Published online before print October 4, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
B through genome-wide survey of monocyte adherence-induced genes
,
,,,,,,,1
* Attagene Inc., Research Triangle Park, North Carolina;
Comprehensive Center for Inflammatory Disorders, and
Thurston Arthritis Center, University of North Carolina at Chapel Hill
1 Correspondence: Attagene, Inc., 7030 Kit Creek Road, Research Triangle Park, NC 27560. E-mail: smak{at}attagene.com
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B controls the expression of genes involved in inflammation, cell proliferation, apoptosis, and differentiation. Impaired regulation of NF-B has been associated with many diseases; thus, there is significant interest in therapeutic approaches based on modulation of this transcription factor. NF-B activity is controlled by numerous signaling molecules, many of which are potentially to be identified. Monocytes are principal effectors of the immune system, and monocyte adherence is the first step leading to their activation and differentiation. Adherence induces activation of NF-B, resulting in the induction of proinflammatory genes as well as anti-inflammatory genes, which counterbalance and limit the intensity and duration of NF-B activation. Here, to identify novel mediators of NF-B signaling, we used the model of monocyte adherence to perform a systematic, genome-wide survey of adherence-induced genes. Having isolated mRNAs from nonadherent and adherent primary human monocytes, we constructed suppressive subtraction hybridization libraries containing cDNAs, which were differentially regulated by adherence. Of 366 identified differentially expressed genes, most were found to be up-regulated by adherence. Having analyzed a subset of these genes, we found that the library was enriched with inhibitors of NF-B. Three of those (an orphan nuclear receptor NUR77, a guanosine 5'-diphosphate/guanosine 5'-triphosphate exchange factor RABEX5, and a PRK1-associated protein AWP1) were particularly potent inhibitors of NF-B activation. Thus, the collection of monocyte adherence-regulated genes represents a rich source for the identification of novel components of the machinery that controls NF-B activation.
Key Words: transcription factors • cell differentiation • gene regulation • inflammation • SSH subtractive hybridization
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B plays a particularly important role in induction of inflammatory genes [5
6
7
]. Cell adherence to various extracellular matrix components, such as fibronectin, collagen, or laminin, induces the rapid (within minutes) activation of NF-B [8
], which controls the expression of numerous inflammatory cytokines, including interleukin (IL)-1ß, IL-8, tumor necrosis factor 
(TNF-), and members of the growth-related oncogene (GRO) family [1
, 2
].
Along with proinflammatory genes that enhance the defensive potential of monocytes, adherence induces the expression of anti-inflammatory genes, some of which act as inhibitors of NF-B and provide a negative-feedback mechanism that limits the intensity and duration of NF-B activation. Earlier studies about monocyte adherence identified a number of positive and negative regulators of NF-B signaling [1
]. In these studies, a cDNA library from monocytes adhered on plastic was screened by differential hybridization. As a result of the limitations of this technique, only a small number of highly abundant transcripts [including IL-1ß, TNF-, macrophage-inflammatory protein-1 (MIP-1), GRO, IL-1 receptor (IL-1r), superoxide dismutase 1 (SOD 1), and tumor necrosic factor, alpha-induced protein 3 (TNFAIP3/A20) were identified. Additionally, it was in this screen that inhibitor of B (IB), the natural and potent regulatory inhibitor of NF-B, was identified [1
].
During the past decade, it has become increasingly clear that NF-B is a pivotal regulator of the inflammatory response through its role as a transcriptional regulator of genes involved in inflammation, as well as in cell growth, death, and differentiation [5
6
7
, 9
]. Impaired regulation of NF-B has been associated with many diseases, including various chronic inflammatory conditions, such as arthritis, atherosclerosis [6
, 7
, 9
], and diabetes [10
]. Thus, there is much interest in therapeutic approaches based on modulation of activation of this transcription factor. The regulation of NF-B activity is controlled primarily through receptor-induced activation of the IB kinase (IKK) complex (the signalsome), leading to phosphorylation and subsequent ubiquitination and proteasome-dependent degradation of IB [11
]. However, it has become apparent that many other signaling intermediates exist, which may also regulate the NF-B pathway [12
]. In this regard, monocyte adherence may serve as a useful model for the identification of novel mediators of NF-B signaling.
