(HPK1) stress response signaling pathway activates IkB kinases - Nature

ã

Oncogene (1999) 18, 5514 ± 5524 1999 Stockton Press All rights reserved 0950 ± 9232/99 $15.00 http://www.stockton-press.co.uk/onc

Hematopoietic progenitor kinase-1 (HPK1) stress response signaling pathway activates IkB kinases (IKK-a/b) and IKK-b is a developmentally regulated protein kinase Mickey C-T Hu*,1, You-ping Wang1, Wan R Qiu1, Adel Mikhail2, Christian F Meyer3 and Tse-Hua Tan*,3 1

Department of Cell Biology, Amgen, Inc., Thousand Oaks, California, CA 91320, USA; 2Phylogeny, Inc., Columbus, Ohio, OH 43212, USA; 3Department of Microbiology and Immunology, Baylor College of Medicine, Houston, Texas, TX 77030, USA

Nuclear factor kappa-B (NF-kB) is a pleiotropic transcription factor that plays a central role in the immune and in¯ammatory responses, and is also involved in controlling viral transcription and apoptosis. A critical control in the activation of NF-kB is the phosphorylation of its inhibitory factor IkBs by IkB kinases (IKK-a and -b). Here, we present experiments addressing the regulation and global expression of murine IKK-b, and localize the IKK-b gene to mouse chromosome 8A3-A4. IKK-b was expressed primarily in the liver, kidney and spleen, and at lower levels in the other adult tissues. While IKK-b was expressed ubiquitously throughout the mouse embryo at 9.5 days, its expression began to be localized to the brain, neural ganglia, neural tube, and liver in the 12.5-day's embryo. At 15.5 days, the expression of IKK-b was further restricted to speci®c tissues of the embryo, suggesting that IKK-b is a developmentally regulated protein kinase. Interestingly, IKK-b phosphorylated IkB constitutively, whereas IKK-a was not active in the absence of cell stimulation. Moreover, both IKK-a and -b were activated by hematopoietic progenitor kinase-1 (HPK1) and MAPK/ ERK kinase kinase-1 (MEKK1) speci®cally, suggesting that IkB/NF-kB is regulated through the HPK1MEKK1 stress response signaling pathway. Keywords: NF-kB; IkB; HPK1; MEKK1, protein kinase; signal transduction

Introduction Nuclear factor kB (NF-kB) is a pleiotropic transcription factor that regulates the expression of viral genomes, including the human immunode®ciency virus (HIV), and a variety of cellular genes, especially those involved in the immune and in¯ammatory responses (reviewed in Siebenlist et al., 1994; Verma et al., 1995; Baeuerle and Baltimore, 1996). Recently, it has also been shown that activation of NF-kB can protect cells from cellular apoptosis induced by tumor necrosis factor (TNF) and other cytotoxic compounds (Beg and Baltimore, 1996; Liu et al., 1996; Van Antwerp et al., 1996; Wang et al., 1996). Inhibition of NF-kB activation is required in the E1A-mediated sensitiza-

*Correspondence: MC-T Hu or T-H Tan Received 22 September 1998; revised 27 January 1999; accepted 28 January 1999

tion of radiation-induced apoptosis (Shao et al., 1997). The members of the NF-kB/Rel family in mammalian cells include p50 (NF-kB1), p52 (NF-kB2), p65 (RelA), RelB, and the proto-oncoprotein c-Rel. These proteins contain a conserved *300 amino acid region known as the Rel homology domain (RHD), which is responsible for subunit dimerization, DNA binding, and nuclear translocation of NF-kB (reviewed in Baldwin, 1996). In most cells, the primary form of NF-kB is a heterodimer of p50 and RelA subunits and is retained in the cytoplasm through interaction with its inhibitory proteins, IkBs. In response to a variety of extracellular stimuli, IkBs proteins undergo rapid phosphorylation, and ubiquitination, followed by degradation through the 26S proteasome (Chen et al., 1996). This results in the release of NF-kB which translocates to the nucleus to activate the transcription of target genes containing the decameric kB DNA-binding site (reviewed in Miyamoto and Verma, 1995). A cytokine-induced IkB kinase (IKK) complex has been puri®ed (reviewed in Stancovski and Baltimore, 1997; Verma and Stevenson 1997), and a previously identi®ed 85-kD serine/threonine protein kinase of unknown function termed CHUK (Connelly and Marcu, 1995) has been shown as an IKK and designated IKK-a. Independently, IKK-a has also been identi®ed and isolated by a yeast two-hybrid screening as an interacting protein of NF-kB-inducible kinase (NIK) (Regnier et al., 1997). Subsequently, a second component (87-kD) of the human IKK complex has been isolated and designated IKK-b (Woronicz et al., 1997; Zandi et al., 1997; Mercurio et al., 1997). Strikingly, both proteins contain a conserved serine/threonine kinase domain at their amino-terminal halves, a leucine zipper motif near their central regions and two amphipathic helices of a helix ± loop ± helix domain at their carboxy-terminal portions. Functional studies indicate that an active IKK complex contains at least IKK-a and IKK-b, and both kinases are essential for IkB phosphorylation and NF-kB activation in the NF-kB signaling pathway (Woronicz et al., 1997; Zandi et al., 1997; Mercurio et al., 1997). Recently, two more critical components (NEMO and IKAP) of the IKK complex have been identi®ed (Yamaoka et al., 1998; Cohen et al., 1998). Although IKK-a and IKK-b can be activated by NIK (Regnier et al., 1997; Woronicz et al., 1997), the molecular mechanism underlying the regulation of the IKK complex is still not well de®ned. It has been suggested that mitogen-activated protein kinase (MAPK)/ERK kinase kinase-1 (MEKK1) plays an

