Rafts Increases Shedding of CD30 Depletion of Cellular Cholesterol

Depletion of Cellular Cholesterol and Lipid Rafts Increases Shedding of CD30 Bastian von Tresckow, Karl-Josef Kallen, Elke Pogge von Strandmann, Peter Borchmann, Hans Lange, Andreas Engert This information is current as and Hinrich P. Hansen of October 1, 2021. J Immunol 2004; 172:4324-4331; ; doi: 10.4049/jimmunol.172.7.4324 http://www.jimmunol.org/content/172/7/4324 Downloaded from

References This article cites 46 articles, 16 of which you can access for free at: http://www.jimmunol.org/content/172/7/4324.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 1, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Depletion of Cellular Cholesterol and Lipid Rafts Increases Shedding of CD301

Bastian von Tresckow,* Karl-Josef Kallen,† Elke Pogge von Strandmann,* Peter Borchmann,* Hans Lange,† Andreas Engert,* and Hinrich P. Hansen2*

CD30, a lymphoid activation marker, is shed into the cell environment after endoproteolytic cleavage of its ectodomain. Soluble (s)CD30 is able to suppress the Th1-type immune response. Because high serum levels of sCD30 and cholesterol-lowering drugs seem to be beneﬁcial in some Th1-type autoimmune diseases, we focused on a link between CD30 shedding and the amount of cellular cholesterol. Cholesterol depletion of human Hodgkin lymphoma- and non-Hodgkin lymphoma-derived cell lines by methyl-␤-cyclodextrin led to a down-regulation of membrane-bound CD30 and increased release of sCD30. Additionally, the cholesterol-interfering drugs lovastatin, cholesterol oxidase, and ﬁlipin increased CD30 shedding. Both the down-regulation of mem-

brane-anchored CD30 and the release of sCD30 were dependent on metalloproteinases. Using specific inhibitors, we detected Downloaded from TNF-␣ converting enzyme (TACE) as the leading enzyme responsible for cholesterol-dependent CD30 shedding. A Triton X-100- based method for lipid raft isolation revealed that CD30 was partially present in lipid rafts, whereas TACE was localized in the nonraft fractions. Disintegration of lipid rafts by cholesterol depletion might therefore lead to dynamic interactions of CD30 with TACE, resulting in enhanced shedding of CD30. Our results suggest a possible role of cholesterol-dependent shedding of CD30 in the pathogenesis of immune diseases. The Journal of Immunology, 2004, 172: 4324–4331. http://www.jimmunol.org/ 120-kDa type I transmembrane glycoprotein, CD30 is matoid arthritis, systemic sclerosis, atopic dermatitis, Wegener’s normally found on a small subset of lymphocytes, espe- granulomatosis, Graves’ disease, and Hashimoto’s thyroiditis also cially Th2 cells (1). Due to its typical cysteine-rich do- show increased serum levels of sCD30 (7). A ϩ mains, CD30 was classified as a member of the TNFR superfamily There are inconsistent data on the contribution of CD30 cells (2). This family includes receptors like the TNFR1 and -2 and sCD30 in inflammatory processes. On the one hand, CD30 (CD120a/b), CD95/Fas, CD40, and CD27, collectively involved in expression and its release seems to correlate with the extent of a wide spectrum of cellular responses, such as induction of gene inflammation in diseases such as atopic dermatitis (8), systemic expression or proliferation but also initiation of apoptosis (3). lupus erythematosus (9), and Wegener’s granulomatosis (10). In Due to its overexpression on the malignant cells of Hodgkin’s contrast, there is evidence for a counterregulatory role of CD30ϩ by guest on October 1, 2021 disease (HD)3 and anaplastic large cell lymphoma (ALCL), CD30 T cells at the site of inflammation in the Th1-mediated rheumatoid is currently evaluated as a target for the immunotherapy of certain arthritis (11). Recently, it has been shown that this effect is also ϩ cancers (4). In this respect, it is a severe obstacle that the CD30 caused by sCD30, because a viral sCD30 homolog (vCD30) is able cells release a 90-kDa soluble form of CD30 (sCD30) (5), which to inhibit the Th1-, but not Th2-type inflammation in vivo (12). In serves as a serum competitor. The sCD30 concentrations in the line with this finding, elevated serum levels of sCD30 have been serum of patients suffering from HD and ALCL are ϳ20 and 1500 found in multiple sclerosis patients in remission (13) and corre- times higher, respectively, as compared with healthy donors (6). lated negatively with inflammatory markers in rheumatoid arthritis Increased serum levels were also detected in infectious and allergic (14). Because both diseases are regarded as Th1 mediated (15, 16), conditions such as EBV-induced mononucleosis, hepatitis virus B the elevated sCD30 levels might be beneficial in readjusting the infection, and immunodeficiency virus type 1 infection. Some au- Th1/Th2 balance. toimmune diseases, such as systemic lupus erythematosus, rheu- The mechanisms that regulate sCD30 generation are poorly un- derstood. Like many other substrates, the shedding of CD30 is catalyzed by TNF-␣ converting enzyme (TACE) (17). Recently, *Department of Internal Medicine I, University Hospital Cologne, Cologne, Ger- lipid rafts were found to be critical in the shedding regulation of many; and †Department of Biochemistry, University of Kiel, Kiel, Germany some membrane proteins (18Ð20). Rafts are microdomains of the Received for publication October 17, 2003. Accepted for publication January 16, 2004. membrane that are enriched in cholesterol and sphingolipids (21). The costs of publication of this article were defrayed in part by the payment of page They are supposed to form ordered domains that float in the liquid- charges. This article must therefore be hereby marked advertisement in accordance disordered matrix of the cellular membrane and play an important with 18 U.S.C. Section 1734 solely to indicate this fact. role in signal transduction, cell-cell interaction, and endocytosis of 1 This work was supported in part by Deutsche Forschungsgemeinschaft Grant certain proteins. Because cholesterol-depleting agents such as TFB30. methyl-␤-cyclodextrin (M␤CD) lead to a disintegration of lipid 2 Address correspondence and reprint requests to Dr. Hinrich P. Hansen, Depart- ment of Internal Medicine I, University Hospital Cologne, LFI, Ebene 4, Room rafts, these agents play a role in investigating the microdomains. 