PDE4 inhibition drives resolution of neutrophilic inflammation by inducing apoptosis in a...

PDE4 inhibition drives resolution ofneutrophilic inflammation by inducing

apoptosis in a PKA-PI3K/Akt-dependent andNF-�B-independent manner

Lirlandia P. Sousa,*,† Fernando Lopes,†,‡ Douglas M. Silva,† Luciana P. Tavares,†

Angelica T. Vieira,† Barbara M. Rezende,† Aline F. Carmo,† Remo C. Russo,†

Cristiana C. Garcia,† Claudio A. Bonjardim,§ Ana L. Alessandri,†, Adriano G. Rossi,Vanessa Pinho,†,‡ and Mauro M. Teixeira†,1

*Setor de Patologia Clınica-Colegio Tecnico, †Imunofarmacologia-Departamento de Bioquımica e Imunologia, ‡Departamentode Morfologia, and §Departamento de Microbiologia, Instituto de Ciencias Biologicas, Universidade Federal de Minas Gerais,

Belo Horizonte, Brazil; and ��Medical Research Council Centre for Inflammation Research, Queen’s Medical Research Institute,University of Edinburgh, Edinburgh, United Kingdom

RECEIVED AUGUST 10, 2009; REVISED DECEMBER 30, 2009; ACCEPTED JANUARY 2, 2010. DOI: 10.1189/jlb.0809540

ABSTRACTPDE4 inhibitors are effective anti-inflammatory drugswhose effects and putative mechanisms on resolutionof inflammation and neutrophil apoptosis in vivo are stillunclear. Here, we examined the effects of specificPDE4 inhibition on the resolution of neutrophilic inflam-mation in the pleural cavity of LPS-challenged mice.LPS induced neutrophil recruitment that was increasedat 4 h, peaked at 8–24 h, and declined thereafter. Suchan event in the pleural cavity was preceded by in-creased levels of KC and MIP-2 at 1 and 2 h. Treatmentwith the PDE4 inhibitor rolipram, at 4 h after LPS admin-istration, decreased the number of neutrophils and in-creased the percentage of apoptotic cells in the pleuralcavity in a PKA-dependent manner. Conversely, de-layed treatment with a CXCR2 antagonist failed to pre-vent neutrophil recruitment. Forskolin and db-cAMPalso decreased the number of neutrophils and in-creased apoptosis in the pleural cavity. The proapop-totic effect of rolipram was associated with decreasedlevels of the prosurvival protein Mcl-1 and increasedcaspase-3 cleavage. The pan-caspase inhibitor zVAD-fmk prevented rolipram-induced resolution of inflamma-tion. LPS resulted in a time-dependent activation of Akt,which was blocked by treatment with rolipram or PI3Kand Akt inhibitors, and PI3K and Akt inhibitors also en-hanced apoptosis and promoted neutrophil clearance.Although LPS induced NF-�B activation, which wasblocked by rolipram, NF-�B inhibitors did not promote

resolution of neutrophil accumulation in this model. Inconclusion, our data show that PDE4 inhibition resolvesneutrophilic inflammation by promoting caspase-de-pendent apoptosis of inflammatory cells by targeting aPKA/PI3K/Akt-dependent survival pathway. J. Leukoc.Biol. 87: 000–000; 2010.

IntroductioncAMP is an important intracellular second messenger pro-duced after adenylate cyclase activation that regulates differentcellular processes by cAMP effectors [1, 2]. PDEs control intra-cellular cAMP levels by catalyzing their hydrolysis and inactivat-ing these second messengers. PDE4 isoenzymes are predomi-nant isoenzymes in inflammatory cells [3]. Indeed, PDE4 in-hibitors induce an increase in the intracellular levels of cAMPand have potent anti-inflammatory activity on various leuko-cytes [4–7]. Intracellular actions of cAMP can be mimicked byadministration of the cell-permeable analog db-cAMP or theadenylate cyclase activator forskolin. Similar to rolipram, a pro-totypic PDE4 inhibitor, these compounds inhibit leukocytefunction and possess significant anti-inflammatory effects invivo [8]. A few studies have shown that inhibition of PDE4 en-zymes and increase of intracellular levels of cAMP may modifythe survival of human neutrophils in vitro [9–13]. However,the effects of PDE4 inhibitors or other cAMP-elevating/mimet-ics on neutrophil apoptosis and survival in vivo are not wellestablished.

Neutrophils, a major type of blood leukocytes, have a keyrole in the host defense against bacterial, fungal, and viral in-fections. Neutrophils are terminally differentiated cells that

1. Correspondence: UFMG, Bioquimica e Imunologia/ICB, Av Antonio Car-los 6627, Pampulha, Belo Horizonte, MG 31270-901, Brazil. E-mail:[email protected]

Abbreviations: B-CLL�B cell chronic lymphocytic leukemia, db-cAMP�dibutyryl cAMP, EPAC�exchange protein directly activated by cAMP,i.pl.�intrapleural, KC�keratinocyte chemoattractant, Mcl-1�myeloidcell leukemia 1, P�phospho, PDE4�phosphodiesterase 4, PDTC�pyrrolidine dithiocarbamate, PKB�protein kinase B, SRE�serum re-sponse element, zVAD-fmk�benzyloxycarbonyl-Val-Ala-Asp-fluorom-ethylketone

Article

0741-5400/10/0087-0001 © Society for Leukocyte Biology Volume 87, May 2010 Journal of Leukocyte Biology 1

Epub ahead of print January 26, 2010 - doi:10.1189/jlb.0809540

Copyright 2010 by The Society for Leukocyte Biology.

have a limited lifespan [14]. The apoptosis of neutrophils hasbeen characterized as an important factor in the resolution ofinflammation [14–17]. Apoptotic neutrophils in tissues arerecognized and phagocytosed by macrophages, which limittissue injury caused by neutrophils at the sites of inflammation[18, 19]. In contrast, the permanence of neutrophils in tissuesexacerbates inflammation by releasing proteases, reactive oxy-gen species, and proinflammatory mediators [19]. Therefore,the induction of neutrophil apoptosis could be of potentialbenefit in the control of inflammatory diseases [14–17]. Al-though the apoptotic program in neutrophils is an intrinsiccell process, the rate of apoptosis can be altered dramaticallyby a number of agents [20]. Therefore, understanding themechanisms involved in neutrophil recruitment, activation,and survival in inflammatory sites may be useful for the devel-opment of novel pharmacological therapies to control inflam-matory diseases.

PI3K and the downstream serine/threonine kinase Akt/PKBare involved in the regulation of major leukocyte events, in-cluding growth, proliferation, recruitment, activation, and sur-vival [21]. The PI3K/Akt pathway has a role in mediating theantiapoptotic and proinflammatory signals triggered by LPS,GM-CSF, and TNF-�, mainly by influencing proteins and tran-scription factors that produce antiapoptotic signals in granulo-cyte apoptosis [22–25]. In this study, we have investigated themechanisms by which PDE4 inhibition and other cAMP-elevat-ing agents or mimetics influenced the resolution of neutro-philic inflammation in vivo.

MATERIALS AND METHODS

AnimalsAll procedures described here had prior approval from the Animal EthicsCommittee of Universidade Federal de Minas Gerais (Brazil). MaleBALB/c mice (8–10 weeks), obtained from the Bioscience Unit of Institutode Ciencias Biologicas (Brazil), were housed under standard conditionsand had free access to commercial chow and water.

