TrafficVolume 16, Issue 8 p. 871-892 Toolbox Free Access A Cell-Based Assay Reveals Nuclear Translocation of Intracellular Domains Released by SPPL Proteases Torben Mentrup, Biochemical Institute, Christian Albrechts University of Kiel, Otto-Hahn-Platz 9, D-24118 Kiel, GermanySearch for more papers by this authorRobert Häsler, Institute of Clinical Molecular Biology, Christian Albrechts University of Kiel, Schittenhelmstr. 12, D-24105 Kiel, GermanySearch for more papers by this authorRegina Fluhrer, Biomedizinisches Centrum (BMC), Ludwig Maximilians University of Munich, Feodor-Lynen-Strasse 17, D-81377 Munich, Germany DZNE – German Center for Neurodegenerative Diseases, Feodor-Lynen-Strasse 17, D-81377 Munich, GermanySearch for more papers by this authorPaul Saftig, Biochemical Institute, Christian Albrechts University of Kiel, Otto-Hahn-Platz 9, D-24118 Kiel, GermanySearch for more papers by this authorBernd Schröder, Corresponding Author Biochemical Institute, Christian Albrechts University of Kiel, Otto-Hahn-Platz 9, D-24118 Kiel, GermanyCorresponding author: Bernd Schröder, baschroeder@biochem.uni-kiel.deSearch for more papers by this author Torben Mentrup, Biochemical Institute, Christian Albrechts University of Kiel, Otto-Hahn-Platz 9, D-24118 Kiel, GermanySearch for more papers by this authorRobert Häsler, Institute of Clinical Molecular Biology, Christian Albrechts University of Kiel, Schittenhelmstr. 12, D-24105 Kiel, GermanySearch for more papers by this authorRegina Fluhrer, Biomedizinisches Centrum (BMC), Ludwig Maximilians University of Munich, Feodor-Lynen-Strasse 17, D-81377 Munich, Germany DZNE – German Center for Neurodegenerative Diseases, Feodor-Lynen-Strasse 17, D-81377 Munich, GermanySearch for more papers by this authorPaul Saftig, Biochemical Institute, Christian Albrechts University of Kiel, Otto-Hahn-Platz 9, D-24118 Kiel, GermanySearch for more papers by this authorBernd Schröder, Corresponding Author Biochemical Institute, Christian Albrechts University of Kiel, Otto-Hahn-Platz 9, D-24118 Kiel, GermanyCorresponding author: Bernd Schröder, baschroeder@biochem.uni-kiel.deSearch for more papers by this author Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URLShare a linkShare onEmailFacebookTwitterLinked InRedditWechat Abstract During regulated intramembrane proteolysis (RIP) a membrane-spanning substrate protein is cleaved by an ectodomain sheddase and an intramembrane cleaving protease. A cytoplasmic intracellular domain (ICD) is liberated, which can migrate to the nucleus thereby influencing transcriptional regulation. Signal peptide peptidase-like (SPPL) 2a and 2b have been implicated in RIP of type II transmembrane proteins. Even though SPPL2a might represent a potential pharmacological target for treatment of B-cell-mediated autoimmunity, no specific and potent inhibitors for this enzyme are currently available. We report here on the first quantitative cell-based assay for measurement of SPPL2a/b activity. Demonstrating the failure of standard Gal4/VP16 reporter assays for SPPL2a/b analysis, we have devised a novel system employing β-galactosidase (βGal) complementation. This is based on detecting nuclear translocation of the proteolytically released substrate ICDs, which results in specific restoration of βGal activity. Utilizing this potentially high-throughput compatible new setup, we demonstrate nuclear translocation of the ICDs from integral membrane protein 2B (ITM2B), tumor necrosis factor (TNF) and CD74 and identify secreted frizzled-related protein 2 (SFRP2) as potential transcriptional downstream target of the CD74 ICD. We show that the presented assay is easily adaptable to other intramembrane proteases and therefore represents a valuable tool for the functional analysis and development of new inhibitors of this class of enzymes.Intramembrane proteolysis involves the processing of membrane-spanning substrate proteins within their hydrophobic transmembrane segments by so-called I-CLIPs (intramembrane cleaving proteases) 1. Usually, this process is part of a proteolytic cascade that is initiated by processing of the substrate\'s ectodomain and thereby is involved in the turnover of a variety of membrane proteins 1. Beyond that, cleavage fragments generated by intramembrane proteolysis can be actively involved in cellular signal transduction after their release from the membrane involving translocation to the nucleus as it is the case in Notch signaling 2, 3. The signal peptide peptidase/signal peptide peptidase-like (SPP/SPPL) family of aspartyl I-CLIPs is constituted by SPP and four homologous SPPL proteins SPPL2a, -b, -c and SPPL3, which differ in their subcellular localizations 4, 5. SPP/SPPL proteases show selectivity for substrate proteins in type II orientation and exhibit a catalytic center similar to that of the presenilins, which are the catalytically active part of the γ-secretase complex. The family member SPPL2a is primarily localized in membranes of lysosomes and late endosomes 6 and was shown to cleave N-terminal fragments (NTFs) of tumor necrosis factor (TNF) 7, 8, Fas ligand 9, integral membrane protein 2B (ITM2B) 10 and TMEM106B 11 in cell-based setups. So far, the invariant chain (CD74), a chaperone of the MHCII complex, represents the only in vivo validated substrate of this protease 12-14. CD74 binds MHCII in the endoplasmic reticulum (ER) and facilitates targeting of the complex to endosomal compartments where CD74 undergoes step-wise proteolytic degradation starting from the luminal domain 15. In B lymphocytes of SPPL2a-deficient mice, significant amounts of uncleaved CD74 NTF accumulate and disturb homeostasis of these cells. This leads to a block of B-cell maturation at the transitional stage 1 (T1), a loss of dendritic cells and severely impaired humoral immune responses in Sppl2a−/− mice 12-14. Though this phenotype is clearly driven by the CD74 NTF as documented by its alleviation in Sppl2a−/− Cd74−/− mice 12, the fate and putative downstream function of the CD74 intracellular domain (ICD) that is released by SPPL2a is insufficiently resolved. In contrast to SPPL2a, SPPL2b was reported to reside primarily at the plasma membrane 5, 6. Although the substrate spectra of SPPL2a and SPPL2b were found to overlap significantly in cell-based systems, SPPL2b was described to be dispensable for CD74 proteolysis under endogenous conditions, indicating that both proteases exhibit distinct functions in vivo 5. Based on the described immunological phenotype of Sppl2a−/− mice 12, pharmacological inhibition of SPPL2a activity has been proposed as a strategy for depleting and/or inhibiting B cells and dendritic cells in a therapeutic context. Depletion of B cells has been shown to be beneficial in a variety of autoimmune disorders including systemic lupus erythematosus and is currently achieved with antibody-based therapeutics 16. Thus, a small-molecule-based approach utilizing SPPL2a inhibitors would exhibit significant advantages. As expected from the catalytic center homology, several γ-secretase inhibitors, like inhibitor X, also target SPP/SPPL proteases. To date, (Z-LL)2-ketone represents the only SPP/SPPL inhibitor that spares γ-secretase activity but acts unselectively on several members of the SPP/SPPL protease family 4. Because in vitro reconstitution of recombinantly expressed SPP/SPPL proteases as it is routinely employed for rhomboid I-CLIPs 17 has not been achieved yet, functional analysis of SPPL2 proteases relies on cell-based systems followed by western blot determination of uncleaved or cleaved forms of substrate proteins. Thus, a sensitive, quantitative and potentially high-throughput-compatible assay for monitoring activity of SPPL2a and other SPPL2 proteases is currently not available. This would, however, represent a prerequisite not only to further analyze the function and regulation of this protease but also to search for novel inhibitors and to optimize the available compounds. For analysis of γ-secretase, cell-based reporter assays have been successfully employed, especially for monitoring cleavage of the amyloid precursor protein (APP) 18-21 and Notch1 19, 22. In several of these studies, a reporter composed of the yeast DNA-binding protein Gal4 and the transcriptional activator VP16 was fused to the cytosolic C-terminus of the γ-secretase substrates 18-20. Upon proteolytic release of the ICD, the Gal4/VP16 reporter induces transcription of a luciferase via a Gal4-dependent promotor from a co-transfected plasmid. Luciferase activity is determined luminometrically and normalized via activity of a co-expressed reporter enzyme. Here, we report on a novel cell-based reporter assay system for determining the activity of SPPL2 proteases, which is also applicable to other I-CLIPs. We show that conventional Gal4/VP16-based reporter systems interfere with the trafficking of type II-oriented SPPL substrate proteins preventing their use for this class of proteases. Therefore, the novel assay utilizes β-galactosidase (βGal) enzyme fragment complementation (βGEFC) in order to detect nuclear translocation of the ICDs released from the substrate proteins by the intramembrane cleavage. Using this system, we assessed the ability of the ICDs from individual SPPL2a and SPPL2b substrates for nuclear localization. With respect to the putative therapeutic impact of the SPPL2a substrate CD74, we specifically validated our readout system for this substrate. Furthermore, we provide first insights into transcriptional regulation exerted by the CD74 ICD. Cell-based assays utilizing the Gal4/VP16 reporter system have been successfully applied to determine the activity of several I-CLIPs. We therefore tried to expand this system to SPPL2 substrates. For this purpose, murine TNF, a substrate of SPPL2a, was tagged at its cytoplasmic N-terminus with the Gal4/VP16 reporter and then transiently expressed in HeLa cells to verify the subcellular localization of this construct by indirect immunofluorescence (Figure 1A). In comparison to a protein carrying an hemagglutinin (HA) epitope tag in the same position, the Gal4/VP16-TNF fusion protein demonstrated a significantly altered subcellular distribution and was predominantly trapped in perinuclear structures not colocalizing with any organelles as demonstrated here for the Golgi marker protein GM130 (Figure 1A). Presumably, these structures represent aggregates of misfolded proteins. In addition to this, western blot analysis revealed rather low expression of the Gal4/VP16-tagged TNF in comparison to the HA-tagged protein (Figure 1B) and only very minor amounts of the TNF NTF, which is generated by ectodomain shedding and represents the actual SPPL2a substrate, were detected. Furthermore, no decrease in the TNF NTF levels was observed after co-expression of active SPPL2a. This was in sharp contrast to the HA-tagged TNF NTF, which was clearly processed by SPPL2a (Figure 1B) that could be blocked by application of the SPP/SPPL inhibitor X. Figure 1Open in figure viewerPowerPoint Gal4/VP16 assays are not applicable to SPPL2a/b substrates. A) The 25-kDa Gal4/VP16 tag interferes with trafficking of murine TNF. For reporter assays based on Gal4/VP16-driven expression of a firefly luciferase a TNF construct was generated