Here, we performed a systematic, genome-wide survey of monocyte adherence-induced genes to identify novel mediators of the NF-B signaling pathway. To this end, we used a model of monocyte adherence in which primary human monocytes were allowed to adhere to a tissue-culture plastic [13
, 14
]. To construct the differential cDNA libraries, we used the suppressive subtraction hybridization (SSH) technique, which allows for identification of high- and low-abundant genes [15
]. As this technique does not rely on gene sequence, it can identify known as well as yet-to-be identified genes. We have constructed two SSH cDNA libraries: one, containing genes that were up-regulated by monocyte adherence, and the other one, containing down-regulated genes. The subtraction libraries contained 366 nonredundant, differential cDNA clones, most of which were found to be up-regulated by adherence.
To assess the relevance of the differential genes to NF-B signaling, we randomly selected 25 clones from the SSH libraries, and their full-length cDNAs were expressed and assessed in a NF-B reporter assay. For a comparison, we used cDNA clones, which were randomly selected from the mammalian gene collection (MGC) IMAGE clone collection [16
]. We found that genes that encoded inhibitors of NF-B activation significantly enriched the monocyte-adherence cDNA library. Among those clones, we characterized three cDNAs (an orphan nuclear receptor NUR77, a guanosine 5'-diphosphate/guanosine 5'-triphosphate (GDP/GTP) exchange factor RABEX5, and a PRK1-associated protein AWP1), which were particularly strong inhibitors of NF-B. The expression of these cDNAs provided more than a tenfold inhibition of NF-B activation in response to stimulation with TNF- or with IL-1ß. It is important that none of these genes has been implicated previously in the regulation of NF-B. Thus, our data indicate that the differential library of primary monocyte adherence-induced genes represents a rich source for the discovery of novel mediators of NF-B signaling.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Construction of SSH libraries
Total RNAs were extracted from adherent monocytes with Clontech’s NucleoSpin nucleic acid purification kit (Palo Alto, CA). PolyA+ RNA was purified by using Oligotex RNA purification kit (Qiagen, Germantown, MD). The cDNA subtraction was done by using a Clontech cDNA subtraction kit following the manufacturer’s instructions [15
]. To construct the SSH library of genes up-regulated by adherence, one-half of monocytes was allowed to adhere to a plastic for a brief period of time (10 min). These cells were considered as nonadherent cells, and their RNA was used as a driver. The other half of cells was allowed to adhere for 45 min (the adherent population), and the RNA isolated from these cells was used as a tester. To construct the library of down-regulated genes, RNA, which was isolated from adherent cells, was used as a driver, and RNA from nonadherent cells was used as a tester. Both libraries were subcloned into a pT-Adv cloning vector (Clontech) and were used to transform DH5a bacterial cells. To assess a quality of the subtraction, 96 bacterial cDNA clones from each library were selected randomly and assessed by differential screening, as described [17
]. Briefly, the cDNA inserts were polymerase chain reaction (PCR)-amplified, arrayed on a nylon membrane, and hybridized with radiolabeled probes, which were prepared from the SSH-subtracted cDNA libraries by using the PCR-Select Differential Screening kit (Clontech) [17
]. The cDNA was amplified by PCR by using the Advantage 2 PCR enzyme system (Clontech). The Rediprime II [32P]deoxy-cytidine 5'-triphosphate labeling kit (Amersham Biosciences, Amersham, UK) was used for preparing radioactive probes. The intensities of hybridization signals were assessed quantitatively with a phosphoimager (Storm PhosphoImager, Molecular Dynamics, Amersham Biosciences AB, Uppsala, Sweden). The intensities of hybridization signals were normalized on the intensities of background hybridization with "empty" clones, i.e., the clones with no inserts (for example, see Fig. 1
, clone H12). The ratios of hybridization signals were calculated for each individual clone; the clones for which the ratio was larger than 3.0 were considered as differentially expressed ones.