HPK1 activates IkB kinases and characterization of mIKK-b MC-T Hu et al

important role in the TNF-a-induced NF-kB activation (Hirano et al., 1996; Meyer et al., 1996). Recently, a large IKK complex (*700-kD) has been shown to be activated by MEKK1 (Lee et al., 1997). However, the detailed mechanism by which MEKK1 activates the IKK complex leading to IkB degradation remains to be elucidated. MEKK1 is a critical regulator of the c-Jun amino-terminal kinases (JNKs)/stress-activated protein kinases (SAPKs) (Minden et al., 1994; Yan et al., 1994; Derijard et al., 1995) and hematopoietic progenitor kinase-1 (HPK1) is an upstream activator of MEKK1 (Hu et al., 1996), thereby raising the possibility that the IKK complex may be regulated by the HPK1-MEKK1 stress response pathway. Here, we isolate the full-length cDNA clone encoding the murine homolog of human IKK-b. Using In situ hybridization, we show that expression of IKK-b mRNA is regulated in dierent murine developmental stages, suggesting that IKK-b is a developmentally regulated protein kinase that may play a role in embryonic development. While IKK-a phosphorylates IkB in a cell-stimulation-dependent manner, IKK-b phosphorylates IkB constitutively. Strikingly, both IKK-a and -b are activated by MEKK1 and HPK1 speci®cally, suggesting that IkB/ NF-kB is regulated through the HPK1-MEKK1 stress response signaling pathway. Thus, these data demonstrate the ®rst signi®cant role for the HPK1-MEKK1 kinase cascade in the IkB/NF-kB signaling pathway.

Results Molecular cloning and structure of murine IKK-b cDNA We were motivated to search for the IKK-a-like kinases by recent identi®cation of the ®rst component of the IKK complex (IKK-a) (DiDonato et al., 1997; Regnier et al., 1997). A 504-bp partial cDNA sequence with signi®cant homology (*63% amino acid identity) to the kinase domain of human IKK-a cDNA was identi®ed from the Amgen expressed sequence tag (EST) database of a mouse bone cell cDNA library. Using this IKK-a-like cDNA as a probe, we have isolated a full-length cDNA clone from a mouse liver cDNA library. The nucleotide sequence of 4245 bp contains a 380 bp 5' untranslated region followed by a single open reading frame (ORF) of 2214 bp encoding a polypeptide of 738 amino acids, and followed by a 1651-bp 3' untranslated region that contains the polyadenylation signal at position 4198 (Figure 1a). The initiation codon is preceded by ®ve in-frame termination codons and presumably represents the true translation start site. The calculated molecular mass of the deduced amino acid sequence is about 82 kD. A homology search of the available databases revealed that the coding sequence of this clone is highly similar to that of the recently published human IKK-b (Woronicz et al., 1997; Zandi et al., 1997; Mercurio et al., 1997) and murine IKK-b (Nakano et al., 1998), and designated as mouse IKK-b (mIKK-b). Sequence alignment showed that the overall amino acid sequences of murine and human IKK-b exhibit 93% identity, while the extreme C-terminal domains are not conserved. (Figure 1b). Both polypeptides contain a highly conserved kinase domain (amino acids 15 ± 300,

96% identity) at their N-terminal halves, a potential leucine zipper motif near their central regions and two amphipathic helices of a candidate helix ± loop ± helix domain at their C-terminal portions. Expression of IKK-b is regulated in dierent developmental stages The expression of IKK-b was examined in a variety of mouse tissues by Northern blot analysis. A tissue Northern blot was probed with the mIKK-b cDNA, and a major miKK-b transcript (approximately 4.2 kb) was identi®ed in most tissues tested and a minor transcript (approximately 7.4 kb) was also detectable in some tissues (Figure 2). The mIKK-b transcript was expressed primarily in the liver, kidney and spleen, and at lower levels in the other adult murine tissues. The same blot was re-probed with a b-actin cDNA to check the integrity and loading of the RNAs (Figure 2, lower panel). Furthermore, we examined the expression of mIKK-b mRNA in various days of mouse embryos and adult tissues by in situ hybridization using a 35S-labeled antisense mIKK-b RNA probe, followed by autoradiography. While mIKK-b was expressed ubiquitously throughout the mouse embryo at 9.5 days (Figure 3a), its expression began to be localized to the brain, neural ganglia, neural tube, and liver in the 12.5 day embryo (Figure 3b). At 15.5 days, the expression of mIKK-b was further restricted to speci®c tissues of the embryo, primarily in the brain. The strongest signal for mIKK-b was found in the neurocortex and ventricular zone of the thalamus (Figure 3c). Low level signals were also detected in the thymus, lung, liver, and kidney, whereas mIKK-b signal had decreased to background levels in the other tissues (data not shown). In the adult mouse, signi®cant mIKK-b signal was localized to the areas of hippocampus and dentate gyrus in the brain (Figure 3d). mIKK-b was expressed above background levels in seminiferous tubules of the testis (Figure 3e), and in the thymus and liver (data not shown). The negative control hybridization with a 35Slabeled sense mIKK-b RNA probe showed the level of background in a sagittal section of a 15.5 day embryo (Figure 3f). Taken together, these results indicate that the IKK-b gene expression was modulated in dierent developmental stages, suggesting that IKK-b is a developmentally regulated protein kinase. The IKK-b gene is localized to mouse chromosome 8A3-A4 To determine the chromosomal localization of the mIKK-b gene in the mouse genome, the biotinylated mIKK-b cDNA probe was used to map the mouse chromosome, using the ¯uorescence in situ hybridization (FISH) technique (Heng et al., 1992; Heng and Tsui, 1993). A speci®c region of one chromosome showed the FISH positive with the mIKK-b probe (Figure 4a). Under the condition used, the hybridization eciency was approximately 72% for this probe (among 100 checked mitotic ®gures, 72 of them revealed positive signals on one pair of the chromosomes). Since the DAPI banding was used to identify the speci®c chromosome, the assignment between signal from the probe and the mouse chromosome 8

5515


5516

was established (Figure 4b). The detailed position was further determined to region A3-A4 based on the summary of ten photographs. There were no other positive loci detectable under the condition used; therefore, the gene of mIKK-b was mapped to mouse chromosome 8, region A3-A4 (Figure 4c).

