| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print June 22, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,
,1
* The Inflammation Program, Division of Infectious Diseases,
Departments of Internal Medicine, and Physiology and Biophysics and Graduate Programs in
Immunology and
Molecular Biology, University of Iowa Carver College of Medicine and Iowa City Veterans Affairs Medical Center, Iowa City, Iowa, USA
1 Correspondence: Inflammation Program, Division of Infectious Diseases, Department of Internal Medicine, University of Iowa Carver College of Medicine, 200 Hawkins Dr., SW 54-8, GH, Iowa City, IA 52242, USA. E-mail: david-kusner{at}uiowa.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: cell activation • signal transduction • antigen presentation/processing • transcription factors
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have demonstrated recently that the multifunctional serine/threonine kinase, calmodulin kinase II (CaMKII), is activated in DC during infection by gram-negative and gram-positive bacteria, as well as by specific bacterial ligands for TLRs [4
]. Inhibition of CaMKII in primary human myeloid DC or the human DC line KG-1 results in reductions in antigen-induced surface expression of MHC Class II, CD40, CD83, and CD86; secretion of IL-12, IFN-
, and IL-2; and MHC Class II-restricted T cell proliferation [4
]. These data strongly suggest that CaMKII is an important regulator of the maturation and function of DC. The activity of CaMKII is regulated by its binding to Ca+2-CaM and the resultant autophosphorylation of stimulatory and inhibitory sites [5
]. Specifically, increases in cytosolic Ca2+ result in the binding of Ca+2-CaM to a site near the autoinhibitory domain of the enzyme, leading to autophosphorylation of Thr286 and stimulation of CaMKII activity [5
, 6
]. It is important that CaMKII is dysregulated in several disease states, including tuberculosis, in which multiple abnormalities in MHC Class II antigen presentation occur [7
, 8
].
MHC Class II is a heterodimeric, transmembrane protein, which presents extracellularly derived antigens to CD4+ T cells. Its constitutive expression is limited to APC (DC, macrophages, B cells) and thymic epithelial cells, although several additional cell types have been reported to express MHC Class II under inflammatory conditions, including fibroblasts, astrocytes, and endothelial and epithelial cells. The expression of MHC Class II is tightly regulated at the mRNA and protein levels in APC [9
, 10
]. Immature DC constitutively synthesize new MHC Class II molecules, which have a t1/2 of 
10 h. Upon encounter of inflammatory stimuli and/or a danger signal, DC transiently (4–6 h) increase de novo synthesis of MHC Class II protein. Shortly thereafter, synthesis of MHC Class II protein is halted, but its t1/2 increases to >52 h within the maturating DC [10
11
12
13
]. Stimulation of APC by antigen or cytokines results in up-regulation of the surface expression of MHC Class II [14
].
MHC Class II is central to a number of immune processes, including selection of the CD4+ T cell repertoire, homeostasis of mature T cells, preservation of self-tolerance, and initiation and regulation of the adaptive immune response [9 , 11 , 15 ]. Expression of MHC Class II mRNA is regulated at the transcriptional and post-transcriptional levels. MHC Class II gene expression is controlled primarily by the MHC Class II transactivator CIITA [16 ], which is a non-DNA-binding coactivator that functions as a master regulator of MHC Class II mRNA expression by coordinating the multiple protein–protein and protein–DNA interactions required for formation of the MHC Class II enhancesome and initiation of transcription [16 17 18 ]. CIITA is the major transcriptional regulator of all MHC Class II genes, functioning as an integrator for the assembly of the transcriptome complex [18 19 20 21 ]. CIITA itself is regulated at multiple levels, including transcription, translation, and post-translational events [17 , 22 23 24 25 ]. Antigen-induced inhibition of MHC Class II gene transcription in DC is linked to decreases in transcription of CIITA [16 , 17 , 26 ]. The objective of this study was to determine the mechanisms by which CaMKII regulates the levels and subcellular localization of MHC Class II and its transcriptional regulator, CIITA, in human DC.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were obtained from R&D Systems (Minneapolis, MN, USA). Rabbit polyclonal antibody [27
] to CIITA was from AbCam (Cambridge, MA, USA). Murine mAb [28
] to CaMKII (which recognizes unphosphorylated and phosphorylated CaMKII, Pan-CaMKII) and phosphospecific CaMKII mAb were from Affinity Bioreagents (Golden, CO, USA), mAb to MHC Class II (HLA-DR) from Monosan (Uden, Netherlands), and mAb to GAPDH from Cell Signal Technologies (Danvers, MA, USA). TO-PRO-3 nuclear stain and Oregon Green- and Texas Red (TR)-conjugated secondary antibodies were from Molecular Probes (Eugene, OR, USA). Glass chamber slides were from Fisher (Hampton, NH, USA). E64 (cysteine protease inhibitor) was purchased from Sigma Chemical Co. (St. Louis, MO, USA). RNeasy kits were from Qiagen (Valencia, CA, USA). One-Step Sybr Green real-time, quantitative PCR (qPCR) kits were from Invitrogen (Carlsbad, CA, USA), and the ABI PRISM 7700 sequence detection system was from Perkin Elmer (Foster City, CA, USA). The transcriptional inhibitor, d-ribofuranosylbenzimidazole (DRB; 50 µM) was from Calbiochem (San Diego, CA, USA), and the Zenon rabbit antibody labeling kit was from Invitrogen.