that harbored the Gal4/VP16 reporter at its N-terminus. HeLa cells seeded on coverslips were transiently transfected with HA-TNF-V5 or Gal4/VP16-TNF-V5, and the subcellular distribution of the TNF fusion proteins was analyzed by indirect immunofluorescence using an antibody against an N-terminal epitope of TNF. In parallel, endogenous GM130 was visualized. B) The Gal4/VP16-TNF fusion protein is not processed by SPPL2a. HeLa cells were transfected with either of the two TNF constructs alone or in combination with SPPL2a or the catalytically inactive SPPL2a D416A mutant (D/A). In order to inhibit endogenous SPPL2a, cells were treated for 6 h with the SPPL-inhibitor inhibitor X (1 µm) as indicated. Total cell lysates were analyzed by western blotting with an antibody against an N-terminal epitope detecting the TNF full-length protein as well as different NTFs derived from the HA- and Gal4/VP16-tagged TNF. C) Transcriptional activation induced by the Gal4/VP16-TNF-V5 construct does not depend on SPPL protease activity. HeLa cells were transiently transfected with Gal4/VP16-TNF-V5, a response plasmid with a firefly luciferase under the control of a Gal4-dependent promotor (pL8G5-Luc) and a constitutively expressed Renilla luciferase (pRL-TK). Cells were treated for 6 h with 1 µm inhibitor X and processed for determination of firefly and Renilla luciferase activities. The normalized ratio of firefly/renilla activities is shown. Bars indicate mean values of n = 4 experiments ± SEM. One-way ANOVA with Tukey post hoc testing was performed. ***p 0.001; ns, not significant. D) An N-terminal GFP tag disturbs subcellular trafficking of CD74. HeLa cells were transfected using CD74-HA or CD74 tagged with the 24-kDa GFP tag at its N-terminus (GFP-CD74) and subsequently analyzed for the distribution of the expressed proteins by indirect immunofluorescence using the In-1 antibody directed against the N-terminus of CD74. Co-staining for endogenous PDI was performed. E) NTFs derived from GFP-tagged CD74 are not processed by SPPL2a. Processing of CD74-HA or GFP-CD74 was compared upon co-expression of both constructs with active or inactive (D/A) SPPL2a in transiently transfected HeLa cells. In (B) and (E) β-actin served as loading control. Nevertheless, we tried to employ the Gal4/VP16-tagged TNF reporter construct for a luciferase-based SPPL activity assay and transfected this construct and a corresponding reporter plasmid into HeLa cells (Figure 1C). Despite the disturbed trafficking and poor expression of the fusion protein, transfection of this construct was associated with a significant increase in the firefly/Renilla luciferase ratio when compared with vector-transfected control cells. However, this ratio could not be modulated by application of the SPP/SPPL inhibitor X. This strongly indicates that the luciferase expression induced by the Gal4/VP16 TNF reporter construct was not related to SPPL-mediated intramembrane proteolysis, but rather to unrelated processes like the ERAD pathway. Because TNF could be expressed with small N-terminal tags like the HA epitope but not with the large Gal4/VP16 tag, which comprise 9 and ∼230 amino acids, respectively, we wondered whether the size of a cytosolically fused reporter might also critically influence expression and subcellular sorting of other SPPL2a/b substrates. For this purpose, we compared processing and trafficking of CD74 devoid of any N-terminal tag (CD74-HA) with a GFP-CD74 fusion protein (Figure 1D) because GFP is of a similar size as the Gal4/VP16 reporter. A major function of CD74 is to mediate trafficking of MHC II complexes to antigen-processing endosomal compartments, where CD74 is subjected to proteolytic degradation starting from its C-terminal luminal domain 15. As expected, CD74 was localized primarily in vesicular compartments after heterologous expression in HeLa cells. However, under the same conditions, GFP-CD74 was predominantly retained in the ER as revealed by colocalization with protein disulfide-isomerase (PDI) (Figure 1D). Correspondingly, SPPL2a-mediated processing of the NTFs that were produced from the GFP-tagged variant was not observed. In contrast, upon expression of CD74-HA (Figure 1E) the full-length protein as well as different C-terminally truncated degradation intermediates were detected under steady-state conditions. As we showed previously 5, 12, co-expression of SPPL2a significantly reduced the CD74 NTF (Figure 1E). Thus, both tested type II-oriented SPPL2a/b substrates TNF and CD74 do not tolerate larger tags at their short cytosolic termini in order to allow trafficking to the plasma membrane or the endocytic pathway, which is a prerequisite for their proteolytic processing. The βGal α-tag is compatible with correct sorting and processing of SPPL2a/b substrates Because these findings precluded the use of Gal4/VP16-based reporter assays for the analysis of SPPL2a/b-dependent proteolysis we considered alternative approaches. Therefore, we sought for a reporter system that would require only the addition of a short tag to the N-terminus of the substrate protein, which was tolerated by CD74 and TNF (Figure 1) and also by other substrates like TMEM106B 11 and ITM2B 10. This essential selection criterion was fulfilled by βGal enzyme fragment complementation (βGEFC) 23, 24. We therefore devised the system illustrated in Figure 2A in order to measure the proteolytic activity of SPPL2 proteases based on determining the nuclear translocation of the released substrate ICDs. To achieve this, we aimed to express the large βGal ω-fragment, which is on its own catalytically inactive, in the nucleus of a reporter cell line. The complementing βGal α-peptide, which has a limited size of only ≈50 aa and is required to restore enzymatic activity, would be fused to the N-terminus of CD74 or other SPPL2a/b substrate proteins. Then, βGal activity reconstitution could occur only after proteolytic liberation and nuclear translocation of the ICD. Figure 2Open in figure viewerPowerPoint The βGal α-tag for βGEFC does not interfere with trafficking and processing of SPPL2a/b substrates. A) Design of the βGEFC assay for quantitative detection of the CD74 ICD after its liberation by intramembrane proteolysis. In endosomes, proteolytic processing of CD74 generates an NTF that is subjected to SPPL2a intramembrane cleavage. In the assay, a CD74 fusion protein with the N-terminally appended βGal α-peptide is expressed in a reporter cell line that stably expresses the catalytically inactive βGal ω-fragment localized to the nucleus. After proteolytic release, the CD74 ICD is released to the cytosol and able to traffic to the nucleus, thereby restoring βGal activity that serves as sensitive and quantitative readout. B) To test application of the small α-tag to the SPPL2a substrate CD74, two constructs were designed that differed in the presence or absence of an extrinsically added NLS between the βGal α-peptide and the CD74 N-terminus. C) Endosomal/lysosomal targeting of CD74 was not impaired by the N-terminally appended α-peptide. The subcellular localization of α-CD74 and α-NLS-CD74 in comparison to CD74 devoid of any N-terminal tag was analyzed by immunofluorescence using the In-1 antibody in transiently transfected HeLa cells. Colocalization with the lysosomal/late endosomal protein LAMP-2 was determined with the 2D5 antibody. Scale bars = 10 µm. D) Processing of the different α-tagged CD74 reporter constructs by co-expressed active, but not inactive (D/A) SPPL2a was evaluated by western blot analysis of transiently transfected HEK293 cells stably expressing the βGal ω-fragment in the nucleus (ωnuc cells) in comparison to CD74 without any N-terminal tag. E) Proteolytic cleavage of the α-tagged CD74 as expressed from the assay constructs by endogenous protease levels was validated by application of SPP/SPPL inhibitors to ωnuc cells transiently transfected with the indicated CD74 constructs. Cells were treated with inhibitor X (InX, 1 µm) or (Z-LL)2-ketone (ZLL, 20 µm) for 6 h prior to cell harvest. To test if the βGal α-tag would be compatible with trafficking and proteolysis of CD74, we created an N-terminally α-tagged CD74 construct (α-CD74) (Figure 2B). For the case that the CD74 ICD generated by SPPL2a would not intrinsically localize to the nucleus, we generated a second construct harboring an SV40 NLS sequence (α-NLS-CD74) between the α-tag and the CD74 N-terminus (Figure 2B). We analyzed trafficking and proteolytic processing of these CD74 fusion proteins in comparison to wild-type CD74 by indirect immunofluorescence (Figure 2C) and western blotting (Figure 2D,E), respectively. Neither targeting to endosomal/lysosomal compartments nor processing by SPPL2a was altered by the described tags. The latter applied to co-expressed (Figure 2D) as well as endogenous SPPL proteases (Figure 2E) as revealed by the NTF stabilization induced by the SPP/SPPL inhibitors (Z-LL)2-ketone and inhibitor X. These findings strongly support the applicability of βGEFC to monitor intramembrane cleavage of CD74. On the basis of a previous report 25, we generated a reporter construct (ωnuc) of the βGal ω-fragment fused C-terminally to three SV40 nuclear localization sequences (NLSs), a nuclear retention sequence from SC35 (NRS) and an additional NLS from c-Myc (Figure 3A). Using this construct, we produced a stable HEK293T ωnuc reporter cell line that showed homogenous nuclear expression of the ω-fragment (Figure 3A). As a positive control for the devised assay, we employed a GFP carrying the βGal α-tag in addition to three NLS at its N-terminus (α-eGFPnuc) in order to facilitate constitutive transport of the expressed protein to the nucleus, which was confirmed by fluorescence microscopy (Figure 3B). In contrast, transmembrane protein 192 (TMEM192) is a multispanning integral protein of the lysosomal membrane 26, 27 for which proteolytic cleavage has not been observed, thus precluding that the N-terminally fused α-tag reaches the ω-reporter in the nucleus (Figure 3C). The α-eGFPnuc and α-TMEM192 control constructs were transiently expressed in the ωnuc-reporter cell line and βGal activity was determined using a luminescent substrate (Figure 3D). In order to account for differences in transfection efficiency, the βGal activity was normalized to the activity of a co-expressed firefly luciferase. While significant βGal activity was detected in cells expressing the nuclear-localized α-eGFPnuc, almost no βGal reconstitution was observed following expression of α-TMEM192. This confirms suitability of the described system to determine nuclear translocation of α-tagged proteins. Figure 3Open in figure viewerPowerPoint Establishment of a βGEFC assay to measure nuclear translocation. A) Design of a galactosidase-based reporter construct. The ωnuc reporter construct is based on E. coli βGal with deletion of the α-peptide (amino acids 12–42) resulting in a catalytically inactive ω-fragment. Nuclear localization of the construct was enabled by addition of three SV40 NLSs, a NRS from SC35 protein and a c-Myc-derived NLS. Nuclear transport of the reporter protein in a subcloned stable cell line was validated by indirect immunofluorescence analysis using an anti-βGal antibody. B) As a positive control