Dendritic cells (DCs) effectively process exogenous and endogenous Ag and present peptide in the context of both class I and class II molecules. We have demonstrated that peripheral blood DCs efficiently degrade synthetic class I peptides at their cell surface within minutes as determined by analyzing DC supernatants by HPLC. Fragments were verified as bona fide cleavage products by direct sequencing using collision-induced dissociation tandem mass spectrometry. The predominant degradative activities were 1) not secreted but associated with activity at the plasma membrane, 2) ecto-orientated, 3) not induced by peptide-specific interactions, and 4) not associated with nonspecific uptake. Sequence analysis indicated that both N- and C-terminal as well as endoproteolytic events were occurring at the cell surface. The primary exoproteolytic event was identified as CD13 or CD13-like activity through inhibition studies and could be inhibited by ubiquitin and metal-chelating agents. Endoproteolytic events could be inhibited in the presence of DTT, but the precise nature of this enzyme is still undetermined. Compared with the starting monocyte population, DCs cultured in the presence of granulocyte-macrophage CSF/IL-4 exhibited the highest degradative rate (4.3 nmol/min), followed by cultured monocytes (2.9 nmol/min) and freshly isolated monocytes (1.0 nmol/min). In addition to increased enzymatic activity, a change in substrate specificity was noted. Results are discussed with respect to APC loading, and alternatives are offered for circumventing such degradation.

Dendritic cells (DCs)3 represent are highly efficient APCs that are the most efficient for primary sensitization in both in vivo and in vitro immune responses. They are found associated with T cell-dependent areas of lymphoid tissue, within tissues at primary areas of Ag entry, and as well at sites of chronic inflammation. DCs possess several important characteristics that are essential for efficient Ag presentation to T cells, including their abilities to 1) express high levels of MHC molecules, 2) express the appropriate costimulatory molecules, 3) traffick to the appropriate lymphoid organs to initiate T cell responses, 4) secrete cytokines that recruit and mature cells involved in the immune response, and 5) degrade proteins for efficient Ag processing and presentation (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14). Their ability to influence tumor growth has also made them extremely attractive candidates for APC/peptide vaccine therapy (15, 16, 17).

In addition to possessing the appropriate intracellular Ag-processing machinery, these highly efficient APCs also express cell surface markers that are shared with cells of the myeloid and lymphoid lineages (e.g., CD13), whose function has been associated with ectoenzymatic activity. Until recently, the biologic role of these ectoenzymes was poorly understood, but evidence is emerging that supports their role as markers for cellular activation and differentiation, signal-transducing molecules, molecules involved in cellular adhesion and migration, molecules that inactivate soluble mediators, and molecules involved in Ag processing (18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41). To date, there is no information concerning the existence or nature of cellular ectoenzymes on DCs other than the absence or the presence of the specific Ag as determined by phenotypic analysis. As DCs are increasingly appreciated as tools used in peptide vaccines, the analysis of DC ectoenzyme function becomes a critical limiting parameter for proper loading and presentation. In this study we report on the rapid and efficient degradation of synthetic class I peptides by endo- and exoproteases on the surface of human myeloid dendritic cells and offer alternatives to inhibit unwanted peptide degradation.

PBMCs were obtained from a leukocyte research product (Leukopak, a by-product of platelet donation, Central Blood Bank, Pittsburgh, PA) by first washing the cells extensively in PBS to remove residual platelets, followed by centrifugation over Ficoll (Sigma, St. Louis, MO). The PBMC layer was washed resuspended in RPMI 1640 (Life Technologies, Grand Island, NY) and plated at a density of 5 × 106 cells/ml in tissue culture flasks. Monocytes from this population were allowed to adhere for 2 h at 37 C in a humidified incubator with 5% CO2. After 2 h, the nonadherent cells were washed in PBS several times, and the adherent cells were recultured in serum-free medium (AIM V, Life Technologies) containing 1000 U/ml of human recombinant GM-CSF and IL-4 (Schering-Plough, Kenilworth, NJ). Loose clumps of DCs became apparent after 2 or 3 days and peaked after days 5 to 7 in culture. The cells exhibited a typical DC morphology by light microscopy. Cell populations were consistently >80% DCs by FACS analysis. Cells were typically CD3, CD14, CD20, HLA-DR+, CD40+, CD80+, and CD86+. These cells also displayed high levels of LFA-1 and ICAM-1 on their surface.

The following peptides were synthesized at the University of Pittsburgh Peptide Synthesis Facility: Melan-A/MART-1(27, 28, 29, 30, 31, 32, 33, 34, 35), AAGIGILTV; tyrosinase(368–376), YMDGTMSQV; gp100(280–288), YLEPGPVTA; and angiotensin, DRVYIHPFHL (Sigma).

DCs were washed several times in PBS and resuspended to a concentration of 1 × 106/250 μl. They were preincubated in PBS at 37°C for 15 min before the addition of peptide. Peptide was added at a final concentration of 48 nmol and incubated for varying lengths of time. At the indicated time periods, supernatants were harvested by microcentrifugation and immediately frozen on dry ice and acetone or were loaded directly onto an analytical C18 column.