Preparation of human DC
PBMC were isolated from healthy adult volunteers by Ficoll-Paque density gradient centrifugation [4
] in accordance with a protocol approved by the Human Subjects Institutional Review Board of the University of Iowa (Iowa City, IA, USA). PBMC were incubated in a 225-cm2 flask in RPMI 1640 for 1.5 h in a humidified incubator at 37°C. Nonadherent cells were removed by washing with warm RPMI and adherent monocytes cultured in RPMI 1640 with 20% FSC, L-glutamine, essential and nonessential amino acids, HEPES (10 mM), sodium pyruvate (100 mM), GM-CSF (100 ng/ml), and IL-4 (20 ng/ml). On Day 5, immature DC were harvested using 4 ml 0.25% Trypsin plus EDTA (Invitrogen). To assess the purity and phenotype of DC preparations, 106 cells were incubated with mAb to CD1a, CD11c, CD14, HLA-DR, or isotype-matched control antibody at 10 µl antibody/105 cells in PBS, 5% FBS for 40 min at 4°C. Stained cells were analyzed by flow cytometry on a FACSCalibur (Becton Dickinson, San Diego, CA, USA). Consistent with previous studies, CD1a was detected uniformly on immature DC, whereas CD14 was down-regulated by Day 5. The DC preparations exhibited >95% purity (Wright staining) and viability (exclusion of trypan blue) [4
]. All experiments, which used primary human DC, represent data from cells derived from three to five different donors, as specified in the figure legends. Primary human monocyte-derived macrophages and neutrophils [polymorphonuclear leukocytes (PMN)] were isolated as described [29
30
31
].
Preparation of antigen-stimulated DC
Immature, human, myeloid DC were activated with highly purified tetanus toxoid (TT) adsorbed onto 3 µ diameter polystyrene beads {particle-bound TT (PB-TT) [4
]}. This system is designed to model antigen processing and presentation from phagosomal compartments, as the effects of CaMKII inhibition are especially marked with particulate antigens compared with soluble antigens [4
, 32
33
34
]. For assessment of the percent change in surface expression of MHC Class II (HLA-DR), the total cell count in the M1 region (i.e., the area of the histogram considered positive when compared with the isotype control) was compared between control and treated samples. The mean fluorescence of each sample was also examined to verify results obtained using M1 gating.
Differentiation of KG-1 cells
The human DC line KG-1 (gift from Kelvin P. Lee, University of Miami, Miami, FL, USA), which exhibits an immature phenotype, was cultured in modified Eagle’s medium with L-glutamine, supplemented with 20% FCS, penicillin (100 U/ml), and streptomycin (100 µg/ml) [27
]. Cells (2x105/ml) were differentiated to a mature phenotype with 100 ng/ml GM-CSF, 20 ng/ml IL-4, and 10 ng/ml TNF- for 5 days [4
, 35
].
Cell fractionation and immunoblotting
DC (2.5x105) were in 60 µl lysis buffer A [50 mM Tris HCl (pH 6.8), 300 mM NaCl, and 0.1% SDS], plus 1% Tween detergent. For immunoblot analyses, 30 µl of each sample was separated on 12% SDS-polyacrylamide gels, transferred to polyvinylidene difluoride (PVDF) membrane, blocked for 1 h at 25°C in 5% dry milk in PBS, and then incubated with antibodies to human HLA-DR or GAPDH overnight at 4°C. Blots were then washed in TBS with 0.1% Tween (TBST) for 20 min and then incubated for 1 h at 25°C with goat anti-rabbit or rabbit anti-mouse HRP (Cell Signal Technologies) and visualized using Supersignal West Pico chemiluminescent substrate (Pierce, Rockford, IL, USA). Blots were then washed again for 20 min in TBST followed by HRP substrate, and chemiluminescence was analyzed using a Kodak X-OMAT 2000A developer.