Drugs, reagents, and antibodiesRolipram (Biomol, Plymouth Meeting, PA, USA); wortmannin, forskolin,and Akt inhibitor IV (from Calbiochem, San Diego, CA, USA); H89 (TocrisBioscience, Ellisville, MO, USA); and LY294002 (Alamone Labs, Jerusalem,Israel) were diluted in DMSO and further in PBS. db-cAMP and PDTC(Sigma-Aldrich, St. Louis, MO, USA) were diluted in PBS. Rabbit anti-P-Akt(Ser 473), anti-Akt, and anti caspase-3 were from Cell Signaling Technology(Beverly MA, USA). Rabbit anti Mcl-1 or secondary anti-rabbit peroxidase-conjugate antibodies were purchased from Santa Cruz Biotechnology(Santa Cruz, CA, USA). Anti �-actin, anti-mouse peroxidase-conjugate anti-bodies, zVAD-fmk, and LPS (from Escherichia coli serotype O:111:B4) werefrom Sigma-Aldrich.

Leukocyte migration into the pleural cavity inducedby LPSMice received an i.pl. administration of LPS (250 ng/cavity) or PBS. Thisdose of LPS was shown previously to be optimal to induce efficient recruit-ment of neutrophils in the pleural cavity of mice (data not shown). Cellspresent in the pleural cavity were harvested at different times after adminis-tration of LPS by washing the cavity with 2 ml PBS and total cell countsperformed in a modified Neubauer chamber using Turk’s stain. For theexperiments evaluating leukocyte apoptosis, infiltrating leukocytes were ex-

amined 2 h (caspase-3 cleavage and Mcl-1 expression) and 4 h (morpho-logic apoptosis) after drug treatment. Differential cell counts were per-formed on cyto-centrifuge preparations (Shandon III) stained with May-Grunwald-Giemsa using standard morphological criteria to identify celltypes. The results are presented as the number of cells/cavity. To analyzethe effect of drugs on early events that precede resolution, such as signaltransduction pathways (Akt, NF-�B, Mcl-1, and caspase-3), compounds weregiven 4 h after LPS challenge and cells in the pleural wash harvested 2 hlater (4 h of LPS�2 h of drugs).

Measurement of chemokinesThe levels of KC and MIP-2 were measured in frozen supernatants ob-tained from pleural cavity washes after different time-points of LPS chal-lenge by ELISA using commercially available antibodies, according to theprocedures supplied by the manufacturer (R&D Systems, Minneapolis, MN,USA).

Treatment with drugsInitial experiments evaluated the optimal dose of rolipram (0.6–6.0 mg/kg, 15–150 �g/mouse) administered systemically (i.p.) 4 h after i.pl. LPSchallenge. The dose of 150 �g was found to be optimal for the resolutionof inflammation and was used in all subsequent experiments. This dose hasbeen shown to be effective in other experimental systems [26]. Otherdrugs used in this study were: forskolin (0.4 mg/kg, 10 �g/mouse), db-cAMP (4 mg/kg, 100 �g/mouse), LY294002 and Wortmannin (1.0 mg/kg,30 �g/mouse,) Akt inhibitor IV (0.04 and 0.12 mg/kg, 1 and 3 �g/mouse), H89 (2.6 mg/kg, 80 �g/mouse), and SN-50 (10 �g/mouse),which were given in the pleural cavity in a volume of 100 �l, 4 h after i.pl.LPS challenge. The zVAD-fmk (1.0 mg/kg, 30 �g/mouse) was given sys-temically (i.p.) 15 min before rolipram. PDTC (100 mg/kg) was given sys-temically (i.p.) 4 h after LPS challenge. Drugs were dissolved in DMSO anddiluted further in PBS. Control mice received drug vehicle only.

Assessment of leukocyte apoptosisApoptosis was assessed morphologically as described previously by us [26,27]. Briefly, cells (5�104) collected 8 h after LPS administration were cyto-centrifuged, fixed, and stained with May-Grunwald-Giemsa and countedusing oil immersion microscopy (�100 objective) to determine the propor-tion of cells with distinctive apoptotic morphology (cells presented chroma-tin condensation, nuclear fragmentation, and formation of apoptotic bod-ies out or inside macrophages). Twenty-five fields (at least 500 cells) werecounted/slide, and results are expressed as the percentage of neutrophilswith apoptotic morphology. The cleavage of caspase-3, as evaluated byWestern blot analysis, was used as a biochemical marker of apoptosis.

Lysate preparation and Western blot analysisInflammatory cells harvested from the pleural cavity were washed with PBS,and whole cell extracts were prepared as described [26, 28]. Proteinamounts were quantified with the Bradford assay reagent from Bio-Rad(Hercules, CA, USA). Extracts (40 �g) were separated by electrophoresison a denaturing 10–15% polyacrylamide-SDS gel and electrotransferred tonitrocellulose membranes, as described [26, 28]. Membranes were blockedovernight at 4°C with PBS containing 5% (w/v) nonfat dry milk and 0.1%Tween-20, washed three times with PBS containing 0.1% Tween-20, andthen incubated with specific primary antibodies (anti-P-AktSer 473, anti-Akt,anti-caspase-3, anti-Mcl-1, or anti-�-actin) using a dilution of 1:1000 in PBScontaining 5% (w/v) BSA and 0.1% Tween-20. After washing, membraneswere incubated with appropriate HRP-conjugated secondary antibody (1:3000). Immunoreactive bands were visualized by using an ECL detectionsystem, as described by the manufacturer (GE Healthcare, Piscataway, NJ,USA).

EMSABand shift assay was carried out of 10 �g nuclear extracts essentially as de-scribed [26, 29], using a 5� [32P]-end-labeled, double-stranded probe (only

2 Journal of Leukocyte Biology Volume 87, May 2010 www.jleukbio.org

one strand is shown) corresponding to the consensus binding site of NF-�B(5�-AGT TGA GGG GAC TTT CCC AGG C-3�). Heterologous competitionassays were performed with a 100-fold molar excess of cold oligonucleotidecorresponding to c-fos SRE.

Statistical analysisAll results are presented as the mean � sem. Normalized data were ana-lyzed by one-way ANOVA, and differences between groups were assessedusing the Student-Newman-Keuls post-test. A P value �0.05 was consideredsignificant. Calculations were performed using the Prism 4.0 software pro-gram for Windows (GraphPad Software, San Diego, CA, USA).

RESULTS

Rolipram diminishes neutrophil accumulation into thepleural cavity of LPS-challenged mice in a PKA-dependent mannerThe i.pl. injection of LPS (250 ng) induced a time-dependentinflux of neutrophils into the pleural cavity of mice that wasdetectable at 2 h and reached maximal at 8–24 h. Neutrophilswere diminished at 48 h (Fig. 1A), and complete resolutionoccurred only at 72 h (data not shown). There was no in-crease in number of mononuclear cells at 4–24 h after LPSchallenge (data not shown). The cellularity of the pleural cav-ity was 98–100% of mononuclear cells in PBS-injected miceand 50–60% of neutrophils and 40–50% mononuclear cells inLPS-injected mice. We also evaluated the levels of KC andMIP-2, two major CXC chemokines involved with neutrophilrecruitment. Levels of both chemokines increased 1 and 2 h

after LPS challenge and declined to basal levels thereafter(Fig. 1B).

The next experiments were designed to investigate whetherrolipram, a selective PDE4 inhibitor [4, 5], could interferewith neutrophil accumulation in the pleural cavity. To thisend, we treated mice with rolipram 4 h after LPS challenge,when inflammatory cell influx was established already but hadnot peaked yet. The pleural lavage was performed 4 h afterrolipram treatment, i.e., 8 h after LPS administration (as indi-cated by arrows in Fig. 1A). As seen in Figure 1C, rolipramdose-dependently reduced the number of neutrophils in thepleural cavity, without changing the number of mononuclearcells (LPS�vehicle: 9.6�1.2�105 mononuclear/cavity;LPS�rolipram 150 �g/mouse: 8.8�1.7�105 mononuclear/cavity; n�5).

