Transcriptional profiling of human osteoblast differentiation.

Osteoblast differentiation is a key aspect of bone formation and remodeling. To further our understanding of the differentiation process, we have developed a collection of conditionally immortalized adult human osteoblast cell lines representing discrete stages of differentiation. To evaluate changes in gene expression associated with differentiation, polyA((+)) RNA from pre-osteoblasts, early and late osteoblasts, and pre-osteocytes was subjected to gene chip analysis using the Affymetrix Hu6800 chip in conjunction with an Affymetrix custom chip enriched in bone and cartilage cDNAs. Overall, the expression of 47 genes was found to change threefold or more on both chips between the pre-osteoblastic and pre-osteocytic stages of differentiation. Many of the observed differences, including down-regulation of collagen type I and collagen-processing enzymes, reflect expected patterns and support the relevance of our results. Other changes have not been reported and offer new insight into the osteoblast differentiation process. Thus, we observed regulation of factors controlling cell cycle and proliferation, reflecting decreased proliferation, and increased apoptosis in pre-osteocytic cells. Elements maintaining the cytoskeleton, extracellular matrix, and cell-cell adhesion also changed with differentiation reflecting profound alterations in cell architecture associated with the differentiation process. We also saw dramatic down-regulation of several components of complement and other immune response factors that may be involved in recruitment and differentiation of osteoclasts. The decrease in this group of genes may provide a mechanism for controlling bone remodeling of newly formed bone. Our screen also identified several signaling proteins that may control osteoblast differentiation. These include an orphan nuclear receptor DAX1 and a small ras-related GTPase associated with diabetes, both of which increased with increasing differentiation, as well as a high mobility group-box transcription factor, SOX4, that was down-regulated during differentiation. In summary, our study provides a comprehensive transcriptional profile of human osteoblast differentiation and identifies several genes of potential importance in controlling differentiation of osteoblasts.

Oligonucleotide arrays for expression monitoring.

This unit provides protocols for the amplification and labeling of mRNA (and the necessary controls) for hybridization to oligonucleotide arrays. It also describes methods for processing and normalizing the raw gene expression data in preparation for clustering and further analysis.

Preparation of mRNA for expression monitoring.

The ability to construct comprehensive gene expression profiles comprising hundreds to thousands of genes whose RNA levels are monitored simultaneously represents an exciting new capability in molecular biology. This is accomplished by hybridizing mRNA, which has been quantitatively amplified and labeled with biotin, to DNA chips that display thousands of nucleotides complementary to the mRNAs of interest. In this unit, rationale for starting with poly(A(+)) versus total RNA is discussed, and strategies for choosing oligonucleotides for chip design is presented. Protocols on RNA amplification and labeling, and purifying and quantifying the cDNA and in vitro transcription products are included.

Preparation of mRNA for expression monitoring.

The ability to construct comprehensive gene expression profiles comprising hundreds to thousands of genes whose RNA levels are monitored simultaneously represents an exciting new capability in molecular biology. This is accomplished by hybridizing mRNA, which has been quantitatively amplified and labeled with biotin, to DNA chips that display thousands of nucleotides complementary to the mRNAs of interest. In this unit, rationale for starting with poly(A(+)) vs. total RNA is discussed, and strategies for choosing oligonucleotides for chip design is presented. Protocols on RNA amplification and labeling, and purifying and quantifying the cDNA and in vitro transcription products are included.

Evaluation of normalization procedures for oligonucleotide array data based on spiked cRNA controls.

BACKGROUND: Affymetrix oligonucleotide arrays simultaneously measure the abundances of thousands of mRNAs in biological samples. Comparability of array results is necessary for the creation of large-scale gene expression databases. The standard strategy for normalizing oligonucleotide array readouts has practical drawbacks. We describe alternative normalization procedures for oligonucleotide arrays based on a common pool of known biotin-labeled cRNAs spiked into each hybridization. RESULTS: We first explore the conditions for validity of the 'constant mean assumption', the key assumption underlying current normalization methods. We introduce 'frequency normalization', a 'spike-in'-based normalization method which estimates array sensitivity, reduces background noise and allows comparison between array designs. This approach does not rely on the constant mean assumption and so can be effective in conditions where standard procedures fail. We also define 'scaled frequency', a hybrid normalization method relying on both spiked transcripts and the constant mean assumption while maintaining all other advantages of frequency normalization. We compare these two procedures to a standard global normalization method using experimental data. We also use simulated data to estimate accuracy and investigate the effects of noise. We find that scaled frequency is as reproducible and accurate as global normalization while offering several practical advantages. CONCLUSIONS: Scaled frequency quantitation is a convenient, reproducible technique that performs as well as global normalization on serial experiments with the same array design, while offering several additional features. Specifically, the scaled-frequency method enables the comparison of expression measurements across different array designs, yields estimates of absolute message abundance in cRNA and determines the sensitivity of individual arrays.