703, Joseph-Stelzmann-Strasse 9, 50924 Cologne, Germany. E-mail address: Besides other effects, the cholesterol-lowering drugs induce met- [email protected] alloproteinase-dependent shedding in some cases (18, 22). In this 3 Abbreviations used in this paper: HD, Hodgkin’s disease; ALCL, anaplastic large respect, the localization of Alzheimer’s amyloid precursor protein cell lymphoma; s, soluble; TACE, TNF-␣ converting enzyme; M␤CD, methyl-␤- cyclodextrin; APP, amyloid precursor protein; TIMP, tissue inhibitor of metallopro- (APP) inside or outside of rafts seems to be decisive in terms of teinases; DAG, diacylglycerol. whether amyloidogenic or nonamyloidogenic APP is generated

(18, 19). Moreover, the shedding regulation of the IL-6R, L-se- mg/ml in distilled water) and incubated for 20 min at 25¡C. The solution lectin, and the L1 adhesion molecule was shown to be connected was dialyzed overnight against sodium acetate buffer (1 mM; pH 4.4) at to cellular cholesterol and lipid rafts, but so far, the exact mech- 4¡C. One milliliter of Ab solution (2 mg/ml) was added, and the reaction was started after addition of 40 ␮l of sodium carbonate buffer (0.5 M; pH anisms remain unclear (20, 23, 24). 9.5). After2hofincubation at 25¡C, 0.1 ml of sodium borohydride solu- The aim of this study was to examine how the cellular choles- tion (4 mg/ml in distilled water) was added and incubated for another 2 h terol content and lipid rafts influence the shedding of CD30. Our at 4¡C. For stabilization, BSA (final concentration, 0.1 mg/ml) and a few results shed light on a novel mechanism of the generation of grains of thymol were added. Then, the Abs were stored at Ϫ80¡C. sCD30 and provide a possible link between cholesterol and CD30- Determination of sCD30 based immune regulation. Flexible 96-well microtiter plates (Maxisorb, polar surface; Nunc; Life Technologies, Karlsruhe, Germany) were coated with 50 ␮l of Ki-2 mAb Materials and Methods (50 ␮g/ml in sodium carbonate buffer (50 mM; pH 9.2)) by overnight Cell culture and reagents incubation at 4¡C. The plates were washed, blocked with PBS containing 5% milk powder for1hatroom temperature, and subsequently stored at The mAbs Ki-1, Ki-2, Ki-3, and Ki-4 (25) were used for the detection of Ϫ20¡C until use. After triplicate washing with PBS containing 0.1% CD30 in flow cytometry, ELISA, immunoprecipitation, and Western blot. Tween 20, serial dilutions of both the sCD30 standard (1320 U/ml) and the Sodium orthovanadate, Triton X-100, M␤CD, filipin, BSA, lovastatin, per- sCD30-containing samples were added and incubated for2hatroom tem- oxidase-labeled choleratoxin subunit B, HRP, FITC isomer I-celite, protein perature. Plates were washed three times, and 50 ␮l of peroxidase-coupled A-agarose beads, cholesterol, and cholesterol oxidase were purchased from Ki-3 mAb was added and incubated for1hatroom temperature. After Sigma-Aldrich (Taufkirchen, Germany). Complete protease inhibitor tab- washing three times with PBS containing 0.1% Tween 20 and one time in lets were purchased from Roche Diagnostics (Mannheim, Germany). The PBS, 50 ␮l of ABTS solution (Roche Diagnostics) was added. After1hof anti-TACE mAb MAB9302, the recombinant human TACE ectodomain incubation, the plates were evaluated at 405 nm in an ELISA reader. Sam- Downloaded from Western blotting standard, and tissue inhibitor of metalloproteinases ples were compared with the internal sCD30 standard. (TIMP)-1 and TIMP-3 were purchased from R&D Systems (Wiesbaden- Nordenstadt, Germany). The inhibitor BB-3644 was a kind gift from Brit- Immunoprecipitation of sCD30 ish Biotech Pharmaceuticals (Oxford, U.K.). ECL detection reagent was purchased from Amersham Pharmacia Biotech (Freiburg, Germany). TLC For immunoprecipitation of sCD30 from culture supernatants, protein A- plates were purchased from Merck (Darmstadt, Germany). The HD-de- agarose beads were coupled with Ki-1 mAb in 30 mM Tris-HCl (pH 8.5), rived cell line L428 was kindly donated by Dr. V. Diehl (University Hos- 120 mM NaCl, and 0.1% Triton X-100 for 120 min at 4¡C. After washing pital Cologne, Cologne, Germany). The large cellular anaplastic lymphoma (three times) in the buffer, beads were incubated for 120 min at 4¡C with http://www.jimmunol.org/ cell line Karpas 299 was kindly provided by Dr. A. Karpas (Department of the supernatants and washed again. The beads were centrifuged, and the Hematology, Cambridge University, Cambridge, U.K.). The cells were cul- supernatant was discarded. Beads were then heated with nonreducing SDS- tivated in RPMI 1640 supplemented with 10% FCS. PAGE loading buffer for 5 min at 95¡C and subsequently run on SDS- PAGE (4Ð15% acrylamide). Gels were blotted on nitrocellulose membrane Alterations of the cellular cholesterol content and stained with a mixture of peroxidase-labeled