Potential cleavage fragments were separated on an analytical C18 column (Waters Associates, Millipore, Bedford, MA) and separated with a linear gradient (3–60% B, 60 min) using a buffer system consisting 0.1% trifluoroacetic acid/H2O (buffer A) and 100% acetonitrile containing 0.1% trifluoroacetic acid using a Rainin HPLC system (Rainin, Emeryville, CA). 1.0 ml fractions were collected. Absorbance was at 214 nm.

Portions of the UV-absorbing HPLC fractions were concentrated and directly injected into a Fisons Quattro II triple quadrupole mass spectrometer (Fisons, Loughborough, U.K.) equipped with an electrospray ionization source. The electrospray needle was operated at a voltage differential of 3.5 keV using a sheath flow of 5 μl/min consisting of 50% acetonitrile/H2O containing 1% acetic acid. Mass spectra were obtained by scanning the range of mass to charge values of 300 to 1350 every 2.7 s and summing the individual spectra. Collision-induced dissociation spectra were obtained by selecting the appropriate mass ion and scanning at 500 amu/s using 3 mtorr Ar in the collision chamber.

DASA was prepared as previously described (42). Briefly, NaNO2 was dissolved in cold H2O and sulfanilic acid added (1 × 10−4 mol) until dissolved. Concentrated HCl was added, and the mixture was chilled on ice. The product precipitated upon standing. The product was washed, and the concentration was determined by the addition of excess resorcinol and monitoring the OD at 385 nm of the colored product. Cells were treated with 3.0 mM DASA for 20 min, and the cells were washed and processed for enzyme assays as described above.

Cells were incubated in the presence of a variety of enzyme inhibitors, including soybean trypsin inhibitor, PMSF, pepstatin A, EDTA, phophoramidon, aprotinin, ortho-phenanthroline, ubiquitin, 5,5′-dithiobis-2-nitrobenzoic acid, N-ethylmaleimide, iodoacetamide, DTT, leupeptin, benzamidine iodoacetic acid, N-α-p-tosyl-l-lysine chloromethyl ketone, and l-p-tosylamino-2-phenylethyl chloromethyl ketone (Sigma) at varying concentrations for 15 to 30 min at 37°C before the addition of peptide. For the purposes of the figures presented in this study, the final concentration of inhibitors was 1 mM unless stated otherwise.

DC membranes were isolated using an aqueous two-phase polymer system as previously described (43). Purity was assessed by monitoring the activity of the following marker enzymes: aminopeptidase, (plasma membranes), n-acetyl-β-d-glucosaminidase (lysosomes), lactate dehydrogenase (cytosol), cytochrome oxidase (mitochondria), thiamine pryophosphatase (Golgi), and NADH diaphorase (endoplasmic reticulum). A purity of 85 to 90% or better was estimated by marker enzyme analysis.

When necessary, peptides were methylated by dissolving the peptide in a solution of 3 N methanolic HCl prepared by the addition of acetyl chloride to dry methanol. Methylated peptides were purified by reverse phase HPLC. Methylation was verified by mass increases of 14 amu or multiples thereof.

A2+ dendritic cells (1 × 106) were preincubated with ubiquitin (1 mM) and DTT (0.1 μM) or without inhibitors for 30 min before the addition of peptide. Tyrosinase peptide (48 nmol) was added, and the cells were incubated for an additional 60 min at 37°C. Supernatants were then pulsed onto chromium-labeled Jurkat A2/Kb target cells in the presence of β2m (Sigma) for 2 h. A tyrosinase-specific effector cell line (CAM) was added (E:T cell ratio, 5:1), and the percent specific lysis was determined by a standard chromium release assay after 4 h at 37°C.

We wondered whether the synthetic peptides derived from tyrosinase would persist in culture with DCs. The interaction of GM-CSF/IL-4-cultured DCs with the tyrosinase peptide yielded a rapid loss in area of the parent peak and the appearance of five additional peaks upon HPLC analysis (Fig. 1,A). These peaks were not present when supernatants of DCs were analyzed in the absence of peptide (Fig. 1,B) or when the peptide was incubated with DC-conditioned medium (Fig. 1,C). The retention times for the products are shown in Table I. The nature and identity of these species were initially not clear. Possibilities included degradation of the added peptide from ectoproteases present on the cell surface or from internalization and secretion. Alternatively, it was possible that peptides were derived from displacement of endogenous class I or class II peptides. When a portion of the material was analyzed at 280 nm, only peaks with retention times at 4.7 and 15.7 min displayed strong absorbance, indicating the presence of the tyrosine group (data not shown).

FIGURE 1.

Rapid degradation of tyrosinase peptide by human DCs. DCs (day 7) were incubated with peptide (A), without peptide (B), or with peptide and DC-conditioned medium (C) for 4 min, and the supernatants were harvested and fractionated by RP-HPLC. The y-axis shows absorbance at 214 nm; the x-axis indicates time (minutes).

FIGURE 1.

Rapid degradation of tyrosinase peptide by human DCs. DCs (day 7) were incubated with peptide (A), without peptide (B), or with peptide and DC-conditioned medium (C) for 4 min, and the supernatants were harvested and fractionated by RP-HPLC. The y-axis shows absorbance at 214 nm; the x-axis indicates time (minutes).

Close modal
Table I.

