Published online before print June 3, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
-induced apoptosis by activation of cAMP-dependent protein kinase, independently of exchange protein directly activated by cAMP (Epac)
Cell Biology Research Group, Department of Biomedicine, Section of Anatomy and Cell Biology, University of Bergen, Norway
1 Correspondence: University of Bergen, Department of Biomedicine, Section of Anatomy and Cell Biology, Jonas Lies vei 91, N-5009 Bergen, Norway. E-mail: stein.doskeland{at}iac.uib.no
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
against apoptosis. This analog acted more rapidly than forskolin (which increases the endogenous cAMP production) and allowed us to demonstrate that cA-PK must be activated during the first 10 min after TNF- challenge to protect against apoptosis. The protective effect was mediated solely through cA-PK activation, as it was abolished by the cA-PKI-directed inhibitor Rp-8-Br-cAMPS and the general cA-PK inhibitor H-89. Neutrophils not stimulated by cAMP-elevating agents showed increased apoptosis when exposed to the cA-PK inhibitors Rp-8-Br-cAMPS and H-89, suggesting that even moderate activation of cA-PK is sufficient to enhance neutrophil longevity and thereby contribute to neutrophil accumulation in chronic inflammation.
Key Words: apoptosis • inflammation • cytokine
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B, which is believed to act by enhancing the transcription of genes coding for survival proteins [15
16
17
]. Enhanced apoptosis of granulocytes at inflamed sites promotes the resolution of chronic inflammation [1 , 18 ]. Elucidating the molecular mechanisms that regulate granulocyte apoptosis has therefore significant therapeutic implication. Particularly interesting drug targets are pathways such as the cAMP-signaling system, which serve to extend neutrophil life at the inflamed site. Knowledge of the mediator of cAMP action will not only help future studies aimed at pinpointing the details of the pathway downstream of the primary mediator but will also provide opportunity to target directly the cAMP mediator. Most effects of cAMP are mediated through activation of cAMP-dependent protein kinase (cA-PK) or the newly discovered cAMP-regulated guanosine 5'-triphosphate exchange protein directly activated by cAMP (Epac) family of exchange proteins [19 20 21 ]. The mediator of the antiapoptotic effect of cAMP is not known with certainty [6 , 22 ]; for a recent review, see ref. [3 ]. It has recently been proposed that cAMP protects against neutrophil apoptosis by a non-cA-PK mechanism [3 , 22 ].
We have developed novel cAMP analogs that selectively activate Epac without activating cA-PK in vitro and in vivo [23 , 24 ] and have shown that six modified cAMP analogs preferentially activate cA-PK [19 , 23 ]. These analogs, together with Rp-8-bromo (8-Br)-cAMPS, which we previously showed to be a preferential inhibitor of cA-PK type I (cA-PKI) isozyme [25 ], provide powerful tools to study the possible involvement of cA-PK or Epac in neutrophil apoptosis.
In the present study, we show that cA-PK-specific analogs can substitute for the second messenger cAMP to inhibit TNF-- and anti-Fas (a-Fas)-induced, as well as "spontaneous" neutrophil apoptosis. The Epac-specific activator failed to protect against apoptosis. Additionally, we show that cA-PK must be active during early stages of the apoptotic process to prevent the cell from dying. Thus, the death-inhibitory effect of cAMP is mediated via the cA-PK signaling pathway through cA-PK activation in the preapoptotic stage.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were from Sigma Chemical Co. (St. Louis, MO). Rp-8-Br-cAMPS, 8-(4-chlorophenylthio; 8-pCPT)-2'-O-methyl (Me)-cAMP, N6-monobutyryl-cAMP (N6-MB-cAMP), 2-Cl-8-methylamino-cAMP (2-Cl-8-MA-cAMP), 8-Br-cAMP-acetoxymethyl (8-Br-cAMP-AM), and 8-pCPT-cAMP were from Biolog Life Science Institute (Bremen, Germany). Determination of the binding affinity for sites A and B and the equilibrium-binding inhibition constant of analog relative to cAMP (K'i) values for 2-Cl-8-MA-cAMP for cA-PK regulatory subunits I and II as well as Epac were performed as described [23 ].