mAbs Ki-2 and Ki-4. Cells undergoing cholesterol depletion or enrichment were washed and Immunoreactive bands were detected by chemiluminescence with ECL de- suspended with HBSS containing 0.1% BSA and then treated with M␤CD tection reagent. for depletion or M␤CD-cholesterol complex for enrichment at 37¡C. The M␤CD-cholesterol complex was prepared as previously described (18). Isolation and analysis of lipid rafts ␤ Briefly, a solution of M CD (40 mg/ml) in HBSS was added to cholesterol Membrane rafts were prepared as previously described with minor modi- by guest on October 1, 2021 ␤ 7 (2.5 mg/ml M CD solution) in an Eppendorf tube and overlaid with N2 fications (27). Briefly, 3 ϫ 10 cells were washed in PBS and solubilized followed by overnight incubation at 30¡C under gentle shaking. The solu- in 0.5 ml of TNE buffer (25 mM Tris-HCl (pH 7.5), 150 mM NaCl, 5 mM ␮ Ϫ tion was sterile filtered (0.2 m) and stored at 20¡C. EDTA, complete protease inhibitor, and 1 mM sodium orthovanadate) con- Lipid analysis taining 1% Triton X-100 for 30 min on ice. Cells were thoroughly suspended with a yellow 200-␮l tip, and an equal volume of 80% sucrose (w/v Cellular lipid composition was determined using methods previously de- in TNE) was mixed with the lysate. It was overlaid with 6 ml of 35% scribed (26). Lipids were analyzed using charring densitometry. Briefly, sucrose (w/v in TNE) followed by 3.5 ml of 5% sucrose (w/v in TNE). The samples were dissolved in 20 ␮l of chloroform. Defined dilutions of cho- gradient was centrifuged with a SW41 rotor at 4¡C for 20 h at 40,000 rpm. lesterol were used as standards. The samples were applied to TLC plates Eleven gradient fractions (1 ml each) were harvested from the top. Frac- that were developed in benzene/diethylether/acetic acid (50:40:0.1, v/v/v). tions 1 and 2 were discarded, and 100 ␮l of fractions 3Ð11 were heated for The plates were dried for 10 min at 180¡C, allowed to cool, and then 5 min at 95¡C with 100 ␮l of double-concentrated nonreducing SDS- exposed to a solution of 10% copper sulfate in 8% aqueous phosphoric PAGE buffer (0.25 M Tris-HCl (pH 6.8), 40% glycerol, 0.02% bromphenol acid. The plates were subsequently rinsed, dried for 2 min at 110¡C, and blue, and 2% SDS). After protein separation by SDS-PAGE, gels were charred at 175¡C for 10 min until lipid bands became visible. After scan- blotted on nitrocellulose membranes and stained with HRP-labeled anti- ning the TLC plates, lipids were identified by using an internal standard TACE Ab (MAB9302) or a mixture of HRP-labeled anti-CD30 Abs (Ki-2 ϭ ϭ mAb and Ki-4 mAb). Immunoreactive bands were detected by chemilu- (cholesterol; Rf 0.36) (diacylglycerol (DAG); Rf 0.51). Lipids were quantified by measuring the ODs using the PCBAS program. Results are minescence with ECL detection reagent. Two microliters of each fraction either given as OD ratios of cholesterol/DAG, or given as micrograms per were dot blotted on a polyvinylidene difluoride membrane and stained with 106 cells after comparison with an internal cholesterol standard. peroxidase-coupled cholera toxin subunit B as a control of successful raft isolation. Determination of membrane-bound CD30 and CD70 by flow cytometry Results FITC labeling was performed as follows: Abs (1 mg/ml; 0.5 ml) in 0.1 M Cholesterol depletion with M␤CD induces shedding of CD30 in

NaHCO3/0.1 M NaCl (pH 9.2) were incubated with 1.5 mg of FITC/celite lymphoma cells for 30 min. The pellet-free supernatant was applied to a G-25 desalting To study the influence of the cellular cholesterol level on CD30 column to remove unconjugated material. Then, cells were stained with ϩ saturating amounts of FITC-conjugated anti-CD30 or anti-CD70 mAb (Ki- shedding, we treated the CD30 ALCL-derived cell line Karpas 3-FITC or Ki-24-FITC, respectively) by 30-min incubation on ice in PBS 299 with M␤CD, which is known to decrease the cellular choles- containing BSA (0.1%) and propidium iodide (25 ng/ml) for the recogni- terol level and to disrupt lipid rafts. The influence of M␤CD on the tion and gating of dead cells. Cells were analyzed using a FACSCalibur flow cytometer (BD Biosciences, Heidelberg, Germany). surface expression of CD30 was measured by flow cytometry. As shown in Fig. 1A, surface expression of CD30 decreased upon a Peroxidase labeling of Abs 30-min M␤CD treatment (3 mM). The decrease continued when For labeling of Abs with HRP, 0.2 ml of an aqueous sodium periodate M␤CD was removed after 30 min, and cells were further incubated solution (21.4 mg/ml) was added to 1 ml of a solution of peroxidase (1 (Fig. 1B). To investigate whether this loss of membrane-bound 4326 CHOLESTEROL-DEPENDENT CD30 SHEDDING Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 1. Stimulation of CD30 shedding by M␤CD in Karpas 299 and L428 cells. AÐC, Karpas 299 cells (2 ϫ 105/ml) were incubated in the absence or presence of M␤CD. After 30 min of incubation with M␤CD (3 mM) membrane-bound CD30 was determined (A), or cells were washed and further incubated for another 240 min (B). Membrane-bound CD30 was determined by ﬂow cytometry. C, Cells were incubated for 60 min with M␤CD as indicated. sCD30 was determined in the cell supernatants. The results show units per milliliter as means Ϯ SD for quadruplicate determinations. DÐF, L428 cells (106 per milliliter) were incubated with M␤CD as indicated for 60 min. Membrane-bound CD30 (D) or CD70 (as a control) (E) was determined by ﬂow cytometry. F, For determination of sCD30, supernatants of M␤CD-treated cells were immunoprecipitated and immunoblotted.