Tumor necrosis factor alpha-induced E-selectin expression is activated by the nuclear factor-kappaB and c-JUN N-terminal kinase/p38 mitogen-activated protein kinase pathways.

E-selectin expression by endothelium is crucial for leukocyte recruitment during inflammatory responses. Transcriptional regulation of the E-selectin promoter by tumor necrosis factor alpha (TNFalpha) requires multiple nuclear factor-kappaB (NF-kappaB) binding sites and a cAMP-responsive element/activating transcription factor-like binding site designated positive domain II (PDII). Here we characterize the role of the stress-activated family of mitogen-activated protein (MAP) kinases in induced expression of this adhesion molecule. By UV cross-linking and immunoprecipitation, we demonstrated that a heterodimer of transcription factors ATF-2 and c-JUN is constitutively bound to the PDII site. TNFalpha stimulation of endothelial cells induces transient phosphorylation of both ATF-2 and c-JUN and induces marked activation of the c-JUN N-terminal kinase (JNK1) and p38 but not extracellular signal-regulated kinase (ERK1). JNK and p38 are constitutively present in the nucleus, and DNA-bound c-JUN and ATF-2 are stably contacted by JNK and p38, respectively. MAP/ERK kinase kinase 1 (MEKK1), an upstream activator of MAP kinases, increases E-selectin promoter transcription and requires an intact PDII site for maximal induction. MEKK1 can also activate NF-kappaB -dependent gene expression. The effects of dominant interfering forms of the JNK/p38 signaling pathway demonstrate that activation of these kinases is critical for cytokine-induced E-selectin gene expression. Thus, TNFalpha activates two signaling pathways, NF-kappaB and JNK/p38, which are both required for maximal expression of E-selectin.

Transcriptional regulation of endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible enhancers.

Transcription of endothelial-leukocyte adhesion molecule-1 (E-selectin or ELAM-1), vascular cell adhesion molecule-1 (VCAM-1), and intercellular adhesion molecule-1 (ICAM-1) is induced by the inflammatory cytokines interleukin-1 beta (IL-1 beta) and tumor necrosis factor-alpha (TNF alpha). The positive regulatory domains required for maximal levels of cytokine induction have been defined in the promoters of all three genes. DNA binding studies reveal a requirement for nuclear factor-kappa B (NF-kappa B) and a small group of other transcriptional activators. The organization of the cytokine-inducible element in the E-selectin promoter is remarkably similar to that of the virus-inducible promoter of the human interferon-beta gene in that both promoters require NF-kappa B, activating transcription factor-2 (ATF-2), and high mobility group protein I(Y) for induction. Based on this structural similarity, a model has been proposed for the cytokine-induced E-selectin enhancer that is similar to the stereospecific complex proposed for the interferon-beta gene promoter. In these models, multiple DNA bending proteins facilitate the assembly of higher order complexes of transcriptional activators that interact as a unit with the basal transcriptional machinery. The assembly of unique enhancer complexes from similar sets of transcriptional factors may provide the specificity required to regulate complex patterns of gene expression and correlate with the distinct patterns of expression of the leukocyte adhesion molecules.

A striking similarity in the organization of the E-selectin and beta interferon gene promoters.

Transcription of the endothelial leukocyte adhesion molecule 1 (E-selectin or ELAM-1) gene is induced by the inflammatory cytokines interleukin-1 beta and tumor necrosis factor alpha (TNF-alpha). In this report, we identify four positive regulatory domains (PDI to PDIV) in the E-selectin promoter that are required for maximal levels of TNF-alpha induction in endothelial cells. In vitro DNA binding studies reveal that two of the domains contain novel adjacent binding sites for the transcription factor NF-kappa B (PDIII and PDIV), a third corresponds to a recently described CRE/ATF site (PDII), and a fourth is a consensus NF-kappa B site (PDI). Mutations that decrease the binding of NF-kappa B to any one of the NF-kappa B binding sites in vitro abolished cytokine-induced E-selectin gene expression in vivo. Previous studies demonstrated a similar correlation between ATF binding to PDII and E-selectin gene expression. Here we show that the high-mobility-group protein I(Y) [HMG I(Y)] also binds specifically to the E-selectin promoter and thereby enhances the binding of both ATF-2 and NF-kappa B to the E-selectin promoter in vitro. Moreover, mutations that interfere with HMG I(Y) binding decrease the level of cytokine-induced E-selectin expression. The organization of the TNF-alpha-inducible element of the E-selectin promoter is remarkably similar to that of the virus-inducible promoter of the human beta interferon gene in that both promoters require NF-kappa B, ATF-2, and HMG I(Y). We propose that HMG I(Y) functions as a key architectural component in the assembly of inducible transcription activation complexes on both promoters.