Cathepsin D activity assay
Primary, human, immature DC (2x105) were cultured with PB-TT (5 mg/ml) at 37°C after pretreatment with buffer or inhibitor peptide (IP; 1 µM). Cells were then harvested, and whole-cell extracts were prepared by use of lysis buffer A [50 mM Tris HCl (pH 6.8) and 300 mM NaCl] plus 1% Triton X-100 detergent to generate a nondenaturing lysis buffer. A 1:10 dilution using PBS was then made from these DC lysates and used in an activity assay to quantitate levels of mature Cathepsin D according to the manufacturer’s instructions (Calbiochem). Specifically, cathepsin D activity was determined with a flourogenic assay, which uses immunocapture of the enzyme, followed by addition of a specific, quenched fluorescent peptide. Cathepsin D-specific cleavage of the peptide results in emission of fluorescence at 328 nm. Relative fluorescence unit (RFU) values are converted to cathepsin D activity via reference to a standard curve.
Caspase 3 activity assay
Primary, human, immature DC (105) were cultured with buffer or PB-TT (5 mg/ml) with 75 µl PhiPhiLux peptide substrate, which is specific for caspase 3 (cleaved PhiPhilux emits at 488 nm) at 37°C after pretreatment with buffer or IP (1 µM). As a positive control for caspase 3 activity, apoptosis was induced in primary, resting, immature DC using etoposide (1:1000 dilution). At 4 h, media were removed, and cells were washed twice with PBS (150 µl), which was then added to each well, and samples were analyzed on a fluorescent plate reader (Molecular Devices, Sunnyvale, CA, USA) at 488 nm excitation.
Real-time qPCR
DC were preincubated with or without the CaMKII IP (1 µM), derived from the autoinhibitory domain of CaMKII (Met281-Ala302, with Thr286
Ala) [36
, 37
]. We have shown previously that in human DC, IP exhibits greater potency and efficacy than other inhibitors of CaMKII (e.g., KN93, KN62) [4
]. Cells were washed twice with warm RPMI and incubated with media, PB-TT (5 mg/ml), or the calcium ionophore A23187 (10 µM) for 1 or 4 h at 37°C [4
]. RNA was prepared from whole cells using the RNeasy kit, per the manufacturer’s instructions. RNA was quantified for real-time qRT-PCR using primer sequences reported previously for MHC Class II and CIITA [19
, 38
]. Thermal cycler parameters were 45 min at 50°C, 2 min at 95°C, and 40 cycles involving denaturation at 95°C for 15 s and annealing/extension at 60°C for 1 min. Values were calculated based on cycle threshold for each condition. These were normalized to GAPDH, using the cycle threshold [19
, 38
]. Data were expressed as the percentage of the normalized ration in resting, immature DC. To determine mRNA t1/2, cells were pretreated with IP (1 µM) and DRB (50 µM). At –1 (1 h of pretreatment), 0, 1, 2, 3, 4, 12, 24, and 36 h, cells were harvested, and mRNA t1/2 was normalized to levels in untreated, immature DC (i.e., not normalized to GAPDH).
Confocal microscopy
Primary, immature DC (5x104) were incubated for 2 h at 37°C on collagen-coated slides with buffer, IP, or IP and E64 (20 µM). Cells were washed twice with warm RPMI and incubated with media, PB-TT (5 mg/ml), or A23187 (10 µM) for 1 or 4 h at 37°C. Cells were washed in ice-cold PBS, fixed in 10% formalin for 15 min, and permeabilized in methanol:acetone (1:1) for 5 min at 4°C . Samples were incubated sequentially with blocking buffer (PBS, 5% BSA, 10% horse serum), primary antibodies [phosphor-specific CaMKII mAb, Pan-CaMKII mAb (which detects phosphorylated and unphosphorylated forms of the protein), and MHC Class II mAb], and secondary goat anti-murine IgG-Oregon Green, all for 1 h at 25°C. TO-PRO-3 was added during the last 10 min of the secondary antibody incubation for nuclear staining. For analysis of lysosomal trafficking, cells were incubated with IP or buffer for 2 h, washed, and then incubated with 25 µg/ml TR-dextran for 1 h as described [39
, 40
]. The media were removed, and cells were washed until no more TR could be detected in the media. Cells were then incubated for an additional 1 h in media, then fixed, and stained for MHC Class II. For analysis of BSA localization, FITC-BSA (150 µg/ml) was added to each sample for 30 min. Cells were washed to remove excess BSA, chased for 30 min in buffer, and then incubated with buffer or PB-TT for 30 min. Confocal microscopy was performed on a Zeiss Laser Scan Inverted 510 microscope. Quantitation of differences in fluorescent staining between samples, including the degree of colocalization, was performed using ImageJ software.