CD30 was due to CD30 shedding, we analyzed sCD30 in the su- natant (Fig. 2C). M␤CD treatment of Karpas 299 cells had similar pernatant of M␤CD-treated cells. In accordance with the flow cy- effects on cellular lipids (Fig. 2D). tometry, there was a dose-dependent M␤CD-induced increase of As a control, we treated Karpas 299 and L428 cells with M␤CD, sCD30 in the supernatants (Fig. 1C), corroborating that this effect saturated with cholesterol. In this case, we did not observe any was caused by shedding. effect on CD30 shedding as shown by sCD30 ELISA (Fig. 2E). To Next, we studied the effect of M␤CD on the Hodgkin-derived find out whether the cholesterol-dependent CD30 shedding was cell line L428. The cell surface expression of CD30 decreased in restricted to lymphoma cells, HEK 293 cells were transfected with a time- (data not shown) and dose-dependent way (Fig. 1D). As a CD30 cDNA. However, in this respect, HEK 293 cells behaved control, we measured the surface expression of CD70, which was like lymphoma cells, because M␤CD, but not the M␤CD-choles- not altered by M␤CD treatment (Fig. 1E). To investigate the in- terol complex, was able to induce CD30 shedding (data not tegrity of the shedding product, sCD30 from the culture superna- shown). tant was immunoprecipitated and evaluated by Western blotting (Fig. 1F). In this study, the sCD30 generated in response to M␤CD treatment had the same molecular mass as constitutively generated Inhibition profile of the cholesterol-dependent CD30 sheddase 90-kDa sCD30. To exclude nonspecific effects of M␤CD, we measured cellular TACE has been identified as the main sheddase for phorbol ester- lipid levels of L428 cells after treatment with M␤CD. As shown in stimulated shedding of CD30 (17). To identify the responsible pro- Fig. 2, A and B, cellular cholesterol could be removed by treatment teinase for cholesterol-dependent CD30 shedding, we tried to de- with M␤CD in a dose-dependent way. As a control, we measured termine the inhibition profile of M␤CD-induced shedding in L428 cellular DAG, which was not affected. The loss of cellular cho- cells. TIMPs were used to pursue this objective. As shown by lesterol correlated with an increase of sCD30 in the culture super- sCD30 ELISA, the shedding was efficiently blocked by TIMP-3 The Journal of Immunology 4327 Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 2. Cellular cholesterol level regulates CD30 shedding. AÐC, L428 cells (106 per milliliter) were incubated with M␤CD as indicated for 60 min, and the cell pellets of 106 cells/sample were used for determination of cellular cholesterol as described in Materials and Methods. A, OD ratios of cholesterol/DAG. B, Cholesterol was quantiﬁed by comparison with the internal standard. C, sCD30 was determined in the cell supernatants. The results show units per milliliter as means Ϯ SD for quadruplicate determinations. D, Karpas 299 cells (106 per milliliter) were incubated with M␤CD as indicated for 60 min, and the cell pellets of 106 cells/sample were used for determination of cellular cholesterol as described in Materials and Methods. E, L428 cells or Karpas 299 cells (2 ϫ 105/ml) were treated either with M␤CD (8 mM) or with M␤CD-cholesterol complex (6.15 mM), and sCD30 was determined in the cell supernatants. The results show units per milliliter as means Ϯ SD for quadruplicate determinations.

(Fig. 3A), whereas TIMP-1 showed no effect on the M␤CD-de- rupts lipid rafts. As shown by flow cytometry, a down-regulation pendent release of sCD30 (B). Monitoring the cell surface expres- of membrane-anchored CD30 (Fig. 4A) could be measured after sion of CD30 by flow cytometry revealed the same results (data filipin treatment. As a control, surface expression of CD70 re- not shown). Because this is the inhibition profile of TACE (28), it mained unchanged (Fig. 4B). Because the broad spectrum metal- is very likely that TACE is the main enzyme of cholesterol-de- loproteinase inhibitor BB-3644 showed inhibition on the filipin- pendent CD30 shedding. dependent generation of sCD30 as shown in sCD30-ELISA (Fig. 4C), we concluded that metalloproteinase-dependent shedding was Filipin, lovastatin, or cholesterol oxidase induces CD30 responsible for this effect. shedding For the inhibition of cellular cholesterol synthesis, cells were We also examined the effect of filipin on CD30 shedding in L428 cultured with lovastatin, a potent inhibitor of 3-hydroxy-3-meth- cells. Filipin is a sterol-binding polyene antibiotic (18) that dis- ylglutaryl-CoA reductase. Forty-eight-hour incubation with 15 4328 CHOLESTEROL-DEPENDENT CD30 SHEDDING Downloaded from http://www.jimmunol.org/

