinase (ERK) Kinase-dependent MAP Kinase Activation Needed
`to Elicit HL-60 Cell Differentiation and Growth Arrest1
`
`Andrew Yen,2 Mark S. Roberson, Susi Varvayanis, and Amy T. Lee
`
`Ctincer ßioloi>\Laboratory. Department of Piilh(>log\ ¡A.Y., S. V., A. T. L.j. and Department of Plivxiologv IM. S. R.I. College of Veterinary Medicine. Cornell Universitv.
`New York 14853
`
`Ithaca,
`
`ABSTRACT
`
`kinase
`signal-regulated
`Retinoic acid (RA) activated the extracellular
`(ERK)
`2 mitogen-activated
`protein kinase
`(MAPK)
`of HL-60 human
`myeloblastic
`leukemia cells before causing myeloid differentiation and cell
`cycle arrest associated with hypophosphorylation
`of the retinoblastoma
`(RB)
`tumor suppressor
`protein. ERK2 activation by mitogen-activated
`protein/ERK kinase (MEK) was necessary for RA-induced differentiation
`in studies using PD98059
`to block MEK phosphorylation. G0 growth
`arrest and RB tumor suppressor protein hypophosphorylation
`(which is
`typically associated with induced differentiation
`and (•„arrest),
`two pu-
`tatively RB-regulated
`processes,
`also depended on ERK2 activation by
`MEK. Activation of ERK2 by RA occurred within hours and persisted
`until
`the onset of RB hypophosphorylation,
`differentiation,
`and arrest.
`ERK2 activation was probably needed early, because delaying the addi
`tion of PD98059 relative to that of RA restored most of the RA-induced
`cellular response.
`In contrast
`to RA (which activates RA receptors
`(RARs)
`and retinoid X receptors
`in HL-60 cells with its metabolite
`retinoids), a
`retinoid that selectively binds RAR-y, which is not expressed in HL-60
`cells, was relatively ineffective
`in causing ERK2 activation. This is con
`sistent with the need for a nuclear
`retinoid receptor
`function in RA-
`induced ERK2 activation. RA reduced the amount of unphosphorylated
`RAR-a, whose activation is necessary for RA-induced differentiation
`and
`arrest. This shifted the ratio of phosphorylated:unphosphorylated
`RAR-a
`to predominantly
`the phosphorylated
`form. Unlike other steroid thyroid
`hormone
`receptors
`susceptible
`to phosphorylation
`and activation
`by
`MAPKs, RAR-a was not phosphorylated
`by the activated ERK2 MAPK.
`The results thus show that RA augments MEK-dependent ERK2 activa
`tion that
`is needed for subsequent RB hypophosphorylation.
`cell differ
`entiation,
`and G0 arrest. The process
`seems
`to be nuclear
`receptor de
`pendent
`and an early
`seminal
`component
`of RA signaling
`causing
`differentiation
`and growth arrest.
`
`INTRODUCTION
`RA3 regulates cell cycle progression and differentiation in a variety
`
`(1). The molecular mode of action of RA may
`of cellular contexts
`involve various potential mechanisms
`reflecting the cellular activities
`of RA. These
`include:
`(a)
`retinoid activation
`of RAR and RXR
`transcription
`factors (2); (b) CBP-dependent
`inhibition of activator
`protein 1 (AP-1) activation (3-6);
`(c) histone acetylation (7-11);
`(d)
`retinoylation
`of cellular proteins
`(12-14);
`and (e) autocrine
`loops
`using growth inhibitors
`transforming growth factor ßor insulin-like
`growth factor binding protein 3 (15-20). A variety of potential
`sig-
`
`Rcceived 12/15/97; accepted 5/13/98.
`The costs of publication of this article were defrayed in part by the payment of page
`charges. This article must
`therefore be hereby marked advertisement
`in accordance with
`18 U.S.C. Section 1734 solely to indicate this fact.
`1Supported in pan by grants from the NIH/National Cancer Institute (USPHS) and the
`United States Department of Agriculture.
`2 To whom requests for reprints should he addressed, at Cancer Biology Laboratory.
`Department of Pathology. College of Veterinary Medicine. Cornell University,
`Ithaca, NY
`14853.
`used are: RA. retinole acid; RAR. RA receptor; RXR. retinoid X
`' The abbreviations
`receptor; RB. retinoblastoma; ERK. extracellular
`signal-regulated
`kinase; MAPK, mito
`gen-activated protein kinase: MEK. mitogen-activated
`protein/ERK kinase; PDGF, plate
`let-derived growth factor: EOF. epidermal growth factor: TPA. 12-O-tetradecanoylphor-
`bol-13-acetate; GST. glutathione S-transferase.