IKK-b phosphorylates IkB constitutively, whereas IKK-a phosphorylates Ik in a cell-stimulation-dependent manner To investigate whether IKK-b could phosphorylate IkB constitutively, we transfected a N-terminal HAtagged IKK-b (HA-IKK-b) cDNA into 293T cells and

Figure 1 Nucleotide and amino acid sequences of mouse IKK-b cDNA and sequence alignment. (a) The nucleotide and predicted amino acid sequences of mouse IKK-b are shown. The deduced amino acid sequence is indicated below the ®rst nucleotide of each codon and the termination codon is marked with an asterisk. The polyadenylation signal is underlined, and the 5' end in-frame stop codons are highlighted in boxes. GenBank accession numbers: AF088910 and AF026524. (b) Alignment of the mouse and human IKK-b (mIKK-b and hIKK-b) amino acid sequences using the single letter code. The sequences were aligned with the Best®t program of the GCG sequence analysis software package. Identical amino acids are boxed in black, and the locations of the kinase domains are indicated by arrows. A putative leucine zipper region is marked by asterisks, and two amphiphathic helices of a candidate helix ± loop ± helix domain are highlighted by horizontal dash lines


prepared the cell lysates from the transfected cells without extracellular stimulation. After immunoprecipitation with anti-HA mAb, the mIKK-b kinase activity was determined by immunocomplex kinase assays, using GST-IkB-aWT, GST-IkB-aAA, and GST-IkB-aTT as substrates. The immunocomplex kinase assay detected strong phosphorylation of GSTIkB-aWT but no or very little phosphorylation of IkB mutant substrates GST-IkB-aAA and GST-IkB-aTT by mIKK-b (Figure 5a), indicating that IKK-b phosphorylates IkB constitutively in the absence of cell stimulation. As a negative control, vector alone was transfected into 293T cells, and no detectable IKK-b kinase activity was found. As a control for mIKK-b expression, equal amounts of each cell lysate were resolved by SDS ± PAGE, immunoblotted with anti-HA mAb (data not shown). Since IKK-a has been shown to be a cytokineactivated IkB kinase (DiDonato et al., 1997), we transfected a C-terminal Flag-tagged IKK-a (IKK-aFlag) cDNA into 293T cells, with or without extracellular stimulation, for a comparison. After immunoprecipitation with anti-Flag M2 mAb, the IKK-a kinase activity was determined by immunocomplex kinase assays as described above. As shown in Figure 5b, GST-IkB-aWT was phosphorylated strongly by IKK-a in the presence of TNF-a stimulation, whereas GST-IkB-aWT was not phosphorylated by

Figure 2 Expression pattern of mouse IKK-b mRNA. Poly(A)+ RNAs from the indicated mouse adult tissues were prepared for Northern blot analysis and probed with the mouse IKK-b cDNA (upper panel). As a control, the same blot was re-probed with a bactin cDNA to check the integrity of the RNA (lower panel)

IKK-a in the absence of cell stimulation. As expected, IkB mutant substrates GST-IkB-aAA and GST-IkBaTT were not phosphorylated by IKK-a, with or without TNF-a stimulation. As a control for IKK-a protein expression, equal amounts of each cell lysate were resolved by SDS ± PAGE, immunoblotted with anti-Flag M2 mAb (data not shown). Additionally, we noted that the eects of immunoprecipitations or Western blots with anti-HA and anti-Flag M2 mAbs were very comparable under our experimental conditions (Hu et al., 1997), and the autoradiographs were exposed exactly in the same manner. Thus, the discrete activities between IKK-b and IKK-a were not attributed by the dierent antibodies. To con®rm that serine residues were the major sites of phosphorylation of IkB-a by IKK-b and IKK-a, we isolated the 32P-labeled band of GST-IkB-a from the polyacrylamide gel of the immunocomplex kinase assays as described above (Figure 5a and b) and performed phosphoamino acid analysis (Boyle et al., 1991). Phosphoamino acid analysis of the in vitro phosphorylated GST-IkB-a detected P-Ser only (Figure 5c and d), indicating that both IKK-b and IKK-a phosphorylate serine residues of IkB-a predominantly. IKK-a and -b are activated by HPK1 and MEKK1 speci®cally It has been shown previously that HPK1 activates MEKK1 (Hu et al., 1996), which in turn regulates the IkB kinase complex and phosphorylation of IkB (Lee et al., 1997); hence, we tested whether IKK-a and IKK-b could be regulated by HPK1 and MEKK1. Either IKK-a-Flag or HA-IKK-b cDNA was cotransfected with an empty vector (negative control), HPK1, HPK1-M(46) mutant, and MEKK1 expression plasmids into 293T cells. The cell lysates were prepared from the transfected cells without extracellular stimulation. As positive controls, either IKK-a-Flag or HAIKK-b cDNA alone was transfected into 293T cells with TNF-a stimulation. After immunoprecipitation with either anti-Flag or anti-HA mAb, the kinase activity of IKK-a-Flag or HA-IKK-b was determined by an immunocomplex kinase assay, using GST-IkB-a as a substrate. As shown in Figure 6a, IKK-a was strongly activated by TNF-a stimulation, HPK1, and MEKK1, but not by the catalytically inactive HPK1M(46) mutant (lane 4). On the other hand, as expected, IKK-b was constitutively active in the absence of cell stimulation or activators (Figure 6b, lane 1). The discrete activities between IKK-b and IKK-a were not attributed by the dierent antibodies as described above. While the kinase activity of IKK-b was further augmented by TNF-a stimulation or HPK1 or MEKK1, its activity was not elevated by HPK1M(46) mutant (Figure 6b). As controls for IKK-a-Flag or HA-IKK-b expression, equals amounts of each cell lysate were resolved by SDS ± PAGE, and immunoblotted with either anti-Flag or anti-HA mAb. The data showed that comparable levels of IKK-a-Flag and HA-IKK-b proteins were expressed in the corresponding samples (Figure 6a and b, bottom panels), indicating that the eects of HPK1 and MEKK1 on activation of IKK-a or IKK-b were not due to higher level expression of IKK-a-Flag or HA-IKK-b. Taken