 View larger version (73K): [in a new window] |
Figure 1. Differential screening of monocyte-adherent, subtracted libraries. The inserts from 96 clones randomly picked from the subtracted libraries were amplified, and 100 ng PCR products were arrayed on the nylon membrane. Both subtracted libraries, containing cDNA clones of the genes down-regulated by the adherence (A and B) and containing cDNA clones of the genes up-regulated by the adherence (C and D), were prepared. Each of the two membranes was hybridized with [32P]deoxy-adenosine 5'-triphosphate (dATP)-labeled subtracted probes specific for cDNA from nonadherent cells (A and C) and specific for cDNA from adherent cells (B and D).
|
The nonredundant set of the identified genes was classified using gene ontology (GO) annotation. As only 300 nonredundant genes were analyzed, we chose the higher hierarchies, less-specialized terms for each category. The terms were chosen on the bases of biological relevance for the process of monocyte adherence and NF-B activation.
Reverse transcriptase (RT)-PCR
First-strand cDNAs were synthesized from 1 µg total RNA by using the Moloney-murine leukemia virus RT (Invitrogen, Carlsbad, CA) following the manufacturer’s protocol. One-tenth of the cDNA was amplified by PCR by using the Advantage cDNA polymerase mix (Clontech). Gene-specific primers were designed by using a Primer3 program [18
]. PCR products were analyzed by gel electrophoresis at the exponential phase of amplification. The intensities of gene-specific PCR products were normalized on the intensity of the PCR product amplified by the primers specific for the housekeeping glyceraldehyde 3-phosphate dehydrogenase gene. The signals were digitized using the integrated Alpha Imaging Station software (Alpha Innotech Inc., San Leandro, CA).
Full-length cDNA collection
The full-length IMAGE cDNA clones were from the National Institutes of Health (NIH; Bethesda, MD) MGC [16
]. The MGC collection was purchased from Incyte (Palo Alto, CA) in June 2000 and contained 
25,000 clones corresponding to putative, full-length human and mouse cDNAs. Only clones in the mammalian expression vector promoter cytomegalovirus (pCMV)Sport6 (Invitrogen) were used.
NF-B reporter gene assay
The modified human embryonic kidney (HEK)293 cell line was generous gift from Dr. George Stark (Cleveland Clinic Foundation, Cleveland, OH). The IL-1rß was introduced to the original HEK293 cell line to enhance IL-1ß-dependent NF-B activation [19
]. To assess NF-B reporter-gene expression, cells (105 cells per six-well plate) were transfected with a 4x B luciferase reporter vector containing cDNA of firefly luciferase driven by a synthetic promoter containing four tandem copies of NF-B binding sites from major histocompatibility complex type 1 gene and a minimal TATA box from the c-fos gene [20
]. The specificity of reporter gene expression was assessed by using a mut 4x B luciferase construct containing mutant NF-B binding sites [21
]. Each transfection contained 10 µl SuperFect reagent (Qiagen) and 2 µg each reporter and expression vector. The amount of DNA was kept constant by addition of pCMVSport6 vector with no insert. Each plasmid used for transfection was purified by the Endo-Free plasmid purification kit (Qiagen). Luciferase activity was determined using luciferase systems (Promega, Madison, WI), according to the vendor’s protocols. Luciferase activity was then normalized for transfection efficiency by measuring the ß-galactosidase activity in each lysate. ß-Galactosidase activity was determined using a ß-galactosidase enzyme system (Promega), according to the supplier’s protocol. To ensure that we were able to detect NF-B activators and inhibitors in our assays, each experiment included cotransfection of luciferase reporter vector with expression vector with IB, as control for NF-B inhibition, and IKKß, as control for NF-B activation.