Human neutrophil granulocytes were obtained from peripheral blood of normal, nonsmoking donors. EDTA blood was diluted 1:1 in phosphate-buffered saline (PBS) containing serum albumin (5 mg/ml) and glucose (1 mg/ml), and the neutrophils isolated by sedimentation through a lymphoprep/polymorphprep (Nycomed Pharma, Norway) gradient. Erythrocytes were removed by hypotonic lysis in ammonium chloride (8 mg/ml), and the remaining cells (95% neutrophils) were washed in PBS.
Induction and evaluation of neutrophil apoptosis
Neutrophils were diluted in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal calf serum (FCS) and were seeded in 96-well culture plates at a concentration of 40,000 cells/well. For apoptosis induction, cells were incubated with TNF- (25 ng/ml) in combination with the translation inhibitor CHX (1 µg/ml) at 37°C for up to 3 h. Alternatively, cells were treated with a-Fas receptor-activating antibody (500 ng/ml) for 3 h. In some cases, the cells were treated with CHX alone or without any apoptogens and left for up to 24 h to study CHX-induced and spontaneous apoptosis.
To study the modulatory effect of cAMP analogs on apoptosis, the cells were pretreated routinely with analogs for 30 min before the addition of apoptosis inducer. In one series of experiments, the analogs were added at various time-points before and after apoptogen addition to determine the critical time-point of action. The incubations were terminated by fixing the cells in 2% glutaraldehyde containing 1 µg/ml Hoechst 33342 (for visualization of chromatin). Apoptosis was determined by Hofmann differential interference contrast and fluorescence microscopy (Nikon Diaphot 300). The percentage of apoptotic cells was determined by counting at least 300 cells/well. Photomicrographs were taken with a digital (Sensys) camera fitted with IPLab Spectrum 3.1a software.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
- and a-Fas-induced neutrophil apoptosis is counteracted by cyclic nucleotides selectively activating cA-PK but not by Epac-specific cAMP analogs/CHX induced neutrophil death rapidly (within 2 h) to at least 90% (Fig. 1
). The commonly used cAMP analog 8-pCPT-cAMP delayed TNF-/CHX-induced and a-Fas-induced apoptosis. It was more efficient against apoptosis induced by TNF-/CHX than against a-Fas (Fig. 1)
.
 View larger version (26K): [in a new window] |
Figure 1. Death receptor-mediated neutrophil apoptosis is inhibited by the nondiscriminating cAMP analog 8-pCPT-cAMP. Freshly isolated human neutrophils cultured in DMEM with 10% FCS were incubated with TNF- (25 ng/ml)/CHX (1 µg/ml) or a-Fas (500 ng/ml) in the absence and presence of the nondiscriminative cAMP agonist 8-pCPT-cAMP (0.7 mM). Apoptosis was determined by fluorescence microscopy. Further details are given in Materials and Methods. The data are given as mean values ± SEM of at least three separate experiments.
|
|
View this table: [in a new window] |
Table 1. Analog Affinity (Rel. to cAMP) for Epac-1 and cA-PKI
|
-induced apoptosis, whereas 8-pCPT-2'-O-Me-cAMP was inefficient (Fig. 2A
and 2B
). This suggested strongly that cA-PK rather than Epac was the mediator of the cAMP effect. To further substantiate this conclusion, we tested whether Rp-8-Br-cAMPS and H-89 could counteract the effect of N6-MB-cAMP. Rp-8-Br-cAMPS is a preferential inhibitor of cA-PK isozyme I [25
], which is the dominating isoenzyme in human neutrophils [6
]. H-89 acts as a competitive inhibitor of adenosine 5'-triphosphate at the active site of the cA-PK [27
]. H-89 (Fig. 2C)
and Rp-8-Br-cAMPS (Fig. 2A
and 2B)
abolished the protective effect of N6-MB-cAMP, indicating that N6-MB-cAMP acted via cA-PK, most probably cA-PKI.