DC-induced cleavage fragments of tyrosinase peptidea

Retention Time (min)Mass IonSequence\E
4.7 182 
9.9 641 MDGTMS 
13.5 737 DGTMSQV\E 
15.7 804 YMDGTMS 
16.9 868 MDGTMSQV\E 
20.2 1031 YMDGTMSQV (parent) 
Retention Time (min)Mass IonSequence\E
4.7 182 
9.9 641 MDGTMS 
13.5 737 DGTMSQV\E 
15.7 804 YMDGTMS 
16.9 868 MDGTMSQV\E 
20.2 1031 YMDGTMSQV (parent) 
a

DCs were incubated with tyrosinase peptide as described in Materials and Methods. Supernatants were fractionated by RP-HPLC, and the peaks were identified and sequenced using collision-induced dissociation tandem mass spectrometry.

The HPLC peaks were concentrated and subjected to mass spectrometry analysis. All the fragment peaks yielded masses smaller than the measured parent mass (Table I). Mass analysis indicated that the peak at 4.7 min contained a number of species in addition to a mass ion at 182, suggesting the presence of a free tyrosine residue. Since the elution time of this product was so early, the other mass ion species may represent species that were carried over from the injection peak. In addition, we confirmed the elution time of free tyrosine by injection of tyrosine standards (not shown). Sequence analysis of the remaining mass ions was performed through collision-induced dissociation tandem mass spectrometry (Fig. 2, A–D, and Table I). Sequence analysis indicated that mass 641 corresponded to MDGTMS, mass 737 to DGTMSQV, mass 804 to YMDGTMS, and mass 868 to MDGTMSQV. These results suggested that N-terminal (exoprotease) as well as C-terminal (exoprotease) and possible endoprotease activities were acting on the peptide.

FIGURE 2.

Collision-induced dissociation (CID) tandem mass spectrometry of the major tyrosinase cleavage products. A, CID of mass ion 641; B, CID of mass ion 737; C, CID of mass ion 804; D, CID of mass ion 868. Numbers above the sequence are the representative b ions, and those below are the y ions. Underlined values indicate those that were found under the experimental conditions.

FIGURE 2.

Collision-induced dissociation (CID) tandem mass spectrometry of the major tyrosinase cleavage products. A, CID of mass ion 641; B, CID of mass ion 737; C, CID of mass ion 804; D, CID of mass ion 868. Numbers above the sequence are the representative b ions, and those below are the y ions. Underlined values indicate those that were found under the experimental conditions.

Close modal

Degradation of the tyrosinase peptide was rapid, with the majority (70%) of the degradation occurring within 10 min (Fig. 3). Degradation rates did vary from donor to donor, but usually not more than 15 to 20% (not shown). Continuous increases were not seen with any of the degradation products, since they also serve as additional substrates. Upon increased incubation times (15 min), masses of 717 (YMDGTM) and 554 (MDGTM) were also identified (not shown).

FIGURE 3.

Degradation of the tyrosinase peptide by human DCs is rapid. DCs were incubated with tyrosinase peptide as described in Materials and Methods, and its degradation curve was determined by RP-HPLC.

FIGURE 3.

Degradation of the tyrosinase peptide by human DCs is rapid. DCs were incubated with tyrosinase peptide as described in Materials and Methods, and its degradation curve was determined by RP-HPLC.

Close modal

The finding of enhanced peptide degradation strongly suggested that the degradative activity was at least cell associated. To evaluate this possibility, DCs were incubated in the presence of the tyrosinase peptide for 4 min, and the cells were rapidly removed by microcentrifugation. The conditioned supernatant was incubated for an additional 56 min, and the amount of tyrosinase remaining was quantitated by HPLC. The results (shown in Fig. 4 A) indicated that degradation ceased upon cell removal. In control cultures containing DCs, the parent peptide was reduced from 48 to 1.3 nmol within 15 min. These results further support the idea of cell-associated enzymatic activity and also indicates that the interaction of the peptide with the DC does not stimulate the secretion of degradative enzymes.

FIGURE 4.

DC-peptide degradative activity is associated with ectoenzymatic activity. A, DCs were incubated in the presence of tyrosinase peptide for varying times, the cells were removed after 4 min in a parallel experiment, and the incubation was continued for an additional 56 min. The parent peptide remaining was quantitated by RP-HPLC. B, DCs were incubated with tyrosinase peptide for 7 min, followed by cell removal and addition of angiotensin with and additional 7-min incubation. The amount of parent peptide was quantitated by RP-HPLC. C, The reverse experiment of B. D, DCs were modified with DASA, and their degradative activities were compared with those of native DCs. Peptide degradation was assessed by RP-HPLC. E, DCs were incubated in the presence of Tyr for 0 to 10 min, and the supernatant was quantitated for the amount of Tyr remaining by RP-HPLC.

FIGURE 4.

DC-peptide degradative activity is associated with ectoenzymatic activity. A, DCs were incubated in the presence of tyrosinase peptide for varying times, the cells were removed after 4 min in a parallel experiment, and the incubation was continued for an additional 56 min. The parent peptide remaining was quantitated by RP-HPLC. B, DCs were incubated with tyrosinase peptide for 7 min, followed by cell removal and addition of angiotensin with and additional 7-min incubation. The amount of parent peptide was quantitated by RP-HPLC. C, The reverse experiment of B. D, DCs were modified with DASA, and their degradative activities were compared with those of native DCs. Peptide degradation was assessed by RP-HPLC. E, DCs were incubated in the presence of Tyr for 0 to 10 min, and the supernatant was quantitated for the amount of Tyr remaining by RP-HPLC.