Statistics
Statistical differences between groups were determined by using the nonparametric ANOVA Kruskal-Wallis test or Fisher’s exact 
2 test. P < 0.01 was selected as the criterion for significance.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Two SSH subtractive libraries were constructed: one, containing cDNAs of genes up-regulated by the adherence (in which RNA from adherent cells was used as a tester, and RNA from nonadherent cells was used as a driver), and the other, containing cDNAs of genes down-regulated by the adherence (in which RNA from adherent cells was used as a driver, and RNA from nonadherent cells was used as a tester). The resulting SSH libraries were propagated in bacteria, and individual bacterial clones from each subtracted SSH library were randomly picked and arrayed in 96-well plates. An initial assessment of the libraries was done by differential screening [17 ]. To do so, the cDNA inserts of 96 bacterial clones from each SSH library were amplified individually by PCR, arrayed on a nylon membrane, and hybridized with the probes that were prepared from the "up-regulated" or from the "down-regulated" SSH library (Fig. 1 ). The individual clones from one library, which preferentially hybridized to the probe prepared from the same library, were considered as target clones, i.e., those differentially regulated by adherence (e.g., Fig. 1A , clones E5, F8, or Fig. 1D , clones A1, H1). Otherwise, we considered the clones as false positives (e.g., Fig. 1A , clone A2). We found that more than 70% of the clones from the up-regulated SSH library were target genes, which appeared differentially regulated. In contrast, there were only a few target genes in the down-regulated SSH library (>5% of the clones in the subtracted library; Fig. 1A ).
To further confirm that the SSH libraries indeed represented differentially expressed genes, we have randomly selected 44 clones from the up-regulated SSH library and analyzed them by hybridizing with probes that were prepared from transcripts, which were isolated from adherent or nonadherent monocytes ("virtual" Northern blotting [22 ]; Fig. 2 ), or by amplifying the reversely transcribed transcripts by PCR by using corresponding pairs of gene-specific primers (data are not shown). Of the analyzed 44 genes, 32 genes (73%) appeared to be more than twofold up-regulated by adherence. Therefore, we concluded that our SSH subtraction mostly yielded monocyte adherence-inducible genes, and most of these library clones represent truly differentially regulated genes.
 View larger version (38K): [in a new window] |
Figure 2. Confirmation of putative differentially expressed cDNAs by virtual Northern blot analysis. Each lane contains 0.5 µg SMART PCR-amplified cDNA from 10 min or 45 min adherent monocytes. Each blot was hybridized to one of the following [32P]dATP-labeled cDNA clones isolated from a subtracted library specific for down-regulated genes (oncogene c-fos) or subtracted library specific for up-regulated genes [neuron-derived orphan receptor-1 (NOR-1), oxidized low-density lipoprotein receptor 1 (OLR1), chemokine (C-C motif) receptor-like 2 (CCRL2), proteoglycan 1 secretory granule (PRG1), pre-B cell colony-enhancing factor (PBEF)]. ß-actin hybridization confirmed that lanes were loaded equally.
|
|
View this table: [in a new window] |
Table 1. The List of the Genes Significantly Regulated by Monocyte Adherence
|
 View larger version (28K): [in a new window] |
Figure 3. Pie chart showing the type of genes enriched in a subtracted library based on their cellular component (A), molecular function (B), and biological process (C).
|
40% of the characterized genes encoded plasma membrane proteins, and 10% encoded extracellular space proteins (Fig. 3A) . In comparison, plasma membrane proteins and extracellular space proteins represent 13% and 1.6% of all annotated gene products in the GO database. Therefore, the monocyte adherence-inducible genes encode a disproportionally large fraction of membrane-associated and secreted proteins. The classification according to a molecular function showed several large groups of genes encoding receptors (12%), transcription factors (12%) and their cofactors (8.4%), kinases (9.3%), guanyl nucleotide-binding proteins (6.1%), GTPases, and associated regulators (5.6%; Fig. 3B ).
The classification according to a biological process has identified a large fraction of genes involved in immune response (12%), cell proliferation (10%), apoptosis (8%), cell motility (7%), cell adhesion (6%), and cell-cycle control (4%; Fig. 3C ).
The monocyte adherence-induced library is enriched by genes encoding inhibitors of NF-B activation
To assess the monocyte adherence-inducible genes in regard to their influence on the NF-B pathway, we arbitrarily selected 22 genes from the up-regulated, subtracted library (Table S1). The full-length cDNAs of those genes in an expression vector were cotransfected in a human HEK293 cell line along with a NF-B luciferase reporter. As a comparison, we assessed 20 full-length human cDNAs, which were in the same expression vector and were randomly selected from the MGC collection, representing a substantial fraction of the human genome (Table S1) [16
]. The transfected cells were stimulated with TNF-, a proinflammatory cytokine considered a prototypical inducer of NF-B.