5517


5518

Figure 3 In situ hybridization (ISH) analysis of IKK-b mRNA expression in mouse embryos and adult tissues. (a) ISH of a sagittal section of a 9.5 day old mouse embryo shows ubiquitous expression of IKK-b (136 magni®cation). (b) ISH of a frontal section of a 12.5 day old mouse embryo (6.56 magni®cation). Intense labeling is observed in the neuroepithelium of the brain (a), the trigeminal ganglion (b), the liver (c), and the neural tube (d). (c) ISH of a sagittal section of a 15.5 day old mouse embryo (6.56 magni®cation). Expression of IKK-b becomes more restricted. Labeling is observed in the neurocortex (e) and ventricular zone of the thalamus (f). (d) ISH of adult mouse brain (6.56 magni®cation) shows that labeling is primarily localized to the hippocampus (g) and the dentate gyrus (h). (e) ISH shows labeling in the seminiferous tubules of the testis (526magni®cation) in the adult mouse. (f) ISH using a sense control probe shows the level of background in a sagittal section of a 15.5 day old mouse embryo (6.56 magni®cation)

together, our results indicate that both IKK-a and -b were activated by HPK1 and MEKK1 speci®cally, suggesting that IkB/NF-kB is regulated through the HPK1-MEKK1 stress response signaling pathway. Discussion We have isolated the full-length cDNA clone encoding mIKK-b, determined its global expression during development, and compared its mechanistic regulation with that of IKK-a. It is intriguing that expression of IKK-b was ubiquitous throughout the early mouse embryo (at 9.5 days) initially, then its expression began to be localized to speci®c areas of the later embryo (after 12.5 days) and adult tissues, suggesting that IKK-b is a developmentally regulated protein kinase. Since IKK-a and IKK-b are structurally and functionally related (Woronicz et al., 1997; Zandi et al., 1997; Mercurio et al., 1997) and co-expressed in adult and embryonic tissues (Hu et al., 1998), it is

possible that expression of IKK-a may also be regulated in dierent developmental stages. Because both IKKs can regulate NF-kB activity, they may have important roles in controlling gene expression during embryonic development. It has been shown that IKK-a and IKK-b can undergo both heterotypic and homotypic interactions by immunoprecipitation experiments, and their dimerization or oligomerization seems to be mediated by the leucine zipper motif (Woronicz et al., 1997; Zandi et al., 1997; Mercurio et al., 1997). Using immunoprecipitation and Western blot analysis, we also found that mIKK-b associated with IKK-a in vivo, with or without extracellular stimulation (data not shown). It has been demonstrated that both IKK-a and IKK-b make crucial contributions to IkB phosphorylation and NF-kB activation in the NF-kB signaling pathway (Woronicz et al., 1997; Zandi et al., 1997; Mercurio et al., 1997). However, it is unknown whether both kinases are regulated by the same mechanism. Here, we showed that IKK-b phosphory-


Figure 4 Chromosomal mapping of mouse IKK-b gene by ¯uorescence in situ hybridization (FISH). (a) FISH signals on a representative metaphase spread. (b) The respective 4',6'diamidino-2-phenylindole (DAPI) banding patterns of the chromosomes. (c) Schematic representation of map assignments for several metaphase spreads. Each dot represents one metaphase spread that showed a signal at the indicated chromosome band

lated IkB strongly and constitutively, whereas IKK-a was not active in the absence of cell stimulation. Upon stimulation with TNF-a, IKK-a became active but its kinase activity was still not as strong as that of IKK-b (data not shown). The distinct activities between IKK-b and IKK-a were not due to the dierent antibodies used in our experiments as described above. Similar results were observed with the endogenous IKK-a and IKK-b in the other cell types using anti-IKK-a and anti-IKK-b antisera (data not shown). Moreover, this phenomenon has also been observed by a previous report (Mercurio et al., 1997). Therefore, these results suggest that IKK-a and IKK-b may be regulated in a distinct manner. Alternatively, IKK-a may require other activators such as TRAFs to achieve optimal activity. Although IKK-b and IKK-a phosphorylate the key serine residues in IkB, it is still unclear whether both kinases directly phosphorylate IkB or they participate in the activation of a novel kinase that has the true IkB speci®city. Perhaps an IKK-a or IKK-b or IKK-