 View larger version (28K): [in a new window] |
Figure 2. The effect of cAMP on TNF--induced neutrophil apoptosis is mediated through cAMP-kinase rather than Epac. To identify the cAMP pathway responsible for inhibition of human neutrophil apoptosis, selected cAMP analogs were added 30 min prior to TNF- (25 ng/ml)/CHX (1 µg/ml) or CHX alone (control). Cells were fixed and stained with fluorescent DNA-binding dye (Hoechst 33342) 2 h thereafter and scored for apoptosis. (A) The left panels show that the apoptotic morphology (note particularly the hypercondensed chromatin) induced by TNF- was counteracted by the cAMP-kinase activator N6-MB-cAMP (1 mM) but not by the specific Epac activator 8-pCPT-2-O'-Me-cAMP (1 mM). The right panels show that the cAMP-kinase antagonist Rp-8-Br-cAMPS (0.7 mM) could abolish the protective effect of N6-MB-cAMP. The apoptotic score, based on microscopy of cells as shown in A, is given in B. The kinase inhibitor H-89 prevents TNF--induced neutrophil apoptosis in a dose-dependent manner (C). For full inhibition of TNF--induced apoptosis, a high and cA-PK-specific concentration (60 µM) is required. The data shown represent the mean of two to four separate experiments. SEM values are given when more than three experiments were conducted.
|
-induced death (Fig. 2A
2B
2C
; data not shown), suggesting that even basal cA-PK activity could provide some protection against cell death. We noted that the ability of H-89 to abolish the effect of N6-MB-cAMP decreased rapidly upon dilution, 60–80 µM being required for full antagonism (Fig. 2C ; data not shown).
Activation of the cA-PKI is responsible for the inhibitory effect of cAMP
To activate preferentially cA-PKI, we exploited the fact that the two cAMP-binding sites of cA-PK (A, B) act synergistically to activate the kinase [28
29
30
]. Among available cAMP analogs, N6-MB-cAMP has the highest relative site AI specificity and 2-Cl-8-MA-cAMP, the highest site BI specificity. The combination of N6-MB-cAMP and 2-Cl-8-MA-cAMP therefore gives a strong synergy for cA-PKI activation. For comparison, the analog pair 8-pCPT-cAMP and 8-AHA-cAMP [6
] has less predicted synergy (Table 2
). Furthermore, N6-MB-cAMP is, unlike 8-pCPT-cAMP, a poor Epac activator (Table 1) . This decreases the possibility that N6-MB-cAMP enhances the action of 2-Cl-8-MA-cAMP through Epac activation. When 2-Cl-8-MA-cAMP was combined with a low concentration (0.05 or 0.10 mM) of N6-MB-cAMP, a marked synergism was noted for protection against TNF- (Fig. 3
). This lent further support to the contention that cA-PKI was responsible for the protective effect of cAMP against apoptosis.
|
View this table: [in a new window] |
Table 2. Predicted cA-PKI Synergy between Pairs of cAMP Analogs
|
 View larger version (19K): [in a new window] |
Figure 3. The site-specific analogs N6-MB-cAMP and 2Cl-8-MA-cAMP act synergistically to activate type I cAMP-kinase. Low concentrations of N6-MB-cAMP strongly enhanced the ability of the complementary cA-PKI activatory analog [28
] 2-Cl-8-MA-cAMP to protect against TNF-/CHX-induced apoptosis. The data shown represent the mean of two to four separate experiments. SEM values are given when more than three experiments were conducted.
|
/CHX-induced apoptosis is phosphorylated at an early, preapoptotic time-point, the cA-PK had to be activated. We decided first to find a highly potent cA-PK activator, which could be used subsequently to activate rapidly the neutrophil cA-PK. We found that 8-Br-cAMP-AM, which is a membrane-permeable precursor of 8-Br-cAMP [31
], was 25-fold more potent than its parent compound 8-Br-cAMP and also more potent than 8-pCPT-cAMP, N6-MB-cAMP, and dibutyryladenosine (DB)-cAMP (Fig. 4
).
