HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 18/7/893 03-0901fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by HANSEN, H. P. Articles by ENGERT, A. Search for Related Content PubMed PubMed Citation Articles by HANSEN, H. P. Articles by ENGERT, A.

The ectodomain shedding of CD30 is specifically regulated by peptide motifs in its cysteine-rich domains 2 and 5 ¹

HINRICH P. HANSEN^*,2, ANDREAS RECKE, ULRICH REINEKE, BASTIAN VON TRESCKOW^*, PETER BORCHMANN^*, ELKE POGGE VON STRANDMANN^*, HANS LANGE, HILMAR LEMKE and ANDREAS ENGERT^*

^* Department of Internal Medicine I, University Hospital Cologne, Cologne, Germany;
Department of Biochemistry, University of Kiel, Kiel, Germany; and
Jerini AG, Berlin, Germany

²Correspondence: Department of Internal Medicine I, University Hospital Cologne, LFI, Ebene 4, Room 703, Joseph-Stelzmann-Str. 9, 50924 Cologne, Germany. E-mail: h.hansen{at}uni-koeln.de

SPECIFIC AIMS

Many transmembrane proteins can selectively be released by the proteolytic action of no more than a few enzymes, often by TNF-converting enzyme (TACE). Since the mechanism of target selection is not understood, we focused on the shedding-influencing role of structural elements of the ectodomain of CD30, a known substrate of TACE.

PRINCIPAL FINDINGS

1. CD30 shedding is independent of the structure at the cleavage site
CD30 shedding is achieved by proteolytic cleavage within the membrane proximal stalk region. To localize shedding-sensitive or regulatory sequences at or near the site of cleavage we deleted three consecutive blocks of 11–13 amino acids together covering the predicted stalk region of CD30. However, shedding was not inhibited in any of the single block deletions, suggesting that shedding of CD30 is not restricted to a defined amino acid sequence at the cleavage site or dependent on regulatory exosites within the stalk region.

2. CD30 shedding inhibition by the anti-CD30 antibodies Ki-4 and Ber-H2
mAbs Ki-4 and Ber-H2 are unique in inhibiting CD30 shedding and binding to cysteine-rich domains (CRD)2 and 5, two regions with almost identical amino acid sequence. The other CD30 antibodies that stimulate CD30 shedding are specific for different epitopes outside these two domains. We therefore investigated the role of CRD2 and 5 in the regulation of CD30 shedding.

3. Antibody-dependent inhibition of CD30 shedding is not mediated by intramolecular cross-linking
The possibility that inhibition of antibody-dependent shedding might arise from intramolecular cross-linking of the duplicated sequences CRD2 and CRD5 was excluded by the finding that monovalent Ber-H2 Fab inhibited shedding of CD30 as the parental bivalent Ber-H2 antibody. We transfected 293T cells with wild-type CD30 or CD30 constructs containing only one (CD30M) or none of the CRD2/5 domains (CD30S) (Fig. 1 A). The constructs were controlled by flow cytometry (Fig. 1B ). Whereas Ki-4 and Ber-H2 did not bind CD30S, the shedding-stimulating mAbs Ki-1 and Ki-2 detected all variants by binding to the amino-terminal CRD1 present in all constructs. Ectodomain shedding was assessed by quantifying the released shedding products after SDS-PAGE (Fig. 1C ). Ki-4 mAb inhibited not only the release of sCD30 from wild-type CD30 but also from the single domain-containing CD30M, thus confirming that intramolecular cross-linking is not involved in shedding inhibition. Due to the lack of a Ki-4 mAb binding site, the release of sCD30S was not inhibited.

View larger version (41K): [in this window] [in a new window]   Figure 1. Shedding of CD30 and different ectodomain deletion constructs. A) Schematic representation of CD30 ectodomain deletions. B) 293T cells were transiently transfected with CD30 wild-type (dark gray), CD30M (light gray), and CD30S (black). CD30 was detected by flow cytometry using FITC-labeled anti-CD30 antibodies. C) Transfected 293 T cells were labeled with [35S]methionine and incubated for 90 min with PMA, PMA + BB-2116, or PMA + Ki-4 mAb (10 µg/mL). Soluble CD30 was isolated with immobilized Ki-1 mAb and analyzed by SDS-PAGE. The quantity of the sCD30 bands was evaluated by density scan.