Close modal

To examine the question of peptide-stimulated release of the degradative enzymes, DCs were incubated in the presence of tyrosinase for 7 min, the cell-free supernatant (after microcentrifugation) was further incubated with another peptide, angiotensin (Fig. 4, B and C), and the amounts of the parent peptides were quantitated by HPLC. Only the peptide that was in contact with the DCs showed substantial amounts of degradation. These results indicate that the major degradative events are not stimulated by peptide-specific interactions that result in the release of enzymes into the extracellular medium. To further investigate whether these degradative events were truly due to ectoenzymes, the DC cell surface was modified with DASA, a nonpermeable membrane-modifying reagent that forms azo, diazoamino, S-azo, or thio ether derivatives with proteins and lipids without passing through the membrane (42, 44). The results shown in Figure 4 D indicate that 70 to 80% of the degradative activity could be abrogated by this treatment. Total inhibition could not be achieved due to a loss in cell viability upon prolonged incubation times (>20 min). The reaction time for DASA modification was kept at 20 min (at concentrations ranging from 0.3–3 mM) to assure viability of 98% or greater and to avoid the potential leaking of intracellular proteases into the extracellular medium.

Additional evidence for the membrane association of DC-degradative enzymes comes from isolated DC membranes that were tested for peptide-degradative activity. Table II compares the tyrosinase degradation products identified by tandem mass spectrometry for intact cells and isolated membranes. All the species identified using intact cells were identified with purified DC membranes. These results strongly support a plasma membrane association and an ecto orientation for the degradative events.

Table II.

Tyrosinase cleavage product detected with intact DCs and isolated DC plasma membranes

Fragment SequenceIntact DCsIsolated Membranes
+\E 
MDGTMS +\E 
DGTMSQV +\E 
YMDGTMS +\E 
MDGTMSQV 
Fragment SequenceIntact DCsIsolated Membranes
+\E 
MDGTMS +\E 
DGTMSQV +\E 
YMDGTMS +\E 
MDGTMSQV 
a

Isolated plasma membranes were prepared as described in Materials and Methods. Tyrosinase peptide was incubated in the presence of purified membranes or intact DCs. Supernatants were fractionated by RP-HPLC, and cleavage products were assessed by CID tandem mass spectrometry.

One other possibility was uptake of the peptide by the DC itself, since DCs are known to be highly active in macropinocytotic activity (45). To determine whether the majority of loss was due to uptake, DCs were incubated with an equivalent amount of tyrosine (an end product of DC/tyrosinase degradation) for varying amounts of time, and the amount of tyrosine in the supernatant was quantitated by HPLC (Fig. 4 E). No loss was detected, suggesting that uptake during the assay (<15 min) was not responsible for the observed loss. If uptake was required, a lag period would have been noted before the detection of products. The literature estimates that DCs can accumulate solute molecules at a rate of 2400 fl/h (45, 46). This would translate into a loss of <1.2% from the DC/peptide mixture, which would fall within the error associated with analytical HPLC and could not account for the loss of parent peptide observed in these studies.

To determine the class of enzyme(s) responsible for the degradation of tyrosinase, cells were incubated with a variety of protease inhibitors (Table III) before the addition of peptide. Only four reagents were capable of affecting DC ectoenzyme cleavage; ubiquitin, EDTA, o-phenanthroline, and DTT. All other inhibitors of serine, aspartic, metallo-, and thiol proteases were without effect. Inhibition of the N-terminal degradative activity could be completely inhibited by ubiquitin, a specific competitive aminopeptidase inhibitor, EDTA, and o-phenanthroline (all >95% inhibition). CD13, aminopeptidase N, is known to exist on a variety of cell types, including DCs and may play a role in migration and Ag presentation processes (41, 47, 48). These data suggest that CD13 or CD13-like enzymes are responsible for the N-terminal degradative activity. The remaining major cleavage product of the tyrosinase peptide is the 7-mer YMDGTMS. The only reagent capable of inhibiting this activity was DTT (83% inhibition). This suggests that this type of enzymatic activity requires intact disulfide bonds for proper function. Studies are ongoing to determine the precise nature of this enzyme. We also investigated the possibility that tyrosinase degradation could be altered by the addition of protein in the form of either serum (5%) or β2m (10 μg/ml), conditions often used during peptide pulsing. Neither manipulation significantly altered tyrosinase peptide degradation by DCs (not shown).

Table III.