`
`naling pathways have thus been implicated in effecting regulation of
`cell growth and differentiation by RA. In addition to these pathways,
`other as yet unspecified signaling routes may also be involved.
`Iden
`tifying these routes defines the earliest events of the mechanism used
`by RA to ultimately change cell growth and differentiation.
`The mechanism of action of RA has been intensely studied using
`various experimentally susceptible model systems. One of these is the
`HL-60 human myeloblastic
`leukemia cell
`line (21, 22). RA induces
`G,, arrest and myeloid differentiation
`in HL-60 cells, whereas other
`inducers,
`including
`1,25-dihydroxy
`vitamin D,,
`induce monocytic
`differentiation.
`In both cases, hypophosphorylation
`of the RB tumor
`suppressor protein, a putative cell cycle and differentiation
`regulator,
`is induced, consistent with RB as a downstream component of growth-
`regulatory
`signals by RA. The duration of the induced metabolic
`cascade culminating in the onset of G0 arrest and myeloid or mono
`cytic differentiation
`equals approximately
`two division cycles. This
`cascade segregates into two sequential segments (23-25). The "early"
`
`in duration to a division cycle, results in cells primed
`segment, equal
`to differentiate, but without
`lineage specificity.
`Its essential events are
`common to both myeloid and monocytic differentiation.
`If cells at this
`point, which is called precommitment,
`are further exposed to inducer
`for a period equal
`to another division cycle,
`they undergo myeloid or
`monocytic differentiation, depending on the character of the inducer.
`This "late" segment
`thus specifies differentiation
`lineage. The HL-60
`cell line has been used as an archetype model to study the RA-induced
`conversion of a differentiatively
`uncommitted,
`proliferatively
`active
`cell to a differentiated, G()-arrested cell.
`vi
`or 1,25-dihydroxy
`RA-induced HL-60 myeloid differentiation
`tamin D,-induced monocytic differentiation depends on the activity of
`a c-FMS-originated
`signal
`transduction cascade (26, 27). c-FMS is a
`transmembrane
`tyrosine kinase receptor of the PDGF subfamily (28,
`29). Signal
`transduction by such transmembrane
`tyrosine kinase re
`ceptors constitutes the prototypical mitogenic signal transduction cas
`cade. Such cascades have been intensively studied (30). Early re
`sponses in this cascade reflect
`the binding domains on the cytosolic
`domain of the receptor
`and are the activation of src-like kinases,
`phosphatidylinositol
`3'-kinase,
`ras-GTPase-activating
`protein,
`and
`phospholipase Cy. In contrast
`to PDGF receptors,
`the c-FMS cytoso
`lic domain apparently does not bind ras-GTPase-activating
`protein or
`phospholipase Cy. although it does bind the adaptor protein Grb-2 that
`forms an adaptor complex with SOS and may thus facilitate RAS
`activation. GTP activation of RAS and the consequent
`activation of
`the RAF serine-threonine
`kinase initiate a kinase cascade. RAF acti
`vates
`the dual-function MEK kinases,
`and these in turn activate
`MAPKs, prominent examples of which are ERK1 and ERK2, by both
`serine and tyrosine phosphorylation. The MAPKs phosphorylate
`and
`activate transcription factors of the Ets family, a prominent example
`of which is ELK1, which with SRF binds SRE DNA sequences
`to
`induce transcriptional
`activation. This is typically a prelude to DNA
`synthesis with cellular
`recruitment
`to S phase.
`It is thus somewhat
`enigmatic
`that
`in the case of
`these RA-treated HL-60 cells,
`the
`c-FMS-originated
`signal promotes G0 arrest and a cessation of DNA
`
`3163
`
`Downloaded from
`
`
`
`on March 18, 2016. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org
`
`Page 1 of 10
`
`Fluidigm
`Exhibit 1006
`
`

`
`RA INDUCED MEK-DEPENDENT MARK ACTIVATION
`
`the entire
`is whether
`itself
`synthesis. The question that presents
`prototypical mitogenic
`signal
`transduction
`cascade originating with
`c-FMS is recapitulated
`in driving induced G() arrest and differentia
`tion. If this is the case, then RA has redirected the mitogenic outcome
`of a prototypical mitogenic
`signal
`transduction
`cascade
`involving
`oncogenes from mitogenesis
`to G„arrest. The alternative to this is that
`c-FMS can also originate
`a novel
`signaling
`pathway
`promoting
`growth arrest
`instead of mitogenesis. One test of these alternatives
`is
`to determine whether the distal components of the putative mitogenic
`pathway, MEK and MAPK,
`are required to support RA-induced
`differentiation
`and G(, arrest.