a/b de®cient (knockout) mouse model will reveal if both kinases are indispensable for IkB phosphorylation and NF-kB activation in the NF-kB signaling pathway. Both transcription factors NF-kB and c-Jun are activated by proin¯ammatory cytokines (e.g. TNF-a and interleukin-1) and environmental stresses (e.g. osmotic shock, UV and g irradiation) (reviewed by Verma et al., 1995; Baeuerle and Baltimore, 1996; Karin, 1995; Kyriakis and Avruch, 1996). Emerging evidence further suggests that both NF-kB and JNK/ SAPK pathways share common signal transduction regulators (Liu et al., 1996; Lee et al., 1997; Song et al., 1997). Previously, we have shown that HPK1 and MEKK1 are critical regulators of the JNK/SAPK signaling pathway (Hu et al., 1996). Here, we provide evidence that both IKK-a and -b are indeed activated by HPK1 and MEKK1 speci®cally, suggesting that HPK1 and MEKK1 are key coordinating regulators of both NF-kB and JNK/SAPK signaling pathways (Figure 7). However, it is unknown whether these parallel signal transduction pathways for HPK1 and MEKK1 are cell-type dependent. Further investigation of bifurcation of the IkB/NF-kB and the JNK/ SAPK pathways may reveal the regulatory network controlling these two critical signal transduction pathways. We showed that the gene of IKK-b was localized to mouse chromosome 8, region A3-A4, which is consistent with its human chromosomal locus on 8p11.2 (Hu et al., 1998). At present, it is unknown whether mutation or defect of IKK-b gene is involved in any diseases. Mutation analysis in mice or human with monogenic disorders that map to mouse chromosome 8A3-A4 or human chromosome 8p11.2, will evaluate this gene's involvement in diseases. Furthermore, targeted disruption (knockout) of this gene in mice may provide crucial evidence for the relationship between its function and diseases. In summary, we have characterized the full-length mIKK-b cDNA, and determined its global tissue distribution, chromosomal localization, and biological activity. More importantly, we have established that both IKK-a and -b are regulated via the HPK1MEKK1 stress response signaling pathway. Further investigation of the control of IKKs may contribute to a better understanding of the roles that IKKs have in normal development and pathological processes.

Materials and methods Isolation of full-length murine IKK-b cDNA A mouse 504-bp expressed sequence tag (EST) cDNA clone with *63% homology to human CHUK (IKK-a) cDNA was used as a probe to screen a mouse liver cDNA library in lTriplEx phage vector (Clontech Laboratories). For hybridization, replicate ®lters were prehybridized for 1 h at 688C in Express hybridization buer (Clontech Laboratories) and hybridized 12 h at 688C in the same solution with the 32PdCTP-labeled probe. After hybridization, the ®lters were washed several times at high stringency, at 658C in 0.1% SDS, 0.26SSC (16SSC: 150 mM NaCl and 15 mM sodium citrate), and subjected to autoradiography. Several positive

5519


5520

Figure 5 IKK-b phosphorylates IkB constitutively, whereas IKK-a phosphorylates IkB depending upon cell stimulation. (a) 293T cells were transfected with either an empty vector alone (negative control) or HA-IKK-b cDNA (10 mg each). pVA1 plasmid (10 mg) containing adenovirus VA1 RNA gene was also included in each transfection to enhance transient protein expression. The cells were harvested 48 h after transfection without stimulation. After immunoprecipitation with anti-HA mAb, IKK-b kinase activity was measured by immunocomplex kinase assays, using GST-IkB-aWT, GST-IkB-aAA, and GST-IkB-aTT as substrates. The autoradiograph was exposed for 10 min without intensifying screen. (b) 293T cells were transfected with IKK-a-Flag cDNA (10 mg) as described above. For cell stimulation, the cells were treated with human TNF-a (20 ng/ml) for 10 min before harvest. After immunoprecipitation with anti-Flag mAb, IKK-a kinase activity was measured by immunocomplex kinase assays, using the same substrates as described above. The autoradiograph was exposed for 10 min without intensifying screen. Phosphoamino acids of the in vitro phosphorylated GST-IkB-a by IKK-b (c) and IKK-a (d) were analysed electrophoretically in two dimensions using TLC with two pH systems. Speci®c phosphoamino acids are indicated. S, serine, T, threonine, Y, tyrosine

clones were picked and puri®ed after screening 46106 phages. The cDNA inserts of these positive phage clones were subsequently converted in vivo into pTripEx plasmid vector, according to the manufacturer's instructions (Clontech Laboratories). After analysis of the inserts, a candidate fulllength cDNA clone was sequenced on both strands, using a PCR procedure employing ¯uorescent dideoxynucleotides and a model 373A automated sequencer (Applied Biosys-

tems). Sequence comparisons were aligned with the Best®t program of the GCG sequence analysis software package (Wisconsin Package Version 9.0). DNA and protein reagents The HA-tagged mouse IKK-b (mIKK-b) cDNA was constructed from the full-length mIKK-b cDNA by the PCR


5521

Figure 6 Activation of IKK-a and IKK-b by HPK1 and MEKK1. (a) 293T cells were co-transfected with IKK-a-Flag cDNA and an empty vector or HPK1 or HPK1-M(46) mutant or MEKK1 expression plasmid (10 mg each). pVA1 plasmid (10 mg) containing adenovirus VA1 RNA gene was also included in each transfection to enhance transient protein expression. The cells were harvested 48 h after transfection without stimulation. As a positive control, IKK-a-Flag cDNA alone was transfected into 293T cells, and the cells were treated with human TNF-a (20 ng/ml) for 10 min before harvest (lane 2). After immunoprecipitation with anti-Flag mAb, IKK-a kinase activity was measured by an immunocomplex kinase assay, using GST-IkB-a as a substrate. The autoradiograph was exposed for 30 min without intensifying screen. As a control for IKK-a-Flag expression, equal amounts of each cell lysate (200 mg) were resolved by 10% SDS ± PAGE, and immunoblotted with anti-Flag mAb (lower panel). (b) 293T cells were co-transfected with HA-IKK-b cDNA (10 mg) and the indicated expression plasmids as described above. As a positive control, HA-IKK-b cDNA alone was transfected into 293T cells and stimulated with TNF-a as described above (lane 2). After immunoprecipitation with anti-HA mAb, IKK-b kinase activity was measured by an immunocomplex kinase assay, using the same substrate as described above. The autoradiograph was exposed for 30 min without intensifying screen. As a control for HA-IKK-b expression, equal amounts of cell lysate (200 mg) were resolved by 10% SDS ± PAGE, and immunoblotted with anti-HA mAb (lower panel)