with constitutive nuclear localization, an N-terminally α-tagged eGFP construct was designed by fusing the βGal α-tag and three NLS to the 5′-end of the eGFP-coding sequence. Fluorescence microscopy showed distinct α-eGFPnuc expression in the nucleus in transiently transfected HeLa cells. C) As a negative control with a definite extranuclear localization, human TMEM192, an integral membrane protein of lysosomes with four transmembrane (TM) segments, was tagged N-terminally with the βGal α-peptide and in addition with an HA epitope to enable detection. Presence of the fusion protein in lysosomes was validated in transiently transfected HeLa cells by indirect immunofluorescence by staining with the HA 3F10 antibody for TMEM192 and the 2D5 antibody for the lysosomal marker protein LAMP-2. Scale bars = 10 µm (A–C). D) Functionality of the assay was validated by expression of the positive and negative control constructs (B and C) in the ωnuc reporter cell line characterized in (A) and quantification of βGal activity using a luminescent assay. The determined βGal signal was divided by the value obtained for activity of the firefly luciferase (Luc) co-transfected with the assay constructs. The βGal/Luc ratios were normalized to the value from the α-eGFPnuc-positive control. Bars indicate mean values of n = 3 experiments ± SEM. One-way ANOVA with Tukey post hoc testing was performed. Significance versus vector-transfected cells is depicted. ***p 0.001; ns, not significant. To verify the applicability of the developed βGEFC-based assay system we focused on CD74 as model substrate. Therefore, we first transfected the α-CD74 and α-NLS-CD74 constructs described in Figure 2 into the ωnuc reporter cell line and analyzed the cells for βGal complementation. Upon application of these constructs to our βGal assay system, we observed a normalized βGal activity that was significantly higher than that of the negative control α-TMEM192 (Figure 4A). This detected βGal activity indicated a translocation of the CD74 ICD to the nucleus following its processing. This effect was further enhanced by an additional NLS sequence. Nevertheless, these data strongly indicate the capability of the α-tagged CD74 ICD to translocate to the nucleus upon its generation by SPPL2a. We also confirmed dose dependency of the βGal activity resulting from transfection of α-CD74 (Figure 4B). Figure 4Open in figure viewerPowerPoint Nuclear translocation of the CD74 ICD can be detected by βGEFC and is dependent on SPPL2a activity. A) The α-CD74 ICD is capable of entering the nucleus as shown by βGEFC assay in the ωnuc reporter cell line. βGal activity was determined in ωnuc cells following transient transfection of α-CD74 and α-NLS-CD74 as well as the α-eGFPnuc-positive and the α-TMEM192-negative controls. To account for differences in transfection activity, βGal activity was normalized using firefly luciferase (Luc). The calculated βGal/Luc ratios were further normalized to that of α-eGFPnuc-positive control. Bars indicate mean values of n = 3 experiments ± SEM. Significance of the CD74-expressing versus the α-TMEM192-transfected cells is depicted. B) Dependency of the determined βGal/firefly Luciferase ratio on the amount of expressed CD74 was verified by transient transfection of different amounts of plasmids coding for α-tagged CD74. Mean values were calculated from n = 3 experiments and error bars indicate SEM. C) Co-transfection of SPPL2a with α-CD74 increases the CD74-associated βGal/Luc activity ratio. Upon concurrent inhibition of SPPL2a activity by application of inhibitor X (InX, 1 µm, 6 h) this effect was abolished. Bars indicate mean values of n = 3 experiments ± SEM. D) Inhibition of endogenous SPPL2a by InX (1 µm) or (Z-LL)2-ketone (ZLL, 20 µm) for 6 h reduced the CD74-associated βGal/Luc signal. A similar reduction of βGal complementation in ωnuc cells transiently transfected with α-CD74 was observed upon interference with the activity of proteases involved in processing of the CD74 luminal domain. This was performed by application of bafilomycin A1 (300 nm), E-64 (10 µm) or leupeptin (2.5 µm). Bars indicate mean values of n = 3–7 experiments ± SEM. Significance of values from inhibitor-treated versus solvent-treated cells is depicted. E) Knockdown of SPPL2a in ωnuc reporter cells reduced the normalized βGal activity following expression of α-CD74. Cells were transfected twice with non-targeting control siRNA or siRNA directed against SPPL2a prior to transfection with the α-CD74 substrate construct. Reduction of cellular SPPL2a levels and an accumulation of the CD74 NTF were confirmed by western blotting. Bars indicate mean values of n = 3 experiments ± SEM. For statistical analysis, one-way ANOVA with Tukey post hoc testing (A, C and D) or an unpaired, two-tailed Student\'s t-test (E) was performed. ***p 0.001; **p 0.01; *p 0.05. In the following, we focused on the α-tagged CD74 wild-type construct (α-CD74) to validate that the βGal reconstitution following expression of this construct was indeed dependent on intramembrane proteolysis by SPPL2a that precedes nuclear translocation of the CD74 N-terminus. Therefore, we co-expressed α-CD74 with SPPL2a and observed that presence of the protease significantly augmented βGal activity (Figure 4C), indicating increased levels of CD74 ICD in the nucleus in the presence of SPPL2a. This effect was reversed when activity of SPPL2a was blocked by concomitant inhibitor X application, an inhibitor of SPP/SPPL proteases and γ-secretase. These findings are in agreement with observations from western blot analysis (Figure 2D), where co-expression of SPPL2a enhanced turnover of CD74 NTF. We also investigated processing of α-tagged CD74 by endogenous proteases in this system using inhibitor treatment (Figure 4D). The SPPL2a inhibitors (Z-LL)2-ketone and inhibitor X significantly reduced α-CD74-associated βGal reconstitution. Processing of the CD74 luminal domain represents a prerequisite of CD74 intramembrane cleavage by SPPL2a 12 because only the NTF and not the full-length protein is a substrate of the protease. We evaluated the effect of preventing lysosomal/endosomal acidification by bafilomycin A1 and of endosomal protease inhibition by E64 or leupeptin on CD74 ICD generation in the βGal system. Both approaches are known to inhibit CD74 degradation 28, 29. Here, all three compounds reduced the detected normalized βGal activity in α-CD74-expressing cells versus solvent-treated controls, indicating that upon interference with CD74 ectodomain proteolysis less CD74 ICD is produced. This agrees well with the observation that co-expressed SPPL2a does not modulate the full-length form of CD74, but only a truncated NTF (Figure 2D) as shown by western blotting. Because (Z-LL)2-ketone and inhibitor X are active against several members of the SPP/SPPL protease family, we sought to confirm the role of endogenous SPPL2a for CD74 ICD release in our βGal reporter cell line by knockdown of SPPL2a. In cells treated with SPPL2a versus non-targeting control siRNA (Figure 4E), normalized βGal reporter activity following α-CD74 transfection was reduced to a similar degree as by treatment with (Z-LL)2-ketone or inhibitor X (Figure 4D). This result validates the described reporter assay as a suitable readout for SPPL2a activity. We wanted to elucidate the pathways involved in turnover of this proteolytic fragment following its release by SPPL2a. Therefore, we treated α-CD74-transfected ωnuc cells with the proteasome inhibitors epoxomicin and bortezomib or a mixture of amastatin, bestatin and AAF-CMK, all inhibitors of cytosolic peptidases. Incubation with the proteasome inhibitors led to a significant increase in normalized βGal activity, whereas treatment with the peptidase inhibitor did not show any significant effect under these conditions (Figure 5A). We scrutinized if the observed effect upon proteosomal inhibition indeed reflected stabilization of the CD74 ICD generated by SPPL2a-mediated intramembrane cleavage. Therefore, we treated the cells with inhibitor X for 2 h before and also during epoxomicin application. The interference with CD74 processing resulted in a significant, however not complete, rescue of the epoxomicin-dependent increase in βGal reconstitution (Figure 5B). Furthermore, we excluded by western blot analysis that significant stabilization of the βGal ωnuc reporter had been induced by the applied proteasomal inhibitors (Figure 5C). Though this does not rule out additional stabilizing effects of the proteasome inhibitor on CD74 fragments generated by putative SPPL2a-independent mechanisms such as ER-associated degradation (ERAD) due to overexpression, these results nonetheless indicate a major role of the proteasome in the degradation of the CD74 ICD. Figure 5Open in figure viewerPowerPoint The CD74 ICD is degraded by the proteasome and activates expression of SFRP2. A) Incubation of α-CD74-transfected ωnuc cells with the proteasome inhibitors epoxomicin (1 µm) and bortezomib (5 µg/mL) clearly increased βGal complementation, while application of a cytosolic peptidase inhibitor mix (Ama/Best/AAF) of amastatin (1 µm), bestatin (20 µm) and AAF-CMK (10 µm) had only minor effects. Bars indicate mean values of n = 3 experiments ± SEM. B) The effect of epoxomicin shown in (A) could be partially blocked by inhibition of SPPL2a by preincubation with inhibitor X (1 µm), respectively. Bars indicate mean values of n = 4 experiments ± SEM. One-way ANOVA with Tukey post hoc testing was performed. C) Application of the proteasomal inhibitors epoxomicin and bortezomib only has minimal effects on the βGal ω-fragment. ωnuc cells were treated for 6 h with epoxomicin or bortezomib and subsequently analyzed for the stabilizing effect of these compounds on the reporter fragment by western blotting. β-Actin served as loading control for normalization of densitometric quantification. D) Design of an inducible system for controlled overexpression of the CD74 ICD (amino acids 1–35). The ICD of CD74 was attached to a membrane-anchored GFP by a TEV cleavage site or an uncleavable stretch of seven alanines (A7). Co-expression of a TEV protease via an IRES element only liberated the CD74 ICD in the presence of the ENLYFQG cleavage site. The membrane-tethered cleavable (CD74-ICD-TEV-GFPmem) and non-cleavable (CD74-ICD-A7-GFPmem) CD74-ICD-GFP fusion proteins were expressed from the depicted constructs that were used for the generation of stable FlipIn™T-Rex™293 cell lines. E) Characterization of the cell lines stably expressing CD74 ICD-TEV-GFPmem or CD74-ICD-A7-GFPmem upon induction with doxycyclin. FlipIn™T-Rex™293 cell lines were generated based on the constructs shown in (D) and treated with 10 µg/mL doxycyclin for 24 h or left untreated. Expression of the fusion proteins was detected by western blotting using anti-CD74 or anti-GFP antibodies. The TEV protease was visualized with anti-Myc. F) SFRP2 is transcriptionally upregulated after overexpression and liberation of the CD74 ICD. The cell lines characterized in (E) were treated for 24 h with 10 µg/mL doxycyclin in order to induce expression of the CD74-ICD-GFP fusion proteins. SFRP2 expression was determined by qRT-PCR in both cell lines and normalized for expression of tubulin. Normalized