Inhibition of fragments by various protease inhibitorsa

InhibitorMDGTMSDGTMSQVYMDGTMSMDGTMSQVYType\E
STI − − − − − S\E 
PMSF − − − − − S\E 
Pepstatin − − − − − A\E 
EDTA Y only − M\E 
Phophoram − − − − − M\E 
Aprotinin − − − − − S\E 
o-Phen Y only − M\E 
Ubiquitin Y only − M, SP\E 
5,5′-dithio − − − − − T\E 
NEM − − − − − T\E 
Iodoacet − − − − − T\E 
DTT QV only − − − T\E 
Leupeptin − − − − − S, T\E 
Benzamidine − − − − − S\E 
TLCK − − − − − 
InhibitorMDGTMSDGTMSQVYMDGTMSMDGTMSQVYType\E
STI − − − − − S\E 
PMSF − − − − − S\E 
Pepstatin − − − − − A\E 
EDTA Y only − M\E 
Phophoram − − − − − M\E 
Aprotinin − − − − − S\E 
o-Phen Y only − M\E 
Ubiquitin Y only − M, SP\E 
5,5′-dithio − − − − − T\E 
NEM − − − − − T\E 
Iodoacet − − − − − T\E 
DTT QV only − − − T\E 
Leupeptin − − − − − S, T\E 
Benzamidine − − − − − S\E 
TLCK − − − − − 
a

STI, soybean trypsin inhibitor; PMSF, phenylmethylsulfonyl fluoride; EDTA, ethylenediamine tetra acetic acid; o-phen, 1,10 phenanthroline; 5,5′-dithio, 5,5′-dithiobis-2-nitrobenzoic acid; NEM, N-ethylmaleimide; iodoacet, iodoacetic acid; DTT, dithiothreitol; TLCK, N-tosyl-l-lysine chloromethyl ketone; TPCK, N-tosyl-l-phenylalanine chloromethyl ketone. Types of protease inhibitors: S, serine; A, aspartic; M, metallo; SP, specific competitive inhibitor of aminopeptidases; T, thiol compounds/reagents for thiol/cysteine proteases. DCs were incubated in the absence or presence of the above protease inhibitors at concentrations within the μM to mM range and were assessed for their ability to degrade tyrosinase peptide as described in Materials and Methods.

DCs were incubated with two other class I-associated synthetic peptides, AAGIGILTV and YLEPGPVTA, as well as with an unrelated 10-mer peptide, angiotensin. Incubation was allowed to proceed for 7 min, the supernatants were harvested, the products were separated by HPLC, and the cleavage sites were mapped and sequenced using collision-induced dissociation tandem mass spectrometry. The cleavage sites are shown in Figure 5. Both exo and endo cleavage sites are present with the majority of the cleavage events taking place with tyrosinase and angiotensin.

FIGURE 5.

Cleavage specificities of DC and monocyte populations. Monocytes and DCs were incubated with a variety of class I synthetic peptides or with angiotensin for 10 min, and the supernatants were harvested and fractionated by RP-HPLC and cleavage fragments sequenced by CID tandem mass spectrometry.

FIGURE 5.

Cleavage specificities of DC and monocyte populations. Monocytes and DCs were incubated with a variety of class I synthetic peptides or with angiotensin for 10 min, and the supernatants were harvested and fractionated by RP-HPLC and cleavage fragments sequenced by CID tandem mass spectrometry.

Close modal

We compared the rates of tyrosinase degradation by DC ectoenzymes with a variety of cell populations, as shown in Figure 6. The nonadherent cell population isolated after the adherence step in the myeloid DC preparation yielded no degradation upon incubation with the tyrosinase peptide during the 15-min assay period. Tyrosinase degradation was detected within the adherent peripheral blood monocyte population, which is used as the starting cell type for myeloid DC preparation. The rate of tyrosinase degradation was 1.0 nmol/min/106 cells, with only limited specificities on the three class I peptides and no activity against angiotensin (Figs. 5 and 6). Upon culture in AIM V medium for 7 days, an increased rate of tyrosinase degradation was observed (2.9 nmol/min/106 cells), but with broader specificity toward tyrosinase and angiotensin substrates. Day 7 DCs cultured in GM-CSF and IL-4 displayed the highest tyrosinase degradation rate (4.3 nmol/min/106) and the broadest peptide specificity. These results indicate that upon maturation in GM-CSF and IL-4, the myeloid DCs not only express an increased amount of their baseline enzyme activities, but additional enzyme class specificities begin to emerge.

FIGURE 6.

Comparative rates of ectoenzymatic activities of DCs and monocytes. Cells from the same individual were processed for monocyte or DC culture. After varying times, the respective cell types were processed for a time-course degradation assay with tyrosinase peptide. The amount of peptide remaining was quantitated by RP-HPLC.

FIGURE 6.

Comparative rates of ectoenzymatic activities of DCs and monocytes. Cells from the same individual were processed for monocyte or DC culture. After varying times, the respective cell types were processed for a time-course degradation assay with tyrosinase peptide. The amount of peptide remaining was quantitated by RP-HPLC.

Close modal

Human DCs (5 × 106) were incubated in the absence and the presence of ubiquitin and DTT (1.0 and 0.1 mM, respectively) for 30 min at 37°C before the addition of tyrosinase peptide. After addition of peptide, incubation was allowed to proceed for an additional 30 min. DCs were then removed by centrifugation, and supernatants containing peptide (0.1 μM) were pulsed onto 51Cr-labeled A2/Kb target cells in the presence of β2m (10 μg/ml). A tyrosinase-specific effector cell line (CAM) was used as the effector cell, and the percent specific lysis was determined after 4 h at 37°C. Figure 7 demonstrates that the tyrosinase-specific effector cells recognized only the targets that had been pulsed with inhibitor-treated DC-peptide supernatants. In conjunction with the above data, we conclude that the lack of recognition of targets pulsed with DC-peptide supernatants is due to efficient degradation by DC ectoenzymes.

