Published online before print July 15, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Center for Immunology and the Departments of Veterinary PathoBiology and Laboratory Medicine and Pathology, University of Minnesota Academic Health Center, University of Minnesota, St. Paul
1Correspondence: University of Minnesota, 295j AS/VM Bldg., 1988 Fitch Ave., St. Paul, MN 55108. E-mail: walch003{at}umn.edu
|
|
|---|
-converting enzyme (ADAM-17)-deficient cells. In this study, we examined the steady-state levels of L-selectin on ADAM-17-deficient cells using a monoclonal antibody to the cytoplasmic region of L-selectin, which allows for the detection of total L-selectin (full-length and the membrane-associated cleavage fragment). We demonstrate that ADAM-17-deficient cells generate a 6-kDa transmembrane fragment of L-selectin. Although inducible L-selectin shedding by phorbol 12-myristate 13-acetate stimulation was not observed by these cells in short-term assays, basal turnover did occur, resulting in the production of soluble L-selectin, as determined by enzyme-linked immunosorbent assay. L-selectin turnover was greatly increased upon ADAM-17 reconstitution. Truncating the juxtamembrane region of L-selectin blocked ADAM-17-independent shedding as did a hydroxymate metalloprotease inhibitor. Together, these findings demonstrate that a metalloprotease activity separate from ADAM-17 can use the cleavage domain of L-selectin. We speculate that separate proteolytic mechanisms of L-selectin shedding may regulate distinct antiadhesive mechanisms, such as inducible shedding for the rapid dissociation of cellcell interactions and constitutive shedding for the homeostatic maintenance of high serum levels of soluble L-selectin, a potential adhesion buffer.
Key Words: inflammation adhesion endoproteolysis
|
|
|---|
L-selectin can be released as a soluble fragment by endoproteolysis [7 8 9 10 ], similar to a number of structurally and functionally diverse membrane proteins that undergo ectodomain shedding upon proteolysis within a membrane-proximal stalk region [11 , 12 ]. The maintenance of high levels of soluble L-selectin in normal serum of humans and mice (1.61.9 µg/ml) indicates that L-selectin shedding is a physiological process [13 14 15 16 ]. Cell activation by various means (e.g., phorbol esters, chemoattractants, cytokines, and stress) results in very rapid L-selectin turnover [7 , 17 , 18 ]. In fact, neutrophils can shed essentially all of their L-selectin molecules in a matter of a few minutes [7 , 17 ]. L-selectin also undergoes a slower basal turnover, which has been described to occur by primary cells and various cell lines expressing recombinant L-selectin [9 , 10 , 19 , 20 ].
L-selectin shedding generates a 6-kDa membrane-associated cleavage fragment that consists of the cytoplasmic region, transmembrane domain, and the first 11 amino acids of the extracellular juxtamembrane region [8
, 10
]. The proteolytic processing of L-selectin within its cleavage domain is relaxed in sequence specificity but requires a certain physical length and secondary structure [10
, 20
, 21
]. Early studies demonstrated that the protease activity involved in inducible L-selectin shedding functioned in cis and could be abrogated by hydroxymate metalloprotease inhibitors [22
23
24
25
]. More recently, Peschon et al. [26
] generated mice lacking functional tumor necrosis factor
(TNF-
)-converting enzyme, also referred to as a disintegrin and metalloprotease (ADAM-17), and demonstrated that ADAM-17-deficient cells were defective in inducible L-selectin shedding. The proteolytic process of basal L-selectin shedding, however, may not be as exclusive. Studies on the structural requirements of L-selectin for basal and inducible shedding indicate that these events may be mediated by separate proteases [10
, 20
]. For instance, a proline substitution at the P1, P2', or P3' position of the L-selectin cleavage site has been shown to block inducible but not basal L-selectin shedding, whereas an alanine substitution at the P1 position enhanced basal but not inducible L-selectin shedding [10
, 20
]. In addition, the epidermal growth factor domain of L-selectin has been shown to be critical for inducible but not basal turnover [10
]. However, because of the ubiquitous expression and predominant role of ADAM-17 in ectodomain shedding, it has been difficult to directly establish whether other proteolytic mechanisms may be involved in L-selectin turnover at the cellular level.