SFRP2 transcript levels were significantly increased in the cell line expressing the cleavable CD74 ICD fusion protein (TEV) versus the control cell line (A7) where the ICD is not liberated. G) Sfrp2 is downregulated in SPPL2a- and CD74-deficient BMDCs compared to wild-type cells. For statistical analysis, one-way ANOVA with Tukey post hoc testing was performed. Significances in (A) are depicted for inhibitor-treated versus solvent-treated cells. Also in (B) significance of values from inhibitor-treated versus solvent-treated cells was depicted. Furthermore, significance of additional inhibitor X treatment in epoxomicin-treated cells was evaluated. In (G) significances of SPPL2a- or CD74-deficient versus wild-type BMDCs are indicated; lines mark significances between single conditions. ***p 0.001; **p 0.01; *p 0.05; ns, not significant. After having documented the capability of the α-CD74 ICD to translocate to the nucleus, we sought to investigate any putative impact of the CD74 ICD on transcriptional regulation. Because direct overexpression of the CD74 ICD has not been successful in our hands, presumably owing to its small size, we devised the system depicted in Figure 5D for controlled overexpression of the CD74 ICD mimicking its proteolytic release from a transmembrane precursor. In the generated constructs (Figure 5D), the 35 N-terminal residues of murine CD74 (ICD) were fused to eGFP followed by the transmembrane segment of human syntaxin 7 anchoring the fusion protein to the membrane. Between the CD74 ICD and the eGFP, a linker was inserted containing either a cleavage site (ENLYFQG) for the NIa protease from the tobacco etch virus (TEV protease) or a stretch of alanines (±TEV site). In both constructs, TEV protease is co-expressed from the same plasmid employing an internal ribosome entry site (IRES) mediating release of the CD74 ICD specifically from the fusion protein with the TEV cleavage site, whereas the CD74 ICD remains membrane bound in the control. Because the TEV protease is highly specific, heterologous expression of this protease is well tolerated in different experimental systems, including mammalian cells 30. We generated stable HEK293 cells inducibly expressing both constructs and validated their functionality regarding membrane localization of the CD74 ICD-GFP fusion proteins (not shown) and the proteolytic processing upon presence of the TEV cleavage site (Figure 5E). Interestingly, the released CD74 ICD was not directly observed (not shown) despite its confirmed production, indicating that intrinsic properties of this fragment impede its western blot-based detection. We determined and compared genome-wide transcriptome profiles of both cell lines. In order to detect changes in gene expression patterns induced by the proteolytically released CD74 ICD, we calculated fold changes between the cell line expressing the cleavable CD74 ICD fusion protein versus the control cell line where the ICD is not liberated. Transcripts that appeared differentially expressed more than twofold between the two cell lines are listed in Table S1, Supporting Information. Among these were 10 protein-coding transcripts, from which 4 appeared to be upregulated and 6 downregulated. We did not further analyze OR514A, TASR30 and KRTAP9-1 owing to their very specific roles as olfactory and taste receptors or in hair follicles, respectively. In contrast, the suggested differential expression of TULP3, ZNF763, SFRP2, DHRSX, LUM, TPTE and CNTN1 was re-evaluated by quantitative real time-polymerase chain reaction (qRT-PCR) (Figure S1). In this analysis, ZNF763 mRNA was not detected and no difference was found for TPTE, TULP3 and DHRSX. However, CD74 ICD-associated upregulation of SFRP2 (Figure 5F) as well the reduction of CNTN1 and LUM expression was confirmed. We asked if this effect of the overexpressed CD74 ICD could also be relevant under endogenous conditions. Sfrp2 mRNA has been detected in bone marrow-derived dendritic cells (BMDC) by transcriptomic profiling 31. Because dendritic cells abundantly express endogenous CD74, we analyzed consequences of a reduced ICD generation in BMDCs. Therefore, we compared Sfrp2 transcript levels in Sppl2a−/− and Cd74−/− BMDCs, in which generation of the CD74 ICD cannot occur owing to the absence of either the protease or the substrate, to BMDCs isolated from wild-type mice (Figure 5G). Secreted frizzled-related protein 2 (SFRP2) expression was significantly downregulated in both the SPPL2a- and the CD74-deficient cells. This agrees well to the described increase of SFRP2 expression upon CD74 ICD overexpression. In conclusion, these findings strongly indicate that the CD74 ICD can have a regulatory impact on gene expression also under endogenous conditions, advocating further investigations especially regarding the underlying mechanism. Following the corroboration of the βGal assay system to monitor intramembrane cleavage of CD74 proteolysis, we sought to evaluate its applicability to other substrates of SPPL proteases. Therefore, we created N-terminally α-tagged expression constructs of TNF and ITM2B (Figure 6A) with or without an inserted additional NLS sequence. We analyzed the subcellular distribution of these α-tagged proteins in comparison to the wild-type proteins with an N-terminal HA tag (Figure 6B). In analogy to CD74, we did not detect any impact of the α-tag on intracellular trafficking of TNF and ITM2B. Both constructs were able to reach the cell surface. Expression of the generated TNF and ITM2B constructs (α-TNF and α-ITM2B) resulted in normalized βGal activities that were significantly higher than those provoked by the negative control α-TMEM192 (Figure 6C). Similar to CD74, βGal reconstitution was enhanced by an additional NLS in the TNF and ITM2B reporter constructs (α-NLS-TNF and α-NLS-ITM2B). Figure 6Open in figure viewerPowerPoint SPPL2-dependent nuclear translocation of the TNF and ITM2B ICDs. A) To test further described SPPL2 substrates for nuclear translocation of their proteolytically liberated ICDs, assay constructs similar to those for CD74 were designed for TNF and ITM2B. In the ITM2B expression constructs also an HA tag was added to allow antibody detection. B) The α-tag does not interfere with sorting of ITM2B and TNF. Comparison of HA-tagged ITM2B and HA-TNF-V5 with the corresponding α-tagged fusion proteins by indirect immunofluorescence revealed no major differences regarding the subcellular distributions of the proteins in HeLa cells. Expression of ITM2B was determined using the HA 3F10 antibody, while TNF localization was detected with a rabbit polyclonal against the N-terminus of the protein. Scale bars = 10 µm. C) βGEFC assays indicate nuclear translocation of TNF and ITM2B ICDs. Normalized βGal activity was determined in ωnuc reporter cells transiently expressing the indicated constructs. Bars indicate mean values of n = 3 experiments ± SEM. D) Co-expression of active SPPL2a and SPPL2b enhanced the βGEFC assay signal in α-TNF-expressing ωnuc reporter cells. This effect of co-expressed SPPL2a/b was blocked by concurrent administration of inhibitor X (InX, 1 µm) for 6 h. Bars indicate mean values of n = 4–5 experiments ± SEM. E) The assay signal induced by α-ITM2B could also be augmented by overexpression of SPPL2b and SPPL2a. Experiments were carried out in analogy to α-CD74 and α-TNF before. Bars indicate mean values of n = 3 experiments ± SEM. One-way ANOVA with Tukey post hoc testing was performed. In (A) significances versus the α-TMEM192-transfected cells were depicted. In (D) and (E) significances for the effect of the SPPL2a/b overexpression alone or in combination with InX application are shown. ***p 0.001; *p 0.05; ns, not significant. We aimed to confirm that the α-TNF- and α-ITM2B-induced assay signals were dependent on intramembrane cleavage by SPPL2a and/or SPPL2b. Therefore, we co-expressed α-TNF and α-ITM2B with both proteases (Figure 6D,E) and observed enhanced α-TNF- and α-ITM2B-associated βGal reconstitution that could be partially blocked by concomitant application of inhibitor X. In conclusion, these results show that the described assay can also be employed to determine the turnover of other substrates than CD74 by SPPL proteases. Furthermore, our findings demonstrate that in this system SPPL2a/b-generated TNF and ITM2B ICDs are able to translocate to the nucleus. While this has been suggested before for the TNF ICD 7, 32, a capability of the ITM2B ICD to access the nucleus has not been reported previously according to our knowledge. Finally, we wanted to assess if the described βGal reporter assay system can be adapted to substrate proteins of different topology and/or of other than SPPL intramembrane proteases. To investigate this, we chose the type 1 transmembrane protein Notch as one of the best studied substrates of intramembrane proteolysis 3. Ligand binding induces proteolytic shedding of the Notch ectodomain that is followed by γ-secretase processing of the remaining C-terminal stub enabling well-documented nuclear translocation of the ICD 3. We used a Notch expression construct devoid of the ectodomain (NotchΔE) (Figure 7A) 33. Therefore, upon expression in HeLa cells this truncated Notch protein was constitutively processed by γ-secretase and the released C-terminal ICD could be detected in the nucleus (Figure 7B). Upon inhibition of γ-secretase by N-[N-(3,5-Difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester (DAPT), generation of the ICD was prevented so that the NotchΔE protein remained membrane-associated (Figure 7B). With respect to applicability in our reporter assay, the employed NotchΔE construct carried the βGal α-tag at its cytosolic C-terminus. In the ωnuc reporter cell line, expression of the NotchΔE-α construct restored βGal activity to about 70% of the level observed with the positive control α-eGFPnuc (Figure 7C). Upon concomitant γ-secretase inhibition by DAPT, significantly less βGal activity was measured. This effect was not observed with the SPP/SPPL inhibitor (Z-LL)2-ketone. Thus, the established βGal reporter assay can be employed to monitor γ-secretase-mediated cleavage of Notch as well as the nuclear translocation of the Notch ICD. This finding substantiates applicability of the system to monitor proteolysis of different substrate topologies as well as by other classes of intramembrane proteases. Figure 7Open in figure viewerPowerPoint The devised βGEFC assay is also applicable to other I-CLIPs-like γ-secretase. A) A Notch1 reporter construct for the βGal assay was designed based on the Notch1ΔE construct, which lacks the extracellular portion of the Notch1 receptor. Therefore, the resulting protein is constitutively processed by γ-secretase to generate the Notch1 ICD (NICD) for which nuclear translocation is well documented. The employed construct exhibited the βGal α-peptide and an additional HA tag at its C-terminus. B) γ-Secretase inhibition prevented generation and nuclear translocation of the NICD and stabilized the uncleaved NotchΔE-α protein at the plasma membrane. HeLa cells transiently expressing NotchΔE-α were treated with 1 µm DAPT for 6 h and analyzed by indirect immunofluorescence using the HA 3F10 antibody. C) Application of the γ-secretase inhibitor