The inducible NF-B activation was determined as fold induction of luciferase reporter by TNF- versus that in cells transfected with an empty expression vector (Table S1). We expected that the cDNA clones selected from the MGC collection would not influence NF-B activation. Indeed, we found that there was an approximate equal number of cDNAs that inhibited and activated NF-B activity (the mean value of the distribution=0.9166±0.080). In contrast, the distribution among the monocyte-induced library was clearly skewed toward NF-B suppression (the mean value of the distribution=0.5755±0.059) and was significantly different from that in the population of random MGC clones (P<0.005; Fig. 4
).
 View larger version (17K): [in a new window] |
Figure 4. Enrichment for inhibitors of NF-B activity in a monocyte-adherent, subtracted library. Modified HEK293 was transiently transfected with two subsets of the cDNA clones in the pCMVSport6 expression vector along with the NF-B reporter vector. The subset of arbitrarily selected 20 clones from the MGC library was considered as the control subset, and the subset of 22 clones, corresponding to genes from the monocyte-adherent library, was considered as the experimental. The cells were treated for 6 h by TNF- after 40 h of transfection. Each experiment has been done in duplicate, and the mean of two measurements has been used in further calculations. Each circle on the figure represents the ratio of NF-B-induced luciferase activity of the cells transfected with the selected gene to that transfected with the control empty vector. *, the mean of each distribution.
|
B inhibitors: a nuclear receptor NUR77, a Rab5 GDP/GTP exchange factor homologue (RABEX5), and AWP1, a protein interacting with a serine/threonine protein kinase PRK1. All three clones inhibited NF-B in a dose-dependent manner (Fig. 5A
) and were able to inhibit IL-1ß-induced NF-B activation as well (Fig. 5B)
. It is important that the efficacy of NF-B suppression by these inhibitors was comparable with that provided by expression of a wild-type IB (Fig. 5C)
, a best-characterized NF-B inhibitor [7
, 9
, 26
].
 View larger version (22K): [in a new window] |
Figure 5. Transient NF-B reporter assay for Nur77, RABEX5, and AWP1 genes. A gene dose effect on inhibition of basal (A) or TNF-induced (B) NF-B activity. Inhibition of basal and TNF-- and IL-ß-induced NF-B activity (C). Each experiment has been done in duplicate, and the means of two measurements are presented.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B, IL-1r, and A20 [1
]. However, the approach used in those earlier studies allowed identifying only few highly abundant transcripts. In this study, we have identified more than 350 monocyte adherence-regulated genes, including six out of the seven genes (an 90% overlap), which were described in the previous study [1
]. Our list of genes partially coincides (a 30% overlap) with genes that were identified in another relevant work, which examined differential gene expression in an adherent human monocytic THP-1 cell line [8
]. In all likelihood, this discrepancy can be accounted for by the differences between primary and cancer cells and in experimental conditions (e.g., different adherence period and adherence substrates). We used the SSH technique, which has several important advantages over other common approaches to studying differential gene expression, e.g., array hybridization. As SSH does not require a previous knowledge of gene sequence, it can identify genes with unknown sequences. Also, as the frequency of clones in a SSH library is not determined by the abundance, but rather by the difference in gene expression, it allows identifying low-abundant genes, which are frequently overlooked by array hybridization.
As we focused on discovery of early genes, we used a brief (45-min) adherence. We constructed two SSH libraries: one, containing genes up-regulated by adherence, and the other, containing down-regulated genes. The vast majority of differentially regulated genes was found in the up-regulated library, suggesting that monocyte adherence results largely in gene induction.
