Biochem. J. (1990) 270, 771-776 (Printed in Great Britain)
771
Modulation of mannose receptor activity by proteolysis Virginia L. -SHEPHERD,*ttII Rasul ABDOLRASULNIA,t Jane STEPHENSON§ and Chenile
CRENSHAW§ TN, and Nashville, University, Vanderbilt $Biochemistry, and *VA Medical Center, and the Departments of tMedicine §VA Medical Center, Memphis, TN, U.S.A.
Macrophages express a receptor on the cell surface that functions to clear glycoproteins from the extracellular milieu. The activity of this receptor is sensitive to treatment with trypsin. In inflammatory situations, macrophages are activated and exposed to increased levels of extracellular proteases. Under these conditions, mannose receptor activity on the macrophages is diminished. We therefore decided to study the effects of trypsin treatment on the structure and activity of cell-associated and purified receptor that might contribute to the activation-associated receptor down-regulation. Trypsin treatment (1 ,tg/ml for 3 h) resulted in the production of a 140 kDa, trypsin-resistant fragment from both intact cells and isolated receptor. This fragment was no longer able to bind ligand. The remaining 35 kDa fragment apparently is further degraded into smaller fragments, since no evidence of this domain was found on Coomassie Blue-stained gels. The 140 kDa fragment retained immunoreactivity and contained at least a portion of the iodinated tyrosine residues following surface labelling with Na'25I. Neither calcium nor ligand protected the receptor from proteolysis. In addition, prior treatment with oxidants did not increase the susceptibility of the receptor to trypsin digestion. We conclude from these results that the macrophage mannose receptor is clipped by the serine protease trypsin at the cell surface, resulting in the release and further degradation of the binding domain, and the production of a membrane-associated 140 kDa fragment. This trypsin-mediated down-regulation of receptor activity might be important in controlling glycoprotein clearance during inflammation.
INTRODUCTION
MATERIALS AND METHODS
Macrophages have on their surface an endocytic receptor that is involved in clearance of mannose-containing glycoproteins [1].
Materials
This receptor binds ligand at the cell surface through
a
Ca2+_
dependent, trypsin-sensitive, process [2]. Receptor-ligand complexes are internalized and dissociate in an acidic endosomal compartment, and receptors recycle to the surface while ligands are delivered to lysosomes [3]. The mannose receptor has been isolated from rat [4], rabbit [5] and human tissue [6,71, and is a 175-180 kDa membrane-associated glycoprotein. Very little is known about the structural domains involved in Ca2+ or ligand binding. In addition, it is not known how the receptor interacts with the membrane, or its orientation at the cell surface. Expression of active receptors is highly sensitive to agents that modulate macrophage function. Glucocorticoids increase total mannose receptor activity, presumably through a direct effect on receptor biosynthesis [8,9]. Neutrophil-derived oxidants rapidly and specifically down-regulate surface receptor activity by preventing the return of internalized receptors to the surface [10]. Macrophage-activating agents such as bacillus Calmette-Guerin [11,12], endotoxin and phorbol esters [13] down-regulate mannose receptor activity through an as yet unknown mechanism. Concomitant with macrophage activation, secretion of proteases by the macrophage is stimulated [14,15]. Since mannose receptor activity is highly sensitive to proteolysis, we hypothesized that receptor inactivation during macrophage activation might be a direct result of receptor proteolysis. In the present study, we report the effect of serine proteases on purified and cell-associated mannose receptor. We have defined a potential structural domain involved in ligand binding, and we speculate on the role of proteases in down-regulation of receptor expression during macrophage activation.
Horseradish peroxidase (HRP), lactoperoxidase, glucose oxidase, yeast mannan, trypsin (Type I from bovine pancreas), and Protein A-Sepharose were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). The mannose receptor was purified from human lung as described by Stephenson & Shepherd [7]. Mannan-2 (Mnn2) was a gift from Dr. Clinton Ballou (University of California, Berkeley, CA, U.S.A.), and was iodinated using chloramine-T [16]. Goat anti-rabbit IgG and its HRP conjugate, and the HRP colour development reagent, were purchased from Bio-Rad Laboratories (Richmond, CA, U.S.A.). Tissue culture media and antibiotics were purchased from Gibco (Grand Island, NY, U.S.A.), and fetal bovine serum was from Hyclone Laboratories (Logan, UT, U.S.A.). Carrier-free Na'251 was purchased from Amersham Corp. (Arlington Heights, IL, U.S.A.). Preparation of macrophages Rat bone marrow-derived macrophages were prepared as previously described [8]. Briefly, bone marrow cells were flushed from rat femurs with phosphate-buffered saline (PBS; 0.01 Msodium phosphate/0. 14 M-NaCI/0.5 mM-MgCl2/ 1 mM-KCl/ 1 mM-CaCI2). Cells were collected by centrifugation and resuspended in Dulbecco's modified medium containing 10% Lcell (mouse fibroblast)-conditioned medium and 10% fetal bovine serum. Cells were plated in 150 mm plastic tissue culture dishes for 4-5 days, and then removed from the plates with cold 5 mM-EDTA and gentle pipetting. The cells were reseeded into 24-well plates for uptake assays, or were used in suspension for radioiodination experiments. Human alveolar macrophages were obtained by bronchoalveolar lavage from human volunteers. The cells were washed,
Abbreviations used: HRP, horseradish peroxidase; Mnn2, mannan-2; PBS, phosphate-buffered saline; HBSS, Hanks' balanced salt solution. 1 To whom correspondence should be sent, at: VA Medical Center/Research Service, 1310 24th Avenue South, Nashville, TN 37212, U.S.A.
Vol. 270
772
V. L. Shepherd and others
resuspended in 10 mM-Hepes, pH 7.4, containing 0.15 M-NaCl, and used immediately for surface labelling as described below. Uptake of HRP by marrow-derived macrophages Receptor activity was assayed by the ability of macrophages to internalize the mannose-containing protein HRP. Cells were seeded into 24-well plates at 5 x 105 cells per well, and allowed to adhere overnight. For measurement of uptake activity via the mannose receptor, HRP was added to the cells at 8 ,ug in 400 #1 of binding medium [Hanks' balanced salt solution (HBSS) plus 1 % BSA]. Non-specific uptake was determined in companion wells by the addition of mannan (1 mg/ml) plus HRP. Cells and ligand were incubated for 60 min at 37 °C, and then washed twice with HBSS and solubilized in 250 ,ul of 0.1 % Triton X-100. Cell-associated HRP activity was measured as described by Rabinovitch et al. [17] as follows. Cell extract (50,l) was added to 0.9 ml of Phenol Red solution (5 mg/50 ml of PBS). H202 (5 ,u) was added to a final concentration of 50 um. The mixture was incubated at room temperature in the dark for 10 min. The reaction was stopped by addition of 10,u of 1 M-NaOH and the absorbance at 610 nm was measured. The protein concentration in the cell extracts (25 u1) was measured using the Bio-Rad dye method. Specific uptake is defined as the amount of HRP uptake normalized to cell protein minus the non-specific uptake in the presence of mannan. Binding of '251-Mnn2 to the purified human mannose receptor Binding of Mnn2 to the purified receptor was measured as described by Townsend & Stahl [16]. Briefly, receptor (2 ,ug) was incubated with 201ul of Mnn2 (3 x 105 c.p.m./0.3 ,ug) at room temperature for 5 min in 50 mM-Tris buffer, pH 7.8, containing 6 mg of BSA/ml and 40 mM-CaCI2. Receptor-Mnn complexes were precipitated with 50% ammonium sulphate and collected on glass fibre filters. The radioactivity on the filters was quantified by gamma counting. Surface iodination of human alveolar macrophages Macrophages were suspended in 10 mM-Hepes buffer, pH 7.4,