 View larger version (24K): [in a new window] |
Figure 4. Comparison of cAMP analogs as antagonists of TNF--induced neutrophil apoptosis. Neutrophils, preincubated with various concentrations of cAMP analogs for 30 min, were treated (2 h) with TNF-. The cells were fixed and scored for apoptosis as described in Materials and Methods. The broadly acting analogs 8-Br-cAMP [inhibitory concentration 50% (IC50)=0.84 mM)] and 8-CPT-cAMP (IC50=0.32 mM) were found to be less potent inhibitors than N6-MB-cAMP (IC50=0.146 mM) and DB-cAMP (IC50=0.096 mM), whereas the caged cAMP agonist 8-Br-cAMP-AM was the most potent inhibitor (IC50=0.034 mM). The data shown represent the mean of two to four separate experiments. SEM values are given when more than three experiments were conducted.
|
(Fig. 5
). To elevate the endogenous cAMP level, the neutrophils were incubated with the adenylate cyclase stimulator forskolin (50 µM) and the phosphodiesterase inhibitor IBMX (300 µM). Cells pretreated with forskolin and IBMX 15 min before the addition of TNF-/CHX were still completely protected against death 2 h later. When forskolin/IBMX was added 10 min after TNF-/CHX, only a marginal protection was achieved (Fig. 5A)
. This shows that stimulants of endogenous cAMP accumulation must be present prior to or close to the point of death receptor activation to delay death. As the stimulation of cAMP accumulation inevitably takes time, we used the extremely cell-permeable AM-ester analog of 8-Br-cAMP to further pinpoint the time when cAMP was required to protect against death. Using this analog, the cells could be nearly completely protected even 3–6 min after the addition of death stimulator (Fig. 5B)
. This suggests that cAMP does not have to be present before the death stimulus but rather that it acts at a step occurring a few minutes downstream of the initial death receptor stimulation. 8-Br-cAMP-AM appeared to give a maximal cAMP effect, as no further protection was observed when it was combined with forskolin, IBMX, and a cocktail of other cAMP analogs (N6-MB-cAMP, 8-pCPT-cAMP, 2-Cl-8-MA-cAMP, and 8-pCPT-2'-O-Me-cAMP), as shown in Figure 5C
. We conclude that cAMP must be present at the latest a few minutes after death receptor stimulation to optimally delay apoptosis.
 View larger version (25K): [in a new window] |
Figure 5. TNF--induced neutrophil apoptosis is counteracted at an early, preapoptotic stage by the caged cAMP agonist 8-Br-cAMP-MA. cAMP-elevating agents or cAMP analogs were added at different time-points relative to the addition of TNF- (25 ng/ml)/CHX (1 µg/ml), and apoptosis was determined 2 h thereafter. Solid bars represent apoptosis induced by TNF- alone. (A) Forskolin (50 µM) and IBMX (300 µM) were added to increase the endogenous cAMP level in the neutrophils. (B) The caged and highly lipophilic AM-ester of 8-Br-cAMP (0.3 mM) was added. (C) To ensure maximal stimulation of potential cAMP effectors, the concentration of 8-Br-cAMP-AM was increased to 0.5 mM and added together with forskolin, IBMX, 8-pCPT-cAMP (0.7 mM), the cA-PKI activators N6-MB-cAMP (1 mM) and 2-Cl-8-MA-cAMP (0.7 mM), and the potent and specific Epac activator 8-pCPT-2'-O-Me-cAMP (1 mM). The data shown represent the mean of two to four separate experiments. SEM values are given when more than three experiments were conducted. Note that 8-Br-cAMP-AM (B) prevented apoptosis at a later time-point than the stimulators of endogenous cAMP accumulation (A) and that additional cAMP agonists failed to further enhance the action of 8-Br-cAMP-AM (C).
|
and CHX in the presence of the cA-PK activator N6-MB-cAMP (and CHX) for 2 and 6 h (Fig. 6
).
 View larger version (18K): [in a new window] |
Figure 6. The cA-PKI-specific activator N6-MB-cAMP prevents neutrophil apoptosis induced by physiological concentrations of TNF-. Human neutrophils were preincubated (0.5 h) with 0.7 mM N6-MB-cAMP before incubation with various concentrations of TNF- for 2 (A) or 6 h (B). Experimental details are as explained in the legend to Figure 1
. The data shown represent the mean of two to four separate experiments. SEM values are given when more than three experiments were conducted.
|
tested (Fig. 6A)
. After 6 h of incubation, N6-MB-cAMP was still able to partially protect against moderate concentrations of TNF-, 50% apoptosis being observed with 0.8 ng/ml in the absence of N6-MB-cAMP and 2 ng/ml in its presence (Fig. 6B)
.