Determination of CIITA levels by flow cytometry
Primary, human DC (2.5x105) were pretreated with IP or buffer and then incubated with PB-TT for 1 or 4 h. Cells were harvested and fixed using 4% paraformaldehyde for 1 h at 4°C. To obtain Oregon Green (488 nm)-labeled CIITA antibody, a Zenon antibody-labeling kit was used, per the manufacturer’s instructions. Briefly, 45 µl CIITA antibody (1 mg/ml) was incubated with 75 µl Zenon antibody for 5 min at room temperature. Blocking reagent (75 µl) was added for another 5 min at room temperature, then samples were placed on ice, and polyclonal antibodies to CIITA (20 µl direct label solution per 2.0x105 cells) were added for 30 min. Human PMN were used as a negative control for CIITA, and human monocyte-derived macrophages served as the positive control [41
, 42
].
Lysosomal degradation of FITC-BSA
Primary, human, immature DC (5x104) were pretreated with IP or buffer for 2 h and washed twice. FITC-BSA (150 µg/ml) was added to each sample for 30 min. Cells were washed to remove excess BSA, chased for 30 min in buffer, and then incubated with PB-TT or buffer for 30 min. The fluorescence of each sample was determined on a M5 Molecular Devices fluorescence plate reader at 488 nm excitation and 515 nm emission [39
, 43
].
Analysis of data
Data from each experimental group were subjected to an analysis of normality and variance. Differences between experimental groups were analyzed for statistical significance using Student’s t-test. Nonparametric evaluation of other datasets was performed using the Wilcoxon Rank Sum test [4
, 44
].
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
DC pretreated with buffer and then stimulated with buffer, A23187, or PB-TT for 1 h expressed abundant levels of MHC Class II (Fig. 1C) , consistent with previous data [45 ]. However, inhibition of CaMKII activity resulted in significant reduction in the total cellular level of MHC Class II protein at 1 h after stimulation with PB-TT (Fig. 1C) . These data confirm and extend the results from confocal microscopy and flow cytometry and support the hypothesis that CaMKII activity is critical for regulation of MHC Class II protein levels in stimulated DC.
Cysteine protease activity contributes to the decreased levels of MHC Class II in CaMKII-inhibited DC
MHC Class II synthesis and trafficking are developmentally regulated in APC via multiple mechanisms, including modulation by cysteine proteases [46
]. In immature DC, cysteine proteases, including those of the cathepsin family, exist predominantly in their immature, proenzyme form. Maturation of DC is accompanied by cleavage of the prodomain, resulting in protease activation [39
, 47
, 48
]. To test the hypothesis that cysteine proteases contribute to decreased levels of MHC Class II in IP-treated DC, we used the cysteine protease inhibitor E64. Cells were pretreated with IP, E64, IP + E64, or buffer for 2 h, washed, and incubated with PB-TT for 1, 4, or 24 h. Expression of MHC Class II was assessed by confocal microscopy and flow cytometry.
Inhibition of cysteine proteases with E64 essentially reversed the reduction of MHC Class II levels in IP-treated DC (Fig. 2 ). The total cellular levels (Fig. 2A) and the surface expression (Fig. 2B) of MHC Class II were restored to control levels at 1 h following antigen stimulation. Similar reversal of the IP-induced reductions in MHC Class II levels was noted in E64-treated cells at 4 h after antigen stimulation (not shown). Immature DC treated with E64 alone, in the absence of antigen or inhibition of CaMKII, increased their cell surface expression of MHC Class II (as determined by flow cytometry, Fig. 2C ), in agreement with previous reports [11 , 49 ]. Taken together, these data support the hypothesis that proteases contribute to the reduction in MHC Class II levels in antigen-stimulated DC, in which CaMKII is inhibited.
|
Multiple cathepsins, including B, C, D, E, H, L, and S, as well as asparagine endopeptidase are essential for antigen presentation [10 , 46 ]. Expression of each cathepsin varies among APC, and human DC express cathepsins B, D, E, H, L, and S, as well as the cathepsin inhibitor Cystatin C. [10 , 46 ]. Several cathepsins regulate MHC Class II function. Cathepsin S, which is localized predominantly in late endosomes, cleaves the invariant chain. Cathepsin D, which is primarily lysosomal, has been demonstrated to function in antigen processing to generate MHC Class II epitopes. Cathepsin D activity was determined via cleavage of a specific, quenched fluorescent peptide by immunoprecipitated Cathepsin D. Inhibition of CaMKII resulted in increased cathepsin D activity in antigen-stimulated DC; p < 0.05 (Fig. 2D , Bars 4 vs. 3).
Caspase 3 is a cysteine protease, which is expressed and active within immature DC and is down-regulated during DC maturation [50 ]. Previous studies have demonstrated that chemical inhibition of caspases is associated with increased surface expression of MHC Class II on immature DC. To determine directly whether inhibition of CaMKII modulates the activity of Caspase 3, we used a highly sensitive and specific fluorescent assay. Inhibition of CaMKII was associated with an increase in the level of caspase 3 activity in DC undergoing antigen-induced maturation; p < 0.05 (Fig. 2E , Bars 5 vs. 4).