containing 0.15 M-NaCl, at 2 x 107 cells per ml. Glucose oxidase (3 units), lactoperoxidase (2 units), Nal'25I (I mCi) and glucose (10 mM) were added in that order, and the mixture was incubated on ice for 30 min. The cells were washed with Hepes/NaCl and resuspended in 3 ml of 25 mM-imidazole buffer, pH 7.5, containing 0.25 M-sucrose. The cells were homogenized, and a crude membrane fraction was collected by centrifugation as described previously [7]. The membranes were solubilized in immunoprecipitation buffer (IP buffer: 20 mM-Tris, pH 7.75, containing 1 % Triton X- 100, 0.5 % deoxycholate, 0.15 M-NaC1 and 0.02 %
NaN3). Immunoprecipitation of the mannose receptor from macrophage membranes Aliquots (100,ul) of 125I-labelled membranes were diluted with 900 #u1 of IP buffer. Protein A-Sepharose (50,1u of a 10% suspension) was added, and the mixture was incubated for 30 min at room temperature. The supernatant was collected after centrifugation, and rabbit antiserum raised against the purified human receptor or preimmune serum was added for overnight incubation at 4 'C. Protein A-Sepharose was added and the pellet was collected after incubation for 30 min. The pellet was washed three times in IP buffer, then analysed by SDS/PAGE. The gel was stained with Coomassie Brilliant Blue, destained and sealed in plastic. The labelled bands were visualized by overnight exposure of Kodak film X-OMAT AR. Immunoblot analysis Samples were electrophoresed under reducing conditions on SDS 7.5 %-polyacrylamide gels. Proteins were electrophoretically transferred to nitrocellulose in transfer buffer (9.7 g of Tris base, 45 g of glycine and 800 ml of methanol in 4000 ml total volume). Transfer was complete after 18 h at 60 V. The nitrocellulose paper was washed and then incubated overnight with a 1:100 dilution of human receptor antiserum. After extensive washing, HRP-conjugated goat anti-(rabbit IgG) was added, and the paper was incubated for 2 h at room temperature. The paper was
(a)
(b) 1001
'5 c 0
O
4-
80
80 80 60
0)
X
40
c0
m
20
20
J L
0
2
4
6
[Trypsin] (pg/ml)
8
1U
0
1
2
3
Time (h)
Fig. 1. Inactivation of the cell surface and isolated macrophage mannose receptor following limited trypsin digestion: concentration and time dependence (a) Rat bone marrow macrophages were plated into 24-well dishes at 500000 cells per well and allowed to adhere overnight in Dulbecco's modified Eagle's medium containing 10 % fetal bovine serum. The medium was removed and I ml of Hanks' balanced salt solution (HBSS) was added. Increasing concentrations of trypsin were added and the cells plus trypsin were incubated for 30 min at 37 'C. The medium was then removed and the cell monolayer was washed twice with HBSS. Mannose receptor activity was measured by quantifying the uptake of HRP as described in the Materials and methods section. Inactivation of purified mannose receptor was measured by incubation of receptor (2 /sg) in elution buffer (20 mMTris, pH 7.8, containing 0.2 M-NaCl, 1 mM-EDTA, 0.2 % NaN3 and 0.5 % Triton X- 100) with increasing concentrations of trypsin for 3 h at 37 'C in a total volume of 30 1il. After the incubation, "251-labelled Mnn2 was added and binding activity was measured as described in the Materials and methods section. Results are expressed as the percentage activit-y remaining compared with an untreated control. *, Macrophages; 0, isolated receptor. (b) Cells were incubated for various periods of time up to 180 min with 1 ,ug of trypsin/ml in HBSS. At the end of each incubation, the cells were washed and HRP uptake was measured as in (a).
1990
Proteolytic modulation of mannose receptor activity washed and the reactive bands were visualized using 4-chloronaphthol.
RESULTS Inactivation of the macrophage mannose receptor by limited trypsin digestion: concentration and time dependence Mannose receptor activity on rat alveolar macrophages is highly sensitive to trypsin treatment. Stahl et al. [2] reported that 0.01 00 trypsin treatment for 30 min at 37°C resulted in loss of > 80 00 of receptor activity. In the present report, we show that the mannose receptor on rat bone marrow-derived macrophages was similarly down-regulated by trypsin (Fig. la, *), with less than 10 0/ of control activity remaining after incubation of 10 of trypsin/ml for 3 h at 37 'C. The loss macrophages with lg of activity was time-dependent, with 50 % of activity lost by 1.5 h (Fig. 1 b). This treatment did not damage the cells, as determined by Trypan Blue exclusion and by the ability of the cells to recover receptor activity by incubation at 37 'C in the absence of trypsin.
773 tissue by affinity chromatography [7]. The receptor has a molecular mass of 175 kDa, and antibodies raised against this protein react specifically with a 175 kDa protein on human macrophages. The isolated receptor was treated with trypsin as above for the intact cells (Fig. la, 0). The isolated receptor lost binding activity in a manner similar to that of the cell surface receptor. Treatment of approx. 2,ug of receptor with 10 ,tg of trypsin/ml for 3 h at 37 °C reduced binding activity to less than 10 % of the control.
Tryptic cleavage of the purified human mannose receptor: analysis of products by SDS/PAGE Purified human receptor was treated for various periods of time with 1 ,ug of trypsin/ml at 37 °C, and the resulting products were analysed by SDS/PAGE. The only protein bands that were visible by either Coomassie Brilliant Blue staining (Fig. 2a) or silver staining (results not shown) were the intact receptor (175 kDa) and a 140 kDa fragment that began to appear by 30 min. As shown in Fig. 2(b), by 30 min approx. 20 % of the intact receptor had been converted into the 140 kDa fragment. Inactivation of purified human mannose receptor by limited By 3 h only a small fraction (approx. 15 %) of the native protein trypsin digestion remained. The square in Fig. 2(b) at 180 min represents the The mannose receptor has been purified from human lung remaining ligand-binding activity. Fig. 2(c) shows the results of an immunoblot analysis of the 140 kDa fragment derived from the purified receptor when added to the anti-(175 kDa protein) (a) antibody. The 140 kDa fragment retained its reactivity, and was the only immunoreactive band seen on the gels. These results - 175 kDa suggest that trypsin clips the receptor at a site which results in 4Ihi- 140 kDa production of a stable inactive 140 kDa fragment. The remaining 35 kDa portion of the molecule was not detected on gels by -cV Coomassie Blue staining or by immunoblotting, suggesting that ~~~~~~~~~~~~(b)this fragment )o 00 Ic either is degraded further by trypsin, or is not Cn detectable by protein staining or antibody reaction. 0 ...
E30
co
0
X
Trypsin treatment of human alveolar macrophages: analysis of products by SDS/PAGE Human alveolar macrophages were surface-labelled with 1251. The cells were homogenized and a crude membrane fraction was collected by centrifugation. The membranes were solubilized and
E
.E 0
440
-
c Q-4
0
20
0
I
:
20 0
30
60
120 90 Time (min)
150
180 k(c)
175 kDa --
Time (min)
...
0
45
90
120
Vol. 270
175 kDa 140 kDa
180
Fig. 2. Time course of inactivation of the purified mannose receptor b) trypsin, and SDS/PAGE and immunoblot analysis of the resulting fragments Purified human mannose receptor (25 ,ul, containing approx. 2 ,sg of protein) was incubated with 1 ,ug of trypsin/ml for various periods of time up to 180 min at 37 'C. Samples at 30 min intervals were analysed by SDS/PAGE, followed by Coomassie Blue staining of the resultant protein bands (a). The amount of protein in each band (175 kDa versus 140 kDa) at each 30 min interval was quantified by laser densitometry. (b) Data expressed as the percentage of the 175 kDa band remaining at each time point. The black square represents the amount of Mnn2 binding activity remaining in a sample treated as above with trypsin for 180 min compared with the untreated receptor. In (c), purified receptor (2 zg) was treated with trypsin for 0, 45, 90, 120 or 180 min. Samples were electrophoresed on SDS/7.5 % acrylamide gels, transferred to nitrocellulose and incubated with rabbit anti-(mannose receptor) antibody as described in the Materials and methods section.