To assess the biological relevance of the identified genes, we used the GO database. About one-half of the identified genes was annotated according to the cellular localization, involvement in biological process, and molecular function. The cellular localization-based annotation revealed an unusually high proportion of genes encoding plasma membrane-bound and secreted proteins (40%). For comparison, this category represents only 6.3% of all genes, charachterized by GO (www.geneontology.org). A large fraction of those genes encoded receptors (12%), lectins (2%), and chemokines (4%; Fig. 3B
), which mediate immune and inflammatory responses (e.g., IL-1ß, IL-8, MIP-1r, Epstein-Barr virus-induced gene 2) and cell adhesion (e.g., coronin, integrin--6, integrin--M, OLR1; Fig. 3C
).
A large group of genes, which were annotated according to the involvement in the biological process, comprised genes controlling distinct aspects of cell differentiation (22% of annotated genes), including proliferation [transforming growth factor-ß1 (TGFBI), von Hippel-Lindau syndrome (VHL), MAX dimerization protein (MAD)], cell cycle [VHL, activator of S-phase kinase, growth arrest and DNA damage (GADD)-153, Janus tyrosine kinase (JAK), N-RAS], oncogenesis [N-RAS, superkiller (SKI)1-like, ubiquitine specific peptidase (Tre-2 oncogene)], and tumor suppression (TGFBI, B-lymphocyte-induced maturation protein 1 (BLIMP1; Fig. 3C )]. Another major gene cluster was comprised of genes regulating cytoskeleton functions, e.g., monocyte spreading, migration, phagocytosis, cell motility, cell polarity, shape, and size (e.g., vanin 1, ezrin, adenylate cyclase-associated protein 1 (CAP), JAK2; 10% of annotated genes; Fig. 3C ). The cluster of pro- and antiapoptotic genes comprised 8% of annotated genes, among them, B cell chronic lymphocytic leukemia/lymphoma 2 (BCL2)-interacting protein, caspase 8, GADD-45B, programmed cell death 4, and MAD-6/A20 (Fig. 3C) .
The annotation based on the molecular function revealed a major fraction of transcription factors and their cofactors (20% of the annotated genes), e.g., transcription factors BLIMP1, MAD, GADD-153, and the cofactors zinc finger protein 136 (ZFP 136), activating transcription factor 3 (ATF3), and SKI1-like (Fig. 3B) . Two other major fractions comprised genes encoding proteins controlling DNA and RNA functions, including RNA processing and stability and DNA structure (7%; Fig. 3B
).
A large cluster of genes was found to be related to the lipid and phospholipid metabolism (6% of genes characterized according to the biological process and 3% according to the molecular function), such as OLR1 and fatty-acid-CoA ligase 2, which were among the most abundant clones in our library. This is in line with an important role that monocyte adherence plays in lipid accumulation and foam cell formation [27 , 28 ].
From numerous studies about differential gene expression, it has been known that for any biological process, cell responses to various stimuli usually result in the expression of gene products that promote the particular biological process and those counteracting it (e.g., refs. [8
, 29
, 30
]). This dichotomy is believed to allow the cell to control the intensity and duration of responses. Consistent with that, our library contains proapoptotic (caspase 8, GADD-45B, Nur77) and antiapoptotic (BCL2-interacting protein, A20 protein, tissue inhibitor of metalloproteinase 1) genes; cell growth-promoting (annexin A2, early growth response factor 3) and cell growth-arresting (GADD-153, GADD-45B) genes; oncogenes (N-RAS, Tre-2 oncogene) and tumor suppressor (VHL, TGFBI, BLIMP1), proinflammatory (IL-1ß, MIP-1), and anti-inflammatory (IL-1r, IB) genes.
Another example of this dichotomy is genes encoding signal transduction intermediates. We were particularly interested in the transcription factor NF-B, which is considered a pivotal regulator of inflammation [7
, 9
, 26
]. The activation of NF-B occurs rapidly upon monocyte adherence and controls the expression of many monocyte adherence-inducible genes [3
]. In turn, many of the inducible genes are likely to affect NF-B activity. In our library, there are several genes that have been implicated previously in the regulation of NF-B in different systems, including NF-B activators (e.g., the ATF3 [31
], arachidonate 5-lipoxygenase [32
, 33
], IL-1ß [33
], phospholipase A2 [33
], TNF receptor-associated factor family member-associated NF-B activator [34
], OLR1 [35
], heat shock protein 90 [36
], and c-mer proto-oncogene tyrosine kinase [37
]) and NF-B inhibitors (e.g., -1-antitrypsin [38
], molecule possessing ankyrin repeats induced by lipopolysaccharide [39
], IB [5
], IL-1r [40
], cold autoinflammatory syndrome 1 gene [41
], and A20 [42
]). The feedback loop may be involved in the control of the duration and intensity of NF-B activation in adherent monocytes.