Treatment with reparixin (an allosteric inhibitor of CXCR2)4 h after LPS challenge failed to affect the accumulation ofneutrophils (Fig. 1D). In the same experiment, post-treatmentwith rolipram greatly decreased neutrophil accumulation inthe cavity (Fig. 1D). The lack of effect of delayed (at 4 h)treatment with reparixin is consistent with the observation thatCXCR2-active chemokines are only detected at 1 and 2 h afterLPS administration (Fig. 1B).

To investigate whether the effect of rolipram in LPS-inducedpleurisy was via PKA, the best-known cAMP effector, we usedthe PKA inhibitor H89. As shown in Figure 1E, H89 preventedresolution of LPS-induced pleurisy induced by rolipram, impli-cating PKA as the cAMP effector in this resolving process. We

A

6

8

10

***Treatment

Cell count

*** ***

utro

phils

vity 750

1000

1250

**150

200

**

/mL)

MIP-2

B

0 1 2 4 8 120

2

4

6

Duration of pleurisy post-LPS (h)

*** #

24 48

Num

ber

of n

eux

105 /c

av

0

250

500

1h 2h 4h 8h 12h

LPS (250ng/cavity)

*

0

50

100

PBS 1h 2h 4h 8h 12hPBS

KC

(pg/ (pg/m

L)

D

C

8

10

12 *

#eutr

ophi

lsav

ity

Duration of pleurisy post LPS (h) LPS (250ng/cavity)

6

8

10

*

utro

phils

vity

PBS vehicle 15 50 1500

2

4

6

Rolipram (µg/mouse)

#

#

Num

ber

of n

ex

105 /c

a

0

2

4

6

PBS

LPS (250ng/cavity)Vehicle Rolipram Reparixin

#

Num

ber

of n

eux

105 /c

av

LPS (250ng/cavity)

6

8

10

*P<0.001

e utr

ophi

lsvi

ty

E

LPS (250ng/cavity)

0

2

4

6

PBS H89-Rolipram

H89-

#

Num

ber

of n

ex

105 /c

av

LPS (250ng/cavity)

Figure 1. Treatment with the PDE4 inhibitor rolip-ram dose-dependently resolves neutrophilic inflam-mation in a PKA-dependent manner. (A and B)Mice were injected with LPS (250 ng/cavity), andnumber of neutrophil (A) or chemokine levels (B)in the pleural lavage was evaluated at various times.(C and D) Mice were injected with LPS and thentreated with rolipram (at doses of 15–150 �g/mouse, i.p.), reparixin (30 mg/kg, i.p.), or drugvehicle 4 h after LPS challenge, as indicated in A.(E) Mice were injected with LPS and then pre-treated with H89 (80 �g/mouse) 30 min before theinjection of rolipram (150 �g/mouse). Results areexpressed as the number of cells/cavity and are

shown as the means � sem of five to six mice in each group. (A and B) *, P � 0.05; **, P � 0.01; ***, P � 0.001, when compared with PBS-injected mice; #, P � 0.05, when compared with 8 h LPS-injected mice. (C–E) *, P �0.001, when compared with PBS-injected mice; #, P �0.01, when compared with vehicle-treated, LPS-injected mice.

Sousa et al. PDE4 inhibition and neutrophil apoptosis

www.jleukbio.org Volume 87, May 2010 Journal of Leukocyte Biology 3

also used other cAMP-elevating agents that are known to act ina PKA-dependent manner. Similarly to rolipram, the adminis-tration of forskolin, an adenylate cyclase activator, or db-cAMP,a cell-permeable cAMP analog, in the pleural cavity when theinflammatory process was established decreased neutrophil

accumulation and increased the number of apoptotic cells inthe pleural cavity (Table 1).

PDE4 inhibition enhances resolution of LPS-inducedpleurisy by inducing caspase-dependent apoptosis ofinflammatory cellsNext, we examined whether apoptosis was the underlying rea-son for the reduction of neutrophil numbers in the pleuralcavity of rolipram-treated mice. The reduction of neutrophilnumber (Fig. 2A) was associated with an increase in the num-ber of apoptotic cells at the pleural cavity, as demonstrated bymorphological criteria (Fig. 2B) and biochemical markers(Figs. 2D and 3C). Annexin-V analysis by flow cytometryshowed a similar result (data not shown). Examples of apopto-tic neutrophils at 4 h after treatment with rolipram are shownin Figure 2C. Of note, although the absolute numbers of apo-ptotic cells in the pleural cavity are similar in rolipram-treatedand untreated mice, the proportion of cells undergoing apo-ptosis is clearly different and is at least fivefold higher in rolip-rarm-treated animals. Neutrophil apoptosis is controlled by acomplex network of signaling pathways that regulates the turn-over of key molecules, including the antiapoptotic proteinMcl-1, and activation of the caspases family of proteases [30–32]. Here, we show that rolipram-induced apoptosis of inflam-matory cells was accompanied by a decrease in the cellular lev-

TABLE 1. Treatment with the Adenylate Cyclase ActivatorForskolin or the Cell-Permeable cAMP Analogue db-cAMP

Resolves Neutrophilic Inflammation and Promotes NeutrophilApoptosis

Number ofneutrophils�105/cavity

% of Neutrophilswith apoptoticmorphology

PBS 0.1 � 0.0 0.06 � 0.06LPS � Vehicle 9.0 � 2.4a 0.75 � 0.15LPS � Forskolin 3.1 � 0.5b 2.9 � 0.4b

LPS � db-cAMP 3.1 � 0.2b 2.8 � 0.5b

Mice were injected with LPS (250 ng/cavity) or PBS and 4 h later,received a local injection of forskolin (10 �g/mouse), db-cAMP (100�g/mouse), or drug vehicle. The number of neutrophils and cellswith distinctive apoptotic morphology was assessed 8 h after LPS injec-tion. Results are expressed as the number of neutrophils cavity or aspercent of neutrophil with apoptotic morphology and are shown asthe mean � sem of at least five mice in each group. aP � 0.001 whencompared with PBS-injected mice; bP � 0.01 when compared with ve-hicle-treated, LPS-challenged mice.

C Vehicle-treated control

∗∗

A

6

8

10

12 *

neut

roph

ilsca

vity

R li

∗∗

∗

∗∗

∗

∗

∗ ∗0

2

4

6

PBSLPS (250ng/cavity)

Vehicle Rolipram

#

Num

ber

of n

x 10

5 / c

Rolipram

3

4

5 #

roph

ils w

ithm

orph

olog

y

B

0

1

2

PBS RolipramVehicleLPS (250ng/cavity)

% o

f Neu

trap

opto

tic m

∗∗

∗

D

Mcl-1

- ROLPBSLPS (250ng)