FIGURE 7.

Addition of protease inhibitors to DC cultures provides protection from tyrosinase degradation and allows for efficient loading onto target cells and recognition by tyrosinase-specific CTLs. DCs were incubated in the presence or the absence of ubiquitin and DTT (1.0 and 0.1 mM, respectively) before the addition of tyrosinase peptide. Tyrosinase peptide was added and incubated for an additional 30 min. DCs were removed, and supernatants were pulsed onto chromium-labeled A2/Kb target cells. Tyrosinase-specific CTLs were added (E:T cell ratio, 5:1), and targets were assayed for percent specific lysis after 4 h at 37°C. −P, no peptide, DC supernatant only; +P, peptide only, no incubation with DCs; DC+P, DC plus peptide, supernatant; DC+P+I, DC, peptide, and inhibitors, supernatant; DC+I, DC plus inhibitors, supernatant.

FIGURE 7.

Addition of protease inhibitors to DC cultures provides protection from tyrosinase degradation and allows for efficient loading onto target cells and recognition by tyrosinase-specific CTLs. DCs were incubated in the presence or the absence of ubiquitin and DTT (1.0 and 0.1 mM, respectively) before the addition of tyrosinase peptide. Tyrosinase peptide was added and incubated for an additional 30 min. DCs were removed, and supernatants were pulsed onto chromium-labeled A2/Kb target cells. Tyrosinase-specific CTLs were added (E:T cell ratio, 5:1), and targets were assayed for percent specific lysis after 4 h at 37°C. −P, no peptide, DC supernatant only; +P, peptide only, no incubation with DCs; DC+P, DC plus peptide, supernatant; DC+P+I, DC, peptide, and inhibitors, supernatant; DC+I, DC plus inhibitors, supernatant.

Close modal

The data presented in this study provide evidence for the expression of cellular ectoenzymes on the surface of the DC. The enzymatic activity is rapid and aggressive toward class I peptides, and substantial degradation can be detected within minutes. Using collision-induced dissociation and tandem mass spectrometry, cleavage products were directly examined and sequenced, allowing us to map the susceptible sites within the peptide sequence. This enzymatic activity was localized to the plasma membrane, was shown to have an ecto orientation, and was not secreted upon peptide-specific interaction.

Sequence analysis by tandem mass spectrometry indicated that the tyrosinase peptide suffers substantial loss at the N-terminus, suggesting possible aminopeptidase activity. Inhibition studies indicated that the N-terminal degradative events could be abrogated by treatment with EDTA, ortho-phenanthroline, and ubiquitin (with >95% inhibition for N-terminal degradation for these three inhibitors). The two former reagents support the presence of a metalloprotease, while the latter inhibitor (a specific competitive inhibitor of aminopeptidases) supports the presence of the CD13 metalloprotease (aminopeptidase N) or CD13-like activity. CD13 is expressed by a number of cell types, including monocytes and dendritic cells and has been suggested to have important physiologic significance (20, 21, 30, 41, 42, 43). We have shown in this study that the overall cell surface enzymatic activity of the DC is increased four- to fivefold upon incubation in culture in the presence of GM-CSF/IL-4 compared with the starting peripheral blood monocyte population. While the enzymatic activities of the day 7 monocytes and the day 7 DCs are similar in substrate specificities, the day 7 DCs exhibit the highest enzymatic rate as well as the appearance of a C-terminal exoproteolytic event that is capable of removing the C-terminal residue from substrate angiotensin. Both the DC population and the day 7 monocyte population exhibit not only increased degradative activities toward neutral and hydrophobic N-terminals, but also the emergence of degradative activities toward acidic N-terminals compared with the starting monocyte population. This suggests the emergence of aminopeptidases of altered specificities (aminopeptidase A). In addition to the alterations in N-terminal degradative activities and specificities, new endoproteolytic activities emerge upon culture. This is clearly seen when angiotensin is used as the substrate.

One of the major cleavage products of the tyrosinase peptide is the hexapeptide YMDGTMS. The possibility exists that cleavage was produced by a carboxypeptidase-like activity. However, typical carboxypeptidases (A, B, P, and Y) are responsive to inhibition with metal chelators, sulfhydryl modifying reagents, or serine protease inhibitors. All these inhibitors were without effect. The only reagent that provided near complete inhibition was DTT. This suggests that the formation of disulfide bonds is critical for this enzymatic activity. This resistance to typical inhibitors is suggestive of a cathepsin-like enzyme, since human, bovine, and rabbit cathepsin D enzymes demonstrate DTT sensitivity (49). However, typical carboxyl proteases such as cathepsin D usually show sensitivity toward pepstatin, which was not seen in these studies. Additional work is in progress to determine the identity of this enzyme.