`then
`If c-FMS promotes RA-induced cell differentiation and arrest,
`an obviously emerging question is how the c-FMS-originated
`signal
`integrates with the RA signal. RARs and RXRs are members of the
`steroid thyroid hormone receptor
`superfamily. A precedent
`for such
`signal
`integration
`has been established
`in the case of the estrogen
`receptor, where the receptor
`is phosphorylated
`by MAPK (31, 32).
`Phosphorylation
`of the estrogen receptor by EGF-stimulated MAPK
`activity results in activation of the estrogen receptor as a transcription
`factor. The activation is sufficiently potent
`that it can occur even for
`unliganded receptor.
`Interestingly,
`estrogen can also activate MAPK
`(33, 34). In the case of the vitamin D and retinoid receptors
`(RARs
`and RXRs). phosphorylation
`in the absence of ligand also results in
`activation.
`In this case, phosphorylation was induced by okadaic acid,
`a phosphatase inhibitor
`that can also activate MAPKs (35). Consistent
`with this, okadaic acid enhances RA-induced HL-60 cell differentia
`tion. In addition to phosphorylation,
`the expression of RAR-a can also
`be stimulated by EOF (36). In contrast,
`the tyrosine kinase inhibitor
`herbimycin A reduced phosphorylation
`as well as expression.
`It is thus
`apparent
`that
`transmembrane
`tyrosine kinase receptor
`signaling can
`phosphorylate
`and activate retinoid receptors. This suggests the pos
`sibility that c-FMS, a PDGF subfamily receptor, might,
`like an EGF
`receptor,
`also phosphorylate
`and activate retinoid receptors.
`If this
`should be the case,
`it would provide an obvious mechanistic rational
`ization for the interaction between the retinoid and c-FMS signals.
`The present studies were undertaken to assess the involvement of
`c-FMS-originated
`signaling in the mechanism of action of RA. They
`address the question of whether the prototypical mitogenic transmem
`brane tyrosine kinase RAS/RAF signaling pathway attributable
`to
`c-FMS is involved.
`In particular,
`the studies ask if ERK2 activation,
`a putative distal component of this signaling pathway,
`is needed for
`RA action. Two specific mechanistic
`aspects of this question were
`also investigated:
`(a) given that RA has potential nonnuclear effects,
`the effects on ERK2 activation of a retinoid without a nuclear receptor
`in HL-60 cells was determined:
`and (b) because
`a precedent
`for
`MAPK activation of the estrogen receptor exists,
`the ability of RA-
`stimulated ERK2 activation to enhance RAR-a phosphorylation was
`determined.
`
`MATERIALS AND METHODS
`
`Cells and Culture Conditions. HL-60 human myeloblastic leukemia cells
`were maintained
`as stock cultures
`in RPMI 1640 (Life Technologies,
`Inc.,
`Grand Island, NY) supplemented with 5% PCS (Intergen. Purchase, NY). The
`cultures were initiated at a density of 0.2 or 0.1 X IO6 cells/ml
`in 10-ml
`
`the medium every 2 or 3 days,
`of
`replacement
`using complete
`cultures
`1-week period consisted
`of
`two successive
`2-day
`respectively. A typical
`cultures
`followed by a 3-day culture. The doubling
`time of the cells was
`approximately
`21 h.
`In experiments where cells were treated with PD98059 (#9900; New Eng
`land Biolabs, Beverly. MA), cells were recultured at 0.5 X IO6 cells/ml using
`30-ml cultures 24 h before the start of measurements
`(defined as 0 h). PD98059
`was added from a 2 mM stock in DMSO to make a final concentration
`of
`2 x I0~6 M 16 h before the start of measurements. At 0 h. PD98059-treated
`
`cells were recultured at 0.2 X IO6 cells/ml using 30-ml cultures. At this time,
`
`by
`in culture
`the final concentration
`to increase
`readded
`PD98059 was
`2 X 10~6 M, and RA (Sigma Chemical Co., St. Louis, MO) was added from
`a 10~3 M stock in ethanol
`to make a final concentration
`of 10~6 M. At 16 h,
`
`in culture by
`the final concentration
`PD98059 was again added to increase
`2 X IO"6 M. In HL-60 cells,
`these three successive
`additions of 2 x 10~6 M
`
`effect, although they eliminated
`PD98059 had no apparent growth-inhibitory
`MAPK activation. This is a relatively low concentration
`compared to typical
`concentrations
`of 50 x 10~6 M added 1 h before treatment with fibroblast
`
`for 30 min to reduce activated MAPK, p42. or p44 below the
`growth factor
`limits detectable by Western blotting in other cell
`lines, as described by the
`manufacturer's
`specifications.
`In tests of the potential
`toxicity of PD98059.