Taken together, the data from experiments using the cysteine protease inhibitor E64, as well as direct determination of the enzymatic activities of cathepsin D and caspase 3, support a model in which inhibition of CaMKII results in increased protease activity, which contributes to reduction in the level of MHC Class II protein.
Inhibition of CaMKII activity increases trafficking of MHC Class II to lysosomes
In immature, human DC, MHC Class II molecules are transported from the endoplasmic reticulum/Golgi complex to the lysosomal network (including the MHC Class II compartment). The relatively low level of surface MHC Class II is recycled via endocytosis and undergoes lysosomal degradation. Antigen-induced maturation results in the exchange of antigenic peptide for CLIP and the transport of mature peptide–MHC complexes from the lysosomal network to the plasma membrane. This is accompanied by decreased endocytosis, leading to enhanced T cell stimulation [39
, 40
]. We hypothesized that the rapid decrease in MHC Class II levels in CaMKII-inhibited cells (within 1 h of antigen stimulation, Figs. 1
and 2
) was associated with their increased trafficking to lysosomes.
To examine this hypothesis, we labeled lysosomes with TR-dextran [39 , 40 ] and used confocal microscopy to determine the localization of MHC Class II in control and CaMKII-inhibited cells following stimulation with PB-TT for 10 min. This duration of stimulation was used to characterize the distribution of MHC Class II prior to the reductions in total cellular levels of this molecule. As demonstrated previously [39 , 51 ], TR-dextran localized to the perinuclear lysosomal region of resting, immature DC following a 1-h incubation and 1-h chase (data not shown). In antigen (PB-TT)-stimulated DC, a minor fraction of MHC Class II molecules localized to dextran-labeled lysosomes (Fig. 3A ). Inhibition of CaMKII activity in maturing DC was associated with a marked increase in the lysosomal localization of MHC Class II (Fig. 3B) . Quantitation by ImageJ software indicated that the level of lysosomal MHC Class II increased 4.7-fold in CaMKII-inhibited cells (range 4.4- to 5.1-fold; P<0.01; n=4). DC stimulated with E. coli or S. aureus (rather than TT) exhibited similar increases in lysosomal localization of MHC Class II (5.1- and 5.7-fold, respectively; p<0.01).
|
Inhibition of CaMKII does not result in a generalized increase in lysosomal protein degradation nor alter the levels of other transmembrane proteins in DC
Given that inhibition of CaMKII activity resulted in a rapid decrease in levels of MHC Class II protein, which was associated with its increased trafficking to lysosomes, we sought to determine whether enhanced lysosomal degradation of proteins was a general property of IP-treated DC. BSA has been used extensively as a soluble, exogenous protein in studies of endocytosis and lysosomal trafficking/degradation [39
]. Primary, human, immature DC were incubated with FITC-labeled BSA for 30 min to permit endocytosis ("loading period"), washed, and then incubated in buffer for an additional 30 min ("chase period"). Confocal microscopy confirmed that the loading and chase phases resulted in a punctuate, perinuclear distribution of BSA, which colocalized with the lysosomal markers lysosome-associated membrane protein-1 and CD63 (Fig. 4A
and data not shown), consistent with previous reports [39
]. Inhibition of CaMKII activity was associated with similar levels and distribution of endocytosed FITC-BSA, although the perinuclear concentration of BSA was somewhat less prominent. IP-treated DC, which were subsequently stimulated with PB-TT, exhibited no differences in the levels or subcellular localization of BSA compared with antigen-stimulated, control cells (Fig. 4A)
.
|
To determine whether inhibition of CaMKII activity is associated with a generalized reduction in expression of transmembrane proteins at the cell surface, we examined the levels of CD34 or CD1b, which exhibit stable surface expression during DC maturation [52 , 53 ]. Flow cytometry demonstrated that the levels of cell surface CD34 and CD1b did not differ between control DC and those in which CaMKII was inhibited (±antigen stimulation; Fig. 4C and 4D ). These data indicate that the rapid reduction in MHC Class II levels is not a general feature of transmembrane proteins in CaMKII-inhibited DC and suggest the hypothesis that there is a degree of selectivity in the effects on endocytosis and lysosomal trafficking of MHC Class II. Additional studies will be required to further define the specificity of this reduction in MHC Class II levels.