To know whether cA-PK activation could also protect against neutrophil apoptosis induced by other agents than TNF-, neutrophils were induced to undergo apoptosis by incubation with a-Fas (2 h) or CHX (20 h). Cells treated with a-Fas (Fig. 7
, left panels) or CHX (Fig. 7
, right panels) in the presence of the cA-PK activator N6-MB-cAMP had normal morphology. The cA-PKI inhibitor Rp-8-Br-cAMPS abolished the protection by N6-MB-cAMP, suggesting that cA-PKI was able to also protect against Fas-mediated and CHX-induced cell death.
 View larger version (35K): [in a new window] |
Figure 7. cA-PK is responsible for the protection by cAMP against apoptosis induced by a-Fas or CHX. Neutrophils were induced to undergo apoptosis by exposure to 500 ng/ml a-Fas for 2 h or 1 µg/ml CHX for 20 h. The cA-PK-activating analog N6-MB-cAMP (1 mM) counteracted the apoptogenic effect of a-Fas or CHX. The protective effect of N6-MB-cAMP could be abolished when coincubating with Rp-8-Br-cAMPS (0.7 mM). The Epac-specific analog 8-pCPT-2-O'-Me-cAMP (1 mM) failed to prevent apoptosis. Values given represent percentage of apoptotic cells. Data shown represent a typical experiment. Values obtained from two other experiments differ from the presented results by less than 5%. For further experimental details, see Materials and Methods and the legend to Figure 2
.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recently, cAMP was reported to protect against neutrophil apoptosis through an uncharacterized cA-PK-independent pathway [3 , 22 ]. To know if this pathway could involve Epac activation, we applied novel cAMP analogs, able to discriminate between cA-PK and Epac. To our surprise, we found that cAMP protected neutrophils against apoptosis through activation of cA-PK alone and not Epac or Epac-like receptors. The cA-PK-preferring activator N6-MB-cAMP [23 , 26 ] protected neutrophils equally well as increased endogenous cAMP, and the Epac-specific activator 8-pCPT-2'-Me-cAMP failed to affect apoptosis. Furthermore, the protection by cAMP or cAMP analogs could be counteracted by the preferential cA-PKI inhibitor Rp-8-Br-cAMPS, as well as by the active, site-directed cA-PK inhibitor H-89. The conclusion that cA-PK mediated the antiapoptotic effect of cAMP was further supported by the synergistic protection obtained by combined treatment with the cAMP analogs 2-Cl-8-MA-cAMP and N6-MB-cAMP, which bind preferentially to site B and site A, respectively, of cA-PKI. The exclusion of cA-PK as protector against neutrophil death by Martin et al. [22 ] was based on the inability of 0.1 mM Rp-8-Br-cAMPS or 10 µM H-89 to reverse the effect of cAMP on neutrophil apoptosis. In the original publication describing Rp-8-Br-cAMPS, we found that it was efficient in concentrations from 0.05 to 0.8 mM, depending on the cell type studied [25 ]. As the mother compound 8-Br-cAMP was unusually weak as a cA-PK agonist in the neutrophils (Fig. 4) , it is not surprising if high concentrations are also required of Rp-8-Br-cAMPS. We found that 0.7 mM Rp-8-Br-cAMPS completely reversed the effect of N6-MB-cAMP. The inhibitor H-89 was, in our hands, inefficient at 10 µM but quite efficient at 30 µM and highly efficient at 60 µM (Fig. 2C) , in line with the results reported in the original publication describing this compound [27 ].