FIGURE 3. Inhibition proﬁle of the cholesterol-dependent CD30 sheddase. L428 cells (106 per milliliter) were incubated with M␤CD and/or different inhibitors as indicated for 60 min. A, After incubation with M␤CD (8 mM) and/or different concentrations of TIMP-3, sCD30 was determined in the cell supernatants. The results show units per milliliter as means Ϯ SD for triplicate determinations. B, After incubation with M␤CD (8 mM) Ϸ ␮

and/or TIMP-1 (485 nM 10 g/ml), sCD30 was determined in the cell by guest on October 1, 2021 supernatants. The results show units per milliliter as means Ϯ SD for FIGURE 4. Stimulation of CD30 shedding by filipin in L428 cells. triplicate determinations. AÐC, L428 cells (106 per milliliter) were incubated in the absence or presence of filipin (400 ng/ml) and/or BB-3644 (5 ␮M) for 60 min. Then, membrane-bound CD30 (A) or CD70 (as a control) (B) was determined by ␮g/ml lovastatin led to a strong down-regulation of surface CD30 flow cytometry. C, sCD30 in the supernatants was measured. The results as determined by flow cytometry, and BB-3644 inhibited this loss show units per milliliter as means Ϯ SD for quadruplicate determinations. (Fig. 5, A and B). We measured triplicate values of cellular lipids as shown in Fig. 5C, which revealed a significant decrease of cholesterol as shown by an unpaired t test ( p ϭ 0.035). Oxidation of cellular cholesterol is catalyzed by cholesterol ox- in detergent-insoluble low-density complexes (Fig. 6A), but TACE idase, which also leads to a degradation of lipid rafts (29). Incu- was totally excluded from lipid rafts (B). As a positive control of bation of L428 cells with cholesterol oxidase (4 U/ml for 20 h) successful raft isolation, we used peroxidase-linked cholera toxin caused a metalloproteinase-dependent down-regulation of CD30 subunit B, which binds to the raft-resident sphingolipid GM1 (Fig. (Fig. 5D) as demonstrated by flow cytometry and coincubation 6C). with BB-3644. A lipid analysis revealed a decrease of cellular Depletion of cellular cholesterol by M␤CD prevents the asso- cholesterol (Fig. 5E). ciation of most proteins with lipid rafts (32). To show a modified distribution of CD30 after raft depletion, we treated L428 cells Localization of CD30 and TACE in detergent-soluble and with M␤CD and separated the detergent-insoluble fractions in a -insoluble membrane compartments sucrose gradient. In untreated L428 cells, CD30 was partially lo- It has been shown that the localization of APP inside or outside of calized in the raft fractions, but after M␤CD treatment, CD30 was lipid rafts plays a crucial role for its shedding. To investigate the almost completely localized in high-density fractions (Fig. 6D), respective localization of CD30 and TACE, we analyzed their sol- suggesting an increased probability of enzyme-to-substrate contact ubility in Triton X-100 at 4¡C, because lipid raft proteins are de- as a mechanism for cholesterol-dependent shedding of CD30. As fined as being detergent insoluble at 4¡C. We detected two bands a control, we examined whether the shedding-stimulating agent corresponding to the precursor and the mature form of CD30 in the PMA had an effect on the distribution of CD30 in rafts and incu- CD30 immunoblots (30) as well as two bands of TACE in the bated cells with PMA for 5 and 60 min. As shown in Fig. 6D,we TACE immunoblot, the latter most likely representing the inactive did not observe any altered distribution. pro-TACE and the smaller activated form of TACE, which lacks Also in K299 cells, TACE was localized in the nonraft fractions, the prodomain (31). After lysis of L428 cells in 1% Triton X-100 whereas CD30 was partially localized in the raft fractions and and separation in a sucrose gradient, CD30 was partially detected could be removed by M␤CD treatment (data not shown). The Journal of Immunology 4329 Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 5. Stimulation of CD30 shedding by lovastatin and cholesterol oxidase in L428 cells. A and B, L428 cells (105 per milliliter) were incubated ␮ ␮ in RPMI 1640 (10% FCS) in the absence or presence of lovastatin (15 g/ml) and/or BB-3644 (5 M) for 48 h at 37¡C/5% CO2. Then membrane-bound CD30 of lovastatin-treated cells (A) and untreated cells (B) was determined by ﬂow cytometry. C, L428 cells (4 ϫ 105 per milliliter) were incubated in ␮ 6 RPMI 1640 (10% FCS) in the absence or presence of lovastatin (15 g/ml) for 48 h at 37¡C/5% CO2, and the cell pellets of 10 cells/sample were used for determination of cellular cholesterol as described in Materials and Methods. D, L428 cells (105 per milliliter) were incubated in RPMI 1640 (10% FCS) ␮ in the absence or presence of cholesterol oxidase (4 U/ml) and/or BB-3644 (5 M) for 20 h at 37¡C/5% CO2. Membrane-bound CD30 was determined. E, L428 cells (6 ϫ 105 per milliliter) were incubated in RPMI 1640 (10% FCS) in the absence or presence of cholesterol oxidase as indicated for 20 h at ϫ 5 37¡C/5% CO2, and the cell pellets of 3 10 cells/sample were used for determination of cellular cholesterol as described in Materials and Methods.