Inhibition of CaMKII activity reduces the level of MHC Class II mRNA in maturing DC
The central role of transcription of MHC Class II genes in adaptive immunity is demonstrated dramatically by patients with bare lymphocyte syndrome, a genetically heterogeneous disorder characterized by minimal-to-absent MHC Class II expression as a result of defects in one of a diverse group of transcriptional regulatory proteins, resulting in near total loss of cell-mediated and humoral immunity [12
, 54
]. To determine whether reductions in MHC Class II mRNA contribute to decreased protein levels in CaMKII-inhibited cells, qRT-PCR was used with normalization to the levels of GAPDH mRNA.
Immature DC were pretreated with IP or buffer for 2 h, washed, and incubated with PB-TT, A23187, or buffer for 1, 4, or 24 h. At 1 h, levels of MHC Class II mRNA were elevated modestly in A23187-treated samples (171±12% of control, P<0.001, Fig. 5A ) and remained stable following antigen stimulation (with or without IP). However, at 4 h, the marked increase in levels of MHC Class II mRNA in DC stimulated with PB-TT (760±164% of control) or A23187 (1489±136%) contrasted sharply with the reduced level in cells in which CaMKII was inhibited (1.0±1%, Fig. 5B ). At 24 h, levels of MHC Class II mRNA in the treated cells were decreased modestly compared with unstimulated controls (Fig. 5C) . Levels of GAPDH mRNA remained stable in all samples for 24 h (not shown). These data support the hypothesis that reduced levels of transcription and/or stability of MHC Class II mRNA contribute to the marked reductions in MHC Class II protein in CaMKII-inhibited cells.
|
24 h (Fig. 5D)
. However, inhibition of CaMKII reduced the t1/2 of MHC Class II mRNA by 70% to 7.5 h. These data are consistent with the hypothesis that CaMKII activity is critical in maintaining MHC Class II mRNA abundance in maturing DC, in part, by positively modulating mRNA stability. The marked reductions in levels of MHC Class II mRNA at 4 h (Fig. 5B)
strongly suggest that inhibition of CaMKII also decreases MHC Class II gene transcription.
Effect of CaMKII inhibition on CIITA mRNA levels and stability
As inhibition of CaMKII was associated with reductions in the level of MHC Class II mRNA, we tested the hypothesis that CIITA mRNA levels would also be decreased. qRT-PCR analysis demonstrated that stimulation with the Ca2+ ionophore, A23187, or antigen (TT) was associated with reduction of CIITA mRNA at 1 h (not shown), 4 h (Fig. 6A
), and 24 h (Fig. 6B)
relative to resting, immature DC. Inhibition of CaMKII resulted in a further reduction in the antigen-induced decline of CIITA mRNA at 4 h (Fig. 6A)
. It is interesting that at 24 h, IP pretreatment in maturing DC resulted in an increased abundance of CIITA mRNA (Fig. 6B)
relative to all other conditions (P<0.001). Analysis of the stability of CIITA mRNA in cells treated with DRB demonstrated that inhibition of CaMKII had no effect on the transcript stability (Fig. 6C)
. Taken together, these data support the hypothesis that CaMKII contributes to the regulation of CIITA gene expression during antigen-induced DC maturation. The direction of the inhibitor-induced modulation of CIITA levels (negative at 4 h, positive at 24 h) parallels its effects on the levels of MHC Class II protein [4
] and mRNA (Fig. 5)
.
|
|
As a complementary method to quantitatively evaluate the cellular levels of CIITA, we used flow cytometry of permeabilized DC stained with anti-CIITA polyclonal antibody. Resting, human PMN served as the negative control, as they do not express CIITA protein, whereas human, monocyte-derived macrophages were used as positive controls (Fig. 8A ) [17 , 41 , 42 ]. Inhibition of CaMKII in resting DC did not alter CIITA levels at 1 or 4 h (Fig. 8. B and 8D) . However, at 4 h following antigen stimulation, CaMKII-inhibited DC exhibited significantly decreased levels of CIITA (Fig. 8E) , in agreement with the confocal microscopy data. Taken together, these data support the hypothesis that inhibition of CaMKII in maturing DC results in reductions in the nuclear localization of CIITA and the total cellular levels of this transcriptional regulator. Further investigation is required to determine the mechanisms by which CaMKII regulates the spatial distribution and levels of CIITA and the contribution of these mechanisms to modulation of MHC Class II gene expression.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, and PGE2 [1
, 3
]. To date, the definition of the downstream signaling pathways activated by Ca2+ in DC has been limited to CaMKII, ERK1/2, MAPK, and NF-
B [4
, 57
, 59
, 60
]. Recent data support the hypothesis that the multifunctional Ser/Thr kinase, CaMKII, regulates multiple phases of DC maturation, including the up-regulation of surface MHC Class II molecules and MHC Class II-restricted T cell proliferation [4 ]. The current study demonstrates that CaMKII regulates MHC Class II levels and functions at several distinct loci, including transcriptional, post-transcriptional, and post-translational levels. Specific inhibition of CaMKII with the autoinhibitory peptide resulted in rapid (60 min) and sustained (24 h) reduction of the levels of MHC Class II protein in antigen-stimulated, primary, human DC and the human DC line KG-1 (ref. [4 ] and this manuscript). This was associated with accelerated colocalization of MHC Class II with lysosomes and was dependent on cysteine protease activity. Inhibition of CaMKII did not result in a general increase in protein degradation, as cell surface expression of the transmembrane proteins, CD34 and CD1b, was unchanged, as was the kinetics of degradation of an exogenous, soluble protein, BSA, following its endocytosis and trafficking to lysosomes. Inhibition of CaMKII also resulted in diminished levels of MHC Class II mRNA and reduced stability of this mRNA. Finally, the total cellular levels of CIITA and its nuclear localization were reduced by inhibition of CaMKII. Taken together, these data support a critical role for CaMKII in the regulation of MHC Class II in human DC via multiple, distinct mechanisms.