In this report, we use ADAM-17-deficient cells to further investigate L-selectin shedding. This system allows for the examination of ADAM-17-independent cleavage mechanisms, as well as for ADAM-17 reconstitution and the expression of L-selectin constructs to study structural relationships. Evidence is provided here that L-selectin undergoes ADAM-17-independent ectodomain shedding proximal to the membrane by a metalloprotease mechanism. An additional L-selectin cleavage mechanism may serve as a novel therapeutic target to regulate L-selectin adhesion events.
|
|
|---|
Cells and transduction
The 300.19 mouse pre-B cell line [27
28
29
] and the ADAM-17-deficient mouse fibroblast cell line (kindly provided by Dr. Jacques Peschon, Amgen Inc., Seattle, WA) [30
] have been described previously. For cell activation, the phorbol ester phorbol 12-myristate 13-acetate (PMA; Sigma Chemical Co., St. Louis, MO) was used at 100 ng/ml. This compound was dissolved in 100% ethanol and diluted more than 1000-fold for all whole cell assays. Human L-selectin and mouse ADAM-17 were stably expressed in these cells by transduction. The human L-selectin constructs used in this report have been described previously [9
]. L-selectin cDNAs were ligated into the pDON-AI retroviral plasmid (Takara, Shiga, Japan). Mouse ADAM-17 cDNA (kindly provided by Dr. Roy Black, Amgen Inc.) was ligated in a murine stem cell virus (MSCV)-based bicistronic retroviral vector coexpressing enhanced green fluorescent protein (EGFP; MigR1, kindly provided by Dr. Warren Pear, University of Pennsylvania, Philadelphia). Retrovirus generation and transduction were performed as described previously [9
]. ADAM-17 expression was estimated based on EGFP fluorescence. L-selectin expression was determined by various immunoassays as described below.
Immunoassays
Flow cytometry, immunoprecipitation, immunoblotting, and enzyme-linked immunosorbent assay (ELISA) were performed as described previously [9
, 17
, 31
]. Antibody-labeled cells were analyzed (10,000 cells/sample) or sorted by flow cytometry on a FACSCalibur instrument (Becton Dickinson Immunocytometry Systems, San Jose, CA). ELISA was performed by chemiluminescence and analyzed on a Fluoroskan Ascent FL luminometric plate reader (LabSystems, Helsinki, Finland). The relative concentration of soluble L-selectin in conditioned tissue-culture media was calculated by comparing the chemiluminescent intensity to a standard curve of titrated, soluble L-selectin of known concentration (R&D Systems, Minneapolis, MN), as per the manufacturers instructions.
|
|
|---|
, TNF receptor (TNFR)I, TNFRII, interleukin (IL)-6R, and TNF-related activation-induced cytokine (TRANCE) [26
, 30
, 33
, 34
]. The cells used here were confirmed to lack ADAM-17 by a polymerase chain reaction screen on genomic DNA using primers specific to the mouse ADAM-17 metalloprotease domain (data not shown). The ADAM-17-deficient cells also did not express detectable levels of endogenous L-selectin; therefore, the cells were transduced with recombinant human L-selectin, resulting in expression levels equivalent to leukocytes. It has been demonstrated that L-selectin does not undergo rapid shedding by ADAM-17-deficient thymocytes upon protein kinase C (PKC) activation with a phorbol ester [26 ]. Similarly, we found that ADAM-17-deficient cells activated with PMA demonstrated little-to-no down-regulation of L-selectin expression (Fig. 1A ). In contrast to the ADAM-17-deficient cells, PMA activation of leukocytes [9 , 17 ] and transduced 300.19 cells (a mouse pre-B cell line) resulted in a considerable down-regulation of L-selectin expression (Fig. 1C) . Upon restoring ADAM-17 expression, we found that L-selectin expression was down-regulated in a spontaneous manner, which was not significantly increased by PMA stimulation, as determined by flow cytometry (Fig. 1B) and ELISA (data not shown). We as well as others have shown that truncation of the juxtamembrane cleavage region of L-selectin abrogates L-selectin shedding [9 , 10 , 20 , 21 ]. As anticipated, such a mutation also prevented the shedding of L-selectin by the ADAM-17-reconstituted cells (Fig. 1B) . In contrast to the ADAM-17-deficient cells, 300.19 cells when transduced with ADAM-17 did not demonstrate a striking down-regulation in L-selectin expression (Fig. 1C) , indicating that an overexpression of ADAM-17 alone may not have accounted for the spontaneous shedding of L-selectin.