Discussion teinase. Thus, cholesterol-dependent shedding of CD30 probably Our results showed that CD30 shedding is stimulated by different involves the same proteolytic enzyme as previously described for agents such as M␤CD, filipin, lovastatin, and cholesterol oxidase. the shedding of CD30 (17). Shedding stimulation was time- and dose-dependent and correlated Cholesterol depletion was shown to stimulate APP cleavage by with the reduction of cellular cholesterol, suggesting that the in- an effect on the ␣-secretase termed a disintegrin and metallopro- creased CD30 shedding is indeed due to cholesterol depletion and teinase-10 (18). Recently, new light was shed on this process, and not to another effect of the drugs. Because this effect could be the existence of two APP pools was suggested. APP inside lipid induced in cell lines of different origin such as the Hodgkin-de- rafts seemed to be cleaved by ␤-secretase, whereas APP outside rived L428, the ALCL-derived Karpas 299, as well as the CD30- rafts was shed by ␣-secretase (19). Both pathways could be stim- transfected HEK 293 cells, it appears to be a general phenomenon, ulated: Raft recruitment of APP by cross-linking led to enhanced which is not restricted to specific cell types. ␤-cleavage, whereas raft depletion by M␤CD increased ␣-cleav- As evidenced by the inhibitory effect of BB-3644, metallopro- age. Therefore, ␤-amyloid peptide generation may depend on in- teinases were identified to cause the cholesterol-dependent shed- teractions of APP with lipid rafts. By separation of cellular lipid ding of CD30, and by evaluating the inhibition profile by the in- rafts with a Triton X-100-based method, we demonstrated for the hibitors TIMP-3 and TIMP-1, we suggest TACE as the leading first time that CD30 partially distributes to the raft fractions, CD30 sheddase (33, 34). So far, TACE is the only described met- whereas TACE was excluded from rafts. How the distribution of alloproteinase with this inhibition profile, but we cannot totally CD30 to raft and nonraft fractions is regulated is not clear, but rule out the possibility of the involvement of another metallopro- many surface receptors such as T and B cell receptors dimerize or 4330 CHOLESTEROL-DEPENDENT CD30 SHEDDING

bility of the CD30 cleavage site to TACE or another nonraft protease and thus provides a possible mechanism that participates in cholesterol-dependent CD30 shedding. Another possible explana- tion is provided by the strongly increased phase separation and the large areas of lipid-ordered and -disordered states that have been described in cholesterol-depleted cells (37). For cholesterol-dependent shedding of IL-6R, the formation of larger subdomains of similar lipid order status in cholesterol-depleted cells has been suggested as a mechanism that would lead to a facilitating association of substrate and enzyme (20). A conceivable mechanism may also be the existence of an accessory raft-dependent protein that is re- quired to present CD30 to the sheddase, which would become available following raft depletion. It has been described that cholesterol may stabilize receptors in defined conformations related to their biological functions (38). Therefore, it is also possible that the sheddase is intrinsically more active in the absence of cholesterol. Statins, known as inhibitors of 3-hydroxy-3-methylglutaryl- CoA reductase and therefore being cholesterol-lowering drugs, are used for the investigation of cholesterol-dependent shedding (18). Downloaded from In addition, they may be beneficial in the treatment of inflammatory diseases (39), especially on Th1-related diseases. Oral atorvastatin reversed paralysis in a mouse model for multiple sclerosis and promoted a Th2 bias (40, 41). Simvastatin possessed anti- inflammatory properties in a Th1-driven model of murine inflammatory arthritis (42). The pathogenesis of atherosclerosis, which is http://www.jimmunol.org/ treated with statins, seems to be Th1 linked (43, 44). Recently, sCD30 has been shown to be important for the regulation of the immune system in vivo and in vitro (12, 45) and seems to play a counterregulatory role in the Th1-type immune response (11). Be- cause serum levels of sCD30 were elevated in remittent phases of two Th1-driven diseases, such as multiple sclerosis and rheumatoid arthritis, it is tempting to speculate that the beneficial effect of cholesterol-lowering drugs on Th1-mediated diseases is connected by guest on October 1, 2021 to an increased generation of sCD30 in addition to other immu- nosuppressive effects of statins that have already been described (46). Whether or not cholesterol may become a direct target for modulation of the immune system remains to be elucidated in the FIGURE 6. Localization of CD30 in membrane rafts. AÐC, Membrane future. rafts of 3 ϫ 107 L428 cells were isolated by applying cell lysates to a sucrose gradient and immunoblotted under nonreducing conditions. The indicated molecular mass markers were internal sCD30 standard (90 kDa) Acknowledgments for determination of CD30 and commercially available TACE ectodomain We thank Mrs. G. Schoen for excellent technical assistance. standard (75 kDa) for determination of TACE. A, Proteins of fractions 3Ð11 were separated by nonreducing SDS-PAGE and immunoblotted with References a mixture of HRP-labeled anti-CD30 Ki-2 and Ki-4 mAb. B, Proteins of 1. Del Prete, G., M. De Carli, F. Almerigogna, C. K. Daniel, M. M. D’Elios, fractions 3Ð5 (raft fractions) and 9Ð11 (nonraft fractions) were separated G. Zancuoghi, F. Vinante, G. Pizzolo, and S. Romagnani. 