We hypothesized that our subtracted library may also contain novel, yet-to-be-identified modulators of NF-B activity. To test this hypothesis, we randomly selected a small set of 22 genes, which have not been implicated in NF-B regulation. For comparison, we randomly selected 20 cDNAs from the MGC collection of human genes [16
] (Fig. 4)
. The selected cDNAs were assessed in a reporter gene assay in a model 293 cell line. As expected, the genes of the control group (from the MGC collection), on average, did not affect TNF--inducible activation of NF-B (an average value of fold induction in the group was very close to 1.0; that is, there was about the same number of weak activators and inhibitors). In contrast, an average NF-B fold induction in the group transfected with genes selected from the monocyte adherence library was about twofold lower (Fig. 4)
. This indicated that inhibitors of NF-B activation dominated in the SSH monocyte adherence library. To further characterize these inhibitors, three most potent inhibitors, NUR77, RABEX5, and AWP1, were examined in more detail. We found that the expression of these cDNAs inhibited TNF-- as well as IL-1ß-inducible NF-B activation and that the potency of these inhibitors was comparable with that of IB, a natural NF-B inhibitor (Fig. 5C)
.
The mechanisms whereby these proteins inhibit NF-B activity are unknown. The strongest NF-B inhibition in our experiments was provided by Nur77, an orphan nuclear receptor, which belongs to the NR4A subgroup of nuclear receptors [43
]. Two other members of the NR4A subgroup, Nurr1 and NOR-1, are also represented in our SSH library (Table 1)
. The Nur77 is considered a proapoptotic protein [44
]; we speculate that the proapoptotic proclivity of Nur 77 may, at least partially, be attributed to its ability to inhibit NF-B activation, which in most systems, plays an antiapoptotic function [5
, 7
, 9
, 12
, 45
].
Another novel NF-B inhibitor is Rabex5, a GDP/GTP exchange factor of the GTPase Rab5a. The Rabex5 forms a complex with Rabaptin-5 and Rab5a, which plays a role in endocytic traffic and membrane fusion and is involved in endocytosis and phagocytosis [46
]. It is interesting that Rabex5 contains an A20-like zinc finger domain, which was first found in a C-terminal region of A20 protein, a known inhibitor of NF-B [42
, 47
]. The inhibitory function of A20 was attributed to a direct interaction of its A20-like zinc finger domains with IKK
, a regulatory subunit of the IKK signalsome [47
]. In the NCBI conserved domain database, we found five human genes containing the A20-like zinc finger domain: A20, Cezanne, Rabex5, AWP1, and an AWP1-related protein ZNF216 (http://www.sanger.ac.uk). Four out of these proteins are presented in our library. Along with Rabex5, AWP1 was found in our study to be a strong NF-B inhibitor (Fig. 5)
. We speculate that the A20-like zinc finger domains of Rabex5 and AWP1 may play a role similar to that in A20-mediated NF-B inhibition. AWP1 is also known to interact directly with a serine/threonine protein kinase PRK1, a lipid-activated serine/threonine kinase of the protein kinase C superfamily [48
]. As these protein kinases have been implicated broadly in NF-B regulation [9
, 26
], this interaction may also play a role in the AWP1-mediated NF-B inhibition.
In summary, in our study, we performed a systematic, genome-wide identification of genes induced by primary human monocyte adherence. We have identified more than 350 genes and annotated half of these genes according to their localization, involvement in biological processes, and molecular function. Our study offers a better understanding of monocyte adherence, a fundamental aspect of monocyte biology. We also show that the SSH library is highly enriched by genes encoding NF-B inhibitors, and thus, it represents a rich source for the identification of novel intermediates in NF-B signaling, which may serve as targets for the development of anti-inflammatory treatments.