![]() View larger version (23K): [in a new window] |
Figure 1. The reconstitution of ADAM-17-deficient cells results in a considerable down-regulation of L-selectin expression. (A) Effects of PKC stimulation on L-selectin expression by ADAM-17-deficient cells. ADAM-17-deficient cells expressing wild-type L-selectin were treated with PMA to induce cell activation or carrier alone (Resting), as indicated. Cells were stained with DREG-200 to detect L-selectin or an appropriate isotype-negative control mAb (shaded graph; mAb control for activated cells is shown). Cell-staining levels were examined by flow cytometry, and 10,000 cells were examined per sample. (B) Effects of ADAM-17 reconstitution on L-selectin expression. Equivalent numbers of ADAM-17-deficient cells expressing wild-type L-selectin were untreated (WT L-selectin), transduced with a MSCV-based bicistronic retroviral vector (MigR1) that coexpresses ADAM-17 and EGFP (WT L-selectin, ADAM-17), or transduced to express ADAM-17 and also stimulated with PMA (WT L-selectin, ADAM-17, PMA). ADAM-17-deficient cells expressing wild-type L-selectin were also transduced with an empty MigR1 vector (WT L-selectin, MigR1). Finally, ADAM-17-deficient cells expressing cleavage-resistant L-selectin were transduced with ADAM-17 (CR L-selectin, ADAM-17). After transduction (4872 h), the cells were stained with DREG-200. (C) Effects of ADAM-17 overexpression on L-selectin expression. The mouse pre-B cell line 300.19 expressing wild-type L-selectin was untreated (WT L-selectin), activated with PMA (WT L-selectin, PMA), or transduced with ADAM-17 (WT L-selectin, ADAM-17). The cells were then stained with DREG-200. (B and C) All cells were dual-analyzed by flow cytometry for their expression of L-selectin (y-axis) and EGFP (x-axis) as indicated. Nonspecific antibody labeling was determined using the appropriate isotype-negative control mAb (above and data not shown). Both axes represent Log10 fluorescence. PMA activation was performed similarly for all experiments (100 ng/ml; 1x106 cells; 37°C for 30 min). Representative data from multiple repetitions are shown.
|
30 min; see above and ref. [26
]). In this study, we examined the steady-state levels of L-selectin on ADAM-17-deficient cells by immunoprecipitation and immunoblotting using a mAb to the cytoplasmic region of L-selectin. This approach allows for the detection of total L-selectin (i.e., full-length and the membrane-associated cleavage fragment) [8
, 9
, 23
]. It is interesting that a full-length species (75 kDa) and a 6-kDa species of L-selectin were detected (Fig. 2A
). It is known that membrane-proximal cleavage of L-selectin results in a 6-kDa transmembrane fragment [8
], which suggests that L-selectin underwent proteolysis by the ADAM-17-deficient cells within its cleavage region. We investigated whether the cleavage-resistant L-selectin mutant was susceptible to shedding by the ADAM-17-deficient cells. Equal numbers of ADAM-17-deficient cells expressing similar levels of wild-type or cleavage-resistant L-selectin (Fig. 2B)
were detergent-lysed, and the L-selectin was immunoprecipitated with an anticytoplasmic region mAb. Immunoblot analysis revealed that the ADAM-17-deficient cells expressing cleavage-resistant L-selectin did not produce a transmembrane fragment (Fig. 2A)
. In addition, these cells did not generate soluble L-selectin, as determined by ELISA (data not shown).
![]() View larger version (16K): [in a new window] |
Figure 2. L-selectin undergoes proteolysis within its cleavage domain by ADAM-17-deficient cells. (A) Steady-state L-selectin expression by ADAM-17-deficient cells. Equal numbers of ADAM-17-deficient cells expressing wild-type (WT) or cleavage-resistant (CR) L-selectin were detergent-lysed and immunoprecipitated with the mAb CA21, specific to the cytoplasmic region of L-selectin. Immunoprecipitated L-selectin was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis under nonreducing conditions. The resolved proteins were transferred to nitrocellulose and immunoblotted with CA21. The upper bands indicate full-length L-selectin, and the lower band indicates a 6-kDa membrane-associated fragment of L-selectin, as indicated. (B) Wild-type and CR L-selectin expression by the ADAM-17-deficient cells. ADAM-17-deficient cells transduced with wild-type or CR L-selectin cDNA were stained for L-selectin and sorted by flow cytometry to obtain cell populations that expressed similar levels of L-selectin on their cell surface. Immediately before detergent-lysing the transductants for immunoprecipitation, both transductants were stained with DREG-200 to determine the relative expression levels of L-selectin by flow cytometry, as described in Figure 1
.