cAMP and tumor necrosis factor competitively regulate transcriptional activation through and nuclear factor binding to the cAMP-responsive element/activating transcription factor element of the endothelial leukocyte adhesion molecule-1 (E-selectin) promoter.

The cAMP-responsive element/activating transcription factor (CRE/ATF) element (also known as NF-ELAM1) of the endothelial leukocyte adhesion molecule-1 (ELAM-1) promoter is necessary for full cytokine responsiveness. It differs from a consensus cAMP-responsive element (CRE) by 1 nucleotide (G-->A conversion) and does not mediate transcriptional activation in response to cAMP. We reported previously that cAMP actually decreases ELAM-1 synthesis induced by tumor necrosis factor (TNF). We now show that cAMP decreases the ELAM-1 promoter response to TNF in transient transfection assays in bovine aortic endothelial cells and that cAMP-mediated inhibition maps to the CRE/ATF element. Electrophoretic mobility shift assays using the ELAM-1 CRE/ATF DNA sequence reveal three complexes. Antibody supershift assays suggest the slowest migrating form (complex 1) contains ATF2, the middle form (complex 2) contains ATF2 and c-Jun, and the fastest migrating form (complex 3) contains a CRE-binding protein. TNF increases c-Jun-containing complex 2 while diminishing complex 1, whereas cAMP decreases complex 2 and increases complex 1. Complex 3 is unchanged by either treatment, and the CRE-binding protein is not phosphorylated. Our data suggest that a change in the composition of the proteins binding to the CRE/ATF promoter element contributes to the competing effects of TNF and cAMP on ELAM-1 gene expression.

NF-kappa B and I kappa B alpha: an inducible regulatory system in endothelial activation.

Structural analysis of the promoters of several endothelial genes induced at sites of inflammatory or immune responses reveals binding sites for the transcription factor nuclear factor kappa B (NF-kappa B). Endothelial cells express transcripts encoding the p50/p105 and p65 components of NF-kappa B and the rel-related proto-oncogene c-rel; steady state levels of these transcripts are transiently increased by tumor necrosis factor alpha (TNF-alpha). Western blotting revealed that stimulation of endothelial cells with TNF-alpha resulted in nuclear accumulation of the p50 and p65 components of NF-kappa B. Ultraviolet crosslinking and immunoprecipitation demonstrated binding of the p50 and p65 components of NF-kappa B to the E-selectin kappa B site. Endothelial cells express an inhibitor of NF-kappa B activation, I kappa B-alpha (MAD-3). Protein levels of this inhibitor fall rapidly after TNF-alpha stimulation. In parallel, p50 and p65 accumulate in the nucleus and RNA transcript levels for I kappa B-alpha are dramatically upregulated. Recombinant p65 stimulates expression of E-selectin promoter-reporter constructs. I kappa B-alpha inhibits p65 or TNF-alpha-stimulated E-selectin promoter-reporter gene expression in transfected endothelial cells. The NF-kappa B and I kappa B-alpha system may be an inducible regulatory mechanism in endothelial activation.

Distinct IL-4 response mechanisms of the MHC gene A alpha in different mouse B cell lines.