4 4
140 kDa
A
B
Fig. 3. SDS/PAGE analysis of the immunoprecipitated products from surface-labelled human alveolar macrophages after treatment with trypsin Human alveolar macrophages (2 x I07 cells) were iodinated as described in the Materials and methods section. A crude membrane fraction was prepared and solubilized in IP buffer (20 mM-Tris, pH 7.75, 1 % Triton X-100, 0.5% deoxycholate, 0.15 M-NaCl and 0.02 % NaN3). One-half of the membrane fraction was incubated with trypsin (10 jug/ml) for 3 h at 37 °C (lane B), and the other half remained untreated (lane A). Both samples were then subjected to immunoprecipitation as described in the Materials and methods section. The resulting immunoprecipitates were boiled in sample buffer and run on SDS/7.5 %-acrylamide. The resulting bands were visualized by exposure of Kodak X-OMAT AR film.
774
V. L. Shepherd and others
treated or not with trypsin (10 ,ug/ml) for 3 h. The receptor and receptor fragments were immunoprecipitated with polyclonal anti-(mannose receptor) antibodies, and the immunoprecipitate was analysed on SDS/polyacrylamide gels (Fig. 3). A 175 kDa receptor was immunoprecipitated from untreated cells (Fig. 3, lane A), whereas cells treated with trypsin showed the presence of a major band at 140 kDa (Fig. 3, lane B), identical to the product from the purified receptor. Trypsin is apparently releasing a fragment from the extracytoplasmic side of the membrane, leaving an inactive 140 kDa fragment associated with the membrane. Some portion of this 140 kDa fragment must be extracytoplasmic, since it still contains a significant amount of the 1251 surface label. Although the smaller 35 kDa portion remaining was not detected, it is likely that this extracytoplasmic domain contains the binding site, leaving the inactive 140 kDa fragment. Effects of Ca2l and ligand on trypsin treatment of the receptor Ligand binding by both the cell surface and the isolated receptor is dependent on the presence of Ca2. Studies of other Ca2+-dependent proteins have shown that Ca2+ stabilizes the protein structure so that proteolysis is blocked completely or one or more fragments are stabilized against further cleavage [18-20]. Loeb & Drickamer [18] and Turkewitz et al. [20] have shown that the binding domains of the transferrin receptor and the chicken hepatic lectin are stabilized in the presence of ligand. Fig. 4 shows the effects of 10 mM-Ca2' and 50 mM-methyl-a-Dmannoside on the trypsin treatment of the isolated human mannose receptor. Inclusion of Ca2+ (lane 3), ligand (lane 4), or both (lane 5) did not alter the production of the 140 kDa fragment. In addition, a 35 kDa band did not appear, suggesting that this fragment had not been stabilized by Ca2+ or ligand. Effect of H202 treatment on trypsin treatment of the isolated mannose receptor We have previously reported that the macrophage mannose receptor activity is highly sensitive to oxidants [10]. Within 30 min, 1 mM-H202 decreases receptor activity to less than 10O of control levels. Several workers have shown that a,-antiprotease has an increased susceptibility to proteolysis following oxidation [21]. It was therefore possible that inactivation of the mannose receptor might be a result of increased proteolysis following receptor oxidation. However, the results in Fig. S demonstrate that preincubation of purified receptor with I mmH202 does not alter the time course of appearance of the 140 kDa fragment, or change the amount of this fragment that is produced. These results, together with the observation that ligand binding is not affected by oxidant treatment (results not shown), suggest that oxidants do not significantly alter the receptor structure.
-4*.4
1
2
3
4
1 75 kDa 140 kDa
5
Fig. 4. Effects of Ca2" and ligand on trypsin treatment of the purified mannose receptor Purified receptor (2 fig) was incubated with 1 ,g of trypsin/ml for 3 h in the presence and absence of 50 mM-methyl-CX-D-mannoside or 10 mM-CaCl2, and the products were analysed by SDS/PAGE. Lane 1, untreated receptor; lane 2, receptor+ trypsin; lane 3, receptor + trypsin + 20 mM-CaCI2; lane 4, receptor + trypsin + methyl mannoside; lane 5, receptor + trypsin + CaCl2 + methyl mannoside.
mum
[Trypsin]
(jig/ml)
H202
... 0
_bI
Now
0.1
1.0 -
-t
Fig. 5. Effect of oxidant treatment on trypsin treatment of the purified receptor Purified human mannose receptor (2 jig) was incubated for 30 min at 37 °C with and without 1 mM-H202 as indicated in the Figure. Trypsin was then added at the indicated concentrations (0, 0.1 ,Ig/ml and 1.0 jug/ml), and the reaction was continued for an additional 30 min. The reaction was stopped by the addition of sample buffer, and the samples were analysed by SDS/PAGE on 7.5% polyacrylamide gels.
Effect of other serine proteases on the mannose receptor Activation of macrophages in vivo or in vitro results in downregulation of receptor activity [11-13]. One consequence of this activation is the increased secretion of plasminogen activator into the extracellular space [14]. Remold-O'Donnell & Lewandrowski [22] have shown that plasmin cleaves a 160 kDa protein on the surface of guinea pig macrophages to a 130 kDa fragment. In addition, plasminogen activator in the presence ofplasminogen has been shown to down-regulate other plasma membrane receptors [23,24]. We treated macrophages and purified mannose receptor with plasmin levels up to 200 ,ug/ml, and looked for subsequent down-regulation of activity and formation of the 140 kDa fragment. No decrease in receptor activity was observed, and no change in mobility of the 180 kDa receptor on SDS gels was found (results not shown). DISCUSSION We have previously shown that the macrophage mannose receptor is down-regulated following treatment in vitro with oxidants [10] or agents that activate macrophages [13]. In this study we have investigated the structural properties of the isolated and cell surface mannose receptor that may contribute to production of an inactive receptor. The rabbit alveolar macrophage surface receptor is highly susceptible to trypsin, as reported by Stahl et al. [2]. We demonstrate in this paper that receptor inactivation in rat bone marrow-derived macrophages and human alveolar macrophages is accompanied by production of a large-molecular-mass fragment (140 kDa) that remains associated with the plasma membrane. This fragment is recognized by a polyclonal antibody both by immunoprecipitation of surface labelled receptor (Fig. 3) and by immunoblotting of isolated receptor following treatment with trypsin (Fig. 2c). Since the remaining 35 kDa fragment could not be detected by protein staining or reaction with antibody, and since no ligand-binding activity could be demonstrated in the isolated receptor after trypsin treatment, it would appear that the 35 kDa fragment is degraded further to inactive peptides. Since binding activity is lost following removal of an extracellular 35 kDa fragment by trypsin treatment of intact cells, this fragment presumably contains the carbohydrate-binding domain. We suggest from these results that limited proteolysis might be a mechanism for down-regulation of the mannose receptor, specifically under conditions such as inflammation in which protease levels are high. 1990