|
![]() View larger version (17K): [in a new window] |
Figure 3. ADAM-17-deficient cells mediate L-selectin shedding by a metalloprotease activity. (A) ADAM-17-deficient and -reconstituted cells generate soluble L-selectin as determined by ELISA. ADAM-17-deficient and -reconstituted cells expressing L-selectin (Fig. 1B
, top two panels) were each plated at 2.5 x 105 cells/well, media supernatant was collected at the indicated time points (data from the ADAM-17-reconstituted cells are shown only at the 2-h time point), and the levels of soluble L-selectin were determined. The y-axis represents ng/ml soluble L-selectin. (B) Effects of a hydroxymate metalloprotease inhibitor on L-selectin shedding by ADAM-17-deficient cells. Equal numbers of mock (pDON-AI empty vector) and L-selectin-transduced, ADAM-17-deficient cells were plated (2.5x105 cells/well). The ADAM-17-deficient cells were then treated with the hydroxymate metalloprotease inhibitor KD-IX-73-4 (50 µg/ml) or carrier alone (dimethyl sulfoxide). Media supernatant was then collected at 4 h from all samples and compared for levels of soluble L-selectin by ELISA. The y-axis represents the luminescence intensity of soluble L-selectin. For all assays described above, each transductant was plated in triplicate wells, and the supernatant from each well was analyzed in duplicate. Each bar represents the mean ± SD. Representative data from multiple repetitions are shown.
|
|
|
|---|
60% (data not shown). It is interesting that others have shown that inducible L-selectin shedding is blocked by inhibition of the Erk and p38 MAPK pathways [18
, 36
]. We also demonstrate in this study that L-selectin shedding can take place in the absence of ADAM-17. This proteolytic process appears to occur within the cleavage region of L-selectin, as indicated by the production of a 6-kDa transmembrane fragment and that truncating this region blocked shedding. In addition, the production of soluble L-selectin by the ADAM-17-deficient cells was greatly reduced by a hydroxymate metalloprotease inhibitor. These data provide the first direct evidence that L-selectin can undergo ectodomain shedding by an ADAM-17-independent, metalloprotease process.
Similar to L-selectin, various other ADAM-17 substrates can undergo shedding in ADAM-17-deficient cells, such as fractalkine, amyloid precursor protein (APP), proTNF-
, TRANCE, proTNF-
, IL-6R, and tyrosine kinase receptor A [30
, 32
33
34
, 37
38
39
]. The additional L-selectin sheddase appears to be a metalloprotease with the same juxtamembrane spatial requirements as ADAM-17. The ADAM family currently consists of >30 members (www.uta.fi/%7Eloiika/ADAMs/HADAMs.htm), of which five can apparently function as sheddases (ADAM-9, ADAM-10, ADAM-12, ADAM-17, and ADAM-19; refs. [40
41
42
43
44
45
46
47
]). In addition, matrix metalloprotease-7 can also mediate the cleavage of proTNF-
and FasL [48
, 49
]. ADAM-10 is the most similar to ADAM-17 in terms of protein sequence and the structural properties of their catalytic domains [50
], and certain substrates can be cleaved by both ADAMs, including APP, cellular prion protein, IL-6R, and proTNF-
[34
, 41
, 44
, 51
, 52
].
Many of the membrane proteins that undergo ectodomain shedding do so in a basal and inducible manner [11 , 12 ]. For proteins such as IL-6R and cellular prion protein, basal and inducible shedding have been shown to involve separate proteolytic mechanisms [34 , 44 ]. This may also be the case for L-selectin as indicated by findings that basal and inducible shedding require distinct structural features of L-selectin [10 , 20 ]. It is possible that ADAM-17 may play a predominant role in inducible L-selectin shedding, whereas the proteolytic process of steady-state L-selectin turnover may not be as exclusive. A limitation of our studies is that an ADAM-17-deficient fibroblast cell line was examined. However, L-selectin expression has been shown to occur by hematopoietic and nonhematopoietic primary cells [1 , 6 ], and L-selectin shedding has been well validated in hematopoietic and nonhematopoietic cell lines [20 , 21 ]. Of importance is that ADAM-17-independent, L-selectin shedding may primarily account for the serum levels of soluble L-selectin. This is based on recent findings that the serum levels of soluble L-selectin from chimeric mice with ADAM-17-deficient bone marrow were equivalent to that of control animals [53 ].