DAPT, but not the SPP/SPPL inhibitor (Z-LL)2-ketone (ZLL), reduced the normalized βGal activity determined in the ωnuc reporter cell line following transient transfection of NotchΔE-α. Post transfection, cells were incubated overnight and subsequently treated with DAPT (1 µm), ZLL (20 µm) or DMSO as vehicle control for 6 h. Bars indicate mean values of n = 3 independent experiments ± SEM. Scale bars = 10 µm. One-way ANOVA with Tukey post hoc testing was performed. Significance of values from NotchΔE-α-expressing versus TMEM192-transfected cells and of the inhibitor treatments was depicted. ***p 0.001; ns, not significant. In this study, we sought to create a cell-based reporter assay to monitor the activity of the intramembrane proteases SPPL2a and SPPL2b in order to provide an alternative to cumbersome western blot procedure when studying these enzymes. To achieve this, we failed to adapt the Gal4/VP16 reporter system that has been successfully applied to monitor cleavage by γ-secretase 18-22 or the ER protease SPP 34, 35. Apparently, in these studies, the cleavage of type 1 transmembrane proteins or ER proteins in type 2 orientation was analyzed. Here, we showed that N-terminal fusion of the Gal4/VP16 reporter, which comprises about 230 aa, or a protein of a similar size (GFP) was not compatible with regular trafficking and proteolysis of the SPPL2a/b substrates TNF and CD74. These proteins exhibit a type II transmembrane topology with a short cytoplasmic N-terminal domain (∼30 residues) and need to travel via the secretory pathway to the plasma membrane and/or endosomal compartments. In the case of CD74, essential sorting motifs in this N-terminal part of the protein have been identified 36. The Gal4/VP16-TNF construct was expressed at very low levels when compared with the wild-type construct, suggesting that in this case even folding and/or membrane insertion of the fusion protein in the ER were disturbed. Because the other currently known SPPL2a/b substrates ITM2B 10, Fas ligand 9, transferrin receptor 37 and TMEM106 11 exhibit a similar topological layout to the ones examined, it can be concluded that the Gal4/VP16 reporter system is not suitable to monitor intramembrane cleavage by SPPL2a/b. We demonstrate here that a split reporter system based on βGal (βGEFC), which only requires the addition of a short tag to the N-terminus of the substrate protein, can be exploited to create a cell-based reporter assay for SPPL2a/b-mediated intramembrane proteolysis. This system allowed quantitative detection of the ICDs following their release from the membrane-bound substrate proteins. We showed that modulation of protease activity in a positive or negative way by overexpression, knockdown or pharmacological inhibition, respectively, was reflected in a corresponding modulation of the assay signal, thus validating this as a quantitative readout of SPPL2a/b-mediated proteolysis. We especially sought to provide an assay for the intramembrane cleavage of CD74 by SPPL2a based on the documented in vivo relevance of this process and its potential as a therapeutic target 12-14. However, we also confirmed applicability of the βGEFC-based assay system for other SPPL2a/b substrates like TNF and ITM2B and demonstrated that cleavage of the type 1 transmembrane protein Notch1 by γ-secretase could be monitored by this means. Thus, the βGEFC reporter assay is not restricted to a certain substrate topology and may be further extended to other classes of intramembrane proteases. Protein fragment complementation describes the reconstitution of a protein\'s function or enzymatic activity from two separately synthetized fragments when these are brought into proximity 38. A variety of different split-reporter systems have been designed and used in multiple applications, especially for the analysis of protein–protein interactions 38. Among these, a quite unique property of split βGal is the pronounced size asymmetry of the two fragments comprising ∼50 (α) and ∼950 (ω) amino acids 23 that are capable of reconstituting βGal activity, making it especially suited for the purpose of this study. We found that the small βGal α-peptide in contrast to other reporters did not interfere with trafficking and proteolysis of SPPL2a/b substrates. βGal-based readout systems have been successfully employed in high-throughput screening applications 39. Similar to other split-protein reporters, βGEFC has been exploited to monitor protein–protein interactions 40. However, owing to the high-affinity complementation of the two βGal fragments already their presence in the same subcellular compartment can be sufficient to promote assembly of a functional enzyme. This fact has been utilized for designing cell-based assays to detect nuclear translocation of target proteins 24, 25, 41, especially in order to provide a readout for the activation of certain signaling pathways like Wnt 41 and glucocorticoid receptor 25 signaling. In this study, we have extended this application to monitor proteolytic processes at biological membranes by the proteases SPPL2a/b. Adaption of the described system to design assays for other intramembrane proteases is possible and may provide essential advantages over previously employed reporters especially for substrate proteins with short cytoplasmic domains. We demonstrated that the βGEFC assay system is suitable to monitor SPPL2a-mediated cleavage of CD74 because interference with this process led to a significant reduction of the assay signal. However, pharmacological inhibition of SPPL2a activity only blocked about 30% of the signal. Similar effects were observed upon knockdown of the protease, which led to a significant, though incomplete depletion of the SPPL2a protein. Therefore, we hypothesize that also in the inhibitor-treated cells residual SPPL2a activity accounts for the remaining assay signal. This would indicate that either the affinity of the used compounds or their cell permeability or a combination of both is limiting in order to achieve a full inhibition. However, we can also not exclude that part of the CD74-associated assay signal reflects SPPL2a-independent degradative routes that liberate CD74 proteolytic fragments into the cytosol. Furthermore, also minor amounts of the βGal ω-fragment in the cytosol, which we failed to detect with immunofluorescence, could in principle account for an SPPL2a-independent background signal. However, this should have also affected signals for the negative control α-TMEM192, which we used as baseline. Therefore, we consider this as unlikely. When studying intramembrane proteolysis, the fate and function of the intracellular cleavage products is an immanent question related to the technical difficulties to detect these often small, low-abundant protein fragments biochemically. The latter also applies to the ICD released from CD74. We have been unable to detect this released CD74 ICD by biochemical means directly in total cell lysates what is most likely due to its small size of 30–40 amino acids and/or its rapid turnover. Our assay data (Figure 5) point to a degradation of the CD74 ICD by the proteasome. This pathway might be influenced by the putative D box motif (RXXL) in the cytoplasmic region of the CD74 protein 42. However, also concurrent inhibition of the proteasome (bortezomib and epoxomicin) did not enhance biochemical detectability of this cleavage product by western blotting in our hands. This may indicate that primarily its unfavorable properties impair its recovery with this technique, which was not resolved despite significant optimization efforts. In this context, the described βGEFC assay system represents a valuable approach to examine the cell biological fate of the CD74 ICD. Beyond providing a quantitative readout for protease activity it also allows to explore the capability of the respective cleavage fragments for entering the nucleus. Nuclear translocation of CD74 fragments has been suggested before 43; however, the proteolytic machinery for the intramembrane cleavage of CD74 was unknown until recently 12-14. In our βGEFC system, we could demonstrate that the CD74 ICD released by the intramembrane protease SPPL2a has the capability to migrate to the nucleus, which could indicate a physiological function of this process similar to other I-CLIP substrates 44. In this regard, we demonstrated that the CD74 ICD may play a role in influencing transcriptional regulation as exemplified by the identification of SFRP2 as a potential target gene. In different cellular systems, modulation of CD74 ICD generation and release was reflected in corresponding changes of SFRP2 transcript levels. SFRP2 is a secretory protein that has been found to be a negative regulator of Wnt signaling 45, 46. Furthermore, among other functions SFRP2 has been suggested to act as a tumor suppressor because it was shown to induce apoptosis and promote tumorigenesis 46, 47. The physiological relevance of a regulatory circuit between CD74 and SFRP2 in dendritic cells as we showed here needs further insights. Interestingly, SFRP2 was reported to be secreted in pathophysiologically relevant amounts by multiple myeloma, a neoplasm of plasma cells 48. This process was demonstrated to contribute to the bone loss induced by these tumors via suppressing bone formation. At the same time, high expression levels of CD74 are usually observed in multiple myeloma 49. It may be interesting to test if the CD74 ICD contributes to the documented upregulation of SFRP2 in these neoplastic plasma cells. The precise molecular mechanism how the CD74 ICD enhances SFRP2 expression is currently unclear. It has been reported previously that the released CD74 ICD has an impact on NFκB signaling 42. Therefore, we analyzed our microarray dataset for any CD74 ICD-associated expression modulation of established NFκB target genes or components of this pathway (Table S2). However, we could not find any changes arguing for a differential activation of the NFκB pathway in those cells expressing the cleavable versus the non-cleavable CD74 ICD fusion protein. This was also supported by biochemical analysis of IκBα levels (not shown). Therefore, we did not obtain any evidence that in this cell system the effect on SFRP2 involves components of the NFκB pathway. We think that the overexpression system we have utilized to mimic proteolytic liberation of the CD74 ICD in combination with transcriptomic profiling represents a well-controlled, unbiased approach that will also be useful to investigate the effects of other ICDs released by I-CLIPs. In this context, we consider the current dataset merely as a proof-of-principle experiment that provides first indications for a regulatory role of the CD74 ICD. We think that our findings strongly advocate to adapt the system to (murine) immune cells that also endogenously express CD74, which would further enhance the validity of the approach. Using the βGEFC assay, we could also confirm nuclear translocation of the TNF ICD in line with previous reports 7, 32. This ICD was reported to enhance expression of the proinflammatory IL12 after its liberation by SPPL2a/b-mediated proteolysis 7. In contrast to TNF, the molecular function of the ITM2B protein is much less understood to date 50. This protein is