∗ ∗

β-actin

Figure 2. Treatment with the PDE4 inhibitor rolipramresolves neutrophilic inflammation by promoting neu-trophil apoptosis. Mice were injected with LPS (250ng/cavity) or PBS and 4 h later, received a systemicinjection of rolipram (150 �g/mouse, i.p.) or drug ve-hicle. The number of neutrophils (A) and cells withdistinctive apoptotic morphology (B) was assessed 8 hafter LPS injection. Results are expressed as the num-ber of cells/cavity or percent of neutrophil with apo-ptotic morphology and are shown as the mean � semof five mice in each group. * or #, P � 0.01, when com-pared with PBS-injected mice or with vehicle-treated,LPS-challenged mice. (C) Representative figures of no-napoptotic (*) and apoptotic (arrows) neutrophils andapoptotic cells inside macrophages (arrowheads). Vehi-cle (upper panel) and rolipram-treated (lower panels)animals are shown. Original magnifications, �100. (D)Western blotting analysis of Mcl-1. Whole cell extractswere obtained from inflammatory cells harvested fromthe pleural cavity 2 h after rolipram (ROL) treatmentof 4 h LPS-challenged mice (i.e., 6 h after LPS). Mem-branes were probed with anti-Mcl-1 or anti-�-actin anti-bodies. Blots are representative of three independentexperiments in pools of cells from at least five animals.


els of Mcl-1 (Fig. 2D), a key antiapoptotic protein regulatingneutrophil survival [30, 31], and increase of caspase-3 cleavage(Fig. 3C). Importantly, the administration of the broad-spec-trum caspase inhibitor zVAD-fmk prevented the resolution byrolipram of LPS-induced neutrophilic inflammation (Fig. 3A).In addition, the increase in inflammatory cell apoptosis in-duced by rolipram treatment was markedly inhibited by treat-ment with zVAD-fmk (Fig. 3B). The ability of zVAD-fmk to in-hibit caspase-3 activation in the model was confirmed by West-ern blot analysis of inflammatory cell lysates from mice treatedwith rolipram (Fig. 3C).

Resolution of LPS-induced pleurisy by rolipram isassociated with PI3K/Akt inhibitionThe PI3K/Akt pathway has been shown to mediate survivalin many cell types [33]. Recently, we have demonstratedthat the PI3K/Akt pathway is important for survival of eo-sinophils in vivo [26, 27] and is a target of cAMP [26]. Withthis in mind, we examined the levels of Akt phosphorylationafter LPS administration and showed that there was a time-dependent increase of Akt phosphorylation in inflammatorycells recovered from the pleural cavity (Fig. 4A). Of note,the kinetics of Akt phosphorylation paralleled the kineticsof neutrophil influx into the pleural cavity (compare Figs.1A and 4A).

To explore the importance of the PI3K/Akt pathway forneutrophil accumulation/survival in the pleural cavity ofmice after LPS administration, we used the PI3K inhibitorLY294002 and Akt inhibitor IV. Local treatment withLY294002 or with Akt inhibitor V 4 h after LPS injectionreduced the number of neutrophils in the pleural cavity in-duced by LPS administration (Fig. 4, B and D) and in-creased the number of apoptotic cells (Fig. 4, C and E).Akin to the effects of LY294002, the use of wortmannin, an-other structurally distinct PI3K inhibitor, also resolved neutro-philic accumulation induced by LPS (PBS: 0.7�0.7�105 neutro-phils/cavity; LPS�vehicle: 11.6�1.8�105 neutrophils/cavity;LPS�wortmannin: 2.1�0.5�105 neutrophils/cavity; n�5; P �0.01 when comparing vehicle�wortmannin). Interestingly,treatment with LY294002 also decreased LPS-associatedMcl-1 expression (Fig. 4F) and induced caspase-3 cleavage(Fig. 4G). Therefore, PI3K and Akt are important survivalfactors for neutrophils in vivo, and the inhibition of theseenzymes induces resolution of neutrophilic inflammationvia induction of apoptosis.

As the accumulation of neutrophils in the pleural cavity wasinhibited by PI3K and Akt inhibitors, next, we examinedwhether the resolution of inflammation and increase of apo-ptotic cell number after PDE4 inhibition could be a result ofdiminished phosphorylation of Akt, the main PI3K substrate.Treatment with rolipram 4 h after LPS administration inhib-ited Akt phosphorylation to baseline levels (Fig. 4F). The ef-fects of rolipram were quantitatively similar to those of thePI3K inhibitor LY294002 (Fig. 4F), suggesting that control ofAkt phosphorylation may be important for the action of PDE4inhibitors.

The transcription factor NF-�B is a crucial regulator ofgranulocyte apoptosis [17], and several studies have shown

3

4

5

6 *

#

P<0.05

of n

eutr

ophi

ls05 /c

avity

0

1

2

LPS (250ng/cavity)

PBS zVAD-Rolipram

zVAD-

#

Num

ber x 1

LPS (250ng/cavity)

8

10*P<0.001

ls w

ithho

logy

B

0

2

4

6

% o

f Neu

trop

hil

apop

totic

mor

ph

0

LPS (250ng/cavity)

PBS zVAD-Rolipram

zVAD-

LPS (250ng)C

Procaspase-3

Cleaved

-PBS

( g)Rolipram- zVAD

Cleaved Caspase-3

A

Figure 3. Treatment with the PDE4 inhibitor rolipram resolves neutro-philic inflammation in a caspase-dependent manner. Mice were in-jected with LPS (250 ng/cavity) or PBS and 4 h later, received a sys-temic injection of rolipram (150 �g/mouse, i.p.) or drug vehicle. Thepan-caspase inhibitor zVAD-fmk (1 mg/kg, i.p.) was given 15 min be-fore rolipram. The number of neutrophils (A) and cells with distinc-tive apoptotic morphology (B) was assessed 8 h after LPS injection.Results are expressed as the number of cells/cavity or percent of neu-trophil with apoptotic morphology and are shown as the mean � semof five mice in each group. *, P � 0.001, when compared with PBS-injected mice. (C) Western blotting analysis of caspase-3. Whole cellextracts were obtained from inflammatory cells harvested from thepleural cavity 2 h after rolipram treatment of 4 h LPS-challenged mice(i.e., 6 h after LPS). Membranes were probed with anti-caspase-3 anti-body. Full-length caspase-3 and the caspase-3 cleavage product are in-dicated. Blots are representative of two independent experiments inpools of cells from at least five animals.



that blockade of NF-�B induces neutrophil apoptosis invitro and in vivo [34 –37]. Thus, we investigated whether theNF-�B pathway was also implicated in the resolution of neu-trophilic inflammation after PDE4 inhibition. Activation ofNF-�B was evaluated by EMSA of the inflammatory cells re-covered from the pleural cavity. LPS induced significantNF-�B activation in the pleural cavity of mice. Importantly,treatment with rolipram or LY294002 4 h after LPS chal-lenge inhibited NF-�B-DNA-binding activity (Fig. 5A) andp65/RelA nuclear accumulation (data not shown). Neverthe-less, the use of two structurally different NF-�B inhibitors,PDTC (an antioxidant compound reported to inhibit NF-�B

efficiently) and SN-50 (a cell-permeable peptide inhibitor ofNF-�B translocation), given in a similar way as rolipram, didnot alter LPS-induced neutrophil accumulation in the pleu-ral cavity (Fig. 5B). The efficacy of the compounds was con-firmed by their ability to block NF-�B in the system (datanot shown).