Proteolytic modulation of mannose receptor activity Several other endocytic receptors have been shown to be selectively modulated by limited proteolysis. Loeb & Drickamer [18] demonstrated that the chicken hepatic lectin was sensitive to cleavage by subtilisin at a single site in the extracellular portion of the molecule. The binding domain of molecular mass 15 kDa was released. In the presence of Ca2 , this domain was resistant to further proteolysis. Turkewitz et al. [20] demonstrated that trypsin cleaved the transferrin receptor at arginine-121 on the extracellular side, releasing an active 70 kDa monomer into the medium. Goldstein & Brown [25] reported that treatment of the low-density lipoprotein receptor with thrombin, another serine protease, resulted in the release of an active soluble binding domain. Westcott et al. [26] reported that protease-resistant fragments could be produced by trypsin treatment of the purified mannose 6-phosphate receptor. They concluded from these studies that the receptor contained several discrete functional domains. Finally, Lipson et al. [27] have shown that ligand binding to the insulin receptor promotes proteolysis of the a-subunit of the receptor, producing a large 120 kDa fragment. The enzyme in this case is plasma membrane-associated, and apparently proteolysis leads to internalization and intralysosomal degradation of the clipped receptor. Our results in this study suggest that the binding domain of the mannose receptor is proteolytically clipped and degraded extracellularly, whereas the 140 kDa membrane-associated fragment undergoes degradation within the cell. Lipson et al. [27] have shown that following the proteolytic cleavage, degraded insulin receptors are sorted from intact receptors, and are unable to recycle. From preliminary studies, we have found that the membrane-associated 140 kDa fragment of the mannose receptor is lost rapidly from the surface (V. Shepherd, unpublished work). It is interesting to speculate that proteolysis might be a signal to the cell to rapidly internalize the 'damaged' receptors, and to direct them to lysosomes for degradation. We have recently demonstrated that oxidants rapidly cause internalization of the mannose receptor, and prevent recycling [10]. Other workers have demonstrated that direct oxidation of proteins increases their susceptibility to attack by proteolysis [21]. However, we have shown in the present report that prior treatment with H202 does not increase the sensitivity of the mannose receptor to trypsin. In support of the view that the receptor is not directly oxidized is the observation that ligand binding by the isolated receptor is unaffected by treatment with high (5 mM) concentrations of H202 (results not shown). Several known proteolytic systems could be involved in regulation of cell surface proteins. One likely candidate for down-regulation of receptor activity is plasmin, generated locally by activation of plasminogen. The secretion of plasminogen activator increases on treatment of macrophages with activating agents, a condition where mannose receptors are down-regulated [11,14]. Recently Stephens et al. [28] have shown that secreted plasminogen activator binds to cell surface receptors, and this bound ligand activates plasminogen. The resulting plasmin becomes surface-bound and can act at the surface to locally regulate membrane enzymes or receptors. At least two receptor systems have been shown to be regulated by plasmin, i.e. the acetylcholine receptor [23] and the epidermal growth factor receptor [24]. Although we found no effect of plasmin on the mannose receptor, it is possible that our conditions were not optimal for testing its activity. Furthermore, stimulated neutrophils secrete a variety of neutral proteases, including the serine proteases elastase [29] and proteinase III [30], that might act on neighbouring cells such as macrophages at sites of inflammation. FinaIly, cell surface proteases might regulate surface proteins, such as has been described for the insulin receptor [27]. We have previously shown that retinal pigmented epithelial
Vol. 270
775
cells express the mannose receptor on their apical surface [31]. Colley et al. [32] reported that Pronase treatment of these cells in culture removed a 180 kDa band on SDS gels, and a new band appeared at approx. 140 kDa. We also found that Pronase treatment of the isolated mannose receptor resulted in the production of a 140 kDa fragment (results not shown), similar to the fragment observed after trypsin treatment. These results lend further support to the hypothesis that the retinal phagocytic cells have a mannose receptor identical with the macrophage receptor, and that proteolysis may play a role in regulating the function of the retinal epithelium surface proteins. One function of the macrophage mannose receptor appears to be the clearance of unwanted, potentially harmful, enzymes from the extracellular milieu. We have recently proposed a scenario for inflammation and its subsequent resolution that might involve the mannose receptor [10], and can now expand that scenario by the inclusion of proteases. The influx of neutrophils during the initial acute phase ofinflammation might down-regulate mannose receptors on resident macrophages through an oxidant-mediated process [10]. This would eliminate the mechanism for removing extracellular enzymes, and allow inflammation to proceed. Oxidant production is short-lived, and the resident macrophages could potentially synthesize new mannose receptors. For longer down-regulatory control, proteases from stimulated neutrophils or activated macrophages might cleave surface receptors, and allow the negative control of receptor expression to continue. Finally, at 72 h, when monocytes begin to move into the area and differentiate, new mannose receptor expression occurs [33]. By this time, oxidants and proteases are inactive, and mannose receptors can begin to clear the enzymes and debris. This work was supported by National Institutes of Health Grant A122697. We thank Dr. Kevin McCusker for providing human alveolar macrophages.
REFERENCES 1. Shepherd, V. L. & Stahl, P. D. (1984) in Lysosomes in Biology and Pathology (Dingle, J. T., Dean, R. T. & Sly, W., eds.), pp. 83-98, Elsevier Science Publishers, Amsterdam 2. Stahl, P., Schlesinger, P. H., Sigardson, E., Rodman, J. S. & Lee, Y. C. (1980) Cell 19, 207-215 3. Wileman, T., Boshans, R. L., Schlesinger, P. & Stahl, P. (1984) Biochem. J. 220, 665-675 4. Haltiwanger, R. S. & Hill, R. L. (1986) J. Biol. Chem. 261, 7440-7444 5. Wileman, T. E., Lennartz, M. R. & Stahl, P. D. (1986) Proc. NatI. Acad. Sci. U.S.A. 83, 2501-2505 6. Lennartz, M. R., Cole, F. S., Shepherd, V. L., Wileman, T. E. & Stahl, P. D. (1987) J. Biol. Chem. 262, 9942-9944 7. Stephenson, J. D. & Shepherd, V. L. (1987) Biochem. Biophys. Res. Commun. 148, 883-889 8. Shepherd, V. L., Konish, M. G. & Stahl, P. (1985) J. Biol. Chem. 260, 160-164 9. Cowan, H. G. & Shepherd, V. L. (1989) Am. Rev. Resp. Dis. 139, A15 10. Bozeman, P. M., Hoidal, J. R. & Shepherd, V. L. (1988) J. Biol. 11. 12. 13. 14. 15.
16. 17.
Chem. 263, 1240-1247 Ezekowitz, R. A. B., Austyn, J., Stahl, P. D. & Gordon, S. (1981) J. Exp. Med. 154, 60-76 Imber, M. J., Pizzo, S. V., Johnson, W. J. & Adams D. 0. (1982) J. Biol. Chem. 257, 5129-5135 Shepherd, V. L., Abdolrasulnia, R., Garrett, M. & Cowan, H. (1990) J. Immunol., in the press Vassalli, J.-D. & Reich, E. (1977) J. Exp. Med. 145, 429-437 Adams, D. 0. & Hamilton, T. A. (1984) Annu. Rev. Immunol. 2, 283-318 Townsend, R. & Stahl, P. (1981) Biochem. J. 194, 209-214 Rabinovitch, M., Topper, G., Cristello, P. & Rich, A. (1985) J. Leukocyte Biol. 37, 247-255
776 18. Loeb, J. A. & Drickamer, K. (1987) J. Biol. Chem. 262, 3022-3029 19. Shirayoshi, Y., Hatta, K., Hosoda, M., Tsunasawa, S., Sakiyama, F. & Masatoshi, T. (1986) EMBO J. 5, 2485-2488 20. Turkewitz, A. P., Amatruda, J. F., Borhani, D., Harrison, S. C. & Schwartz, A. L. (1988) J. Biol. Chem. 263, 8318-8325 21. Goldberg, A. L. & Boches, F. S. (1982) Science 215, 1107-1109 22. Remold-O'Donnell, E. & Lewandrowski, K. (1982) Cell. Immunol. 70, 85-95 23. Gross, J. L., Krupp, M. N., Rifkin, D. B. & Lane, M. D. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 2776-2780 24. Hatzfeld, J., Miskin, R. & Reich, E. (1982) J. Cell Biol. 82, 176-182 25. Goldstein, J. L. & Brown, M. S. (1974) J. Biol. Chem. 249,5153-5162 26. Westcott, K. R., Searles, R. P. & Rome, L. H. (1987) J. Biol. Chem. 262, 6101-6107
V. L. Shepherd and others 27. Lipson, K. E., Kolhatkar, A. A. & Donner, D. B. (1988) J. Biol. Chem. 263, 10495-10501 28. Stephens, R. W., Pollanen, J., Tapiovaara, H., Leung, K.-C., Sim, P.-S., Salonen, E.-M., Ronne, E., Behrendt, N., Dano, K. & Vaheri, A. (1989) J. Cell Biol. 108, 1987-1995 29. Brower, M. S., Levin, R. I. & Garry, K. (1985) J. Clin. Invest. 75, 657-666 30. Kao, R. C., Wehner, N. G., Skubitz, K. M., Gray, B. H. & Hoidal, J. R. (1988) J. Clin. Invest. 82, 1963-1973 31. Tarnowski, B. I., McLaughlin, B. J. & Shepherd, V. L. (1987) J. Cell Biol. 105, 60a 32. Colley, N. J., Clark, V. M. & Hall, M. 0. (1987) J. Cell Biol. 44, 377-392 33. Shepherd, V. L., Campbell, E. J., Senior, R. M. & Stahl, P. D. (1982) J. Reticuloendothel. Soc. 32, 423-431
Received 7 November 1989/18 April 1990; accepted 24 May 1990
1990
771
Modulation of mannose receptor activity by proteolysis Virginia L. -SHEPHERD,*ttII Rasul ABDOLRASULNIA,t Jane STEPHENSON§ and Chenile
CRENSHAW§ TN, and Nashville, University, Vanderbilt $Biochemistry, and *VA Medical Center, and the Departments of tMedicine §VA Medical Center, Memphis, TN, U.S.A.