genetically linked to the familial British and Danish forms of dementia and has been reported to influence processing of the APP 51. However, little is known about the physiological relevance of the ITM2B intramembrane cleavage by SPPL2a/b following release of the ectodomain by ADAM10 10. Here, we reveal for the first time that the ITM2B ICD can travel to the nucleus following its proteolytic liberation. This could suggest a specific function of the ITM2B ICD in signal transduction and transcriptional regulation that will need to be further investigated. Another interesting question is whether the relative small ICD of this protein as well as that of TNF and CD74 is trafficked to the nucleus by specific transport mechanisms. Because molecules with a size 40 kDa are able to enter this compartment by passive diffusion 52, 53, this most likely also applies to the small substrate ICDs. In general, our findings on the capability of SPPL2a/b substrates to localize to the nucleus are in agreement with findings for many of the numerous substrates of the related γ-secretase complex 44. For those substrates of γ-secretase or other intramembrane proteases where biochemical detection of the intracellular cleavage products is inherently difficult the described βGEFC reporter system will provide a powerful tool to clarify the cellular fates of these fragments. In addition to this, the assay described above will provide a versatile technique not only for elaborating intramembrane cleavage of regulated intramembrane proteolysis (RIP) substrates but also for the preceding and kinetically linked shedding process as we could show for the luminal processing of CD74. Therefore, we are confident that the devised assay system provides a new valuable tool for the functional analysis of SPPL2a/b and also other intramembrane proteases as well as for the search of novel inhibitors of these enzymes. The SPP/SPPL-inhibitors (Z-LL)2-ketone (Peptanova) and inhibitor X (EMD Millipore) were used at final concentrations of 20 and 1 µm, respectively. For selective inhibition of γ-secretase, DAPT (Sigma) was applied at 1 µm final concentration. Proteasome inhibition was performed with bortezomib (Selleck chemicals) and epoxomicin (Peptanova) at concentrations of 5 µg/mL and 1 µm, respectively. Amastatin (1 µm, Peptanova), bestatin (20 µm, Sigma) and AAF-CMK (10 µm, Peptanova) were used to target cytosolic peptidases. To interfere with processing of the CD74 luminal domain, leupeptin (2.5 µm, Roth), E64 (10 µm, Enzo) and bafilomycin A1 (300 nm, Sigma) were utilized as inhibitors of lysosomal proteases and lysosomal acidification, respectively. Except for AAF-CMK, leupeptin and E64 which were dissolved in H2O, stock solutions from all other compounds were prepared in dimethyl sulfoxide (DMSO) (Roth) or DMSO/H2O mixtures. Doxycyclin (used at 10 µg/mL) was purchased from Sigma. Murine TNF was tagged with an N-terminal HA and a C-terminal V5 tag and inserted into the pcDNA3.1 hygro (+) vector backbone (Invitrogen). An N-terminal HA tag was added to the open reading frame of murine ITM2B via PCR using extended primers and the fusion construct was cloned into pcDNA3.1 hygro (+) via its XbaI and BamHI sites. For generation of the Gal4/VP16-mTNF fusion construct the Gal4/VP16 reporter coding was amplified from pMst-GV 18 and fused to mTNF via fusion PCR. The pL8G5-Luc (firefly luciferase) vector, the response plasmid for the Gal4/VP16 reporter, as well as the pMst-GV plasmid were a kind gift of Thomas Südhof and Jacqueline Burré. The pRL-TK vector coding for a Renilla luciferase used as transfection control for Gal4/VP16 assays was from Promega. To furthermore test the applicability of large tags to the N-termini of SPPL2a/b substrate molecules, the p31 isoform of murine CD74 12 was cloned into pEGFP-C1 (Clontech) to fuse its coding sequence with one of GFP. The nuclear βGal reporter construct (ωnuc) was cloned by deletion of the coding sequence for amino acids 12–42 of Escherichia coli βGal. To achieve stable nuclear localization of the βGal reporter, three SV40 NLSs (DPKKKRKV) 54, the SC35 nuclear retention signal (NRS, PPPVSKRESKSRSRSKSPPKSPEEEGAVSSDLKVRKAA, see 55) and an additional NLS derived from the transcription factor c-Myc (DLKVRKAA) 56 were fused by two consecutive PCRs to the 3′ end of the reporter construct according to 25. The final PCR product was inserted via appended BamHI and XbaI sites into pcDNA3.1 hygro (+) that was also used for all expression constructs described below. The resulting construct was termed ωnuc. For each SPPL substrate analyzed, expression constructs with the complementing βGal α-peptide fused to the N-terminus (α-tag) were generated with or without an additional SV40 NLS in order to facilitate nuclear translocation of the released ICD also in the absence of a functional intrinsic nuclear import motif. In a first step, either the α-tag alone (MSSNSLAVVLQRRDWENPGVTQLNRLAHPPFASWRNSEEARTDRPSQQLRSLNGE) or in combination with an additional SV40 NLS was appended to the 5′ end of the coding sequence of the p31 isoform of murine CD74 12 by PCR with a BamHI site in between. The whole fragment was then integrated via introduced NheI and XbaI sites into pcDNA3.1 hygro (+). For the generation of murine TNF and ITM2B assay constructs, the CD74 coding sequence was excised from the α-tagged CD74 constructs by restriction digest with BamHI and XbaI and the respective open reading frames were inserted in frame via corresponding 5′ and 3′-terminal sites added by PCR. A C-terminally α-tagged expression construct of the established γ-secretase substrate Notch1 was generated based on the murine NotchΔE-GFP construct described in 33. This was devoid of the Notch1 extracellular domain allowing constitutive cleavage by γ-secretase independent of ligand binding and ectodomain shedding. To enable detection of this construct, an HA tag was inserted between the NotchΔE coding sequence and the βGal α-tag. As a positive control for nuclear localization, we fused the βGal α-tag and three SV40 NLS to the open reading frame of GFP (from pEGFP-N1, Clontech) resulting in a constitutive nuclear transport of GFP (α-eGFPnuc). As an extranuclear negative control, we created an N-terminally α-tagged construct of the human lysosomal integral transmembrane protein 192 (α-TMEM192) that we have characterized previously 26, 27. Plasmids coding for murine SPPL2a/b and the inactive SPPL2a-D416A mutant have been described before 5, 6, 12. An expression construct of firefly luciferase in a pcDNA3 vector 57 was obtained from Addgene (plasmid no. 18964). To generate a system that allowed for expression of the CD74 ICD (amino acids 1–35) a construct was generated that harbored this protein fragment attached to a GFP and a human syntaxin 7 membrane anchor (amino acids 229–261). Either a TEV protease cleavage site (ENLYFQG) or a stretch of seven alanines (A7) were inserted between the CD74 ICD and the membrane-anchored GFP to allow or inhibit cleavage by a myc-tagged TEV protease co-expressed from the same plasmid via an IRES sequence. While the CD74 ICD coding sequence was amplified from plasmids described above, the nucleotide sequences corresponding to GFP were amplified from pEGFP-N1 (Clontech). The coding sequence of the human syntaxin7 membrane anchor as well as the TEV protease cleavage site or the A7 stretch were attached by consecutive PCRs using extended primers. IRES sequence and the TEV protease open reading frame were amplified from pIRES2-EGFP (Clontech) or pRK1043 (Addgene 8835), respectively. Finally, the sequences were joined by overlap-extension PCRs and cloned into pcDNA5.1/FRT/TO (Life Technologies) via its BamHI and XhoI sites. The constructs were termed CD74 ICD-TEV-GFPmem or CD74-ICD-A7-GFPmem, respectively. HeLa (DSMZ) and HEK293T cells were maintained in DMEM (Sigma) supplemented with 10% FBS (Biochrom) as well as 100 U/mL penicillin (Sigma) and 100 µg/mL streptomycin (Sigma). Cells were cultivated at 37°C in a humidified 5% CO2/95% air atmosphere. All transfections of plasmid DNA were performed with Turbofect (Thermo Scientific) according to the manufacturer\'s instructions. A reporter cell line stably expressing ωnuc was generated by transfection of HEK293T cells with the corresponding plasmid and subsequent selection of positive clones with 250 µg/mL hygromycin (Invivogen). After subcloning, individual clones were evaluated regarding homogeneity of expression and nuclear localization of the reporter construct as determined by indirect immunofluorescence as well as discrimination between positive and negative control under assay conditions. The best-performing cell line according to these criteria was chosen for further experiments and termed ωnuc. To generate a cell line that would allow for the inducible expression of the CD74 ICD, the FlipIn™T-Rex™293 cell line (Life Technologies) was transfected with either the CD74 ICD-TEV-GFPmem or CD74-ICD-A7-GFPmem construct and the pOG44 vector (Life Technologies). Subsequently, cells were selected with 100 µg/mL hygromycin and 10 µg/mL blasticidin (Invivogen). Because the FlipIn™ system facilitates site-directed integration of the transfected constructs and transgene expression in the selected cells was homogenous as determined by indirect immunofluorescence, cells were maintained as a batch without subcloning. In the following, these CD74 ICD-TEV-GFPmem or CD74 ICD-A7-GFPmem cell lines were used for microarray and qRT-PCR analysis. For knockdown of SPPL2A in the HEK293T reporter cell line, a pool of four gene-specific siRNAs (SMARTpool: GUUGUUGCCUGGAGACGUA, GGAGUGGACUAGUUGAAUU, CCUCAUGCCUGUUUCAAUA and GGUAACAGCUAUCAGAUGA) as well as of non-targeting siRNAs (pool #1) were obtained from Thermo Scientific and transfected using INTERFERin (Polyplus transfection) following the supplier\'s recommendations. A final concentration of 20 nm siRNA in the culture medium was used for efficient knockdown of SPPL2A. siRNA transfection was directly performed at the time of cell seeding and repeated after 48-h incubation. HeLa cells (5 × 104) were seeded in 24-well plates and transfected with pRL-TK, pL8G5-Luc and empty vector or Gal4/VP16-TNF as indicated. On the next day, the cells were treated with 1 µm inhibitor X or the corresponding amount of DMSO for 6 h and lysed by applying passive lysis buffer (Promega). Afterwards, 20 μL of the uncleared lysate was analyzed for luciferase activity using the Dual-Luciferase® Reporter Assay System Kit (Promega) according to the manufacturer\'s instructions. Luminescent signals were detected using a GloMax™ 96-well plate luminometer (Promega) applying a two-injections protocol. For evaluation of the experiments, the determined firefly luciferase activity was divided by the corresponding Renilla activity for each sample. This firefly/Renilla luciferase signal ratio was then normalized to that from the substrate-expressing, solvent-treated sample. For βGal enzyme fragment complementation assays, 104 cells of the ωnuc reporter cell line were seeded in 100 μL medium without hygromycin in 96-well plates (Greiner Bio-One) suitable for luminescence assays. On the following day, cells were transfected with α-tagged substrate or control constructs as well as firefly luciferase to