technique using oligonucleotides 5'-TCTAGAGTCGAC GCCACCATG TACCC ATACGATG TTCCGGATTACGCTAGCCTCAGCTG GTCACCGTC CCTC CCAACC-3' and 5'TATTGCGGCCGCTTATCAGGC GGTTACCGTGAAGCTTCTGTCTTGCTCCTTC-3' as primers to incorporate a HA epitope (Tyr-Pro-Tyr-Asp-Val-Pro-Asp-Tyr-Ala-Ser-Leu) at the N-terminus of mIKK-b. The PCR generated product was cloned into the expression vector pCl-neo (Promega, Inc.), and designated pCl-HA-mIKK-b. The Flag-tagged IKK-a cDNA was generated from the full-length human IKK-a cDNA (Hu et al., 1998) by the PCR technique using oligonucleotides 5'TCTAGAGT CGACGCCACCAT GGAGCGGCCCC CGGGGCTG-3' and 5'-TATTGCGGCCGC-TTATCATTTATCATCATC AT CTTTATAATCTTCTG TTAACCAACTCCAATCAAGATTC-3' as primers to incorporate a Flag epitope (Asp-Tyr-Lys-Asp-Asp-Asp-Asp-Lys) at the C-terminus of IKK-a. The IKK-a-Flag cDNA was cloned into the expression vector pCl-neo, designated pCl-IKK-a-Flag. The sequences of these cDNA constructs were con®rmed by DNA

sequencing on both strands as described. The pVA1 (containing the adenovirus VA1 RNA gene) plasmid was obtained as previously described (Hu et al., 1996). Anti-HA mAb HA.11 (16B12) and anti-Flag M2 mAb were purchased from Berkeley Antibody Co. and Kodak Scienti®c Imaging Systems, respectively. The GST-IkB-aWT, GST-IkB-aAA, and GSTIkB-aTT plasmids were kindly provided by Dr C-Z Giam (Uniformed Services University, Bethesda, MD, USA). GSTIkB-aAA and GST-IkB-aTT are mutant GST-IkB-a [1-54], in which Ser32 and Ser36 are mutated to Ala32,36 and Thr32,36, respectively. All GST-fusion proteins were puri®ed by anity chromatography on Glutathione Sepharose 4B (Pharmacia, Inc.) according to the manufacturer's instructions. Human TNF-a was purchased from R&D Systems. Northern blot analysis Poly(A)+RNAs from various mouse tissues were obtained from Clontech Laboratories. Each sample (2 mg) was


5522

hydrolysis to give a mean size of 70 nucleotides. The tissue slides were hybridized overnight at 528C in 50% deionized formamide, 0.3 M NaCl, 20 mM Tris-HCl, pH 7.4, 10 mM NaPO4, 5 mM EDTA, 10% dextran sulfate, 16 Denhardt's, 50 mg/ml total yeast RNA, and 5 ± 7.56 104 c.p.m./ml 35Slabeled cRNA probe. The tissue slides were subjected to stringent washing at 658C in 50% formamide, 26 SSC, 10 mM DTT, and washed in PBS before treatment with 20 mg/ml RNase A at 378C for 30 min. Following washes in 26 SSC and 0.16 SSC at 378C for 10 min, the slides were dehydrated and dipped in Kodak NTB-2 nuclear track emulsion and exposed for 2 ± 3 weeks in light-tight boxes with desiccant at 48C. Photographic development was carried out in Kodak D-19. The tissue slides were counterstained lightly with toluidine blue and analysed using both light and dark ®eld optics of a microscope. Sense control cRNA probes indicate the background levels of the hybridization signal. Cell culture and transfections

Figure 7 Proposed coordinating regulation of the IkB kinases and the JNK/SAPK pathways by HPK1 and MEKK1. In the JNK/SAPK signaling pathway, HPK1 represents an upstream kinase of MEKK1, which in turn coordinately activates MKK4/ SEK, JNK/SAPK and leads to activation of transcription factor c-Jun (Hu et al., 1996). The pathway connecting cell surface receptors to HPK1 are yet to be elucidated. In the IkB/NF-kB signaling pathway, HPK1 regulates MEKK1, which in turn activates the IkB kinases (IKK-a/IKK-b) to phosphorylate IkB while still bound to NF-kB. Phosphorylation of IkB leads to degradation of IkB and thereby releasing NF-kB to the nucleus and activating transcription of many genes. In TNF-mediated kinase cascade, TNF receptor-associated factors (TRAFs) regulate NF-kB-inducible kinase (NIK) which in turn activates the IkB kinases and ultimately leads to activation of NF-kB (Regnier et al., 1997; Woronicz et al., 1997). The ubiquitinproteasome pathway for IkB degradation and potential crosstalks among these parallel signaling systems are not ruled out in this diagram

denatured and electrophoresed on a 1.2% agarose gel containing formaldehyde and then transferred to a HybondN membrane (Amersham) in 206 SSC as described (Sambrook et al., 1989). Mouse IKK-b (mIKK-b) or bactin cDNA was labeled with 32P-dCTP to a speci®c activity of approximately 108 d.p.m./mg. Membranes were hybridized with either the mIKK-b or b-actin cDNA probe, washed at high stringency, at 658C in 0.26 SCC, 0.1% SDS, and subjected to autoradiography. Probes were removed in 0.5% SDS at 95 ± 1008C. In situ hybridization (ISH) ISH was performed as described (Lyons et al., 1990). Brie¯y, fetuses and tissues were ®xed in 4% paraformaldehyde in phosphate-buered saline (PBS) overnight, dehydrated and in®ltrated with paran. Serial sections at thickness of 5 ± 7 mm were mounted on gelatin-coated slides, deparanized in xylene, rehydrated and post ®xed. The tissue sections were digested with proteinase K, post ®xed, treated with triethanolamine/acetic anhydride, washed and dehydrated. The cRNA transcripts were synthesized from linearized cDNA templates to generate antisense and sense probes, according to manufacturer's conditions (Ambion) and labeled with 35S-UTP (41000 Ci/mmol; Amersham). cRNA transcripts larger than 200 nucleotides were subjected to alkali