Macrophages express a receptor on the cell surface that functions to clear glycoproteins from the extracellular milieu. The activity of this receptor is sensitive to treatment with trypsin. In inflammatory situations, macrophages are activated and exposed to increased levels of extracellular proteases. Under these conditions, mannose receptor activity on the macrophages is diminished. We therefore decided to study the effects of trypsin treatment on the structure and activity of cell-associated and purified receptor that might contribute to the activation-associated receptor down-regulation. Trypsin treatment (1 ,tg/ml for 3 h) resulted in the production of a 140 kDa, trypsin-resistant fragment from both intact cells and isolated receptor. This fragment was no longer able to bind ligand. The remaining 35 kDa fragment apparently is further degraded into smaller fragments, since no evidence of this domain was found on Coomassie Blue-stained gels. The 140 kDa fragment retained immunoreactivity and contained at least a portion of the iodinated tyrosine residues following surface labelling with Na'25I. Neither calcium nor ligand protected the receptor from proteolysis. In addition, prior treatment with oxidants did not increase the susceptibility of the receptor to trypsin digestion. We conclude from these results that the macrophage mannose receptor is clipped by the serine protease trypsin at the cell surface, resulting in the release and further degradation of the binding domain, and the production of a membrane-associated 140 kDa fragment. This trypsin-mediated down-regulation of receptor activity might be important in controlling glycoprotein clearance during inflammation.
INTRODUCTION
MATERIALS AND METHODS
Macrophages have on their surface an endocytic receptor that is involved in clearance of mannose-containing glycoproteins [1].
Materials
This receptor binds ligand at the cell surface through
a
Ca2+_
dependent, trypsin-sensitive, process [2]. Receptor-ligand complexes are internalized and dissociate in an acidic endosomal compartment, and receptors recycle to the surface while ligands are delivered to lysosomes [3]. The mannose receptor has been isolated from rat [4], rabbit [5] and human tissue [6,71, and is a 175-180 kDa membrane-associated glycoprotein. Very little is known about the structural domains involved in Ca2+ or ligand binding. In addition, it is not known how the receptor interacts with the membrane, or its orientation at the cell surface. Expression of active receptors is highly sensitive to agents that modulate macrophage function. Glucocorticoids increase total mannose receptor activity, presumably through a direct effect on receptor biosynthesis [8,9]. Neutrophil-derived oxidants rapidly and specifically down-regulate surface receptor activity by preventing the return of internalized receptors to the surface [10]. Macrophage-activating agents such as bacillus Calmette-Guerin [11,12], endotoxin and phorbol esters [13] down-regulate mannose receptor activity through an as yet unknown mechanism. Concomitant with macrophage activation, secretion of proteases by the macrophage is stimulated [14,15]. Since mannose receptor activity is highly sensitive to proteolysis, we hypothesized that receptor inactivation during macrophage activation might be a direct result of receptor proteolysis. In the present study, we report the effect of serine proteases on purified and cell-associated mannose receptor. We have defined a potential structural domain involved in ligand binding, and we speculate on the role of proteases in down-regulation of receptor expression during macrophage activation.
Horseradish peroxidase (HRP), lactoperoxidase, glucose oxidase, yeast mannan, trypsin (Type I from bovine pancreas), and Protein A-Sepharose were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). The mannose receptor was purified from human lung as described by Stephenson & Shepherd [7]. Mannan-2 (Mnn2) was a gift from Dr. Clinton Ballou (University of California, Berkeley, CA, U.S.A.), and was iodinated using chloramine-T [16]. Goat anti-rabbit IgG and its HRP conjugate, and the HRP colour development reagent, were purchased from Bio-Rad Laboratories (Richmond, CA, U.S.A.). Tissue culture media and antibiotics were purchased from Gibco (Grand Island, NY, U.S.A.), and fetal bovine serum was from Hyclone Laboratories (Logan, UT, U.S.A.). Carrier-free Na'251 was purchased from Amersham Corp. (Arlington Heights, IL, U.S.A.). Preparation of macrophages Rat bone marrow-derived macrophages were prepared as previously described [8]. Briefly, bone marrow cells were flushed from rat femurs with phosphate-buffered saline (PBS; 0.01 Msodium phosphate/0. 14 M-NaCI/0.5 mM-MgCl2/ 1 mM-KCl/ 1 mM-CaCI2). Cells were collected by centrifugation and resuspended in Dulbecco's modified medium containing 10% Lcell (mouse fibroblast)-conditioned medium and 10% fetal bovine serum. Cells were plated in 150 mm plastic tissue culture dishes for 4-5 days, and then removed from the plates with cold 5 mM-EDTA and gentle pipetting. The cells were reseeded into 24-well plates for uptake assays, or were used in suspension for radioiodination experiments. Human alveolar macrophages were obtained by bronchoalveolar lavage from human volunteers. The cells were washed,
Abbreviations used: HRP, horseradish peroxidase; Mnn2, mannan-2; PBS, phosphate-buffered saline; HBSS, Hanks' balanced salt solution. 1 To whom correspondence should be sent, at: VA Medical Center/Research Service, 1310 24th Avenue South, Nashville, TN 37212, U.S.A.
Vol. 270
772
V. L. Shepherd and others
resuspended in 10 mM-Hepes, pH 7.4, containing 0.15 M-NaCl, and used immediately for surface labelling as described below. Uptake of HRP by marrow-derived macrophages Receptor activity was assayed by the ability of macrophages to internalize the mannose-containing protein HRP. Cells were seeded into 24-well plates at 5 x 105 cells per well, and allowed to adhere overnight. For measurement of uptake activity via the mannose receptor, HRP was added to the cells at 8 ,ug in 400 #1 of binding medium [Hanks' balanced salt solution (HBSS) plus 1 % BSA]. Non-specific uptake was determined in companion wells by the addition of mannan (1 mg/ml) plus HRP. Cells and ligand were incubated for 60 min at 37 °C, and then washed twice with HBSS and solubilized in 250 ,ul of 0.1 % Triton X-100. Cell-associated HRP activity was measured as described by Rabinovitch et al. [17] as follows. Cell extract (50,l) was added to 0.9 ml of Phenol Red solution (5 mg/50 ml of PBS). H202 (5 ,u) was added to a final concentration of 50 um. The mixture was incubated at room temperature in the dark for 10 min. The reaction was stopped by addition of 10,u of 1 M-NaOH and the absorbance at 610 nm was measured. The protein concentration in the cell extracts (25 u1) was measured using the Bio-Rad dye method. Specific uptake is defined as the amount of HRP uptake normalized to cell protein minus the non-specific uptake in the presence of mannan. Binding of '251-Mnn2 to the purified human mannose receptor Binding of Mnn2 to the purified receptor was measured as described by Townsend & Stahl [16]. Briefly, receptor (2 ,ug) was incubated with 201ul of Mnn2 (3 x 105 c.p.m./0.3 ,ug) at room temperature for 5 min in 50 mM-Tris buffer, pH 7.8, containing 6 mg of BSA/ml and 40 mM-CaCI2. Receptor-Mnn complexes were precipitated with 50% ammonium sulphate and collected on glass fibre filters. The radioactivity on the filters was quantified by gamma counting. Surface iodination of human alveolar macrophages Macrophages were suspended in 10 mM-Hepes buffer, pH 7.4,