DISCUSSION

There has been much recent interest in the understandingof the signal transduction pathways that govern leukocyte

A LPS (250ng)A LPS (250ng)

P-Akt

4h 24h8hPBS

Akt

CB

3

4

5

6

7 *

#

#r of

neu

trop

hils

05pe

r ca

vity

2

3

4

5#

#

eutr

ophi

ls w

itho t

ic m

orph

olog

y

0

1

2

Akt inh-IV (µg/cavity)PBS Vehicle 1

LPS (250ng/cavity)

3

Num

be x 1 0

15 5ED

0

1

Akt inh-IV (µg/cavity)PBS Vehicle 1

LPS (250ng/cavity)

3

% o

f Nap

opto

5

10

5

#

*

umbe

r of

neu

trop

hils

x 10

5 /cav

ity

1

2

3

4

5

#

% o

f Neu

trop

hils

with

apop

totic

mor

phol

ogy

0PBS

LPS (250ng/cavity)LY294002Vehicle

N

0PBS

LPS (250ng/cavity)LY294002Vehicle

% a

GF LPS (250ng) LPS (250ng)

P-Akt

Akt

- ROL LYPBS( g)

Mcl-1

- LYPBS( g)

Procaspase-3

Cl dMcl 1

β-actinCleaved

Caspase-3

Figure 4. PI3K/Akt inhibitors re-solve neutrophilic inflammation,and inhibition of PI3K/Akt medi-ates the effects of rolipram. (A)Mice were injected with LPS (250ng/cavity) and Akt phosphoryla-tion evaluated by Western blottingat various times. (B–E) Mice wereinjected with LPS and then treatedwith the PI3K inhibitor LY294002(30 �g/mouse, i.pl.), Akt inhibitorIV (Akt inh-IV; at doses of 1 and 3�g/mouse, i.pl.), or drug vehicle4 h after LPS injection in mice,and neutrophil accumulation (Band D) and number of apoptoticcells (C and E) were evaluated 4 hafter drug treatment, i.e., 8 h afterLPS challenge. Results are ex-pressed as the number of neutro-phils/cavity or percent of neutro-phil with apoptotic morphologyand are shown as the mean � semof at least five mice in each group.*, P � 0.001, when compared withPBS-injected mice; #, P � 0.01,when compared with vehicle-treated, LPS-challenged mice. (Fand G) Akt phosphorylation,Mcl-1, and caspase-3 cleavage wereevaluated 2 h after rolipram treat-ment of 4 h LPS-challenged mice(i.e., 6 h after LPS) by Westernblot analysis. Data are representa-tive of three independent experi-ments. LY, LY294002.


survival in vivo [16, 26, 27, 38, 39]. The aim of the presentstudy was to understand how signaling through cAMP couldregulate neutrophil survival in vivo. By using a mousemodel of LPS-induced pleurisy, we show that PDE4 inhibi-tion is effective at resolving neutrophilic inflammation afterLPS challenge by promoting caspase-dependent apoptosis ofinflammatory cells in the pleural cavity; and mechanistically,we show that rolipram-induced neutrophil apoptosis wasPKA-dependent, mimicked by cAMP mimetics/inducers, andassociated with inhibition of the PI3K/Akt pathway, de-creased levels of Mcl-1, and increased cleavage of the execu-tioner caspase-3.

Pretreatment of mice with rolipram is clearly associated withdecreased neutrophil recruitment [4–6, 40] and in vitro inhi-bition of chemotaxis [40]. A recent study has found that levelsof KC and MIP-2 were similar in wild-type mice and PDE4Band PDE4B knockout mice [40]. However, neutrophils fromPDE4B and PDE4B knockout mice showed decreased CD18integrin expression and reduced KC-induced migration in

vitro [40]. In the present study, we used a post-treatmentschedule (rolipram was given 4 h after LPS). As shown in Fig-ure 1, this schedule of administration is after the time-pointsin which there is production of neutrophil-active chemokinesin the cavity. Consistent with the early production of chemo-kines and early neutrophil influx, blockade of CXCR2 withreparixin failed to affect neutrophil accumulation at 8 h.These results would suggest that the observed effects of de-layed administration of rolipram are likely not on the produc-tion of neutrophil-active chemokines or on the migration ofneutrophils to the cavity but on events that govern their sur-vival or clearance from the cavity.

In vitro studies using neutrophils have shown that PDE4 in-hibition or an increase of cAMP levels by other means pro-motes neutrophil survival [9–13]. In contrast with these invitro studies, which used mostly human peripheral blood neu-trophils, in our experimental settings, in vivo administration ofa PDE4 inhibitor was clearly associated with resolution of neu-trophilic inflammation by inducing caspase-3-dependent apo-ptosis. In addition to using PDE4 inhibitor rolipram, weshowed that agents that mimic the action of cAMP resolveneutrophilic inflammation by inducing apoptosis. The discrep-ancies between in vitro and in vivo findings are probably a re-sult of differences in the environment, in which the experi-ments are conducted and the fact that mostly in vitro experi-ments have used blood but not migrated neutrophils. Itremains possible that within the far more complex in vivo sig-naling environment, which is not mimicked by in vitro culture,the presence of an inflammatory milieu, cell–cell interaction,or yet, what compartment of cyclic nucleotide signaling is acti-vated selectively (cAMP effector proteins or PDEs) will deter-mine the overall cell fate. Indeed, this is the first report thatshows the cAMP effect on apoptosis of inflammatory neutro-phils in vivo. Furthermore, a recent paper [41] showed thatresolution-phase macrophages possess a unique phenotypethat is controlled by cAMP. As inflammatory cells of the pleu-ral cavity consist of both populations of cells (mononuclearand neutrophils), we cannot exclude the possibility that theobserved effects of rolipram, at least in part, could be medi-ated by such macrophages or T lymphocytes. This might alsohelp to explain the apparently contradictory actions of cAMPon isolated versus in situ neutrophils. Despite these data fromisolated cell systems, our in vivo data show that the net effectof cAMP elevation in the course of neutrophilic inflammationis to resolve neutrophil, but not macrophage, accumulation byinducing apoptosis.

The concept that cAMP is involved in resolution of in-flammation is emerging in the literature. We have shownrecently that cAMP-elevating agents promote resolution ofallergic pleurisy [26]. Bystrom and colleagues [41] demon-strated that cAMP enhanced the development of macro-phages with a phenotype that favored the resolution of in-flammation. Moreover, Rajakariar and colleagues [42] sug-gested that lower levels of cAMP and adenosine mayaccount for the phenotype observed in a murine model ofnonresolving inflammation. Our data add to this body ofevidence by showing that agents that mimic or elevate

LPS (250ng)

Rol LY-NF-κκBCold

SRECold

A

NF-κB

B

8

10

12*

neut

roph

ilsca

vity

0

2

4

6

PBSS ( / )

Vehicle PDTC SN-50

Num

ber

of n

x 10

5 /c

LPS (250ng/cavity)

Figure 5. Inhibition of PDE4 or PI3K prevents activation of NF-�B,but NF-�B inhibitors do not promote resolution of neutrophilic in-flammation. (A) Mice were injected with LPS (250 ng/cavity) or PBSand 4 h later, were treated with rolipram (150 �g/mouse, i.p.) orLY294002 (30 �g/mouse, i.pl.). Nuclear extracts were obtained frominflammatory cells harvested from the pleural cavity 2 h after drugtreatments (i.e., 6 h after LPS challenge) and subjected to EMSA anal-ysis with an end-labeled oligodeoxynucleotide probe containing theconsensus NF-�B site. Specificity of the interactions was confirmed bycompetition of the probe with 100-fold molar excess of the indicatedcold oligodeoxynucleotide. Data are representative of two indepen-dent experiments in pools of cells from at least five animals. (B) Micewere injected with LPS (250 ng/cavity) or PBS and 4 h later, weretreated with rolipram (150 �g/mouse, i.p.), PDTC (100 mg/kg, i.p.),or SN-50 (10 �g/mouse, i.pl.). Neutrophil accumulation was evaluated4 h after drug treatment, i.e., 8 h after LPS challenge. Results are ex-pressed as the number of neutrophils/cavity and are shown as themean � sem of at least five mice in each group. *, P � 0.01, whencompared with PBS-injected mice.



cAMP induce resolution of neutrophilic inflammation viainduction of apoptosis.