293T cells were grown in Dulbecco's modi®ed Eagle's (DME) medium supplemented with 10% fetal bovine serum (FBS) (Gibco/BRL). Cells to be transfected were plated the day before transfection at a density of 26106 cells per 100 mm dish. 293T cells were co-transfected with expression plasmids (10 mg each plasmid per dish) as indicated with pVA1 (10 mg per dish) to enhance transient protein expression, using the calcium phosphate precipitation protocol (Specialty Media, Inc.). The transfected 293T cells were harvested 48 h after transfection. For cell stimulation, 293T cells were treated with human TNF-a (20 ng/ml) for 10 min before harvest. Immunoprecipitation and Western blot analysis Cells were lysed in WCE lysis buer [20 mM HEPES, pH 74.2 mM EGTA, 50 mM b-glycerophosphate, 1% Triton X100, 10% glycerol, 1 mM DTT, 2 mg/ml leupeptin, 5 mg/ml aprotinin, 1 mM Pefabloc (Boehringer Mannheim) or PMSF, 1 mM sodium orthovanadate]. Soluble lysates were prepared by centrifugation at 10 000 g for 30 min at 48C. The lysates were precleared using Pansorbin cells (Calbiochem) and then incubated with speci®c antibodies. After 16 h of incubation, immunocomplexes were recovered with the aid of GammaBind Sepharose beads (Pharmacia, Inc.) and then washed four times with lysis buer. Subsequently, immunoprecipitates were analysed by Western blotting after SDS ± PAGE (10%), electroblotted onto PVDF membranes (Novex, Inc.), and probed with the corresponding rabbit antiserum or mouse monoclonal antibody. Immunocomplexes were visualized by enhanced chemiluminescence (ECL) detection (Amersham) using goat anti-rabbit or anti-mouse antisera conjugated to horseradish peroxidase as a secondary antibody (PIERCE). Immunocomplex kinase assays Immunocomplex kinase assays were carried out as described previously (Hu et al., 1997). Speci®cally, cellular HA-IKK-b or IKK-a-Flag proteins were immunoprecipitated by incubation with anti-HA or anti-Flag mAbs and protein A-agarose beads (Bio-rad) in WCE lysis buer. After 3 h of incubation at 48C, the immunoprecipitates were collected and washed twice with WCE lysis buer, twice with LiCl buer (500 mM LiCl, 100 mM Tris-Cl, pH 7.6, and 0.1% Triton X-100), and twice with kinase buer (20 mM Mops, pH 7.6, 2 mM EGTA, 10 mM MgCl2, 1 mM DTT, 0.1% Triton X-100, and 1 mM Na3VO4). Pellets were then mixed with 5 mg of substrate, 20 mCi of [g-32P]ATP, and 15 mM of unlabeled ATP in 30 ml of kinase buer. The substrates included GST-IkB-aWT, GST-IkB-aAA, and GST-IkB-aTT. The kinase reaction was performed for 30 min at 308C and terminated by boiling in


an equal volume of Laemini sample buer, and the products were resolved by SDS ± PAGE (10%). The gel was dried and subjected to autoradiography. Phosphoamino acid analysis The phosphorylated proteins obtained from immunocomplex kinase assays were transferred electrophoretically to PVDF membranes. The spots containing phosphoproteins on the membranes were excised according to the bands on autoradiograms, and then hydrolysed in 50 ml 6 N HCl for 1 h at 1108C. The supernatant was lyophilized and dissolved in 6 ml pH 1.9 buer (2.2% formic acid and 7.8% acetic acid) containing cold phosphoamino acids as markers. The phosphoamino acids were resolved electrophoretically in two dimensions using a thin-layer cellulose (TLC) plate with two pH systems as described (Boyle et al., 1991). The markers were visualized by staining with 0.2% ninhydrin in acetone and the 32P-labeled residues were detected by autoradiography. Lymphocyte culture and microscope slides preparation Lymphocytes isolated from human blood were cultured in aminimal essential medium (MEM) supplemented with 10% FBS and phytohemagglutinin at 378C for 68 ± 72 h. The lymphocyte cultures were treated with BrdU (0.18 mg/ml, Sigma) to synchronize the cell population. The synchronized cells were washed three times with serum-free medium to release the block and recultured at 378C for 6 h in a-MEM with thymidine (2.5 mg/ml, Sigma). The cells were harvested and the cell slides were prepared by using standard

procedures including hypotonic treatment, ®xation, and airdry. Chromosome mapping by ¯uorescence in situ hybridization (FISH) The procedure for FISH detection was performed as previously described (Heng et al., 1992; Heng and Tsui, 1993). Brie¯y, the cell slides were baked at 558C for 1 h. After RNase treatment, the slides were denatured in 70% formamide in 26 SSC for 2 min at 708C followed by dehydration with ethanol. DNA probes were labeled with biotinylated dATP at 158C for 1 h, using the BRL BioNick labeling kit (Gibco/BRL). Probes were denatured at 758C for 5 min in a hybridization buer containing 50% formamide and 10% dextran sulphate, and loaded onto the denatured chromosomal slides. After 16 ± 20 h hybridization, the slides were washed and incubated with ¯uorescein isothiocyanate (FITC)-conjugated avidin (Vector Laboratories), and the signal was ampli®ed as described (Heng et al., 1992). FISH signals and the 4',6'-diamidino-2-phenylindole (DAPI) banding patterns were recorded separately by taking photographs, and the assignment of the FISH mapping data with chromosomal bands was achieved by superimposing FISH signals with the DAPI-banded chromosomes (Heng and Tsui, 1993). Acknowledgements We thank B Sutton for DNA sequencing; Dr H Heng for assisting in FISH; D Paulin for technical illustration; and Drs L Souza and R Bosselman for their support.