Since DCs are now regarded as the APC of choice and since they possess a repertoire of highly active enzymes on their cell surface, protein Ag loading becomes a critical issue, especially when low levels of previously processed peptides are being pulsed onto an APC. This problem can be circumvented in two ways. With respect to synthetic peptides, one can effectively overload the system such that sufficient peptide remains in an intact state for proper loading. However, the critical concentration may be different for each peptide, since degradation is enzyme specific and therefore sequence dependent. It is suggested that with synthetic peptides, degradative events should be investigated and controlled for. N- and C-terminal modifications of synthetic peptides can be performed to avoid unwanted degradation; however, once modified, the peptides will lose their ability to bind to class I molecules. One potential modification that will retain the peptide’s natural structure would be the substitution of d-amino acids at critical positions (i.e., at the N-terminus), since aminopeptidase activity is greatly reduced toward d-substituted substrates. However, such modifications may or may not be tolerated at the level of the class I molecule.

A more critical problem comes into play when low levels of eluted natural peptides and peptides isolated from tumor cell class I molecules are used for pulsing. These peptides are often in the femtomole to near picomole range, and efficient loading of these peptides will be dependent on several factors. First, differences in degradative rates of peptides by ectoenzymes may be attributable to species differences. Second, since some of these ectoenzymes have a somewhat ubiquitous nature, the cell type becomes a critical parameter (i.e., those that are highly active in migration and/or Ag processing displaying the broadest and most rapid degradative activities). Third, the state of cellular maturation and activation (including the effects of cytokines) may influence the degree and specificity of ectoenzymatic activity. Finally, degradation will be sequence dependent. The latter statement suggests that in instances where peptides are presented to the APC as extremely dilute mixtures in small volumes used for pulsing, only those peptides with the most resistant sequences will survive. This will allow for the APC system to become biased toward certain peptides and the T cells they elicit. It is suggested that for these conditions, loading in the presence of protease/peptidase inhibitors may be advantageous.

In this study near complete inhibition of tyrosinase degradation could be attained in the presence of DTT and ubiquitin. Since DTT is in the family of thiol compounds, similar results should be obtained with 2-ME. The latter compound has been used as a typical additive to cell culture media and therefore presents itself as a nontoxic additive during the loading process, although it remains to be seen whether the concentration used in cell culture is sufficient for inhibition of tyrosinase degradation. The remaining compound critical for the inhibition of tyrosinase degradation is ubiquitin. This competitive inhibitor is also nontoxic and has been used in vitro and in vivo for over a decade. The addition of these two inhibitors to DC cultures inhibited tryrosinase peptide degradation and allowed efficient pulsing onto target cells and recognition by tyrosinase-specific CTL. Predicting the outcome of peptide loading will ultimately depend on the affinity of peptide for class I molecules vs the rate of peptide degradation.

From a biologic point of view, the role of ectoenzymes on the surface of the DC may have important physiologic significance. First, upon cellular maturation in the presence of cytokines, an enhanced cell surface enzyme repertoire with increased activity may be necessary for the increased ability to inactivate soluble or inflammatory mediators (50). Second, this enhanced ectoenzymatic activity could enhance DC migration, since CD13 activity has been shown to play a role in adhesion and tumor cell infiltration (48). Third, the DC ectoenzymatic activity could function in extracellular Ag processing, since CD13 has already been implicated in the trimming of class II molecules (41). Finally, since CD13 has been described as the receptor for coronaviruses, and CD13-like surface structures have been shown to be involved in CMV infection (51, 52, 53), similar DC surface enzymes may serve as portals necessary for viral entry.

We thank Drs. C. Mosse, T. Bullock, and V. Engelhard for supplying the tyrosinase-specific CAM cell line and for their helpful suggestions for its maintenance, and Dr. W. J. Storkus for the target cells.

1

This work was supported in part by funds awarded by the Comprehensive Research Fund of the University of Pittsburgh Medical Center and National Institutes of Health Grant RO11RO1CA73816-01.

3

Abbreviations used in this paper: DC, dendritic cell; GM-CSF, granulocyte-macrophage CSF; DASA, diazotized sulfanilic acid.