1995. Preferential by nonreducing SDS-PAGE and immunoblotted with HRP-labeled anti- expression of CD30 by human CD4ϩ T cells producing Th2-type cytokines. TACE MAB9302 mAb. As a control, the TACE-ectodomain standard was FASEB J. 9:81. used. C, Two microliters of each fraction of L428 cells were dot blotted 2. Durkop, H., U. Latza, M. Hummel, F. Eitelbach, B. Seed, and H. Stein. 1992. Molecular cloning and expression of a new member of the nerve growth factor and stained with peroxidase-labeled cholera toxin subunit B as a positive receptor family that is characteristic for Hodgkin’s disease. Cell 68:421. control for lipid rafts. In fractions 3Ð5, rafts were isolated. D, L428 cells 3. Baker, S. J., and E. P. Reddy. 1998. Modulation of life and death by the TNF (1.5 ϫ 107 per milliliter) were incubated in prewarmed RPMI 1640 con- receptor superfamily. Oncogene 17:3261. taining 0.1% BSA in the absence or presence of M␤CD (10 mM) or PMA 4. Schnell, R., P. Borchmann, H. Schulz, and A. Engert. 2002. Current strategies of ϫ 7 antibody-based treatment in Hodgkin’s disease. Ann. Oncol. 13:57. (5 ng/ml) at 37¡C for 5 or 60 min as indicated. Membrane rafts of 3 10 5. Josimovic-Alasevic, O., H. Durkop, R. Schwarting, E. Backe, H. Stein, and cells were prepared, and proteins of fractions 3Ð5 (raft fractions) and 9Ð11 T. Diamantstein. 1989. Ki-1 (CD30) antigen is released by Ki-1-positive tumor (nonraft fractions) were separated by nonreducing SDS-PAGE and immu- cells in vitro and in vivo. I. Partial characterization of soluble Ki-1 antigen and noblotted with peroxidase-labeled anti-CD30 Ki-2/Ki-4 mAb. detection of the antigen in cell culture supernatants and in serum by an enzyme- linked immunosorbent assay. Eur. J. Immunol. 19:157. 6. Zinzani, P. L., S. Pileri, M. Bendandi, M. Buzzi, E. Sabattini, S. Ascani, F. Gherlinzoni, M. Magagnoli, P. Albertini, and S. Tura. 1998. Clinical impli- oligomerize after ligand binding, which has been shown to in- cations of serum levels of soluble CD30 in 70 adult anaplastic large-cell lymphoma patients. J. Clin. Oncol. 16:1532. crease association with rafts (35). For CD30, self-association and 7. Horie, R., and T. Watanabe. 1998. CD30: expression and function in health and oligomerization have been demonstrated (36) and might be impor- disease. Semin. Immunol. 10:457. tant for regulating the raft association of CD30. We showed that, 8. Folster-Holst, R., T. Henseler, J. Wehde, H. Lemke, M. Weichenthal, E. Christophers, and H. P. Hansen. 2002. Soluble CD30 plasma concentrations after cholesterol depletion, CD30 was no longer present in the correlate with disease activity in patients with atopic dermatitis. Acta Derm. low-density lipid rafts. This might possibly increase the accessi- Venereol. 82:245. The Journal of Immunology 4331

9. Caligaris Cappio, F., M. T. Bertero, M. Converso, A. Stacchini, F. Vinante, 29. Samsonov, A. V., I. Mihalyov, and F. S. Cohen. 2001. Characterization of cho- S. Romagnani, and G. Pizzolo. 1995. Circulating levels of soluble CD30, a lesterol-sphingomyelin domains and their dynamics in bilayer membranes. Bio- marker of cells producing Th2-type cytokines, are increased in patients with phys. J. 81:1486. systemic lupus erythematosus and correlate with disease activity. Clin. Exp. 30. Froese, P., H. Lemke, J. Gerdes, B. Havsteen, R. Schwarting, H. Hansen, and Rheumatol. 13:339. H. Stein. 1987. Biochemical characterization and biosynthesis of the Ki-1 antigen 10. Wang, G., H. Hansen, E. Tatsis, E. Csernok, H. Lemke, and W. L. Gross. 1997. in Hodgkin-derived and virus-transformed human B and T lymphoid cell lines. High plasma levels of the soluble form of CD30 activation molecule reflect dis- J. Immunol. 139:2081. ease activity in patients with Wegener’s granulomatosis. Am. J. Med. 102:517. 31. Slack, B. E., L. K. Ma, and C. C. Seah. 2001. Constitutive shedding of the 11. Gerli, R., C. Lunardi, F. Vinante, O. Bistoni, G. Pizzolo, and C. Pitzalis. 2001. ϩ amyloid precursor protein ectodomain is up-regulated by tumour necrosis fac- Role of CD30 T cells in rheumatoid arthritis: a counter-regulatory paradigm for tor-␣ converting enzyme. Biochem. J. 357:787. Th1-driven diseases. Trends Immunol. 22:72. 32. Simons, K., and R. Ehehalt. 2002. Cholesterol, lipid rafts, and disease. J. Clin. 12. Saraiva, M., P. Smith, P. G. Fallon, and A. Alcami. 2002. Inhibition of type 1 Invest. 110:597. cytokine-mediated inflammation by a soluble CD30 homologue encoded by ec- 33. Beck, G., G. Bottomley, D. Bradshaw, M. Brewster, M. Broadhurst, R. Devos, tromelia (mousepox) virus. J. Exp. Med. 196:829. C. Hill, W. Johnson, H. J. Kim, S. Kirtland, et al. 2002. (E)-2(R)-[1(S)-(Hydroxy- 13. McMillan, S. A., G. V. McDonnell, J. P. Douglas, A. G. Droogan, and carbamoyl)-4-phenyl-3-butenyl]-2Ј-isobutyl-2Ј-(methanesulfonyl)-4-methyl- S. A. Hawkins. 1998. Elevated serum and CSF levels of soluble CD30 during valerohydrazide (Ro 32-7315), a selective and orally active inhibitor of tumor clinical remission in multiple sclerosis. Neurology 51:1156. necrosis factor-␣ convertase. J. Pharmacol. Exp. Ther. 302:390. 14. Gerli, R., C. Lunardi, and C. Pitzalis. 2002. Unmasking the anti-inflammatory cytokine response in rheumatoid synovitis. Rheumatology (Oxford) 41:1341. 34. Amour, A., C. G. Knight, A. Webster, P. M. Slocombe, P. E. Stephens, 15. Giovannoni, G., and D. H. Miller. 1999. Multiple sclerosis and its treatment. V. Knauper, A. J. Docherty, and G. Murphy. 2000. The in vitro activity of J. R. Coll. Physicians Lond. 33:315. ADAM-10 is inhibited by TIMP-1 and TIMP-3. FEBS Lett. 473:275. 