4. CRD2/5-specific antibodies cause partial inhibition of CD30 shedding
Whereas the hydroxamic acid-based metalloproteinase inhibitor BB-2116 caused a nearly complete inhibition of CD30 shedding, binding of Ki-4 mAb to CRD2/5-containing CD30 constructs was only partially effective (Fig. 1C ). The Ki-4-entailed inhibition of the shedding reached maximal values of 50%. Complete inhibition could not be achieved by superior antibody concentrations. In agreement, a constitutive shedding occurred in the absence of CRD2/5 but at reduced levels, as demonstrated by comparing sCD30M and sCD30S release. Hence, CRD2 and CRD5 are seemingly not essential for CD30 shedding, but their presence in the CD30 substrate is facilitating this process.

5. Shedding of membrane-anchored CD30 is stimulated by CRD2/5-containing sCD30
We explored the possibility that the releasing enzyme might be regulated in a feedback fashion by the soluble product through competitive inhibition. However, flow cytometry revealed that purified sCD30 could not inhibit the phorbol myristate acetate (PMA) -induced down-modulation of CD30 from Hodgkin-derived L428 cells; in contrast, alone it caused a loss of membrane-anchored CD30 from nonstimulated cells. Soluble CD30 without CRD2/5 (sCD30S) was ineffective.

6. Release of sCD30 can be induced by CRD2/5-containing sCD30 or CRD2/5 peptides
Since the CRD2/5-containing soluble forms of CD30 could directly induce a loss of membrane-bound CD30, as shown by flow cytometry, we examined whether there was a concomitant enhancement of sCD30 in cell supernatants. Indeed, an increased release of sCD30 from [³⁵S]methionine-labeled L428 cells was stimulated by 20000 U/mL of purified sCD30WT and, to a lesser degree, by identical concentrations of sCD30M containing only one CRD2/5 motif; sCD30S without CDR2/5 motif was ineffective (Fig. 2 B).

View larger version (43K): [in this window] [in a new window]   Figure 2. Modulation of sCD30 release by the cleavage product. A) CD30-derived peptides used for shedding stimulation. B) L428 cells were metabolically labeled with [35S]methionine, then incubated for 90 min with peptides (1 mM) or sCD30 from CD30(WT), CD30M, or CD30S (20000 U/mL). Radiolabeled sCD30 from the samples was immunoprecipitated by Ki-1 mAb and analyzed by SDS-PAGE. The quantity of the sCD30 bands was evaluated by density scan and depicted as % of untreated control. Results of 3 experiments ± SD are shown. The inhibitor BB-2116 (BB) was applied as a control. C) U937 cells were incubated for 90 min with PMA, RKQ (1 mM), or TAC (1 mM). TNF was determined in the supernatants with a sandwich ELISA.

We tested whether CRD2/5-derived peptides (Fig. 2A ) could also induce sCD30 release. The Ber-H2 epitope-derived linear peptide TDC (1 mM), cyclic RKQ (1 mM), and the downstream cyclic peptide TAC (1 mM) induced a substantial increase of sCD30. Since this effect could not be caused by the CRD1 control peptide YRC (1 mM; Ki-2 epitope), these data confirm the shedding-stimulatory effect of CRD2/5 sequences.

The peptide-induced shedding could be inhibited with the hydroxamate inhibitor BB-2116 (Fig. 2B ), indicating that this cleavage was catalyzed by metalloproteinases. TACE is not only responsible for the release of sCD30, but also cleaves membrane-bound TNF. We performed another specificity control and examined whether the cyclic peptides RKQ and TDC (both 1 mM) could also induce the release of TNF from U937 cells. However, this was not the case (Fig. 2C ). Hence, these data suggest that the metalloproteinase-dependent CD30 shedding is selectively stimulated by CRD2/5 sequences localized distant from the cleavage site.