Stimulators of endogenous cAMP or conventional cAMP analogs acted too slowly to pinpoint the critical time when cAMP no longer could protect the cells. The highly cell-permeable "caged" analog 8-Br-cAMP-AM protected most of the cells when added 6–10 min after TNF-, i.e., 
50 min before the onset of cell death. TNF- is known to induce an oxidative burst in neutrophils with a latency of 15 min, which is prevented by cAMP-elevating agents given early relative to TNF- [14
]. Similarly, apoptosis induced by the reactive oxygen species (ROS) generator NaN3 is only inhibited by cA-PK activators [19
], when they are given early (C. Krakstad and S. O. Døskeland, unpublished data). The time-frame is therefore similar for cAMP protection against apoptosis and against ROS formation. One possible cA-PK target is cytochrome c oxidase. When this enzyme is phosphorylated by cA-PK, it becomes more resistant to stress-induced uncoupling of mitochondrial respiration and the ensuing ROS formation [35
]. Another plausible target is Bad [36
, 37
]. When phosphorylated by cA-PK, the p-Bad will release antiapoptotic Bcl-2 family members, which may prevent generation of ROS [38
].
The main importance of the present study is to identify the cA-PK as the only mediator of the action of cAMP to protect neutrophils against TNF-- and a-Fas-induced as well as spontaneous apoptosis. Mechanistically, it means that emphasis must be on identifying substrates of cA-PK rather than downstream targets of other cAMP receptors, such as the Epac family members, ion-channels directly activated by cAMP, or yet other putative cAMP mediators [39
]. From a therapeutic standpoint, it means that agents directed at counteracting the protection by cAMP in chronic inflammation can be directed at cA-PK. We noted that the cA-PKI-directed inhibitor Rp-8-Br-cAMP blocked the protective effect of elevated cAMP. We noted also that it enhanced slightly the death of isolated neutrophils. Inhibitors of the antiapoptotic pathways in granulocytes are drug candidates to combat chronic inflammation [2
, 40
]. In inflammation, the neutrophil is exposed to several cAMP-elevating agents such as prostaglandin and will therefore have an elevated level of cAMP. Inhibition of cA-PK will therefore be expected to help resolve chronic inflammation.
Received January 6, 2004; revised April 6, 2004; accepted May 4, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
induced neutrophil apoptosis by cyclic AMP: involvement of caspase cascade Eur. J. Pharmacol. 371,59-67[CrossRef][Medline]-induced oxidative burst in neutrophils adherent to fibronectin: effects of cyclic AMP-elevating agents Br. J. Haematol. 91,566-570[Medline]-mediated apoptosis requires Ras and the activation of mitogen-activated protein kinase J. Biol. Chem. 271,14560-14566
in neutrophil apoptosis Cell. Signal. 15,225-233[CrossRef][Medline]-induced apoptosis Mol. Cell. Biol. 20,5680-5689
J. Biol. Chem. 274,37329-37334This article has been cited by other articles:
 |
|
 |
|
  K. R. Vaughan, L. Stokes, L. R. Prince, H. M. Marriott, S. Meis, M. U. Kassack, C. D. Bingle, I. Sabroe, A. Surprenant, and M. K. B. Whyte Inhibition of Neutrophil Apoptosis by ATP Is Mediated by the P2Y11 Receptor J. Immunol., December 15, 2007; 179(12): 8544 - 8553. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Bryn, M. Mahic, J. M. Enserink, F. Schwede, E. M. Aandahl, and K. Tasken The Cyclic AMP-Epac1-Rap1 Pathway Is Dissociated from Regulation of Effector Functions in Monocytes but Acquires Immunoregulatory Function in Mature Macrophages. J. Immunol., June 15, 2006; 176(12): 7361 - 7370. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Goichberg, A. Kalinkovich, N. Borodovsky, M. Tesio, I. Petit, A. Nagler, I. Hardan, and T. Lapidot cAMP-induced PKC{zeta} activation increases functional CXCR4 expression on human CD34+ hematopoietic progenitors Blood, February 1, 2006; 107(3): 870 - 879. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-L. C. Jin, L. Lan, M. Zoudilova, and M. Conti Specific Role of Phosphodiesterase 4B in Lipopolysaccharide-Induced Signaling in Mouse Macrophages J. Immunol., August 1, 2005; 175(3): 1523 - 1531. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