allow normalization of transfection efficiency. After 1 day of culture post transfection, cells were either used directly for luminescent determination of βGal and luciferase activities or following a 6-h treatment with inhibitors as indicated. Activities of the reporter enzymes were measured employing the Dual-Light® Luciferase & β-Galactosidase Reporter Gene Assay System (Life technologies) and a GloMax™ 96-well plate luminometer E6521 (Promega). Measurements were performed at least in technical duplicates based on two independent transfections. For evaluation, βGal/firefly luciferase (βGal/Luc) activity ratios were calculated for each individual sample. These ratios were normalized as indicated either to positive control samples (α-eGFPnuc) that were part of each assay or, when protease activity was modulated, to solvent-treated or mock-transfected samples. In both cases, the individual sample ratios were divided by the mean value of the respective control samples that were used as basis for the normalization. Figures show mean values and standard error of the mean (SEM) of at least three independent experiments each carried out with at least two technical duplicates per condition. Protein was extracted from harvested cells as described previously 26. Protein concentration of total lysates was determined with a bicinchoninic acid protein assay (Thermo Scientific). Following electrophoretic separation by SDS–PAGE with tris-glycine 58 or tris-tricine 59 buffer systems as applicable, semi-dry transfer of protein to nitrocellulose membranes was performed as reported in 26. For immunodetection of murine CD74, the established monoclonal antibody In-1 directed against the CD74 ICD was employed (BD Biosciences). To detect cleavage fragments of TNF, a rabbit polyclonal antibody was generated against a synthetic peptide corresponding to amino acids 4–22 of murine TNF (ESMIRDVELAEEALPQKMG). A rabbit monoclonal anti-GFP antibody was obtained from Cell Signaling. The monoclonal 3F10 antibody recognizing the HA epitope and the monoclonal 9B11 antibody against the Myc epitope were purchased from Roche and Cell Signaling, respectively. The generation of antibodies against murine and human SPPL2a has been reported before 6, 60. Expression of the βGal ω-fragment was detected using an anti-βGal antibody (Promega). To confirm equal protein loading, a rabbit polyclonal antibody against β-actin (Sigma) was used. Horseradish peroxidase-conjugated secondary antibodies for chemiluminescent detection were from Dianova. For indirect immunofluorescence analysis, 5 × 104 HeLa cells were seeded on coverslips in 12-well plates the day before transfection. One day post transfection cells were fixed with 4% (wt/v) paraformaldehyde and permeabilized with 0.2% (wt/v) saponine, both in PBS, according to the protocol described earlier 26. Primary antibodies used included anti-LAMP2 (2D5) 61, anti-GM130 (BD Biosciences) and anti-PDI (Abcam) as well as the already mentioned anti-CD74, anti-βGal, anti-TNF Nterm and anti-HA. For visualization, Alexa-488- and -594-coupled secondary antibodies were utilized (MobiTec). After mounting in Mowiol supplemented with DABCO [1,4-diazobicyclo-(2.2.2) octane] and DAPI (4-,6-diamidino-2-phenylindole) photographs from optical sections were acquired with a FV1000 confocal laser scanning microscope (Olympus) equipped with an U Plan S Apo 64x oil immersion objective and Olympus Fluoview Software (3.0a). Adobe Photoshop was applied for further processing. To evaluate potential transcriptional effects due to liberation of the overexpressed CD74 ICD, an exemplary microarray experiment was carried out using the CD74 ICD-TEV-GFPmem and CD74-ICD-A7-GFPmem cell lines described above as well as a corresponding stably empty vector-transfected cell line. In order to induce expression of the CD74 ICD fusion proteins, cells were treated for 24 h with 10 µg/mL doxycyclin (Sigma). Subsequently, RNA was isolated using the Nucleospin RNA® Kit (Macherey Nagel). Genome-wide transcriptome profiles were generated employing Human Gene 2.0 ST Arrays (Affymetrix) according to the manufacturer\'s guidelines and further processed as previously described 62, while transcripts with a fold change 2.0 or −2.0 were considered as differentially expressed. SPPL2a- and CD74-deficient mice were described before 12, 63. For differentiation of BMDCs according to 64, the bone marrow of three animals per genotype was prepared by flushing tibia and femur bone marrow using 10 mL BMDC medium consisting of RPMI medium (GIBCO) supplied with 10% FBS, 100 U/mL penicillin, 100 µg/mL streptomycin and 50 µm β-mercaptoethanol (GIBCO). A total of 5 × 106 bone marrow cells were seeded in 10-cm dishes in 10 mL of RPMI medium supplied with 20 ng/mL granulocyte–macrophage colony-stimulating factor (GM-CSF) (Immunotools). Fresh medium (10 mL) supplemented with the same amount of GM-CSF was added after 3 days. At day 6, 10 mL of medium were replaced by fresh BMDC medium containing 10 ng/mL GM-CSF. At day 7, 1.5 × 106 cells were seeded and harvested for RNA extraction at the following day. RNA was extracted using the Nucleospin RNA® Kit according to the manufacturer\'s recommendations. Afterward, 1 µg RNA was transcribed into cDNA using the RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific) in a 20 μL volume using random hexamer oligonucleotides. Finally, 0.5 μL cDNA per reaction was analyzed via qRT-PCR using the Universal Probe Library System (Roche) and a Lightcycler 480II (Roche). The following primers were used to determine expression of SFRP2 and α-tubulin for normalization: hSFRP2-fw, GCTAGCAGCGACCACCTC; hSFRP2-rv, TTTTTGCAGGCTTCACATACC; hTuba1c-fw, CCCCTTCAAGTTCTAGTCATGC; hTuba1c-rv, GCATTGCCAATCTGGACAC; mSFRP2-fw, AAACCCTTTGTAAAAATGACTTCG; mSFRP2-rv, CAGCTTGTAAATGGTCTTGCTC; mTuba1a-fw, CTGGAACCCACGGTCATC; mTuba1a-rv, GTGGCCACGAGCATAGTTATT. Expression of murine and human SFRP2 was normalized to mTuba1a or hTuba1c, respectively. Data shown represent means ± SEM. For statistical analyses, unpaired two-tailed t-test or one-way ANOVA followed by Tukey post hoc testing were used as indicated. Significant levels of p 0.05 (*), p 0.01 (**) and p 0.001 (***) were applied. We thank Dr William Kaelin, Harvard Medical School, for providing a firefly luciferase expression construct via Addgene, Dr David Waugh, National Cancer Institute, Rockville, for a plasmid coding for TEV protease made accessible via Addgene and Dr Christoph Kaether, Leibniz Institute for Age Research, Jena, Germany, for the NotchΔE construct. Furthermore, we thank Jacqueline Burré and Thomas Südhof, Stanford School of Medicine, for plasmids for the Gal4/VP16 reporter assay. We are grateful for excellent technical assistance from Sebastian Held. This work was supported by the Deutsche Forschungsgemeinschaft as part of the SFB877 (project B7, to B. S.) and the Cluster of Excellence ‘Inflammation at Interfaces’ as well as grant SCHR 1284/1-1 (to B. S.). tra12287-sup-0003-SuppMatt.pdfPDF document, 931.5 KB Table S1: Full list of putative CD74 ICD target genes. Microarray analysis of the inducible TREX CD74 ICD-TEV-GFPmem and CD74-ICD-A7-GFPmem cell lines was performed. The relative abundance of transcripts was determined and the fold change (TEV/A7) was calculated. The table comprises all identified targets with a fold change 2.0 or −2.0. Protein-coding transcripts are marked in red. This table is associated with Figure 5. Table S2: NFκB target genes are not differentially regulated by the liberated CD74 ICD. Gene expression profiles of the inducible TREX CD74 ICD-TEV-GFPmem and CD74-ICD-A7-GFPmem T-Rex™ were obtained by microarray analysis. The expression of documented NFκB target genes (http://www.bu.edu/nf-kb/gene-resources/target-genes/) as well as genes associated with the NFKB-signaling pathway (KEGG hsa04064) was compared based on the calculated fold change (TEV/A7). No significant effect that would support a significant impact of the liberated CD74 ICD on the NFκB pathway in this cell system was observed. This table is associated with Figure 5. Figure S1: Validation of transcriptional targets identified in the microarray experiment. The inducible TREX CD74 ICD-TEV-GFPmem and CD74-ICD-A7-GFPmem cell lines were treated for 24 h with 10 µg/mL doxycyclin in order to induce expression of the CD74-ICD-GFP fusion proteins. Expression of TULP3, ZNF763, SFRP2, DHRSX, LUM, TPTE and CNTN1 was determined by qRT-PCR in both cell lines and normalized for expression of tubulin. The following primers in addition to those specified in the Materials and Methods section were used: hTULP3-fw, CCAGATAGTCCACAAAAATGACC; hTULP3-rv, GCCTGTACTGCACAAAGTGG; hZNF763-fw, GCTGTGCCCTTCTGTAGTCA; hZNF763-rv, CCACAGGGTCCTGAAACATC; hDHRSX-fw, CGGCTGGTTGCTTTTCAA; hDHRSX-rv, TCTCTTTCTCGTTGTATAGGTAATGG; hLUM-fw, GAAAGCAGTGTCAAGACAGTAAGG; hLUM-rv, GGCCACTGGTACCACCAA; hTPTE-fw, CCCCAGGCTCTTCTGACC; hTPTE-rv, AGAATCTGGTAGTGAATCCATAACTCT; hCNTN1-fw, GATGACTGGAAAGATGCAAAGA; hCNTN1-rv, CATGGGATTAAGTCCACTGCT. This is associated with Figure 5. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. 1Wolfe MS, Kopan R. Intramembrane proteolysis: theme and variations. Science 2004; 305: 1119– 1123. 2Urban S, Freeman M. Intramembrane proteolysis controls diverse signalling pathways throughout evolution. Curr Opin Genet Dev 2002; 12: 512– 518. 3De Strooper B, Annaert W, Cupers P, Saftig P, Craessaerts K, Mumm JS, Schroeter EH, Schrijvers V, Wolfe MS, Ray WJ, Goate A, Kopan R. A presenilin-1-dependent gamma-secretase-like protease mediates release of Notch intracellular domain. Nature 1999; 398: 518– 522. 4Voss M, Schröder B, Fluhrer R. Mechanism, specificity, and physiology of signal peptide peptidase (SPP) and SPP-like proteases. Biochim Biophys Acta 1828; 2013: 2828– 2839. 5Schneppenheim J, Hüttl S, Mentrup T, Lüllmann-Rauch R, Rothaug M, Engelke M, Dittmann K, Dressel R, Araki M, Araki K, Wienands J, Fluhrer R, Saftig P, Schröder B. The intramembrane proteases signal peptide peptidase-like 2a and 2b have distinct functions in vivo. Mol Cell Biol 2014; 34: 1398– 1411. 6Behnke J, Schneppenheim J, Koch-Nolte F, Haag F, Saftig P, Schröder B. Signal-peptide-peptidase-like 2a (SPPL2a) is targeted to lysosomes/late endosomes by a tyrosine motif in its C-terminal tail. FEBS Lett 2011; 585: 2951– 2957. Wiley Online Library 7Friedmann E, Hauben E, Maylandt K, Schleeger S, Vreugde S, Lichtenthaler SF, Kuhn PH, Stauffer D, Rovelli G, Martoglio B. SPPL2a and SPPL2b promote intramembrane proteolysis of TNFalpha in activated dendritic cells to trigger IL-12 production. Nat Cell Biol 2006; 8: 843– 848. 8Fluhrer R, Grammer G, Israel L, Condron MM, Haffner C, Friedmann E, Bohland C, Imhof A, Martoglio B, Teplow DB, Haass C. A gamma-secretase-like intramembrane cleavage of TNFalpha by the GxGD aspartyl protease SPPL2b. Nat Cell Biol 2006; 8: 894– 896. 9Kirkin V, Cahuzac N, Guardiola-Serrano F, Huault S, Luckerath K, Friedmann E, Novac N, Wels WS, Martoglio B, Hueber AO, Zornig M. The Fas ligand intracellular domain is released by ADAM10 and SPPL2a cleavage in T-cells. Cell Death Differ 2007; 14: 1678– 1687. 10Martin L, Fluhrer R, Reiss K, Kremmer E, Saftig P, Haass C. Regulated intramembrane proteolysis of Bri2 (Itm2b) by ADAM10 and SPPL2a/SPPL2b. J Biol Chem 2008; 283: 1644– 1652. 