Caspases are a family of aspartic acid-specific proteases thatare the central effectors of the apoptotic pathway [43]. Here,we show that rolipram-induced apoptosis and its anti-inflam-matory effects were reversed in vivo by the pan-caspase inhibi-tor zVAD-fmk. Our finding that rolipram-mediated apoptosis iscaspase-dependent is in accordance with previous in vitro stud-ies showing a caspase-dependent apoptosis of B-CLL by rolip-ram [44, 45]. Moreover, the loss of Mcl-1 may contribute tothe induction of apoptosis by rolipram, as Mcl-1 is highly in-ducible and has been described as the main antiapoptotic pro-tein of the Bcl-2 family expressed in neutrophils [30–32]. In-deed, Mcl-1 down-regulation contributes to the proapoptoticeffects of R-roscovitine, a cyclin-dependent kinase inhibitor, inhuman neutrophils [38]. However, as Mcl-1 has a short half-life in mature neutrophils, the drop in Mcl-1 levels may simplyreflect the loss of neutrophils from the cavity and may not bemechanistically related to the effects of the PDE4 inhibitor.Further experiments in alternative systems will be necessary toconfirm the latter possibility, as it was not feasible to purifyneutrophils in adequate numbers from the pleural cavity inour experiments, especially in rolipram-treated animals, mak-ing it impossible to identify levels of Mcl-1 in neutrophilsthemselves.

The binding of cAMP to proteins such as PKA and EPACexplains most of its functional activities, but there are addi-tional, less well-characterized effector proteins [1]. Al-though nonspecific effects of H89 may exist [46], this is awidely used tool to assess the role of PKA in in vitro and invivo systems. In our model system, PKA inhibition by H89limited cAMP-mediated neutrophil clearance, suggestingthat PKA may be the cAMP effector in this resolving pro-cess. Indeed, a previous study [47] showed that elevatedcAMP levels in B-CLL cells could have proapoptotic or anti-apoptotic effects, depending on whether PKA or EPAC, re-spectively, was activated. As EPAC is not expressed in neu-trophils and mononuclear cells [47], and we show here thatinhibition of PKA by H89 reversed rolipram-induced effects,we suggest that PKA is the major cAMP effector in this re-solving process (Fig. 6).

The PI3K pathway has also been shown to be an importantsurvival-signaling pathway in inflammatory cells [33]. Previouswork from our group has demonstrated that the PI3K/Aktpathway plays a role for survival of eosinophils in vivo [26, 27].Here, we find that LPS promoted Akt phosphorylation in in-flammatory cells with a time-course that paralleled the influxof neutrophils in the pleural cavity. The importance of the Aktpathway for neutrophil survival was evidenced by experimentsusing PI3K and Akt inhibitors. Treatment with rolipram inhib-ited LPS-induced Akt phosphorylation and increased activationof caspase-3, suggesting that Akt is indeed relevant for neutro-phil survival in vivo and is a site for the action of cAMP-elevat-ing agents. In keeping with previous data demonstrating thatincreased Mcl-1 expression depends on the PI3K/Akt pathway[48], we found that that LPS-induced Mcl-1 elevation was re-versed by treatment with LY294002 and rolipram, reinforcing

the importance of the PI3K/Akt pathway in rolipram-inducedapoptosis (Fig. 6).

Our results are consistent with studies that demonstrate across-talk between cAMP-dependent and PI3K pathways[49 –51]. Particularly, the studies of Smith and colleagues[50] showed that cAMP-mediated apoptosis in a diffuse,large B cell lymphoma was associated with marked inhibi-tion of the PI3K/Akt pathway. Although it is not clear howcAMP modifies Akt activity, a recent report suggests thatcAMP-dependent inhibition of Akt in thyroid cells is medi-ated by phosphatase 2A, involving EPAC and PKA [51]. An-other study has also shown that stimulation of PKA inhibitsAkt activity [49]. Thus, cAMP may mediate its survival/pro-apoptotic effects by modifying PI3K/Akt. Consistent withthese observations, our results with H89 show that PKA ap-pears to be the cAMP effector that drives the inhibitory ef-fects of rolipram.

We have found previously that the NF-�B pathway is in-volved in the resolution of the eosinophilic inflammation in-duced by treatment with rolipram [26]. Here, we show thatalthough rolipram promotes inhibition of NF-�B activation,this pathway appears not to be relevant for rolipram-inducedresolution of neutrophilic inflammation. Indeed, NF-�B inhibi-tors failed to resolve neutrophil accumulation in our LPS-in-duced pleurisy model. In a model of carrageenan-inducedneutrophilic inflammation, Lawrence and colleagues [52]showed that inhibition of NF-�B protracted the inflammatoryresponse and prevented apoptosis. Therefore, and in contrastto the role of NF-�B in mediating eosinophil survival [26],blockade of NF-�B has no effect (present findings) or pro-tracts [52] neutrophilic inflammation when blockers are givenafter inflammation has been established.

ATP

AcForskolin Rolipram

cAMP

Ac

5’-AMPPDE4db-AMP

LY294002PI3K

PKA AktH89 Akt inh-IV

C 3VAD

???

Resolution of neutrophilicinflammation by apoptosis Cell survival

Mcl-1 (?)Caspase-3zVAD

Figure 6. Model for the resolution-inducing effects of the treatmentwith cAMP-elevating agents during neutrophilic inflammation. Increasein the intracellular levels of cAMP by means of the treatment with anadenylate cyclase activator (forskolin), a PDE4 inhibitor (rolipram), ordb-cAMP (db-AMP) promotes activation of a cAMP effector (PKA),which then inhibits signaling through the PI3K/Akt pathway. Down-regulation of the Akt phosphorylation causes a decrease of Mcl-1 ex-pression, the most important prosurvival protein of the Bcl-2 familyexpressed in neutrophils and activation of the executioner caspase-3leading to neutrophil apoptosis and consequent resolution of inflam-mation.


In conclusion, our data demonstrate that PDE4 inhibitionpromotes PKA-dependent resolution of established neutro-philic inflammation by inducing apoptosis of inflammatorycells mediated by activation of caspase-3 and correlated withinhibition of the PI3K/Akt/Mcl-1 pathway (Fig. 6). To ourknowledge, this is the first observation that rolipram promotesapoptosis of neutrophils in vivo via inhibition of a PI3K/Aktpathway. Hence, we suggest that elevation of cAMP in vivo mayrepresent a powerful anti-inflammatory strategy for the treat-ment of diseases in which neutrophil accumulation is thoughtto play a relevant role.

AUTHORSHIP

Author contributions: L. P. S., V. P., A. G. R., and M. M. T.designed research, analyzed data, and wrote the paper.L. P. S., F. L., D. M. S., L. P. T., B. M. R., A. T. V., A. F. C.,R. C. R., C. C. G., and A. L. A. carried out experiments.A. T. V. and A. L. A. carried out animal injections. C. A. B.supplied essential experimental tools.

ACKNOWLEDGMENTS

This work was supported by grants from: Fundacao de Amparoa Pesquisa do Estado de Minas Gerais (FAPEMIG/Brazil),Conselho Nacional de Desenvolvimento Cientıfico e Tecno-logico (CNPq/Brazil), and Pro-Reitoria de Pesquisa da Univer-sidade Federal de Minas Gerais (PRPq-UFMG). We thankValdineria Borges and Ilma Marcal for technical assistance.