CaMKII is encoded by four genes (, ß, , 
), and each mRNA undergoes alternate splicing, resulting in a spectrum of distinct isozymes. As the functional enzyme complex is a heterododecamer, the range of structural diversity is enormous. Finally, the functional diversity is multiplied by numerous specific protein–protein interactions in the assembly and disassembly of spatiotemporally regulated, macromolecular signaling complexes [61
, 62
]. The isoforms of CaMKII expressed within DC are unknown, limiting the use of specific molecular modulation via dominant-negative mutants or RNA interference. The inhibitory peptide, derived from the autoinhibitory domain and modified via Ala substitution for the activation-associated phosphorylation site Thr286, is a potent (inhibitor constant=0.2 µM) and specific inhibitor of CaMKII activity [63
]. Autoinhibitory, domain-derived peptides have been characterized extensively as specific inhibitors of numerous protein kinases, including protein kinases C and A (PKC and PKA, respectively) [64
]. We are currently characterizing the isoform expression of CaMKII in immature and mature DC, which will enable complementary molecular approaches to assess CaMKII function.
Alterations in vesicular trafficking likely contribute to the rapid reduction in MHC Class II protein levels in CaMKII-inhibited DC, as demonstrated by increased lysosomal localization (Fig. 3) . The established roles of CaMKII in endosome-endosome fusion, phagosome maturation, and granule secretion are consistent with this hypothesis. In conjunction with data supporting increased activity of specific proteases (cathepsin D and caspase 3) and partial reversal via the cysteine protease inhibitor E64 (Fig. 2) , we hypothesize that inhibition of CaMKII promotes increased degradation of MHC Class II within lysosomes of DC undergoing antigen-induced maturation. This hypothesis is consistent with demonstrations of developmental regulation of lysosomal cysteine proteases, including cathepsins [10 , 50 , 65 66 67 68 ]. Maturation of DC is associated with conversion of inactive cathepsin proforms to the active proteases, reduction in natural inhibitors of the cathepsin family, progressive lysosomal acidification, and decreased activity of several caspases [39 , 40 , 46 , 50 ]. Integrated regulation of these steps is essential to the ability of DC to link innate and adaptive immunity, as demonstrated by the deleterious consequences of alterations of these processes [69 ]. Further studies will be required to identify the full range of proteases modulated by CaMKII and the mechanism(s) of this regulation.
Three independent reports have demonstrated recently that developmentally regulated ubiquitination of the cytoplasmic tail of the MHC Class II ß-chain regulates its endocytosis and level of surface expression [70 71 72 ]. This covalent modification could explain some of the selective effects of CaMKII inhibition on MHC Class II molecules (e.g., the lack of a generalized increase in lysosomal protein degradation). We hypothesize that antigen-induced activation of CaMKII inhibits ubiquitination of MHC Class II during DC maturation. Conversely, inhibition of CaMKII in stimulated DC may enhance ubquitination of MHC Class II and resultant lysosomal degradation.
Recent work by Blander and Medzhitov [73 ] has identified a critical role for TLR-specific, phagosome-restricted regulation of antigen presentation during DC maturation. This study demonstrated that membrane-trafficking events are the major regulatory locus for MHC Class II-restricted antigen presentation and the fundamental requirement to distinguish self from harmful nonself. Thus, definition of the signal transduction pathways by which TLRs promote productive, phagosome-specific interactions between microbial protein antigens and host MHC Class II molecules is crucial to our understanding of the elicitation of adaptive immunity and its potential therapeutic modulation. As the effects of CaMKII inhibition are largely limited to antigen-stimulated DC as opposed to resting cells and are quantitatively most significant with particulate (phagosomal) antigens [4 ], we hypothesize that CaMKII functions in this phagosome-autonomous pathway of antigen presentation. As phagosome-directed selection of TLR agonists and antigen occurs early after encountering pathogenic stimuli, it is worth noting that activated CaMKII localizes to nascent phagosomes within the first 15 min of ingestion [4 , 7 ].