11Brady OA, Zhou X, Hu F. Regulated intramembrane proteolysis of the frontotemporal lobar degeneration risk factor, TMEM106B, by signal peptide peptidase-like 2a (SPPL2a). J Biol Chem 2014; 289: 19670– 19680. 12Schneppenheim J, Dressel R, Hüttl S, Lüllmann-Rauch R, Engelke M, Dittmann K, Wienands J, Eskelinen EL, Hermans-Borgmeyer I, Fluhrer R, Saftig P, Schröder B. The intramembrane protease SPPL2a promotes B cell development and controls endosomal traffic by cleavage of the invariant chain. J Exp Med 2013; 210: 41– 58. 13Beisner DR, Langerak P, Parker AE, Dahlberg C, Otero FJ, Sutton SE, Poirot L, Barnes W, Young MA, Niessen S, Wiltshire T, Bodendorf U, Martoglio B, Cravatt B, Cooke MP. The intramembrane protease Sppl2a is required for B cell and DC development and survival via cleavage of the invariant chain. J Exp Med 2013; 210: 23– 30. 14Bergmann H, Yabas M, Short A, Miosge L, Barthel N, Teh CE, Roots CM, Bull KR, Jeelall Y, Horikawa K, Whittle B, Balakishnan B, Sjollema G, Bertram EM, MacKay F, et al. B cell survival, surface BCR and BAFFR expression, CD74 metabolism, and CD8- dendritic cells require the intramembrane endopeptidase SPPL2A. J Exp Med 2013; 210: 31– 40. 15Neefjes J, Jongsma ML, Paul P, Bakke O. Towards a systems understanding of MHC class I and MHC class II antigen presentation. Nat Rev Immunol 2011; 11: 823– 836. 16Townsend MJ, Monroe JG, Chan AC. B-cell targeted therapies in human autoimmune diseases: an updated perspective. Immunol Rev 2010; 237: 264– 283. Wiley Online Library 17Dickey SW, Baker RP, Cho S, Urban S. Proteolysis inside the membrane is a rate-governed reaction not driven by substrate affinity. Cell 2013; 155: 1270– 1281. 18Biederer T, Cao X, Sudhof TC, Liu X. Regulation of APP-dependent transcription complexes by Mint/X11s: differential functions of Mint isoforms. J Neurosci 2002; 22: 7340– 7351. 19Karlstrom H, Bergman A, Lendahl U, Naslund J, Lundkvist J. A sensitive and quantitative assay for measuring cleavage of presenilin substrates. J Biol Chem 2002; 277: 6763– 6766. 20Bakshi P, Liao YF, Gao J, Ni J, Stein R, Yeh LA, Wolfe MS. A high-throughput screen to identify inhibitors of amyloid beta-protein precursor processing. J Biomol Screen 2005; 10: 1– 12. 21Florean C, Zampese E, Zanese M, Brunello L, Ichas F, De GF, Pizzo P. High content analysis of gamma-secretase activity reveals variable dominance of presenilin mutations linked to familial Alzheimer\'s disease. Biochim Biophys Acta 2008; 1783: 1551– 1560. 22Li J, Fici GJ, Mao CA, Myers RL, Shuang R, Donoho GP, Pauley AM, Himes CS, Qin W, Kola I, Merchant KM, Nye JS. Positive and negative regulation of the gamma-secretase activity by nicastrin in a murine model. J Biol Chem 2003; 278: 33445– 33449. 23Langley KE, Villarejo MR, Fowler AV, Zamenhof PJ, Zabin I. Molecular basis of beta-galactosidase alpha-complementation. Proc Natl Acad Sci USA 1975; 72: 1254– 1257. 24Wehrman TS, Casipit CL, Gewertz NM, Blau HM. Enzymatic detection of protein translocation. Nat Methods 2005; 2: 521– 527. 25Fung P, Peng K, Kobel P, Dotimas H, Kauffman L, Olson K, Eglen RM. A homogeneous cell-based assay to measure nuclear translocation using beta-galactosidase enzyme fragment complementation. Assay Drug Dev Technol 2006; 4: 263– 272. 26Schröder B, Wrocklage C, Hasilik A, Saftig P. Molecular characterisation of \'transmembrane protein 192\' (TMEM192), a novel protein of the lysosomal membrane. Biol Chem 2010; 391: 695– 704. 27Behnke J, Eskelinen EL, Saftig P, Schröder B. Two dileucine motifs mediate late endosomal/lysosomal targeting of transmembrane protein 192 (TMEM192) and a C-terminal cysteine residue is responsible for disulfide bond formation in TMEM192 homodimers. Biochem J 2011; 434: 219– 231. 28Landsverk OJ, Ottesen AH, Berg-Larsen A, Appel S, Bakke O. Differential regulation of MHC II and invariant chain expression during maturation of monocyte-derived dendritic cells. J Leukoc Biol 2012; 91: 729– 737. Wiley Online Library 29Landsverk OJ, Barois N, Gregers TF, Bakke O. Invariant chain increases the half-life of MHC II by delaying endosomal maturation. Immunol Cell Biol 2011; 89: 619– 629. Wiley Online Library 30Wehr MC, Laage R, Bolz U, Fischer TM, Grunewald S, Scheek S, Bach A, Nave KA, Rossner MJ. Monitoring regulated protein-protein interactions using split TEV. Nat Methods 2006; 3: 985– 993. 31Pletinckx K, Stijlemans B, Pavlovic V, Laube R, Brandl C, Kneitz S, Beschin A, De BP, Lutz MB. Similar inflammatory DC maturation signatures induced by TNF or Trypanosoma brucei antigens instruct default Th2-cell responses. Eur J Immunol 2011; 41: 3479– 3494. Wiley Online Library 32Domonkos A, Udvardy A, Laszlo L, Nagy T, Duda E. Receptor-like properties of the 26 kDa transmembrane form of TNF. Eur Cytokine Netw 2001; 12: 411– 419. 33Huenniger K, Kramer A, Soom M, Chang I, Kohler M, Depping R, Kehlenbach RH, Kaether C. Notch1 signaling is mediated by importins alpha 3, 4, and 7. Cell Mol Life Sci 2010; 67: 3187– 3196. 34Nyborg AC, Jansen K, Ladd TB, Fauq A, Golde TE. A signal peptide peptidase (SPP) reporter activity assay based on the cleavage of type II membrane protein substrates provides further evidence for an inverted orientation of the SPP active site relative to presenilin. J Biol Chem 2004; 279: 43148– 43156. 35Dev KK, Chatterjee S, Osinde M, Stauffer D, Morgan H, Kobialko M, Dengler U, Rueeger H, Martoglio B, Rovelli G. Signal peptide peptidase dependent cleavage of type II transmembrane substrates releases intracellular and extracellular signals. Eur J Pharmacol 2006; 540: 10– 17. 36Bonifacino JS, Traub LM. Signals for sorting of transmembrane proteins to endosomes and lysosomes. Annu Rev Biochem 2003; 72: 395– 447. 37Zahn C, Kaup M, Fluhrer R, Fuchs H. The transferrin receptor-1 membrane stub undergoes intramembrane proteolysis by signal peptide peptidase-like 2b. FEBS J 2013; 280: 1653– 1663. Wiley Online Library 38Michnick SW, Ear PH, Manderson EN, Remy I, Stefan E. Universal strategies in research and drug discovery based on protein-fragment complementation assays. Nat Rev Drug Discov 2007; 6: 569– 582. 39Olson KR, Eglen RM. Beta galactosidase complementation: a cell-based luminescent assay platform for drug discovery. Assay Drug Dev Technol 2007; 5: 137– 144. 40Rossi F, Charlton CA, Blau HM. Monitoring protein-protein interactions in intact eukaryotic cells by beta-galactosidase complementation. Proc Natl Acad Sci USA 1997; 94: 8405– 8410. 41Verkaar F, Blankesteijn WM, Smits JF, Zaman GJ. beta-Galactosidase enzyme fragment complementation for the measurement of Wnt/beta-catenin signaling. FASEB J 2010; 24: 1205– 1217. Wiley Online Library 42Matza D, Kerem A, Medvedovsky H, Lantner F, Shachar I. Invariant chain-induced B cell differentiation requires intramembrane proteolytic release of the cytosolic domain. Immunity 2002; 17: 549– 560. 43Becker-Herman S, Arie G, Medvedovsky H, Kerem A, Shachar I. CD74 is a member of the regulated intramembrane proteolysis-processed protein family. Mol Biol Cell 2005; 16: 5061– 5069. 44Haapasalo A, Kovacs DM. The many substrates of presenilin/gamma-secretase. J Alzheimers Dis 2011; 25: 3– 28. 45Mii Y, Taira M. Secreted Wnt \"inhibitors\" are not just inhibitors: regulation of extracellular Wnt by secreted Frizzled-related proteins. Dev Growth Differ 2011; 53: 911– 923. Wiley Online Library 46Esteve P, Bovolenta P. The advantages and disadvantages of sfrp1 and sfrp2 expression in pathological events. Tohoku J Exp Med 2010; 221: 11– 17. 47Lee JL, Lin CT, Chueh LL, Chang CJ. Autocrine/paracrine secreted Frizzled-related protein 2 induces cellular resistance to apoptosis: a possible mechanism of mammary tumorigenesis. J Biol Chem 2004; 279: 14602– 14609. 48Oshima T, Abe M, Asano J, Hara T, Kitazoe K, Sekimoto E, Tanaka Y, Shibata H, Hashimoto T, Ozaki S, Kido S, Inoue D, Matsumoto T. Myeloma cells suppress bone formation by secreting a soluble Wnt inhibitor, sFRP-2. Blood 2005; 106: 3160– 3165. 49Stein R, Mattes MJ, Cardillo TM, Hansen HJ, Chang CH, Burton J, Govindan S, Goldenberg DM. CD74: a new candidate target for the immunotherapy of B-cell neoplasms. Clin Cancer Res 2007; 13: 5556s– 5563s. 50Del CM, Teunissen CE. Role of BRI2 in dementia. J Alzheimers Dis 2014; 40: 481– 494. 51Matsuda S, Giliberto L, Matsuda Y, Davies P, McGowan E, Pickford F, Ghiso J, Frangione B, D\'Adamio L. The familial dementia BRI2 gene binds the Alzheimer gene amyloid-beta precursor protein and inhibits amyloid-beta production. J Biol Chem 2005; 280: 28912– 28916. 52McLane LM, Corbett AH. Nuclear localization signals and human disease. IUBMB Life 2009; 61: 697– 706. Wiley Online Library 53Paine PL, Moore LC, Horowitz SB. Nuclear envelope permeability. Nature 1975; 254: 109– 114. 54Kalderon D, Roberts BL, Richardson WD, Smith AE. A short amino acid sequence able to specify nuclear location. Cell 1984; 39: 499– 509. 55Cazalla D, Zhu J, Manche L, Huber E, Krainer AR, Caceres JF. Nuclear export and retention signals in the RS domain of SR proteins. Mol Cell Biol 2002; 22: 6871– 6882. 56Saphire AC, Bark SJ, Gerace L. All four homochiral enantiomers of a nuclear localization sequence derived from c-Myc serve as functional import signals. J Biol Chem 1998; 273: 29764– 29769. 57Safran M, Kim WY, O\'Connell F, Flippin L, Gunzler V, Horner JW, Depinho RA, Kaelin WG Jr. Mouse model for noninvasive imaging of HIF prolyl hydroxylase activity: assessment of an oral agent that stimulates erythropoietin production. Proc Natl Acad Sci USA 2006; 103: 105– 110. 58Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227: 680– 685. Wiley Online Library 60Voss M, Fukumori A, Kuhn PH, Kunzel U, Klier B, Grammer G, Haug-Kroper M, Kremmer E, Lichtenthaler SF, Steiner H, Schröder B, Haass C, Fluhrer R. Foamy virus envelope protein is a substrate for signal peptide peptidase-like 3 (SPPL3). J Biol Chem 2012; 287: 43401– 43409. 61Radons J, Faber V, Buhrmester H, Volker W, Horejsi V, Hasilik A. Stimulation of the biosynthesis of lactosamine repeats in glycoproteins in differentiating U937 cells and its suppression in the presence of NH4Cl. Eur J Cell Biol 1992; 57: 184– 192. 62Geismann C, Grohmann F, Sebens S, Wirths G, Dreher A, Hasler R, Rosenstiel P, Hauser C, Egberts JH, Trauzold A, Schneider G, Sipos B, Zeissig S, Schreiber S, Schafer H. Arlt A c-Rel is a critical mediator of NF-kappaB-dependent TRAIL resistance of pancreatic cancer cells. Cell Death Dis 2014; 5: e1455. 63Bikoff EK, Huang LY, Episkopou V, van Meerwijk J, Germain RN, Robertson EJ. Defective major histocompatibility complex class II assembly, transport, peptide acquisition, and CD4+ T cell selection in mice lacking invariant chain expression. J Exp Med 1993; 177: 1699– 1712. 64Lutz MB, Kukutsch N, Ogilvie AL, Rossner S, Koch F, Romani N, Schuler G. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods 1999; 223: 77– 92. The full text of this article hosted at iucr.org is unavailable due to technical difficulties. Please check your email for instructions on resetting your password. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. Can\'t sign in? Forgot your username? Enter your email address below and we will send you your username If the address matches an existing account you will receive an email with instructions to retrieve your username