`the
`growth of cultures
`initiated with 0.1. 1. 2. 10. 20. 50. and 100 x 10~6 M
`PD98059 was assayed over 72 h. A concentration
`of 20 X 10~" M caused G,
`arrest and toxicity. Only a slight G,
`retardation was detectable
`at 10 X 10~6
`
`(a) control cells cultured
`cases were also studied:
`M. Three other experimental
`as described above, but without PD98059 or RA; (b) cells cultured and treated
`with PD98059 as described above, but without RA; and (c) cells cultured and
`treated with RA, but without PD98059. Cells were harvested
`from these
`cultures
`for assays of cell growth, cell cycle distribution,
`cell differentiation,
`RB expression,
`activated MAPK expression,
`total MAPK protein expression,
`RAR-a
`expression,
`and in vitro MAPK kinase
`activity
`against ELK1 or
`RAR-a. All data shown are typical of three repeats.
`In the case of delayed addition of PD98059.
`cultures were initiated at
`0.2 X IO6 cells/ml
`in 30-ml cultures in the absence or presence of I0~6 M RA.
`
`of 2 /J.Mand added
`PD98059 was added at 16 h to make a final concentration
`again at 32 h to increase the final concentration
`by 2 /AM.Cells were harvested
`at the indicated times to assay cell differentiation,
`cell cycle arrest, cell density,
`and RB expression.
`the retinoid
`the effects of the RAR--y ligand.
`to determine
`In experiments
`ligand CD437 was used. Cultures were initiated at a cell density of 0.2 X IO6
`cells/ml using 30-ml cultures with no addition (control).
`10~6 M RA. or IO"6
`M CD437. CD437 is a RAR--y selective ligand with IC50 displacement
`values
`for 9-c/i-RA-bound
`RAR-a, RAR-/3, and RAR-y of 7300. 3900. and 90 nM,
`respectively. The CD437 was added to cultures from a stock solution of 5 mM
`in ethanol.
`stored at —¿(cid:3)86°C,and protected from light. CD437 was a generous
`gift of Drs. Art Levin and Micheal Klaus (Hoftman-LaRoche.
`Inc.. Nutley, NJ
`and Basel, Switzerland). At indicated times, cells were harvested for Western
`analysis of activated MAPK and total MAPK as well as cell density and
`differentiation.
`identical cultures
`In the case of in vitro kinase activity assays for MAPK,
`manipulated with or without PD98059 or RA as described above were initiated
`at staggered times to allow their simultaneous
`harvest. The kinase assays were
`thus performed
`simultaneously
`for all
`time points being compared. For
`the
`assay of MAPK activity on an ELK1 substrate, HL-60 cells were cultured as
`described above with or without PD98059 or RA.
`In the case of assays of
`MAPK activity in TPA-treated cells. TPA from a 0.5 mg/ml stock was added
`to 4 X IO6 exponentially
`growing cells resuspended
`in 10 ml of medium to
`make a final concentration
`of 1 X 10~6 M TPA. After 15 min of treatment,
`the
`
`out of the suspension. For the assay of
`cells were harvested by centrifugation
`MAPK activity on a RAR-a
`substrate, RAR-a was immunoprecipitated
`from
`exponentially
`proliferating HL-60 cells and used in lieu of ELK1 in the kinase
`assay.
`Assays of cell growth by meas
`and Differentiation.
`Assays of Growth
`uring cell density and distribution
`in the cell cycle and assays of cell differ
`entiation detected by inducible oxidative metabolism were performed as de
`scribed previously (25, 37). Briefly, cell density in experimental
`cultures was
`measured by repeated counts with a hemacytometer. Viability was assessed by
`the exclusion of 0.2% trypan blue dye and was routinely at least 95% in all
`cultures. The distribution
`of cells
`in the cell cycle was determined
`by flow
`cytometry
`using propidium iodide-stained
`nuclei. A total of 0.5 X IO6 cells
`
`in 0.5 ml of hypotonie
`was harvested at each indicated time and resuspended
`propidium iodide solution (50 mg/liter propidium iodide.
`1 g/liter
`sodium
`citrate, and 0.1 % Triton X-100) and stored, refrigerated,
`and protected from the
`light until analyzed. Flow cytometric analysis was done with a multipararneter
`dual-laser
`fluorescence-activated
`cell sorter
`(EPICS: Coulter Electronics, Hi-
`aleah, FL) using 200 mW of 488 nm excitation from a tunable argon ion laser.
`Functional differentiation
`to a mature myelomonocytic
`phenotype
`capable of
`inducible
`oxidative metabolism was
`assayed
`by
`phorbol
`12-myristate
`
`3164
`
`Downloaded from
`
`
`
`on March 18, 2016. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org
`
`Page 2 of 10
`
`Fluidigm
`Exhibit 1006
`
`

`
`RA INDUCED MEK-DEPENDENT MARK ACTIVATION
`
`resulting in
`oxidative metabolism,
`13-acetate (Sigma Chemical Co.)-inducible
`an intrucellular
`reduction of nitroblue tetrazolium to formazan by Superoxide.