REFERENCES

1. Serezani, C. H., Ballinger, M. N., Aronoff, D. M., Peters-Golden, M.(2008) Cyclic AMP: master regulator of innate immune cell function.Am. J. Respir. Cell Mol. Biol. 39, 127–132.

2. Soderling, S. H., Beavo, J. A. (2000) Regulation of cAMP and cGMP sig-naling: new phosphodiesterases and new functions. Curr. Opin. Cell Biol.12, 174–179.

3. Torphy, T. J. (1998) Phosphodiesterase isozymes: molecular targets fornovel antiasthma agents. Am. J. Respir. Crit. Care Med. 157, 351–370.

4. Souness, J. E., Aldous, D., Sargent, C. (2000) Immunosuppressive andanti-inflammatory effects of cyclic AMP phosphodiesterase (PDE) type 4inhibitors. Immunopharmacology 47, 127–162.

5. Teixeira, M. M., Gristwood, R. W., Cooper, N., Hellewell, P. G. (1997)Phosphodiesterase (PDE)4 inhibitors: anti-inflammatory drugs of the fu-ture? Trends Pharmacol. Sci. 18, 164–171.

6. Souza, D. G., Cassali, G. D., Poole, S., Teixeira, M. M. (2001) Effects ofinhibition of PDE4 and TNF-� on local and remote injuries followingischaemia and reperfusion injury. Br. J. Pharmacol. 134, 985–994.

7. Bopp, T., Dehzad, N., Reuter, S., Klein, M., Ullrich, N., Stassen, M.,Schild, H., Buhl, R., Schmitt, E., Taube, C. (2009) Inhibition of cAMPdegradation improves regulatory T cell-mediated suppression. J. Immunol.182, 4017–4024.

8. Diaz, B. L., Serra, M. F., Alves, A. C., Cordeiro, S. B., Martins, M. A.,Silva, P. M. (1996) Local exposure to salbutamol or Bt2 cyclic AMP inhib-its pleural exudation and leukocyte influx caused by antigen in rats. Eur.J. Pharmacol. 296, 173–180.

9. Rossi, A. G., Cousin, J. M., Dransfield, I., Lawson, M. F., Chilvers, E. R.,Haslett, C. (1995) Agents that elevate cAMP inhibit human neutrophilapoptosis. Biochem. Biophys. Res. Commun. 217, 892–899.

10. Martin, M. C., Dransfield, I., Haslett, C., Rossi, A. G. (2001) Cyclic AMPregulation of neutrophil apoptosis occurs via a novel protein kinase A-in-dependent signaling pathway. J. Biol. Chem. 276, 45041–45050.

11. Parkkonen, J., Hasala, H., Moilanen, E., Giembycz, M. A., Kankaanranta,H. (2008) Phosphodiesterase 4 inhibitors delay human eosinophil andneutrophil apoptosis in the absence and presence of salbutamol. Pulm.Pharmacol. Ther. 21, 499–506.

12. Krakstad, C., Christensen, A. E., Doskeland, S. O. (2004) cAMP protectsneutrophils against TNF-�-induced apoptosis by activation of cAMP-de-

pendent protein kinase, independently of exchange protein directly acti-vated by cAMP (Epac). J. Leukoc. Biol. 76, 641–647.

13. Conran, N., Almeida, C. B., Lanaro, C., Ferreira, R. P., Traina, F., Saad,S. T., Costa, F. F. (2007) Inhibition of caspase-dependent spontaneousapoptosis via a cAMP-protein kinase A dependent pathway in neutrophilsfrom sickle cell disease patients. Br. J. Haematol. 139, 148–158.

14. Savill, J. (1997) Apoptosis in resolution of inflammation. J. Leukoc. Biol.61, 375–380.

15. Rossi, A. G., Hallett, J. M., Sawatzky, D. A., Teixeira, M. M., Haslett, C.(2007) Modulation of granulocyte apoptosis can influence the resolutionof inflammation. Biochem. Soc. Trans. 35, 288–291.

16. Hallett, J. M., Leitch, A. E., Riley, N. A., Duffin, R., Haslett, C., Rossi,A. G. (2008) Novel pharmacological strategies for driving inflammatorycell apoptosis and enhancing the resolution of inflammation. Trends Phar-macol. Sci. 29, 250–257.

17. Ward, C., Walker, A., Dransfield, I., Haslett, C., Rossi, A. G. (2004) Regu-lation of granulocyte apoptosis by NF-�B. Biochem. Soc. Trans. 32, 465–467.

18. Michlewska, S., Dransfield, I., Megson, I. L., Rossi, A. G. (2009) Macro-phage phagocytosis of apoptotic neutrophils is critically regulated by theopposing actions of pro-inflammatory and anti-inflammatory agents: keyrole for TNF-�. FASEB J. 23, 844–854.

19. Savill, J. S., Wyllie, A. H., Henson, J. E., Walport, M. J., Henson, P. M.,Haslett, C. (1989) Macrophage phagocytosis of aging neutrophils in in-flammation. Programmed cell death in the neutrophil leads to its recog-nition by macrophages. J. Clin. Invest. 83, 865–875.

20. Ward, C., Dransfield, I., Chilvers, E. R., Haslett, C., Rossi, A. G. (1999)Pharmacological manipulation of granulocyte apoptosis: potential thera-peutic targets. Trends Pharmacol. Sci. 20, 503–509.

21. Hawkins, P. T., Anderson, K. E., Davidson, K., Stephens, L. R. (2006) Sig-naling through class I PI3Ks in mammalian cells. Biochem. Soc. Trans. 34,647–662.

22. Yang, K. Y., Arcaroli, J., Kupfner, J., Pitts, T. M., Park, J. S., Strasshiem,D., Perng, R. P., Abraham, E. (2003) Involvement of phosphatidylinositol3-kinase � in neutrophil apoptosis. Cell. Signal. 15, 225–233.

23. Lindemans, C. A., Coffer, P. J. (2004) Regulation of granulocyte apopto-sis by phosphatidylinositol 3-kinase. Biochem. Soc. Trans. 32, 480–484.

24. Cowburn, A. S., Deighton, J., Walmsley, S. R., Chilvers, E. R. (2004) Thesurvival effect of TNF-� in human neutrophils is mediated via NF-� B-dependent IL-8 release. Eur. J. Immunol. 34, 1733–1743.

25. Rane, M. J., Klein, J. B. (2009) Regulation of neutrophil apoptosis bymodulation of PKB/Akt activation. Front. Biosci. 14, 2400–2412.

26. Sousa, L. P., Carmo, A. F., Rezende, B. M., Lopes, F., Silva, D. M., Ales-sandri, A. L., Bonjardim, C. A., Rossi, A. G., Teixeira, M. M., Pinho, V.(2009) Cyclic AMP enhances resolution of allergic pleurisy by promotinginflammatory cell apoptosis via inhibition of PI3K/Akt and NF-�B. Bio-chem. Pharmacol. 78, 396–405.

27. Pinho, V., Souza, D. G., Barsante, M. M., Hamer, F. P., De Freitas, M. S.,Rossi, A. G., Teixeira, M. M. (2005) Phosphoinositide-3 kinases criticallyregulate the recruitment and survival of eosinophils in vivo: importancefor the resolution of allergic inflammation. J. Leukoc. Biol. 77, 800–810.

28. Sousa, L. P., Silva, B. M., Brasil, B. S., Nogueira, S. V., Ferreira, P. C.,Kroon, E. G., Kato, K., Bonjardim, C. A. (2005) Plasminogen/plasminregulates �-enolase expression through the MEK/ERK pathway. Biochem.Biophys. Res. Commun. 337, 1065–1071.