containing 0.15 M-NaCl, at 2 x 107 cells per ml. Glucose oxidase (3 units), lactoperoxidase (2 units), Nal'25I (I mCi) and glucose (10 mM) were added in that order, and the mixture was incubated on ice for 30 min. The cells were washed with Hepes/NaCl and resuspended in 3 ml of 25 mM-imidazole buffer, pH 7.5, containing 0.25 M-sucrose. The cells were homogenized, and a crude membrane fraction was collected by centrifugation as described previously [7]. The membranes were solubilized in immunoprecipitation buffer (IP buffer: 20 mM-Tris, pH 7.75, containing 1 % Triton X- 100, 0.5 % deoxycholate, 0.15 M-NaC1 and 0.02 %
NaN3). Immunoprecipitation of the mannose receptor from macrophage membranes Aliquots (100,ul) of 125I-labelled membranes were diluted with 900 #u1 of IP buffer. Protein A-Sepharose (50,1u of a 10% suspension) was added, and the mixture was incubated for 30 min at room temperature. The supernatant was collected after centrifugation, and rabbit antiserum raised against the purified human receptor or preimmune serum was added for overnight incubation at 4 'C. Protein A-Sepharose was added and the pellet was collected after incubation for 30 min. The pellet was washed three times in IP buffer, then analysed by SDS/PAGE. The gel was stained with Coomassie Brilliant Blue, destained and sealed in plastic. The labelled bands were visualized by overnight exposure of Kodak film X-OMAT AR. Immunoblot analysis Samples were electrophoresed under reducing conditions on SDS 7.5 %-polyacrylamide gels. Proteins were electrophoretically transferred to nitrocellulose in transfer buffer (9.7 g of Tris base, 45 g of glycine and 800 ml of methanol in 4000 ml total volume). Transfer was complete after 18 h at 60 V. The nitrocellulose paper was washed and then incubated overnight with a 1:100 dilution of human receptor antiserum. After extensive washing, HRP-conjugated goat anti-(rabbit IgG) was added, and the paper was incubated for 2 h at room temperature. The paper was
(a)
(b) 1001
'5 c 0
O
4-
80
80 80 60
0)
X
40
c0
m
20
20
J L
0
2
4
6
[Trypsin] (pg/ml)
8
1U
0
1
2
3
Time (h)
Fig. 1. Inactivation of the cell surface and isolated macrophage mannose receptor following limited trypsin digestion: concentration and time dependence (a) Rat bone marrow macrophages were plated into 24-well dishes at 500000 cells per well and allowed to adhere overnight in Dulbecco's modified Eagle's medium containing 10 % fetal bovine serum. The medium was removed and I ml of Hanks' balanced salt solution (HBSS) was added. Increasing concentrations of trypsin were added and the cells plus trypsin were incubated for 30 min at 37 'C. The medium was then removed and the cell monolayer was washed twice with HBSS. Mannose receptor activity was measured by quantifying the uptake of HRP as described in the Materials and methods section. Inactivation of purified mannose receptor was measured by incubation of receptor (2 /sg) in elution buffer (20 mMTris, pH 7.8, containing 0.2 M-NaCl, 1 mM-EDTA, 0.2 % NaN3 and 0.5 % Triton X- 100) with increasing concentrations of trypsin for 3 h at 37 'C in a total volume of 30 1il. After the incubation, "251-labelled Mnn2 was added and binding activity was measured as described in the Materials and methods section. Results are expressed as the percentage activit-y remaining compared with an untreated control. *, Macrophages; 0, isolated receptor. (b) Cells were incubated for various periods of time up to 180 min with 1 ,ug of trypsin/ml in HBSS. At the end of each incubation, the cells were washed and HRP uptake was measured as in (a).
1990
Proteolytic modulation of mannose receptor activity washed and the reactive bands were visualized using 4-chloronaphthol.
RESULTS Inactivation of the macrophage mannose receptor by limited trypsin digestion: concentration and time dependence Mannose receptor activity on rat alveolar macrophages is highly sensitive to trypsin treatment. Stahl et al. [2] reported that 0.01 00 trypsin treatment for 30 min at 37°C resulted in loss of > 80 00 of receptor activity. In the present report, we show that the mannose receptor on rat bone marrow-derived macrophages was similarly down-regulated by trypsin (Fig. la, *), with less than 10 0/ of control activity remaining after incubation of 10 of trypsin/ml for 3 h at 37 'C. The loss macrophages with lg of activity was time-dependent, with 50 % of activity lost by 1.5 h (Fig. 1 b). This treatment did not damage the cells, as determined by Trypan Blue exclusion and by the ability of the cells to recover receptor activity by incubation at 37 'C in the absence of trypsin.
773 tissue by affinity chromatography [7]. The receptor has a molecular mass of 175 kDa, and antibodies raised against this protein react specifically with a 175 kDa protein on human macrophages. The isolated receptor was treated with trypsin as above for the intact cells (Fig. la, 0). The isolated receptor lost binding activity in a manner similar to that of the cell surface receptor. Treatment of approx. 2,ug of receptor with 10 ,tg of trypsin/ml for 3 h at 37 °C reduced binding activity to less than 10 % of the control.
Tryptic cleavage of the purified human mannose receptor: analysis of products by SDS/PAGE Purified human receptor was treated for various periods of time with 1 ,ug of trypsin/ml at 37 °C, and the resulting products were analysed by SDS/PAGE. The only protein bands that were visible by either Coomassie Brilliant Blue staining (Fig. 2a) or silver staining (results not shown) were the intact receptor (175 kDa) and a 140 kDa fragment that began to appear by 30 min. As shown in Fig. 2(b), by 30 min approx. 20 % of the intact receptor had been converted into the 140 kDa fragment. Inactivation of purified human mannose receptor by limited By 3 h only a small fraction (approx. 15 %) of the native protein trypsin digestion remained. The square in Fig. 2(b) at 180 min represents the The mannose receptor has been purified from human lung remaining ligand-binding activity. Fig. 2(c) shows the results of an immunoblot analysis of the 140 kDa fragment derived from the purified receptor when added to the anti-(175 kDa protein) (a) antibody. The 140 kDa fragment retained its reactivity, and was the only immunoreactive band seen on the gels. These results - 175 kDa suggest that trypsin clips the receptor at a site which results in 4Ihi- 140 kDa production of a stable inactive 140 kDa fragment. The remaining 35 kDa portion of the molecule was not detected on gels by -cV Coomassie Blue staining or by immunoblotting, suggesting that ~~~~~~~~~~~~(b)this fragment )o 00 Ic either is degraded further by trypsin, or is not Cn detectable by protein staining or antibody reaction. 0 ...
E30
co
0
X
Trypsin treatment of human alveolar macrophages: analysis of products by SDS/PAGE Human alveolar macrophages were surface-labelled with 1251. The cells were homogenized and a crude membrane fraction was collected by centrifugation. The membranes were solubilized and
E
.E 0
440
-
c Q-4
0
20
0
I
:
20 0
30
60
120 90 Time (min)
150
180 k(c)
175 kDa --
Time (min)
...
0
45
90
120
Vol. 270
175 kDa 140 kDa
180
Fig. 2. Time course of inactivation of the purified mannose receptor b) trypsin, and SDS/PAGE and immunoblot analysis of the resulting fragments Purified human mannose receptor (25 ,ul, containing approx. 2 ,sg of protein) was incubated with 1 ,ug of trypsin/ml for various periods of time up to 180 min at 37 'C. Samples at 30 min intervals were analysed by SDS/PAGE, followed by Coomassie Blue staining of the resultant protein bands (a). The amount of protein in each band (175 kDa versus 140 kDa) at each 30 min interval was quantified by laser densitometry. (b) Data expressed as the percentage of the 175 kDa band remaining at each time point. The black square represents the amount of Mnn2 binding activity remaining in a sample treated as above with trypsin for 180 min compared with the untreated receptor. In (c), purified receptor (2 zg) was treated with trypsin for 0, 45, 90, 120 or 180 min. Samples were electrophoresed on SDS/7.5 % acrylamide gels, transferred to nitrocellulose and incubated with rabbit anti-(mannose receptor) antibody as described in the Materials and methods section.