Received April 19, 2005; revised July 6, 2005; accepted August 1, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B and I B proteins: new discoveries and insights Annu. Rev. Immunol. 14,649-683[CrossRef][Medline]B: a pivotal transcription factor in chronic inflammatory diseases N. Engl. J. Med. 336,1066-1071 B in rheumatoid arthritis: a pivotal regulator of inflammation, hyperplasia, and tissue destruction Arthritis Res. 3,200-206[CrossRef][Medline]B as a therapeutic target in chronic inflammation: recent advances Mol. Med. Today 6,441-448[CrossRef][Medline]B-dependent fashion J. Clin. Invest. 101,1905-1915[Medline]]B activity Annu. Rev. Immunol. 18,621-663[CrossRef][Medline]B puzzle Cell 109(Suppl.),S81-S96B activation provides the potential link between inflammation and hyperplasia in the arthritic joint Proc. Natl. Acad. Sci. USA 95,13859-13864B activation Curr. Drug Targets Inflamm. Allergy 1,137-149[CrossRef][Medline] B and function in the activation of the E-selectin promoter in response to cytokines Mol. Cell. Biol. 13,7180-7190 B activation and NF B-dependent gene expression by tepoxalin, a dual inhibitor of cyclooxygenase and 5-lipoxygenase J. Cell. Biochem. 57,299-310[CrossRef][Medline]
/
PKC conveys 5-lipoxygenase/leukotriene B4-mediated cross-talk between phospholipase A2s regulating NF- B activation in response to tumor necrosis factor- and interleukin-1ß J. Biol. Chem. 276,35344-35351B activation Genes Dev. 10,963-973 B activation in endothelial cells treated with oxidized high-density lipoprotein Biochem. Biophys. Res. Commun. 303,313-319[CrossRef][Medline]B Mol. Cell. Biol. 19,1171-1181-1-antitrypsin inhibits human immunodeficiency virus type 1 FASEB J. 15,115-122B-
, a new anti-inflammatory nuclear protein induced by lipopolysaccharide, is a negative regulator for nuclear factor-B J. Endotoxin Res. 9,187-191[CrossRef] B through its type I but not its type II receptor in lymphocytes J. Biol. Chem. 267,15836-15841B suppressive properties J. Immunol. 171,6329-6333 B-dependent gene expression and apoptosis Biochem. Pharmacol. 60,1143-1151[CrossRef][Medline]B is a target of AKT in anti-apoptotic PDGF signaling Nature 401,86-90[CrossRef][Medline]) upon receptor stimulation Immunity 12,301-311[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  B. You, Y.-Y. Jiang, S. Chen, G. Yan, and J. Sun The Orphan Nuclear Receptor Nur77 Suppresses Endothelial Cell Activation Through Induction of I{kappa}B{alpha} Expression Circ. Res., March 27, 2009; 104(6): 742 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kalesnikoff, E. J. Rios, C.-C. Chen, M. Alejandro Barbieri, M. Tsai, S.-Y. Tam, and S. J. Galli Roles of RabGEF1/Rabex-5 domains in regulating Fc{epsilon}RI surface expression and Fc{epsilon}RI-dependent responses in mast cells Blood, June 15, 2007; 109(12): 5308 - 5317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Mix, M. G. Attur, H. Al-Mussawir, S. B. Abramson, C. E. Brinckerhoff, and E. P. Murphy Transcriptional Repression of Matrix Metalloproteinase Gene Expression by the Orphan Nuclear Receptor NURR1 in Cartilage J. Biol. Chem., March 30, 2007; 282(13): 9492 - 9504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. I. Bonta, C. M. van Tiel, M. Vos, T. W.H. Pols, J. V. van Thienen, V. Ferreira, E. K. Arkenbout, J. Seppen, C. A. Spek, T. van der Poll, et al. Nuclear Receptors Nur77, Nurr1, and NOR-1 Expressed in Atherosclerotic Lesion Macrophages Reduce Lipid Loading and Inflammatory Responses Arterioscler. Thromb. Vasc. Biol., October 1, 2006; 26(10): 2288 - 2288. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