`A total of 0.2 x 10'' cells were harvested at the indicated times and resus-
`
`200
`tetrazolium in PBS containing
`pended in 0.2 ml of 2 mg/ml nitroblue
`ng/ml phorbol 12-myristate
`13-acetate. The cell suspension was incubated for
`20 min in a 37"C water bath with occasional vortexing and then scored using
`a hemacytometer
`for the percentage of cells expressing intracellular blue-black
`formazan precipitated
`by superoxide.
`Western Analysis of RB, MAPK, and Activated MAPK. Western blot
`ting of RB. MAPK. and activated MAPK was done using whole-cell
`lysates
`from cells cultured as indicated. At
`the indicated times. 10* cells were har
`vested and fixed in I ml of 90% methanol at —¿(cid:3)80°Cas described previously
`(38. 39). The cells were stored at -20°C until analysis by SDS-PAGE. Cells
`
`[6% SDS. 4 M urea, 4 mM EDTA.
`were solubilized in 50 /¿Iof loading buffer
`blue, and 35 ¿¿I/mlß-mercaptoeth-
`125 mM Tris (pH 6.9). 0.25% bromphenol
`anol] by immersion in a boiling water bath for 5 min. SDS-PAGE was done
`using a 4% stacking
`gel and a 10% resolving
`gel with a 37.5:1 ratio of
`acrylamide:bis.
`Samples were electrophoresed
`for approximately
`1200 V X h
`(typically 75 V for 16 h). A total of I X IO6cells was loaded per lane. Proteins
`were electrotransferred
`(Trans Blot Cell: Bio-Rad Inc., Hercules. CA) from the
`gel
`to a nitrocellulose membrane. Transfer was done at 0.4 A for 2 h. The
`resulting membrane was blocked
`by immersion
`overnight
`at 4°C in 5%
`
`powdered milk and 0.05% Tween 20 in PBS. The membranes were probed
`with antibodies detecting the phosphorylated
`and unphosphorylated
`forms of
`RB. ERK2. and ERK1;
`the activated forms of ERK2 and ERK1 bearing a
`Thr-Glu-Tyr motif with Thr11" and Tyr1*5 phosphorylation:
`or the phospho
`rylated and unphosphorylated
`forms of RAR-a. The antibody to detect RB [RB
`Gene Product
`(mAbl
`) Monoclonal Antibody: Zymed Laboratories. South San
`Francisco. CA] was used at 0.4 /j.g/ml PBS with an overnight
`incubation at
`room temperature. The antibody to detect ERK2 and ERKI
`[C-14 (SC154
`rabbit polyclonal
`antibody): Santa Cruz Biotechnology.
`Inc.. Santa Cruz. CA]
`was used at 0.1 X 10~6 g/ml
`in PBS with a 1-h incubation at room temper
`
`in 300 /*! of lysis buffer containing
`in 1 ml of ice-cold PBS and resuspended
`25 mM HEPES (pH 7.5), 25 mM ß-glycerophosphate. 3 mM EDTA. 3 mM
`EGTA. 250 mM NaCl. 1% Triton X-100, 2 mM sodium orthovanadate.
`2 HIM
`DTT. 0.01 mg/ml aprotinin. 0.01 mg/ml
`leupetin. and I mM phenylmelhylsul-
`fonyl
`fluoride (added freshly before use) for 15 min on ice. The whole-cell
`lysate was clarified by centrifugation
`using a microfuge
`for 5 min at 4°C.and
`
`by adding 10 ^.1of an
`ERK activity in the supernatant was immunoprecipilated
`ERK-specific
`antibody (C-14; Santa Cruz Biotechnology.
`Inc.) and 20 jitl of
`protein A/G Plus-agarose
`beads
`(A/G Plus Agarose
`562(X)2: Santa Cruz
`Biotechnology.
`Inc.). This
`suspension was mixed for 3-4
`h at 4°C.The
`immunoprecipitates were washed three times (3500 rpm. 2.5-min microcen-
`trifugation)
`in 0.5 ml of
`lysis buffer and once in 0.5 ml of kinase buffer
`containing
`20 mM HEPES (pH 7.5). 20 mM magnesium chloride.
`25 mM
`ß-glycerophosphate. 100 JIM sodium orthovanadate.