CONCLUSIONS AND SIGNIFICANCE

TACE is responsible for the release of many membrane proteins such as TNF, the TNF receptors I and II, the IL-1 receptor II, and CD30. It is a common feature that cleavage of the different membrane-anchored substrates occurs in the unfolded juxtamembrane stalk region. Nonselective shedding is stimulated by perturbation of the integrity of the membrane. Agents like PMA, cholesterol-depleting compounds, some bacterial toxins, or C-reactive protein (CRP) stimulate different shedding processes. Despite common releasing enzymes and membrane requirements, the shedding of membrane proteins can be selective, often following specific ligand–receptor interaction. This target-selective shedding might participate in controlling the receptor responsiveness and, as for CD30, in cell–cell interactions. Initial attention focused on possible consensus sequences at the cleavage site that might be exposed to the cleaving enzyme. However, as already described for many substrates of "sheddases," there is hardly any evidence for a cleavage at a specific amino acid sequence or for regulatory exosites within the stalk region of CD30.

Instead, CD30 shedding was influenced by the nearly identical domains CRD2 and/or 5, which are localized distant from the cleavage site (Fig. 3 ). Removal or antibody targeting strongly reduced the CD30 shedding but did not block its cleavage. Such regulatory structures in the distal region of some other shedding substrates have been described. While a distal domain in the angiotensin-converting enzyme is necessary for substrate cleavage, conformations of the distal domains of the TNFR I contribute to tuning the shedding event. The latter effect is comparable to CRD2/5 regulation of CD30 shedding, and substrates TNFR I and CD30 belong to the same family of membrane receptors. A remaining question is the biological impact of CD30 shedding regulation through CRD2/5 sequences, especially since this influence was mild. Families suffering from the dominantly inherited autoinflammatory syndromes often show single exchanges of amino acids within the TNFR I. These mutations they markedly reduced the receptor shedding but did not completely block its cleavage. This attenuated shedding causes an elevated TNFR I density on the cell surface, which is assumed to be responsible the dominantly inherited disease syndromes. This example demonstrates that even a mild influence on the shedding process might cause strong biological effects.

View larger version (36K): [in this window] [in a new window]   Figure 3. Model for the regulation of CD30 cleavage through the cysteine-rich domains 2 and 5 (CRD2/5). The availability of the cysteine-rich domains 2 and 5 is crucial for the shedding of CD30. While abundant soluble CRD2/5 stimulates the shedding of CD30, the masking of these domains by specific antibodies inhibits this process.

Of particular interest is the observation that the shedding product sCD30 itself did not inhibit but stimulated CD30 cleavage (Fig. 3) . As demonstrated with relevant peptides, this effect was specifically dependent on the CRD2/5 motifs of soluble CD30 and restricted to the processing of CD30. To our knowledge, such product activation has not been described before. The underlying mechanism is not understood. There is evidence that an interaction of CD30 ectodomains causes at least a partial ligand-independent self-aggregation and that this cross-linking stimulates CD30 shedding. We speculate that CRD2/5 might participate in CD30 aggregation and binding to the sheddase. It is conceivable that enhanced concentrations of the CRD2/5-containing cleavage product (sCD30) in the surrounding of a cell can substitute for membrane-anchored CD30 in the substrate/enzyme complex and would facilitate the shedding of CD30. This model suggests a pivotal role of CRD2 and CRD5 in aggregation and enzyme binding.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-0901fje;

This article has been cited by other articles:

				S. J. Levine Molecular Mechanisms of Soluble Cytokine Receptor Generation J. Biol. Chem., May 23, 2008; 283(21): 14177 - 14181. [Abstract] [Full Text] [PDF]

				D. A. Eichenauer, V. L. Simhadri, E. P. von Strandmann, A. Ludwig, V. Matthews, K. S. Reiners, B. von Tresckow, P. Saftig, S. Rose-John, A. Engert, et al. ADAM10 Inhibition of Human CD30 Shedding Increases Specificity of Targeted Immunotherapy In vitro Cancer Res., January 1, 2007; 67(1): 332 - 338. [Abstract] [Full Text] [PDF]

				S. Nagata, T. Ise, M. Onda, K. Nakamura, M. Ho, A. Raubitschek, and I. H. Pastan Cell membrane-specific epitopes on CD30: Potentially superior targets for immunotherapy PNAS, May 31, 2005; 102(22): 7946 - 7951. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 18/7/893 03-0901fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by HANSEN, H. P. Articles by ENGERT, A. Search for Related Content PubMed PubMed Citation Articles by HANSEN, H. P. Articles by ENGERT, A.