In summary, ectodomain shedding provides particular growth factors and cytokines with a juxtacrine and paracrine activity. Similarly, L-selectin shedding may have a dual role, and this may require separate proteolytic mechanisms. For instance, rapid L-selectin shedding appears to have an immediate influence on leukocyte accumulation along the vascular wall [23 , 54 ], whereas constitutive shedding may allow for homeostatic regulation of the high serum levels of soluble L-selectin [13 14 15 ]. The soluble form of L-selectin is functional and can influence leukocyte-endothelium interactions [14 , 16 ] and thus may serve as an adhesion buffer to diminish inflammation. Indeed, a reduced level of serum L-selectin has been shown to correlate with susceptibility to inflammatory disease, such as acute respiratory distress syndrome [55 ].
Received April 7, 2003; accepted June 12, 2003.
|
|
|---|
This article has been cited by other articles:
![]() |
Y. Wang, A. H. Herrera, Y. Li, K. K. Belani, and B. Walcheck Regulation of Mature ADAM17 by Redox Agents for L-Selectin Shedding J. Immunol., February 15, 2009; 182(4): 2449 - 2457. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Ahrens, C. Ellwanger, B. K. Smith, N. Bassler, Y. C. Chen, I. Neudorfer, A. Ludwig, C. Bode, and K. Peter Selenium supplementation induces metalloproteinase-dependent L-selectin shedding from monocytes J. Leukoc. Biol., June 1, 2008; 83(6): 1388 - 1395. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. Schaff, P. E. Mattila, S. I. Simon, and B. Walcheck Neutrophil adhesion to E-selectin under shear promotes the redistribution and co-clustering of ADAM17 and its proteolytic substrate L-selectin J. Leukoc. Biol., January 1, 2008; 83(1): 99 - 105. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Gomez-Gaviro, M. Dominguez-Luis, J. Canchado, J. Calafat, H. Janssen, E. Lara-Pezzi, A. Fourie, A. Tugores, A. Valenzuela-Fernandez, F. Mollinedo, et al. Expression and Regulation of the Metalloproteinase ADAM-8 during Human Neutrophil Pathophysiological Activation and Its Catalytic Activity on L-Selectin Shedding J. Immunol., June 15, 2007; 178(12): 8053 - 8063. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Li, J. Brazzell, A. Herrera, and B. Walcheck ADAM17 deficiency by mature neutrophils has differential effects on L-selectin shedding Blood, October 1, 2006; 108(7): 2275 - 2279. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Sengstake, E.-M. Boneberg, and H. Illges CD21 and CD62L shedding are both inducible via P2X7Rs Int. Immunol., July 1, 2006; 18(7): 1171 - 1178. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. Mattila, C. E. Green, U. Schaff, S. I. Simon, and B. Walcheck Cytoskeletal interactions regulate inducible L-selectin clustering Am J Physiol Cell Physiol, August 1, 2005; 289(2): C323 - C332. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Garcia-Touchard, T. D. Henry, G. Sangiorgi, L. G. Spagnoli, A. Mauriello, C. Conover, and R. S. Schwartz Extracellular Proteases in Atherosclerosis and Restenosis Arterioscler. Thromb. Vasc. Biol., June 1, 2005; 25(6): 1119 - 1127. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Rabie, A. Strehl, A. Ludwig, and B. Nieswandt Evidence for a Role of ADAM17 (TACE) in the Regulation of Platelet Glycoprotein V J. Biol. Chem., April 15, 2005; 280(15): 14462 - 14468. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Bergmeier, C. L. Piffath, G. Cheng, V. S. Dole, Y. Zhang, U. H. von Andrian, and D. D. Wagner Tumor Necrosis Factor-{alpha}-Converting Enzyme (ADAM17) Mediates GPIb{alpha} Shedding From Platelets In Vitro and In Vivo Circ. Res., October 1, 2004; 95(7): 677 - 683. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||