`2 mM DTT. and 20 /XM
`ATP. The immunoprecipitated
`kinase complex was then resuspended in 50 ¡JL\
`of kinase buffer with IO /¿Ciof [-y-12P]ATP (BLU502H:
`specific activity. 3000
`
`in 50 mM Tricine; Easy Tides: DuPont New England
`5 mCi/ml
`Ci/mmol:
`Nuclear, Boston, MA), and approximately
`1 fig of GST-ELK 1 fusion protein
`(1 ¿¿g/jil)was added. The kinase reaction proceeded for 30 min at 3()°Cwith
`
`frequent mixing. The reaction was terminated by doubling the volume with 2X
`SDS lysis buffer and immersion
`in a boiling water bath for 2.5 min. The
`reaction mixture was then subjected to SDS-PAGE using a 4% stacking gel
`and a 10% resolving gel run at 150 V for 3-4 h. The reaction products were
`visualized by autoradiography.
`of
`In vitro MAPK activity on RAR-a was measured by the phosphorylalion
`RAR-a
`synthesized
`in a reticulolysate
`expression
`system. RAR-a was ex
`pressed from a RAR-a
`cDNA fragment cloned as an £iV»RIfragment
`into a
`pSG5 expression vector using a T7 bacterial promoter
`for reliculocyte
`expres
`sion (43. 44). Reticulocyte
`expression was performed according to the instruc
`tions of the manufacturer
`(TNT T7 Reticulocyte Lysate System: Promega.
`Inc.).
`
`rabbit
`ature. The antibody used to detect activated ERKI and ERK2 (V667I
`polyclonal
`antibody; Promega.
`Inc.. Madison. Wl) was used at 0.025 10~6
`
`RESULTS
`
`in 0.1 % BSA with a 2-h incubation at room temperature. Detection was
`g/ml
`performed using a horseradish peroxidase-conjugated
`secondary antimurine or
`antirabbit
`antibody and enhanced chemiluminescence
`(ECL Kit; Amersham,
`Arlington Heights.
`IL) following the manufacturer's
`instructions. The rabbit
`
`polyclonal antibody used to detect RAR-a was RPa(F). which was a generous
`gift of Dr. Pierre Chambón (INSERM,
`Facultéde Médecine,Strasbourg.
`France; Ref. 40). It was used at a 1:1000 dilution with a 1-3-h incubation at
`room temperature.
`of UK I or RAR-a.
`In Vitro Kinase Assays of MAPK Phosphorylation
`In vitro MAPK kinase activity was measured by phosphorylation
`of a recom
`binant ELK 1 substrate, one of the in vint targets of MAPK. The transcriptional
`activation domain of human ELK1 (amino acids 307-428) was obtained by
`PCR from HeLa cell cDNA and confirmed by nucleotide
`sequence analysis
`(41). The ELK1 transactivation
`domain was cloned into pGEX3 to produce a
`GST-ELK 1 fusion gene. The GST-ELK 1 fusion protein was expressed
`in
`bacteria and was partially purified as described previously (42).
`The in vitro kinase assay for ERKs was performed as described previously
`(41). Briefly, after the indicated treatments, 4 X IO6 HL-60 cells were washed
`
`RA Causes MEK-dependent ERK2 Activation. RA caused the
`activation of ERK2 detected by its phosphorylation
`at Thr""
`and
`Tyrlx5. HL-60 cells were cultured in the absence (control) or presence
`of 10~6 M RA. At sequential
`times, cells were harvested for Western
`
`analysis. Fig. 1 shows the Western blot analysis for cells at 0, 1,4, 12,
`16, 24, 48, 72, and 96 h using an antibody specific for the phospho
`rylated Thr183 and Tyr'"5 of a Thr-Glu-Tyr motif characterizing
`activated ERKI or ERK2. (As discussed below. HL-60 cells express
`very little ERKI
`relative to ERK2, making ERK2 the predominant
`species.) All
`lanes were loaded with the same number of cells. RA
`increased the amount of activated ERK2 per cell by 4 h. Fig. 1 also
`shows the results when Western blotting is performed with an anti
`body detecting both unphosphorylated
`and phosphorylated
`ERK2.
`Phosphorylation
`causes a mobility shift on SDS-PAGE,
`resulting in
`distinguishable
`faster-migrating
`unphosphorylated ERK2 and slower-
`migrating phosphorylated ERK2. By 4 h of RA treatment,
`there was
`
`1C
`RA I.
`
`KA.4C
`
`RA ('+ RA*12C
`
`R A
`
`( « HA.16<
`
`RA
`
`C* RA +2-1C
`
`H \
`
`(
`
`.iH \-•••••^•H72i
`. K \ .481
`
`
`
`
`
`
`
`K -.
`
`
`
`k v C« R\«•¡•••••••196("K \
`
`(
`
`.
`
`K \
`
`active MAPK
`
`F.rk-2
`
`for cells that were untreated controls (C). treated with RA (RA). treated with PD98059 (C+).