Interleukin-4 (IL-4) is a multipotent cytokine which stimulates proliferation of B and T lymphocytes, induces B lymphocyte expression of major histocompatibility complex (MHC) class II molecules and Fc epsilon R II (CD23) molecules, and promotes immunoglobulin class switching to IgE and IgG1. The mechanisms by which IL-4 induces these changes are unclear. To study the basis for heterogeneity in induction of class II MHC proteins observed in splenic B cells, three mouse B cell lines were treated with IL-4, and the response of MHC class II A alpha mRNA was analyzed. Each of the three cell lines responded with a distinctive profile. In one line, 70Z/3, A alpha mRNA was induced greater than 10 fold by 65 hr of IL-4 stimulation. Additional studies showed that A alpha mRNA was stabilized by IL-4 treatment of 70Z/3 cells, and that changes in gene transcription accounted for little of the increase in mRNA levels. A second line, WEHI.231, was shown to increase A alpha mRNA levels 4 fold after 48 hr of IL-4 treatment. In contrast to 70Z/3, when A alpha mRNA stability in the IL-4 treated WEHI.231 cells was compared to untreated cells, no difference was observed, IL-4 treatment induced A alpha transcription. The third cell line, M12.4.1, expressed high basal levels of A alpha, and these levels increased only slightly following IL-4 stimulation. The small increase correlated with a comparable transcriptional response. These data shown that the nature of the A alpha gene response to IL-4 differs among B cell lines. This heterogeneity of response is consistent with responses in total splenic B cells, and with the existence of functionally distinct subpopulations of B cells.

A common theme in endothelial activation Insights from the structural analysis of the genes for E-selectin and VCAM-1.

An important early step in the development of an inflammatory infiltrate is the induction of new endothelial cell surface proteins. Cytokine-induced expression of endothelial-leukocyte adhesion molecule-1 (E-selectin or ELAM-1) may mediate neutrophil adhesion to endothelium, whereas induced expression of vascular cell adhesion molecule (VCAM-1) may be important in the adhesion of mononuclear cells. Both the corresponding genes are transcriptionally regulated by the inflammatory cytokines. Structural analysis of the promoters of these genes reveals binding sites for the transcription factor nuclear factor κB (NF-κB). Activation of the pleiotropic mediator NF-κB in endothelial cells could coordinate the expression of numerous endothelial products that are important in endothelial activation, including some cell surface adhesion proteins, cytokines, growth factors, and components of the coagulation system.

Functional analysis of the human vascular cell adhesion molecule 1 promoter.

The vascular cell adhesion molecule 1 (VCAM-1) is a 110-kD member of the immunoglobulin gene superfamily expressed on the surface of interleukin 1 beta- or tumor necrosis factor alpha (TNF)-stimulated endothelial cells. The cell surface protein functions as an inducible adhesion receptor for circulating mononuclear leukocytes and some tumor cells. We have previously characterized the genomic organization of the VCAM1 gene and described its chromosomal localization. In this report, the promoter of the VCAM1 gene is characterized. New transcription of the VCAM1 gene occurred when endothelial cells were treated with TNF. Fusion plasmids containing the 5' flanking sequence of the VCAM1 gene and the chloramphenicol acetyltransferase reporter gene were used to identify cis-acting sequences that direct the cytokine-induced transcription. When transfected into bovine aortic endothelial cells, constructs containing 755 bp of the 5' flanking sequence were induced by TNF. Within the cytokine-responsive region of the core promoter were functional NF-kappa B and GATA elements. Upstream of the core promoter, the VCAM1 5' flanking sequence contained a negative regulatory activity. NF-kappa B-mediated activation of VCAM1 gene expression may lead to endothelial expression of a mononuclear leukocyte adhesion molecule associated with initial events in the development of an atherosclerotic lesion.

Non-consensus DNA sequences function in a cell-type-specific enhancer of the mouse class II MHC gene A alpha.

Class II MHC proteins play central roles in controlling immune cell repertoire and responses. These roles depend on precise regulation of the level and cell-type specificity of class II gene expression. Instances of both coordinate and non-coordinate regulation of the multiple class II genes have been described. A 1.3 kb region of the class II MHC gene A alpha has previously been shown to activate transcription in a cell-type specific fashion that correlated with the expression of A alpha. The mouse A alpha gene differs from other class II MHC genes in that its conserved X region also contains the CRE/ATF DNA motif TGACGTCA. Substitution mutations were introduced into the 1.3 kb region such that the CRE/ATF (X2) motif was altered, but not the adjacent X1 or Y box motifs. Controls confirmed that these mutations eliminated the binding of nuclear proteins to the CRE/ATF motif and reduced transcriptional activity as much as mutation of the Y box. In addition, a new positive transcription element was identified far upstream from the conserved X-Y region, centered on position -970. The sequence of this region does not resemble previously described transcription elements or other MHC class II 5' flanking sequences. The activity of this element was absolutely dependent on the presence of the X-Y region. These data are most consistent with a model in which functionally important sequences unique to a single class II MHC gene can be intimately interposed between conserved MHC transcription elements, and non-consensus elements upstream from the conserved region contribute to control of A alpha.