4 4
140 kDa
A
B
Fig. 3. SDS/PAGE analysis of the immunoprecipitated products from surface-labelled human alveolar macrophages after treatment with trypsin Human alveolar macrophages (2 x I07 cells) were iodinated as described in the Materials and methods section. A crude membrane fraction was prepared and solubilized in IP buffer (20 mM-Tris, pH 7.75, 1 % Triton X-100, 0.5% deoxycholate, 0.15 M-NaCl and 0.02 % NaN3). One-half of the membrane fraction was incubated with trypsin (10 jug/ml) for 3 h at 37 °C (lane B), and the other half remained untreated (lane A). Both samples were then subjected to immunoprecipitation as described in the Materials and methods section. The resulting immunoprecipitates were boiled in sample buffer and run on SDS/7.5 %-acrylamide. The resulting bands were visualized by exposure of Kodak X-OMAT AR film.
774
V. L. Shepherd and others
treated or not with trypsin (10 ,ug/ml) for 3 h. The receptor and receptor fragments were immunoprecipitated with polyclonal anti-(mannose receptor) antibodies, and the immunoprecipitate was analysed on SDS/polyacrylamide gels (Fig. 3). A 175 kDa receptor was immunoprecipitated from untreated cells (Fig. 3, lane A), whereas cells treated with trypsin showed the presence of a major band at 140 kDa (Fig. 3, lane B), identical to the product from the purified receptor. Trypsin is apparently releasing a fragment from the extracytoplasmic side of the membrane, leaving an inactive 140 kDa fragment associated with the membrane. Some portion of this 140 kDa fragment must be extracytoplasmic, since it still contains a significant amount of the 1251 surface label. Although the smaller 35 kDa portion remaining was not detected, it is likely that this extracytoplasmic domain contains the binding site, leaving the inactive 140 kDa fragment. Effects of Ca2l and ligand on trypsin treatment of the receptor Ligand binding by both the cell surface and the isolated receptor is dependent on the presence of Ca2. Studies of other Ca2+-dependent proteins have shown that Ca2+ stabilizes the protein structure so that proteolysis is blocked completely or one or more fragments are stabilized against further cleavage [18-20]. Loeb & Drickamer [18] and Turkewitz et al. [20] have shown that the binding domains of the transferrin receptor and the chicken hepatic lectin are stabilized in the presence of ligand. Fig. 4 shows the effects of 10 mM-Ca2' and 50 mM-methyl-a-Dmannoside on the trypsin treatment of the isolated human mannose receptor. Inclusion of Ca2+ (lane 3), ligand (lane 4), or both (lane 5) did not alter the production of the 140 kDa fragment. In addition, a 35 kDa band did not appear, suggesting that this fragment had not been stabilized by Ca2+ or ligand. Effect of H202 treatment on trypsin treatment of the isolated mannose receptor We have previously reported that the macrophage mannose receptor activity is highly sensitive to oxidants [10]. Within 30 min, 1 mM-H202 decreases receptor activity to less than 10O of control levels. Several workers have shown that a,-antiprotease has an increased susceptibility to proteolysis following oxidation [21]. It was therefore possible that inactivation of the mannose receptor might be a result of increased proteolysis following receptor oxidation. However, the results in Fig. S demonstrate that preincubation of purified receptor with I mmH202 does not alter the time course of appearance of the 140 kDa fragment, or change the amount of this fragment that is produced. These results, together with the observation that ligand binding is not affected by oxidant treatment (results not shown), suggest that oxidants do not significantly alter the receptor structure.
-4*.4
1
2
3
4
1 75 kDa 140 kDa
5
Fig. 4. Effects of Ca2" and ligand on trypsin treatment of the purified mannose receptor Purified receptor (2 fig) was incubated with 1 ,g of trypsin/ml for 3 h in the presence and absence of 50 mM-methyl-CX-D-mannoside or 10 mM-CaCl2, and the products were analysed by SDS/PAGE. Lane 1, untreated receptor; lane 2, receptor+ trypsin; lane 3, receptor + trypsin + 20 mM-CaCI2; lane 4, receptor + trypsin + methyl mannoside; lane 5, receptor + trypsin + CaCl2 + methyl mannoside.
mum
[Trypsin]
(jig/ml)
H202
... 0
_bI
Now
0.1
1.0 -
-t
Fig. 5. Effect of oxidant treatment on trypsin treatment of the purified receptor Purified human mannose receptor (2 jig) was incubated for 30 min at 37 °C with and without 1 mM-H202 as indicated in the Figure. Trypsin was then added at the indicated concentrations (0, 0.1 ,Ig/ml and 1.0 jug/ml), and the reaction was continued for an additional 30 min. The reaction was stopped by the addition of sample buffer, and the samples were analysed by SDS/PAGE on 7.5% polyacrylamide gels.
Effect of other serine proteases on the mannose receptor Activation of macrophages in vivo or in vitro results in downregulation of receptor activity [11-13]. One consequence of this activation is the increased secretion of plasminogen activator into the extracellular space [14]. Remold-O'Donnell & Lewandrowski [22] have shown that plasmin cleaves a 160 kDa protein on the surface of guinea pig macrophages to a 130 kDa fragment. In addition, plasminogen activator in the presence ofplasminogen has been shown to down-regulate other plasma membrane receptors [23,24]. We treated macrophages and purified mannose receptor with plasmin levels up to 200 ,ug/ml, and looked for subsequent down-regulation of activity and formation of the 140 kDa fragment. No decrease in receptor activity was observed, and no change in mobility of the 180 kDa receptor on SDS gels was found (results not shown). DISCUSSION We have previously shown that the macrophage mannose receptor is down-regulated following treatment in vitro with oxidants [10] or agents that activate macrophages [13]. In this study we have investigated the structural properties of the isolated and cell surface mannose receptor that may contribute to production of an inactive receptor. The rabbit alveolar macrophage surface receptor is highly susceptible to trypsin, as reported by Stahl et al. [2]. We demonstrate in this paper that receptor inactivation in rat bone marrow-derived macrophages and human alveolar macrophages is accompanied by production of a large-molecular-mass fragment (140 kDa) that remains associated with the plasma membrane. This fragment is recognized by a polyclonal antibody both by immunoprecipitation of surface labelled receptor (Fig. 3) and by immunoblotting of isolated receptor following treatment with trypsin (Fig. 2c). Since the remaining 35 kDa fragment could not be detected by protein staining or reaction with antibody, and since no ligand-binding activity could be demonstrated in the isolated receptor after trypsin treatment, it would appear that the 35 kDa fragment is degraded further to inactive peptides. Since binding activity is lost following removal of an extracellular 35 kDa fragment by trypsin treatment of intact cells, this fragment presumably contains the carbohydrate-binding domain. We suggest from these results that limited proteolysis might be a mechanism for down-regulation of the mannose receptor, specifically under conditions such as inflammation in which protease levels are high. 1990