`l. Western analysis of activated ERK2 (top panels) and lotal ERK2 (bottom panels)
`Fig.
`or treated with RA plus PD98059 (RA + ) at the indicated limes (hours) of RA Ireatment. PD98059 (final concentration.
`2 ¡a»)was added 16 h before the addition of RA at 0 h. when
`it was added again, and 16 h after the addition of RA. Activated ERK2 was detected by Western blotting using an antibody specific for ERK2 dual-phosphorylated
`at the Thr-Glu-Tyr
`motif. ERK2 detected by Western blotting showed the distinct slower-mobility
`band characteristic of phosphorylated F.RK2. which paralleled the occurrence of activated ERK2. and
`showed that a loss of activated ERK2 due to PD98059 reflected a loss of the phosphorylated ERK2 and not the elimination of the total amount of ERK2. Data in this and all other
`figures are typical of three repeats.
`
`3165
`
`Downloaded from
`
`
`
`on March 18, 2016. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org
`
`Page 3 of 10
`
`Fluidigm
`Exhibit 1006
`
`

`
`RA INDUCED MEK-DEPENDENT MARK ACTIVATION
`
`Cell
`
`cycle
`
`arrest
`
`00-90-80-,
`
`._—A
`
`—¿(cid:3)•*—CRAOPD98059RA+PD98059
`
`__.
`
`-A
`
`72
`
`96
`
`time
`
`(hrs)
`
`70-
`
`60
`
`SO
`
`40
`
`O *
`
`Fig. 4. The percentage of cells with G, DNA measured by flow cytometry for untreated
`controls (O), RA-treated HL-60 cells (A), PD98059-treated HL-60 cells (•).or RA plus
`PD98059-treated HL-60 cells (A) at
`the indicated times of RA treatment. Cells were
`treated as described in the Fig. 1 legend.
`
`RA. Cells were pretreated with PD98059 before the addition of RA to
`allow clearance of existing activated MEK and ERK2. Fig. 1 shows
`that
`treatment of control cells with PD98059 resulted in a loss of
`activated ERK2 detected by Western blotting for activated ERK2,
`which was verified by a loss of the lower-mobility
`phosphorylated
`ERK2 detected by blotting for total ERK2. The absence of detectable
`activated ERK2 relative to untreated controls was apparent
`at
`the
`outset of RA addition and thereafter
`up to 72 h. At 96 h,
`the
`attenuation in untreated control cells is probably due to nutritional
`deprivation
`and parallels
`reduced growth and culture degeneration.
`Inhibition of ERK2 activation was also verified by in vitro kinase
`activity as shown in Fig. 2. In RA-treated
`cells,
`the RA-induced
`activation
`of ERK2 is also suppressed
`by PD98059. Assayed by
`Western blotting for either activated ERK2 or phosphorylated ERK2,
`ERK2 activation was suppressed during the 48 h typically needed for
`RA to induce the onset of differentiation and G,,-specific growth arrest
`(Figs. 3 and 4). In vitro kinase assays also corroborated
`that RA-
`induced ERK2 activation was inhibited by PD98059. At later times,
`RA could apparently slightly overcome
`the inhibitor by causing a
`minor amount of activated ERK2, probably reflecting a loss of effec
`tiveness of the drug with prolonged times. For example,
`some RA-
`induced ERK2 in vitro kinase activity is apparent at 48 h (Fig. 2), and
`some activated ERK2 is apparent by Western blotting at 72 h. It
`should be noted that the 2 /J.Mdose of PD98059 used repeatedly here
`is relatively low compared to typical values of 50 /AMused to suppress
`growth factor activation of ERKs. There was no apparent
`loss of
`viability or growth inhibition associated with this treatment.
`HL-60 cells express predominantly
`the M, 42,000 ERK2 and only
`a minor amount of A/r 44,000 ERK1. The Western analysis of whole-
`cell
`lysates in Fig. 1 using an antibody detecting activated ERK1 or
`ERK2 reveals only the Mr 42,000 ERK2. Likewise, Western blotting
`using an antibody detecting total ERK1 or ERK2 indicates only the
`ERK2. Using prolonged autoradiographic
`film exposures,
`it has been
`possible
`to detect
`the M, 44,000 ERK1 in whole-cell
`lysates of
`untreated HL-60 cells and of HL-60 cells treated with RA for 48 h.
`Densitometric
`scanning
`of
`these Western
`blot
`autoradiographs
`showed that approximately 95% of the ERK was ERK2, and 5% was
`ERK1 (data not shown). The ratio explains why only the ERK2
`appears in most films. RA induced no apparent change in the ratio. It
`should be noted that due to the limited number of lanes available per
`3166
`
`band corresponding to the phosphoryl-
`an enhanced slower-migrating
`ated ERK2,
`thereby verifying the Western analysis detecting only
`activated ERK2. The Western blot for total ERK2 protein also indi
`cates that whereas RA increased the amount of activated ERK2,
`there
`was no gross change detectable in the total amount of ERK.2 per cell.