Inhibition of CaMKII was associated with reductions in the total cellular levels of CIITA and its nuclear localization, as defined by two independent techniques (confocal microscopy with quantitation by ImageJ software and flow cytometry of permeabilized cells). To date, we have been unable to achieve sufficient sensitivity to detect endogenous CIITA by Western blot in subcellular fractions from primary, human DC. Of note, many studies about the regulation of CIITA levels and subcellular localization have been conducted in human embryo kidney and COS cells, rather than APC or APC-like cell lines. However, recent work in monocytic cell lines has also demonstrated that PKA-induced phosphorylation of CIITA results in decreased levels of transcriptional activity [74 ]. Several studies have demonstrated that phosphorylation of Ser288 is critical for the export of CIITA from the nucleus into the cytosol [23 , 24 , 74 ]. In addition, it has been hypothesized that Ser288 phosphorylation targets CIITA for ubiquitination and proteosomal degradation [24 , 75 , 76 ]. Further study will be required to determine whether CaMKII modulates the phosphorylation and/or ubiquitination of CIITA during DC maturation.
The sustained reduction in MHC Class II levels in CaMKII-inhibited cells was associated with marked decreases in MHC Class II mRNA levels and stability. Multiple proteins interact with CIITA to modulate transcription of MHC Class II genes, and insights into the regulatory networks are accumulating rapidly. For example, in resting macrophages, histone deactylase (HDAC)-4 and -5 repress transcription of MHC Class II. Antigen-induced stimulation is accompanied by phosphorylation of HDAC-4 and -5, resulting in their nucleus-to-cytosol translocation, thus removing transcriptional inhibition. Recent data implicate CaMKI and CaMKII as HDAC kinases, suggesting a potential mechanism by which CAMKII inhibitors may reduce transcription of MHC Class II genes [77 78 79 ].
MHC Class II and CIITA are regulated in an APC-specific manner. For example, MHC Class II protein expression is consistently lower in macrophages, compared with B cells and DC [80
]. This difference in protein levels is paralleled by differences in protein t1/2 of MHC Class II: 10 h in macrophages versus 52 h in mature DC [68
, 81
]. In contrast, stimulation-induced increases in MHC Class II mRNA levels are prolonged in macrophages (>48 h) and only transient in DC [10
, 16
, 82
]. In addition, macrophages express three isoforms of CIITA—pI, pIII, pIV—with a predominance of pI, whereas DC express only pI and pIII, and the levels of each are approximately equal [20
, 42
]. Finally, macrophage activation results in increased levels of CIITA mRNA levels, whereas the opposite occurs in DC, with mRNA t1/2 of CIITA of 4–8 h versus 1 h, respectively [20
, 42
]. Thus, mechanisms of MHC Class II regulation exhibit common and cell type-specific regulation, and further study of DC is required.
Macrophage phagocytosis of several medically important bacterial pathogens, including Mycobacterium tuberculosis, Salmonella typhimurium, and E. coli, results in reduction in cell-surface expression of MHC Class II, and multiple distinct mechanisms have been proposed [83 ]. In the case of M. tuberculosis, Kirschner and coworkers [8 ] recently conducted a meta-analysis and used mathematical modeling to predict the relative importance of the various proposed mechanisms of reduction in surface MHC Class II expression. Despite the significant, methodological differences between studies, this analysis indicated that the data were compatible with multiple loci of M. tuberculosis-induced inhibition of MHC Class II-restricted antigen presentation, including transcriptional, post-transcriptional, and post-translational (including modulation of vesicular trafficking). It is interesting to note that inhibition of CaMKII results in similar multifocal inhibition of MHC Class II levels and function (this study) and that M. tuberculosis inhibits macrophage CaMKII activity [7 ]. We are currently assessing the quantitative contribution of M. tuberculosis-induced inhibition of CaMKII to MHC Class II dysfunction in human DC.
The regulation and trafficking of MHC Class II have been widely studied and numerous aspects of its biology characterized. However, many of the details of these complex mechanisms remain unknown, particularly in primary human APC. The data reported herein support a model in which antigen-dependent activation of CaMKII regulates MHC Class II and CIITA in human DC. Increased understanding of the molecular mechanisms by which DC direct CD4+ T cells responses may serve to promote therapeutic modulation of the immune system.
|
ACKNOWLEDGEMENTS |
|---|
Received January 19, 2007; revised April 18, 2007; accepted May 14, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ß dimer level through cathepsin S activity in dendritic cells Immunity 23,491-502[CrossRef][Medline]