Proteolytic modulation of mannose receptor activity Several other endocytic receptors have been shown to be selectively modulated by limited proteolysis. Loeb & Drickamer [18] demonstrated that the chicken hepatic lectin was sensitive to cleavage by subtilisin at a single site in the extracellular portion of the molecule. The binding domain of molecular mass 15 kDa was released. In the presence of Ca2 , this domain was resistant to further proteolysis. Turkewitz et al. [20] demonstrated that trypsin cleaved the transferrin receptor at arginine-121 on the extracellular side, releasing an active 70 kDa monomer into the medium. Goldstein & Brown [25] reported that treatment of the low-density lipoprotein receptor with thrombin, another serine protease, resulted in the release of an active soluble binding domain. Westcott et al. [26] reported that protease-resistant fragments could be produced by trypsin treatment of the purified mannose 6-phosphate receptor. They concluded from these studies that the receptor contained several discrete functional domains. Finally, Lipson et al. [27] have shown that ligand binding to the insulin receptor promotes proteolysis of the a-subunit of the receptor, producing a large 120 kDa fragment. The enzyme in this case is plasma membrane-associated, and apparently proteolysis leads to internalization and intralysosomal degradation of the clipped receptor. Our results in this study suggest that the binding domain of the mannose receptor is proteolytically clipped and degraded extracellularly, whereas the 140 kDa membrane-associated fragment undergoes degradation within the cell. Lipson et al. [27] have shown that following the proteolytic cleavage, degraded insulin receptors are sorted from intact receptors, and are unable to recycle. From preliminary studies, we have found that the membrane-associated 140 kDa fragment of the mannose receptor is lost rapidly from the surface (V. Shepherd, unpublished work). It is interesting to speculate that proteolysis might be a signal to the cell to rapidly internalize the 'damaged' receptors, and to direct them to lysosomes for degradation. We have recently demonstrated that oxidants rapidly cause internalization of the mannose receptor, and prevent recycling [10]. Other workers have demonstrated that direct oxidation of proteins increases their susceptibility to attack by proteolysis [21]. However, we have shown in the present report that prior treatment with H202 does not increase the sensitivity of the mannose receptor to trypsin. In support of the view that the receptor is not directly oxidized is the observation that ligand binding by the isolated receptor is unaffected by treatment with high (5 mM) concentrations of H202 (results not shown). Several known proteolytic systems could be involved in regulation of cell surface proteins. One likely candidate for down-regulation of receptor activity is plasmin, generated locally by activation of plasminogen. The secretion of plasminogen activator increases on treatment of macrophages with activating agents, a condition where mannose receptors are down-regulated [11,14]. Recently Stephens et al. [28] have shown that secreted plasminogen activator binds to cell surface receptors, and this bound ligand activates plasminogen. The resulting plasmin becomes surface-bound and can act at the surface to locally regulate membrane enzymes or receptors. At least two receptor systems have been shown to be regulated by plasmin, i.e. the acetylcholine receptor [23] and the epidermal growth factor receptor [24]. Although we found no effect of plasmin on the mannose receptor, it is possible that our conditions were not optimal for testing its activity. Furthermore, stimulated neutrophils secrete a variety of neutral proteases, including the serine proteases elastase [29] and proteinase III [30], that might act on neighbouring cells such as macrophages at sites of inflammation. FinaIly, cell surface proteases might regulate surface proteins, such as has been described for the insulin receptor [27]. We have previously shown that retinal pigmented epithelial
Vol. 270
775
cells express the mannose receptor on their apical surface [31]. Colley et al. [32] reported that Pronase treatment of these cells in culture removed a 180 kDa band on SDS gels, and a new band appeared at approx. 140 kDa. We also found that Pronase treatment of the isolated mannose receptor resulted in the production of a 140 kDa fragment (results not shown), similar to the fragment observed after trypsin treatment. These results lend further support to the hypothesis that the retinal phagocytic cells have a mannose receptor identical with the macrophage receptor, and that proteolysis may play a role in regulating the function of the retinal epithelium surface proteins. One function of the macrophage mannose receptor appears to be the clearance of unwanted, potentially harmful, enzymes from the extracellular milieu. We have recently proposed a scenario for inflammation and its subsequent resolution that might involve the mannose receptor [10], and can now expand that scenario by the inclusion of proteases. The influx of neutrophils during the initial acute phase ofinflammation might down-regulate mannose receptors on resident macrophages through an oxidant-mediated process [10]. This would eliminate the mechanism for removing extracellular enzymes, and allow inflammation to proceed. Oxidant production is short-lived, and the resident macrophages could potentially synthesize new mannose receptors. For longer down-regulatory control, proteases from stimulated neutrophils or activated macrophages might cleave surface receptors, and allow the negative control of receptor expression to continue. Finally, at 72 h, when monocytes begin to move into the area and differentiate, new mannose receptor expression occurs [33]. By this time, oxidants and proteases are inactive, and mannose receptors can begin to clear the enzymes and debris. This work was supported by National Institutes of Health Grant A122697. We thank Dr. Kevin McCusker for providing human alveolar macrophages.
REFERENCES 1. Shepherd, V. L. & Stahl, P. D. (1984) in Lysosomes in Biology and Pathology (Dingle, J. T., Dean, R. T. & Sly, W., eds.), pp. 83-98, Elsevier Science Publishers, Amsterdam 2. Stahl, P., Schlesinger, P. H., Sigardson, E., Rodman, J. S. & Lee, Y. C. (1980) Cell 19, 207-215 3. Wileman, T., Boshans, R. L., Schlesinger, P. & Stahl, P. (1984) Biochem. J. 220, 665-675 4. Haltiwanger, R. S. & Hill, R. L. (1986) J. Biol. Chem. 261, 7440-7444 5. Wileman, T. E., Lennartz, M. R. & Stahl, P. D. (1986) Proc. NatI. Acad. Sci. U.S.A. 83, 2501-2505 6. Lennartz, M. R., Cole, F. S., Shepherd, V. L., Wileman, T. E. & Stahl, P. D. (1987) J. Biol. Chem. 262, 9942-9944 7. Stephenson, J. D. & Shepherd, V. L. (1987) Biochem. Biophys. Res. Commun. 148, 883-889 8. Shepherd, V. L., Konish, M. G. & Stahl, P. (1985) J. Biol. Chem. 260, 160-164 9. Cowan, H. G. & Shepherd, V. L. (1989) Am. Rev. Resp. Dis. 139, A15 10. Bozeman, P. M., Hoidal, J. R. & Shepherd, V. L. (1988) J. Biol. 11. 12. 13. 14. 15.
16. 17.
Chem. 263, 1240-1247 Ezekowitz, R. A. B., Austyn, J., Stahl, P. D. & Gordon, S. (1981) J. Exp. Med. 154, 60-76 Imber, M. J., Pizzo, S. V., Johnson, W. J. & Adams D. 0. (1982) J. Biol. Chem. 257, 5129-5135 Shepherd, V. L., Abdolrasulnia, R., Garrett, M. & Cowan, H. (1990) J. Immunol., in the press Vassalli, J.-D. & Reich, E. (1977) J. Exp. Med. 145, 429-437 Adams, D. 0. & Hamilton, T. A. (1984) Annu. Rev. Immunol. 2, 283-318 Townsend, R. & Stahl, P. (1981) Biochem. J. 194, 209-214 Rabinovitch, M., Topper, G., Cristello, P. & Rich, A. (1985) J. Leukocyte Biol. 37, 247-255
776 18. Loeb, J. A. & Drickamer, K. (1987) J. Biol. Chem. 262, 3022-3029 19. Shirayoshi, Y., Hatta, K., Hosoda, M., Tsunasawa, S., Sakiyama, F. & Masatoshi, T. (1986) EMBO J. 5, 2485-2488 20. Turkewitz, A. P., Amatruda, J. F., Borhani, D., Harrison, S. C. & Schwartz, A. L. (1988) J. Biol. Chem. 263, 8318-8325 21. Goldberg, A. L. & Boches, F. S. (1982) Science 215, 1107-1109 22. Remold-O'Donnell, E. & Lewandrowski, K. (1982) Cell. Immunol. 70, 85-95 23. Gross, J. L., Krupp, M. N., Rifkin, D. B. & Lane, M. D. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 2776-2780 24. Hatzfeld, J., Miskin, R. & Reich, E. (1982) J. Cell Biol. 82, 176-182 25. Goldstein, J. L. & Brown, M. S. (1974) J. Biol. Chem. 249,5153-5162 26. Westcott, K. R., Searles, R. P. & Rome, L. H. (1987) J. Biol. Chem. 262, 6101-6107
V. L. Shepherd and others 27. Lipson, K. E., Kolhatkar, A. A. & Donner, D. B. (1988) J. Biol. Chem. 263, 10495-10501 28. Stephens, R. W., Pollanen, J., Tapiovaara, H., Leung, K.-C., Sim, P.-S., Salonen, E.-M., Ronne, E., Behrendt, N., Dano, K. & Vaheri, A. (1989) J. Cell Biol. 108, 1987-1995 29. Brower, M. S., Levin, R. I. & Garry, K. (1985) J. Clin. Invest. 75, 657-666 30. Kao, R. C., Wehner, N. G., Skubitz, K. M., Gray, B. H. & Hoidal, J. R. (1988) J. Clin. Invest. 82, 1963-1973 31. Tarnowski, B. I., McLaughlin, B. J. & Shepherd, V. L. (1987) J. Cell Biol. 105, 60a 32. Colley, N. J., Clark, V. M. & Hall, M. 0. (1987) J. Cell Biol. 44, 377-392 33. Shepherd, V. L., Campbell, E. J., Senior, R. M. & Stahl, P. D. (1982) J. Reticuloendothel. Soc. 32, 423-431
Received 7 November 1989/18 April 1990; accepted 24 May 1990
1990