16. van Roon, J. A., J. W. Bijlsma, and F. P. Lafeber. 2002. Suppression of inflam- 35. Janes, P. W., S. C. Ley, A. I. Magee, and P. S. Kabouridis. 2000. The role of lipid mation and joint destruction in rheumatoid arthritis may require a concerted ac- rafts in T cell antigen receptor (TCR) signalling. Semin. Immunol. 12:23. tion of Th2 cytokines. Curr. Opin. Investig. Drugs 3:1011. 36. Horie, R., T. Watanabe, Y. Morishita, K. Ito, T. Ishida, Y. Kanegae, I. Saito, 17. Hansen, H. P., S. Dietrich, T. Kisseleva, T. Mokros, R. Mentlein, H. H. Lange, M. Higashihara, S. Mori, and M. E. Kadin. 2002. Ligand-independent signaling G. Murphy, and H. Lemke. 2000. CD30 shedding from Karpas 299 lymphoma by overexpressed CD30 drives NF-␬B activation in Hodgkin-Reed-Sternberg cells is mediated by TNF-␣-converting enzyme. J. Immunol. 165:6703. cells. Oncogene 21:2493. Downloaded from 18. Kojro, E., G. Gimpl, S. Lammich, W. Marz, and F. Fahrenholz. 2001. Low 37. Hao, M., S. Mukherjee, and F. R. Maxfield. 2001. Cholesterol depletion induces cholesterol stimulates the nonamyloidogenic pathway by its effect on the ␣-secre- large scale domain segregation in living cell membranes. Proc. Natl. Acad. Sci. tase ADAM 10. Proc. Natl. Acad. Sci. USA 98:5815. USA 98:13072. 19. Ehehalt, R., P. Keller, C. Haass, C. Thiele, and K. Simons. 2003. Amyloidogenic 38. Burger, K., G. Gimpl, and F. Fahrenholz. 2000. Regulation of receptor function processing of the Alzheimer ␤-amyloid precursor protein depends on lipid rafts. by cholesterol. Cell. Mol. Life Sci. 57:1577. J. Cell Biol. 160:113. 39. Stuve, O., S. Youssef, L. Steinman, and S. S. Zamvil. 2003. Statins as potential 20. Matthews, V., B. Schuster, S. Schutze, I. Bussmeyer, A. Ludwig, C. Hundhausen, therapeutic agents in neuroinflammatory disorders. Curr. Opin. Neurol. 16:393. T. Sadowski, P. Saftig, D. Hartmann, K. J. Kallen, and S. Rose-John. 2003. 40. Youssef, S., O. Stuve, J. C. Patarroyo, P. J. Ruiz, J. L. Radosevich, E. M. Hur, http://www.jimmunol.org/ Cellular cholesterol depletion triggers shedding of the human interleukin-6 re- M. Bravo, D. J. Mitchell, R. A. Sobel, L. Steinman, and S. S. Zamvil. 2002. The ceptor by ADAM10 and ADAM17 (TACE). J. Biol. Chem. 278:38829. HMG-CoA reductase inhibitor, atorvastatin, promotes a Th2 bias and reverses 21. Simons, K., and D. Toomre. 2000. Lipid rafts and signal transduction. Nat. Rev. paralysis in central nervous system autoimmune disease. Nature 420:78. Mol. Cell Biol. 1:31. 41. Aktas, O., S. Waiczies, A. Smorodchenko, J. Dorr, B. Seeger, T. Prozorovski, 22. Kabouridis, P. S., J. Janzen, A. L. Magee, and S. C. Ley. 2000. Cholesterol S. Sallach, M. Endres, S. Brocke, R. Nitsch, and F. Zipp. 2003. Treatment of depletion disrupts lipid rafts and modulates the activity of multiple signaling relapsing paralysis in experimental encephalomyelitis by targeting Th1 cells pathways in T lymphocytes. Eur. J. Immunol. 30:954. through atorvastatin. J. Exp. Med. 197:725. 23. Phong, M. C., P. Gutwein, S. Kadel, K. Hexel, P. Altevogt, O. Linderkamp, and 42. Leung, B. P., N. Sattar, A. Crilly, M. Prach, D. W. McCarey, H. Payne, B. Brenner. 2003. Molecular mechanisms of L-selectin-induced co-localization in R. Madhok, C. Campbell, J. A. Gracie, F. Y. Liew, and I. B. McInnes. 2003. A rafts and shedding. Biochem. Biophys. Res. Commun. 300:563. novel anti-inflammatory role for simvastatin in inflammatory arthritis. J. Immu- 24. Gutwein, P., S. Mechtersheimer, S. Riedle, A. Stoeck, D. Gast, S. Joumaa, nol. 170:1524. H. Zentgraf, M. Fogel, and D. P. Altevogt. 2003. ADAM10-mediated cleavage of by guest on October 1, 2021 L1 adhesion molecule at the cell surface and in released membrane vesicles. 43. Frostegard, J., A. K. Ulfgren, P. Nyberg, U. Hedin, J. Swedenborg, U. Andersson, FASEB J. 17:292. and G. K. Hansson. 1999. Cytokine expression in advanced human atheroscle- 25. Horn-Lohrens, O., M. Tiemann, H. Lange, J. Kobarg, M. Hafner, H. Hansen, rotic plaques: dominance of pro-inflammatory (Th1) and macrophage-stimulating W. Sterry, R. M. Parwaresch, and H. Lemke. 1995. Shedding of the soluble form cytokines. Atherosclerosis 145:33. of CD30 from the Hodgkin-analogous cell line L540 is strongly inhibited by a 44. Benagiano, M., A. Azzurri, A. Ciervo, A. Amedei, C. Tamburini, M. Ferrari, new CD30-specific antibody (Ki-4). Int. J. Cancer 60:539. J. L. Telford, C. T. Baldari, S. Romagnani, A. Cassone, et al. 2003. T helper type 26. Kallen, K. J., P. Quinn, and D. Allan. 1993. Effects of brefeldin A on sphingo- 1 lymphocytes drive inflammation in human atherosclerotic lesions. Proc. Natl. myelin transport and lipid synthesis in BHK21 cells. Biochem. J. 289:307. Acad. Sci. USA 100:6658. 27. Blank, N., C. Gabler, M. Schiller, M. Kriegel, J. R. Kalden, and H. M. Lorenz. 45. Hargreaves, P. G., and A. Al-Shamkhani. 2002. Soluble CD30 binds to CD153 2002. A fast, simple and sensitive method for the detection and quantification of with high affinity and blocks transmembrane signaling by CD30. Eur. J. Immu- detergent-resistant membranes. J. Immunol. Methods 271:25. nol. 32:163. 28. Amour, A., P. M. Slocombe, A. Webster, M. Butler, C. G. Knight, B. J. Smith, 46. Neuhaus, O., S. Strasser-Fuchs, F. Fazekas, B. C. Kieseier, G. Niederwieser, P. E. Stephens, C. Shelley, M. Hutton, V. Knauper, et al. 1998. TNF-␣ converting H. P. Hartung, and J. J. Archelos. 2002. Statins as immunomodulators: compar- enzyme (TACE) is inhibited by TIMP-3. FEBS Lett. 435:39. ison with interferon-␤ 1b in MS. Neurology 59:990.