29. Souza, D. G., Vieira, A. T., Pinho, V., Sousa, L. P., Andrade, A. A., Bon-jardim, C. A., McMillan, M., Kahn, M., Teixeira, M. M. (2005) NF-�Bplays a major role during the systemic and local acute inflammatory re-sponse following intestinal reperfusion injury. Br. J. Pharmacol. 145, 246–254.

30. Edwards, S. W., Derouet, M., Howse, M., Moots, R. J. (2004) Regulationof neutrophil apoptosis by Mcl-1. Biochem. Soc. Trans. 32, 489–492.

31. Dzhagalov, I., St John, A., He, Y. W. (2007) The antiapoptotic proteinMcl-1 is essential for the survival of neutrophils but not macrophages.Blood 109, 1620–1626.

32. Bianchi, S. M., Dockrell, D. H., Renshaw, S. A., Sabroe, I., Whyte, M. K.(2006) Granulocyte apoptosis in the pathogenesis and resolution of lungdisease. Clin. Sci. (Lond.) 110, 293–304.

33. Song, G., Ouyang, G., Bao, S. (2005) The activation of Akt/PKB signalingpathway and cell survival. J. Cell. Mol. Med. 9, 59–71.

34. Ward, C., Chilvers, E. R., Lawson, M. F., Pryde, J. G., Fujihara, S., Farrow,S. N., Haslett, C., Rossi, A. G. (1999) NF-�B activation is a critical regula-tor of human granulocyte apoptosis in vitro. J. Biol. Chem. 274,4309–4318.

35. Blackwell, N. M., Sembi, P., Newson, J. S., Lawrence, T., Gilroy, D. W.,Kabouridis, P. S. (2004) Reduced infiltration and increased apoptosis ofleukocytes at sites of inflammation by systemic administration of a mem-brane-permeable I�B� repressor. Arthritis Rheum. 50, 2675–2684.

36. Cuzzocrea, S., Chatterjee, P. K., Mazzon, E., Dugo, L., Serraino, I., Britti,D., Mazzullo, G., Caputi, A. P., Thiemermann, C. (2002) Pyrrolidine di-thiocarbamate attenuates the development of acute and chronic inflam-mation. Br. J. Pharmacol. 135, 496–510.

37. Maiuri, M. C., Tajana, G., Iuvone, T., De Stefano, D., Mele, G., Ribecco,M. T., Cinelli, M. P., Romano, M. F., Turco, M. C., Carnuccio, R. (2004)



Nuclear factor-�B regulates inflammatory cell apoptosis and phagocytosisin rat carrageenin-sponge implant model. Am. J. Pathol. 165, 115–126.

38. Rossi, A. G., Sawatzky, D. A., Walker, A., Ward, C., Sheldrake, T. A., Riley,N. A., Caldicott, A., Martinez-Losa, M., Walker, T. R., Duffin, R., Gray,M., Crescenzi, E., Martin, M. C., Brady, H. J., Savill, J. S., Dransfield, I.,Haslett, C. (2006) Cyclin-dependent kinase inhibitors enhance the resolu-tion of inflammation by promoting inflammatory cell apoptosis. Nat.Med. 12, 1056–1064.

39. Sawatzky, D. A., Willoughby, D. A., Colville-Nash, P. R., Rossi, A. G.(2006) The involvement of the apoptosis-modulating proteins ERK 1/2,Bcl-xL and Bax in the resolution of acute inflammation in vivo. Am. J.Pathol. 168, 33–41.

40. Ariga, M., Neitzert, B., Nakae, S., Mottin, G., Bertrand, C., Pruniaux,M. P., Jin, S. L., Conti, M. (2004) Nonredundant function of phosphodi-esterases 4D and 4B in neutrophil recruitment to the site of inflamma-tion. J. Immunol. 173, 7531–7538.

41. Bystrom, J., Evans, I., Newson, J., Stables, M., Toor, I., van Rooijen, N.,Crawford, M., Colville-Nash, P., Farrow, S., Gilroy, D. W. (2008) Resolu-tion-phase macrophages possess a unique inflammatory phenotype that iscontrolled by cAMP. Blood 112, 4117–4127.

42. Rajakariar, R., Lawrence, T., Bystrom, J., Hilliard, M., Colville-Nash, P.,Bellingan, G., Fitzgerald, D., Yaqoob, M. M., Gilroy, D. W. (2008) Novelbiphasic role for lymphocytes revealed during resolving inflammation.Blood 111, 4184–4192.

43. Logue, S. E., Martin, S. J. (2008) Caspase activation cascades in apoptosis.Biochem. Soc. Trans. 36, 1–9.

44. Siegmund, B., Welsch, J., Loher, F., Meinhardt, G., Emmerich, B., En-dres, S., Eigler, A. (2001) Phosphodiesterase type 4 inhibitor suppressesexpression of anti-apoptotic members of the Bcl-2 family in B-CLL cellsand induces caspase-dependent apoptosis. Leukemia 15, 1564–1571.

45. Moon, E. Y., Lerner, A. (2003) PDE4 inhibitors activate a mitochondrialapoptotic pathway in chronic lymphocytic leukemia cells that is regulatedby protein phosphatase 2A. Blood 101, 4122–4130.

46. Murray, A. J. (2008) Pharmacological PKA inhibition: all may not bewhat it seems. Sci. Signal. 1, re4.

47. Tiwari, S., Felekkis, K., Moon, E. Y., Flies, A., Sherr, D. H., Lerner, A.(2004) Among circulating hematopoietic cells, B-CLL uniquely expressesfunctional EPAC1, but EPAC1-mediated Rap1 activation does not accountfor PDE4 inhibitor-induced apoptosis. Blood 103, 2661–2667.

48. Francois, S., El Benna, J., Dang, P. M., Pedruzzi, E., Gougerot-Pocidalo,M. A., Elbim, C. (2005) Inhibition of neutrophil apoptosis by TLR ago-nists in whole blood: involvement of the phosphoinositide 3-kinase/Aktand NF-�B signaling pathways, leading to increased levels of Mcl-1, A1,and phosphorylated Bad. J. Immunol. 174, 3633–3642.

49. Mei, F. C., Qiao, J., Tsygankova, O. M., Meinkoth, J. L., Quilliam, L. A.,Cheng, X. (2002) Differential signaling of cyclic AMP: opposing effects ofexchange protein directly activated by cyclic AMP and cAMP-dependentprotein kinase on protein kinase B activation. J. Biol. Chem. 277, 11497–11504.

50. Smith, P. G., Wang, F., Wilkinson, K. N., Savage, K. J., Klein, U., Neu-berg, D. S., Bollag, G., Shipp, M. A., Aguiar, R. C. (2005) The phosphodi-esterase PDE4B limits cAMP-associated PI3K/AKT-dependent apoptosisin diffuse large B-cell lymphoma. Blood 105, 308–316.

51. Hong, K., Lou, L., Gupta, S., Ribeiro-Neto, F., Altschuler, D. L. (2008) Anovel Epac-Rap-PP2A signaling module controls cAMP-dependent Aktregulation. J. Biol. Chem. 283, 23129–23138.

52. Lawrence, T., Gilroy, D., Colville-Nash, W. P. R., Willoughby, D. A.(2001) Possible new role for NF-�B in the resolution of inflammation.Nat. Med. 7, 1291–1297.

KEYWORDS:transcription factor � signaling cascade � cyclic AMP


PDE4 inhibition drives resolution of neutrophilic inflammation by inducing apoptosis in a...

Documents

Transcript of PDE4 inhibition drives resolution of neutrophilic inflammation by inducing apoptosis in a...