References Baeurle PA and Baltimore D. (1996). Cell, 87, 13 ± 20. Baldwin Jr AS. (1996). Annu. Rev. Immunol., 14, 649 ± 681. Beg AA and Baltimore D. (1996). Science, 274, 782 ± 784. Boyle WJ, van der Geer P and Hunter T. (1991). Methods Enzymol., 201, 110 ± 149. Chen ZJ, Parent L and Maniatis T. (1996). Cell, 84, 853 ± 862. Cohen L, Henzel WJ and Baeuerle PA. (1998). Nature, 395, 292 ± 296. Connelly MA and Marcu KB. (1995). Cell Mol. Biol. Res., 41, 537 ± 549. Derijard B, Raingeaud J, Barrett T, Wu IH, Han J, Ulevitch RJ and Davis RJ. (1995). Science, 267, 682 ± 685. DiDonato JA, Hayakawa M, Rothwarf DM, Zandi E and Karin M. (1997). Nature, 388, 548 ± 554. Heng HHQ, Squire J and Tsui L-C. (1992). Proc. Natl. Acad. Sci. USA, 89, 9509 ± 9513. Heng HHQ and Tsui L-C. (1993). Chromosoma, 102, 325 ± 332. Hirano M, Osada S, Aoki T, Hirai S, Hosaka M, Inoue J and Ohno S. (1996). J. Biol. Chem., 271, 13234 ± 13238. Hu MC-T, Qiu WR, Wang X, Meyer CF and Tan T-H. (1996). Genes Dev., 10, 2251 ± 2264. Hu MC-T, Qiu WR and Wang Y-P. (1997). Oncogene, 15, 2277 ± 2287. Hu MC-T and Wang Y-P. (1998). Gene, 222, 31 ± 40. Karin M. (1995). J. Biol. Chem., 270, 16483 ± 16486. Kyriakis JM and Avruch J. (1996). J. Biol. Chem., 271, 24313 ± 24316. Lee FS, Hagler J, Chen ZJ and Maniatis T. (1997). Cell, 88, 213 ± 222. Liu Z-G, Hu H, Goeddel DV and Karin M. (1996). Cell, 87, 565 ± 576. Lyons GE, Schiano S, Sassoon D, Barton P and Buckingham M. (1990). J. Cell Biol., 111, 2427 ± 2436.

Mercurio F, Zhu H, Murray BW, Shevchenko A, Bennett BL, Li JW, Young DB, Barbosa M, Mann M, Manning A and Rao A. (1997). Science, 281, 860 ± 866. Meyer CF, Wang X, Chang C, Templeton D and Tan TH. (1996). J. Biol. Chem., 271, 8971 ± 8976. Minden A, Lin A, McMahon M, Lange-Carter C, Derijard B, Davis RJ, Johnson GL and Karin M. (1994). Science, 266, 1719 ± 1723. Miyamoto S and Verma IM. (1995). Adv. Cancer Res., 66, 255 ± 292. Nakano H, Shindo M, Sakon S, Nishinaka S, Mihara M, Yagita H and Okumura K. (1998). Proc. Natl. Acad. Sci. USA, 95, 3537 ± 3542. Regnier CH, Yeong Song H, Gao X, Goeddel DV, Cao Z and Rothe M. (1997). Cell, 90, 373 ± 383. Sambrook J, Fritsch EF and Maniatis T. (1989). Molecular Cloning: A Laboratory Manual. 2nd ed. Cold Spring Harbor Laboratory: Cold Spring Harbor, New York. Shao R, Karunagaran D, Zhou BP, Li K, Lo S-S, Deng J, Chiao P and Hung M-C. (1997). J. Biol. Chem., 272, 32739 ± 32742. Siebenlist, U, Franzosa G and Brown K. (1994). Annu. Rev. Cell. Biol., 10, 405 ± 455. Song HY, Regnier CH, Kirschning CJ, Goeddel DV and Rothe M. (1997). Proc. Natl. Acad. Sci. USA, 94, 9792 ± 9796. Stancovski I and Baltimore D. (1997). Cell, 91, 299 ± 302. Van Antwerp DJ, Martin SJ, Kafri T, Green DR and Verma IM. (1996). Science, 274, 787 ± 789. Verma IM and Stevenson J. (1997). Proc. Natl. Acad. Sci. USA, 94, 11758 ± 11760. Verma IM, Stevenson JK, Schwarz EM, Van Antwerp D and Miyamoto S. (1995). Genes Dev., 9, 2723 ± 2735. Wang C-Y, Mayo MW and Baldwin Jr AS. (1996). Science, 274, 784 ± 787.

5523


5524

Woronicz JD, Gao X, Cao Z, Rothe M and Goeddel DV. (1997). Science, 281, 866 ± 870. Yamaoka S, Courtois G, Bessia C, Whiteside ST, Weil R, Agou F, Kirk HE, Kay RJ and Israel A. (1998). Cell, 93, 1231 ± 1240.

Yan M, Dai T, Deak JC, Kyriakis JM, Zon LI, Woodgett JR and Templeton DJ. (1994). Nature, 372, 798 ± 800. Zandi E, Rothwarf DM, Delhase M, Hayakawa M and Karin M. (1997). Cell, 91, 243 ± 252.