Essential Role of IκB Kinase α in the Constitutive Processing of NF-κB2 p100*

Processing of NF-κB2 precursor protein p100 to generate p52 is tightly controlled, which is important for proper function of NF-κB. Accordingly, constitutive processing of p100, caused by the loss of its C-terminal processing inhibitory domain due to nfκb2 gene rearrangements, is associated with the development of various lymphomas and leukemia. In contrast to the physiological processing of p100 triggered by NF-κB-inducing kinase (NIK) and its downstream kinase, IκB kinase α (IKKα), which requires the E3 ligase, β-transducin repeat-containing protein (β-TrCP), and occurs only in the cytoplasm, the constitutive processing of p100 is independent of β-TrCP but rather is regulated by the nuclear shuttling of p100. Here, we show that constitutive processing of p100 also requires IKKα, but not IKKβ (IκB kinase β) or IKKγ (IκB kinase γ). It seems that NIK is also dispensable for this pathogenic processing of p100. These results demonstrate a general role of IKKα in p100 processing under both physiological and pathogenic conditions. Additionally, we find that IKKα is not required for the nuclear translocation of p100. Thus, these results also indicate that p100 nuclear translocation is not sufficient for the constitutive processing of p100.

NF-B represents a collection of dimeric transcription factors composed of members of the Rel family with five closely related DNA binding proteins: RelA (p65), RelB, c-Rel, NF-B1/ p50, and NF-B2/p52 (1)(2)(3). Whereas the three Rel proteins are synthesized directly as mature proteins, p50 and p52 are generated by proteolytical processing from their large precursors p105 and p100, respectively (4,5). In resting cells, NF-B dimers are sequestered in the cytoplasm as latent complexes with a family of ankyrin repeat domain-containing inhibitors called IB proteins (1)(2)(3). Interestingly, both p105 and p100 contain ankyrin repeats at their C-terminal regions and function as IB-like inhibitors of NF-B (6,7). Whereas the IB degradation and p100 processing are inducible, the processing of p105 is constitutive (8,9). Accordingly, the signaling leading to NF-B activation can be classified into two major pathways, the canonical and non-canonical NF-B pathways, which are based on inducible IB degradation and p100 processing, respectively (10).
The canonical NF-B pathway is required for fundamental functions of various cells and can be rapidly and transiently activated by a plethora of substances, such as mitogens, cytokines, and microbial components (11). These stimuli ultimately lead to activation of a specific IB kinase (IKK) 1 complex composed of two catalytic subunits, IKK␣ (IKK1) and IKK␤ (IKK2), and a regulatory subunit, IKK␥ (NEMO) (10). Once activated, IKK phosphorylates specific serines within the IB proteins, triggering their ubiquitination by the ␤-TrCP ubiquitin ligase and degradation by the 26 S proteasome, thus allowing the NF-B dimers to move to the nucleus to induce gene expression (10). Whereas IB degradation predominantly depends on IKK␤ and IKK␥, IKK␣ is largely dispensable for this event (10), although the nuclear function of IKK␣ is necessary for the transcription activity of NF-B (12,13).
Although p100 processing is hardly detected in most cell types including T cells, aberrantly persistent processing of p100 has been found in leukemic T cells transformed by the human T cell leukemia virus type I (HTLV-I), in which the processing of p100 is induced by the viral oncoprotein Tax (15). Like NIK, the physiological inducer, Tax also specifically targets IKK␣ to p100, triggering phosphorylation-dependent ubiquitination, and processing of p100 (15,20,24). Constitutive processing of p100 has also been found in various lymphomas associated with nfb2 gene rearrangements (15,25). Interestingly, such genetic alterations always result in generation of C-terminally truncated p100 proteins lacking the processing inhibitory domain (9,25). Importantly, these C-terminal truncation mutants of p100 show oncogenic ability in vitro, and overexpression of p52 in the absence of p100 in p100 knock-in mice causes marked gastric and lymphocyte hyperplasia and early postnatal death of mice (26,27). These findings strongly suggest that deregulated processing of p100 contributes to hu-man malignancies, particularly the lymphomas and leukemia (25).
Recent studies suggested that the physiological and pathogenic processing of p100 is regulated by both common and different mechanisms (20). While NIK-mediated p100 processing depends on ␤-TrCP, constitutive processing of p100 is independent of this protein and is regulated by p100 nuclear shuttling (23,24,28). On the other hand, Tax-induced processing of p100 involves both mechanisms (20,24). However, under all these conditions, the processing of p100 is generally regulated by a ternary domain of p100 consisting of both its N-and C-terminal sequences (20). Here, we provide both genetic and biochemical evidence demonstrating that the constitutive processing of p100 also requires IKK␣ but not IKK␤ or IKK␥, suggesting another common mechanism for p100 processing. However, NIK, the activator of IKK␣ for inducible p100 processing under physiological conditions is not involved in constitutive processing of p100. We also find that IKK␣ and p100 phosphorylation are dispensable for p100 nuclear shuttling. It seems that p100 nuclear expression is not sufficient, though required, for constitutive processing, since the constitutive processing forms of p100 translocate into the nucleus of IKK null cells but defective in processing. Additionally, the phosphorylation-deficient mutants of the p100 constitutive processing forms fail to undergo processing, although they are still expressed in the nucleus.