`RA-induced ERK2 activation persisted 48 h after the addition of
`RA. The Western blots for activated ERK2 and total ERK2 show
`enhanced activated ERK2 by 4 h, but due to saturation of the auto-
`radiograph, which occurred for the strong signal to visualize the weak
`signal,
`it is not clear whether the enhancement persisted by 24 h. The
`in vitro kinase activity of ERK2 immunoprecipitated
`from HL-60
`cells after 0, 24, and 48 h of culture in the absence or presence of RA
`was measured. Kinase activity was measured by phosphorylation
`of a
`recombinant ELK1 fragment, an in vivo substrate for ERK2. Fig. 2
`shows
`the resulting
`autoradiograph. RA induced increased ERK2
`kinase activity present at 24 and 48 h compared with that at 0 h. Thus,
`RA-activated ERK2 persisted until
`the onset of differentiation
`and
`G()-specific growth arrest
`that was apparent by 48 h (Figs. 3 and 4).
`RA-induced ERK2 activation in HL-60 cells depended on MEK
`activation. Treatment with PD98059,
`a specific
`inhibitor of MEK
`phosphorylation
`and activation (45), resulted in the loss of activated
`ERK2 from HL-60 cells. PD98059 suppressed RA-induced ERK2
`activation. HL-60 cells were pretreated 16 h before the addition of RA
`(0 h), at the time of RA addition, and 16 h after the addition of RA.
`Control cells were treated analogously,
`but without
`the addition of
`
`Oh
`
`24h
`
`48h
`
`C C+ C RA C+ RA+ C RA C+ RA+
`
`
`
`•¿(cid:3)«•»-
`
`Elk-1
`
`Fig. 2. In vitro ERK.2 kinase activity on ELK1 for cells that were untreated controls
`(C>. treated with RA iKA>. treated with PD98059 <C+ ), or treated with RA plus PD98059
`(RA + ) at the indicated times of RA treatment, as described in the Fig. 1 legend. Although
`the persistent RA-enhanced ERK2 activation did not appear (o he great at 24 and 48 h due
`to X-ray film signal saturation,
`the kinase activity verifies that ERK2 activity was still
`strongly enhanced by RA at 24 and 48 h.
`
`Differentiation
`
`•¿(cid:3)•«—c
`
`-A— RA
`—¿(cid:3)•OPD98059
`-A—
`RA+PD98059
`
`A
`
`""
`
`A
`
`100 i
`
`80 -
`
`40 -
`
`20 •¿(cid:3)
`
`I
`
`24
`
`48
`
`72
`
`96
`
`time
`
`(hrs)
`
`differentiation marker,
`the functional
`of cells expressing
`Fig. 3. The percentage
`inducihle oxidative metabolism, detected by intracellular
`reduction of nitroblue tetrazo-
`lium (NUT) for untreated controls
`(O). RA-treated HL-60 cells (A), PD98059-lreated
`HL-60 cells (•).or RA plus PD98059-treated HL-60 cells (A) at the indicated times of
`RA treatment. Cells were treated as described in the Fig. 1 legend.
`
`Downloaded from
`
`
`
`on March 18, 2016. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org
`
`Page 4 of 10
`
`Fluidigm
`Exhibit 1006
`
`

`
`RA INDUCED MEK-DEPENDENT MARK ACTIVATION
`
`occurred by 48 h. The fraction of
`induced functional differentiation
`the population that was differentiated
`increased progressively
`there
`after, with approximately 90% of the whole population differentiated
`by 72 h. Untreated control cells showed only low levels of spontane
`ous differentiation
`(approximately
`10%) that did not increase during
`culture.
`In cells where ERK2 activation was suppressed by the addi
`tion of PD98059, RA-induced differentiation was grossly suppressed.
`Only approximately 20% of the cells differentiated by 72 h, and few
`more differentiated thereafter. RA-induced functional differentiation
`thus required MEK-dependent ERK2 activation. PD98059 (2 JU.M),as
`used, had no observable effect on cell viability, which was routinely
`at least 95% in all of the cases (see also "Materials and Methods").
`RA-induced G()-specific growth arrest depended on MEK-induced
`ERK2 activation.
`In these same experiments,
`the percentage of cells
`with G,-G() DNA and the cell density were assayed during culture.
`The percentage of cells with G,-G() DNA was determined by flow
`cytometry of propidium iodide-stained samples harvested at the indi
`cated times.