RESULTS
The IKK␣ Phosphorylation Serines of p100 Are Essential for Its Constitutive Processing-Although we have demonstrated that IKK␣ is required for p100 processing induced by both NIK and Tax (14,15), its role in the constitutive processing of p100 remains to be investigated. For this study, the N-terminal IKK␣ phosphorylation serines within p100, which play an essential role in inducible processing of p100 (22,24), were mutated from four constitutive processing forms of p100: LB40, HuT-78, p100-(1-665), and p100-(1-454) (9, 20) (Fig. 1A). Among them, LB40 and HuT-78 proteins are expressed by the rearranged nfb2 gene in B cell chronic lymphocytic leukemia and T cell lymphoma, respectively (32,33). The constitutive processing of these p100 constructs and their serine-alanine mutants was examined. Consistent with our previous studies (9,20,24), these p100 proteins exhibited constitutive processing (Fig. 1B, odd lanes). Importantly, substitutions of the IKK␣ phosphorylation serines with alanines efficiently prevented their constitutive processing (Fig. 1B, even lanes), suggesting a positive role of IKK␣ in constitutive processing of p100.
NIK Is Not Involved in Constitutive Processing of p100 -We have recently shown that both IKK␣ and its upstream kinase NIK play an essential role in physiological processing of p100, although NIK is not required for Tax-induced p100 processing (9,14,15). It is thus very interesting and important to investigate the role of NIK in the constitutive processing of p100. Since the C-terminal portion of NIK (NIK-c) is known to function as a potent inhibitor of NIK (15,29), p100-(1-665) together with an increasing amount of NIK-c was expressed in 293 cells. Consistent with our previous studies (15), NIK-c could efficiently block NIK-induced p100 processing (Fig. 3A, compare  lane 3 with lane 2). However, NIK-c had no effect on constitutive processing of p100- (1-665) (compare lane 5 with lane 4). To further confirm this result, we also examined the processing of p100- (1-665), HuT-78, and LB40 in NIK Aly/Aly MEFs, which express mutated NIK protein and show defects in inducible p100 processing (9,16). Consistent with the result of the blocking assays above, p100-(1-665) still underwent efficient processing in the NIK Aly/Aly MEFs (Fig. 3B, lane 8). Similarly, HuT-78 and LB40 also showed efficient processing in these mutant MEFs (lanes 6 and 7). These results strongly indicate that NIK is not required for constitutive processing of p100. Nevertheless, these results are consistent with the fact that the constitutive processing of p100 does not require external stimuli.
IKK␣ Is Dispensable for p100 Nuclear Shuttling-A recent FIG. 2. Constitutive processing of p100 requires IKK␣. A, IKK␤ and IKK␥ are not required for constitutive processing of p100 C-terminal mutants. MEFs derived from wild-type (WT), IKK␤ Ϫ/Ϫ and IKK␥ Ϫ/Ϫ mice were transfected with LB40, HuT-78 or p100-(1-665), followed by in vivo p100 processing assays. The p100 mutants, p52 and endogenous p100, are indicated. B, IKK␣ is important for constitutive processing of p100 C-terminal mutants. IKK␣ Ϫ/Ϫ MEFs were transfected with LB40, HuT-78, or p100-(1-665) in the absence or presence of IKK␣, followed by p100 processing assays. The ratio of the processed products, p52, to their precursors, p100 mutants, is indicated in the figure. The expression level of transfected IKK␣ was also examined using an IKK␣specific antibody for IB. The nonspecific background band is labeled with an asterisk in lower panel.

FIG. 3. NIK is not involved in constitutive processing of p100.
A, dominant-negative NIK fails to block the constitutive processing of p100. 293 cells were transfected with indicated constructs, followed by p100 processing assays. The protein expression of NIK and NIK-c was also detected. The expression level of NIK-c in lane 3 is similar to that in lane 5 (date not shown). B, p100 C-terminal mutants undergo constitutive processing in NIK Aly/Aly MEFs. Aly/Aly MEFs were transfected with LB40, HuT-78, or p100-(1-665), followed by p100 processing assays. WT, wild-type.

FIG. 4. The nuclear shuttling of p100 is independent of IKK␣.
Wild-type (WT) and IKK␣ Ϫ/Ϫ MEFs were transfected with p100 or C-terminal HA-tagged p100-(1-665) or p100-(1-665) S/A mutants. The subcellular localization of p100 mutant was detected by immunofluorescence assays using anti-p100C (for p100) or anti-HA (for p100-(1-665) and its S/A mutant) as primary antibodies and FITC-conjugated anti-rabbit or anti-mouse Ig as secondary antibodies (upper panels). The nuclei were counterstained with Hoechst 33258 and visualized with a UV filter (lower panels). study suggested that the constitutive processing of p100 is regulated by p100 nuclear shuttling (28), although the role of IKK␣ in this event is still unknown. For this study, we performed immunofluorescence assays. In agreement with this previous study (28), p100-(1-665) was primarily located in the nucleus (Fig. 4, upper panel 2), although full-length p100 was only expressed in the cytoplasm (Fig. 4, upper panel 1). Disruption of IKK␣ phosphorylation sites did not affect the nuclear translocation of this p100 mutant (upper panel 3). Consistently, this p100 mutant was also able to move into the nucleus of IKK␣-deficient cells (upper panel 4). Similarly, mutation of IKK␣ phosphorylation sites or deficiency of IKK␣ did not alter the nuclear expression of other p100 C-terminal deletion mutants (data not shown). These results clearly indicate that IKK␣ and it-mediated phosphorylation are not required for p100 nuclear shuttling. DISCUSSION Constitutive processing of p100 due to the loss of its Cterminal processing-inhibitory domain contributes to tumorigenesis, it is therefore important to define the molecular mechanism regulating this pathogenic event. In this study, we have demonstrated that the constitutive processing of p100 is specifically regulated by IKK␣ and it-mediated phosphorylation of p100. We have also demonstrated that IKK␤/IKK␥ and NIK, which are essential for canonical and non-canonical NF-B signaling pathways, respectively, are not involved in this pathogenic event. Furthermore, we have shown that IKK␣ is dispensable for p100 nuclear translocation, further suggesting that the nuclear shuttling is not sufficient for its constitutive processing.
Although processing of p100 is usually tightly controlled, this proteolytic event could be efficiently activated at least by NIK, Tax, and the loss of the C-terminal portion of p100. While NIK-mediated p100 processing is dependent on the E3 ligase ␤-TrCP, constitutive processing of p100 C-terminal deletion mutants does not require this ligase but rather is regulated by the nuclear shuttling of p100 (23,24,28). On the other hand, both ␤-TrCP-dependent and -independent mechanisms contribute to Tax-induced processing of p100 (24). Despite these differences, common mechanisms regulating p100 processing also exist. For example, p100 processing under all these situations is controlled by its cis-acting domain (20). In the current study, we have demonstrated an essential role of IKK␣ in the constitutive processing of p100 using both biochemical and genetic approaches ( Figs. 1 and 2). These data clearly indicate another general mechanism for regulation of p100 processing, since IKK␣ is also required for NIK-and Tax-induced p100 processing (14,15).
It seems that IKK␣-mediated phosphorylation of p100 is to trigger recruitment of ubiquitin ligase(s) and subsequent ubiquitination of p100 (20,22,23,24). However, the ␤-TrCP ubiquitin ligase, known to be involved in inducible processing of p100, is not required for constitutive processing of p100 (20,23,24). These findings suggest that another yet-to-be identified ubiquitin ligase may be involved. In support of this idea, we have shown that IKK␣ is also required for ␤-TrCP-independent ubiquitination and processing of p100 induced by Tax (15,24). Furthermore, the constitutive processing of p100 requires proteasome (9), although the ubiquitination and processing of the constitutive processing forms of p100 is weak and slow, compared with the inducible ubiquitination and processing of p100 (9,23,24). It appears that the unidentified ligase is located in the nucleus, since the ␤-TrCP-independent p100 processing is regulated by p100 nuclear shuttling, and ␤-TrCP-dependent processing only occurs in the cytoplasm (20). Clearly, efforts should be focused on identification of this ligase in the future.