1
Romanin, N., E. Kampgen, F. Koch, C. Heufler, G. Schuler.
1990
. Dendritic cell production of cytokines and responses to cytokines.
Int. Rev. Immunol.
6
:
151
2
Sallusto, F., A. Lanzavecchia.
1994
. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by GM-CSF plus IL-4 and downregulated by TNFα.
J. Exp. Med.
179
:
1109
3
Caux, C., C. Massacrier, B. Vanbervliet, B. Dubois, C. Van Kooten, I. Durand, J. Banchereau.
1994
. Activation of human dendritic cells through CD40 cross-linking.
J. Exp. Med.
180
:
1263
4
Roman, N., S. Gruner, D. Brang, E. Kampgen, A. Lenz, B. Konwlinka, G. Fritsch, R. M. Steinman, G. Schuler.
1994
. Proliferating dendritic cell progenitors in blood.
J. Exp. Med.
180
:
83
5
Becker, Y..
1993
. Dendritic cell activity against primary tumors: and overview.
In Vivo
7
:
187
6
Zeid, N. A., H. K. Muller.
1993
. S100 positive dendritic cells in human lung tumors associated with cell differentiation and enhanced survival.
Pathology
25
:
338
7
Furihata, M., Y. Ohtsuki, E. Ido, J. Iwata, H. Sonobe, K. Araki, S. Ogoshi, K. Ohmori.
1992
. HLA-DR antigen and S-100 protein-positive dendritic cells in esophageal squamous cell carcinoma: their distribution in relation to prognosis.
Virchows Arch. B
61
:
409
8
Tsujitani, S., Y. Kakeji, A. Watanabe, S. Kohnoe, Y. Maehara, K. Sugimachi.
1990
. Infiltration of dendritic cells in relationship to tumor invasion and lymph node metastasis in human gastric cancer.
Cancer
66
:
2012
9
Giannini, A., S. Bianchi, L. Messerini, O. Gallo, E. Gallina, G. Asprella-Libonati, P. Olmi, G. Zampi.
1991
. Prognostic significance of accessory cells and lymphocytesin nasopharyngeal carcinoma.
Pathol. Res. Pract.
187
:
498
10
Murphy, G., F. Radu, M. Kaminer, D. Berd.
1993
. Autologous melanoma vaccine induces inflammatory responses in melanoma metastases: relevance to immunologic regression and immunotherapy.
J. Invest. Dermatol.
100
:
335S
11
Knight, S. C., R. Hunt, C. Dore, P. B. Medawar.
1985
. Influence of dendritic cells on tumor growth.
Proc. Natl. Acad. Sci. USA
82
:
4495
12
Flamand, V., T. Sornasse, K. Thielemans, C. Demanet, M. Bakkus, H. Bazin, P. Thielemans, O. Leo, J. Urbain, M. Moser.
1994
. Murine dendritic cells pulsed with tumor antigen induce tumor resistance in vivo.
Eur. J. Immunol.
24
:
605
13
Austen, J. M..
1993
. The dendritic cell system and anti-tumor immunity.
In Vivo
7
:
193
14
Inaba, K..
1992
. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with GM-CSF.
J. Exp. Med.
176
:
1693
15
Mayordomo, J. I., T. Zorina, W. J. Storkus, L. Zivogel, C. Celluzzi, L. D. Falo, C. J. Melief, S. T. Ildstad, W. M. Kast, A. DeLeo, et al
1995
. Bone marrow derived dendritic cells pulsed with synthetic tumor peptides elicit protective and therapeutic anti-tumor immunity.
Nat. Med.
1
:
1297
16
Celluzzi, C. M., J. I. Mayordomo, W. J. Storkus, M. T. Lotze, L. D. Falo.
1996
. Peptide-pulsed dendritic cells induce antigen-specific, CTL-mediated protective immunity.
J. Exp. Med.
283
:
88
17
Zitvogel, L., J. I. Mayordomo, T. Tjandrawan, A. B. DeLeo, M. R. Clarke, M. T. Lotze.
1996
. Therapy for murine tumors with tumor peptide-pulsed dendritic cells: dependence on T cell B7 costimulation and T helper cell-associated cytokines.
J. Exp. Med.
183
:
87
18
Kiyokanan, K., K. Ishimoto, H. Fujita, J. Fujimoto, J.-I. Hata.
1990
. Characterization of the common acute lymphoblastic leukemia antigen (CD10) as an activation molecule on human B cells.
Clin. Exp. Immunol.
79
:
322
19
Bauvois, B..
1990
. Murine thymocytes possess specific cell-surface associated exoaminopeptidase activities: preferential expression by immature CD4 CD8 subpopulations.
Eur. J. Immunol.
20
:
459
20
Look, A. T., R. A. Ashmun, L. H. Shapiro, S. C. Peiper.
1989
. Human myeloid plasma membrane glycoprotein CD13 is identical to aminopeptidase N.
J. Clin. Invest.
83
:
1299
21
Ashmun, R. A., A. T. Look.
1990
. Metalloprotease activity of CD13 aminopeptidase N on the surface of myeloid cells.
Blood
75
:
462
22
Naquet, P., I. Vivier, J.-P. Gorvel, P. Brekemans, M. Barad, A.-M. Bernard, M. Pierres.
1989
. Activation of mouse T-lymphocytes by a monoclonal antibody to a developmentally regulated surface aminopeptidase (THAM).
Immunol. Rev.
11
:
77
23
Wu, Q., J. M. Lahti, P. D. Burrows, M. D. Cooper.
1990
. Molecular cloning of the murine BP-1/6C3 antigen: a member of the zinc-dependent metalloprotease family.
Proc. Natl. Acad. Sci. USA
87
:
993
24
Wu, A., L. Li, M. D. Cooper, M. Pierres, J. P. Gorvel.
1991
. Aminopeptidase A activity of the murine B-lymphocyte differentiation antigen BP-1/6C3.
Proc. Natl. Acad. Sci. USA
88
:
676
25
Halfer, D. A., M. Chofflon, D. Benjamin, N. H. Dang, J. B. Breitmeyer.
1989
. Mechanisms of immune memory T cell activation and CD3 phosphorylation correlates with Ta1 (Cdw26) expression.
J. Immunol.
142
:
2590
26
Dang, N. H., D. A. Halfer, S. F. Schlossman, J. B. Brietmeyer.
1990
. FcR-mediated crosslinking of Ta1 (Cdw26) induces human T lymphocyte activation.
Cell. Immunol.
125
:
42
27
Stamenkovic, I., H. C. Asheim, A. Deggerdac, H. K. Blomhoff, E. B. Smeland, S. Funderud.
1990
. The B-cell antigen CD75 is a cell surface sialyltransferase.
J. Exp. Med.
172
:
641
28
Dang, N. H., Y. Torimoto, K. Deusch, S. F. Schlossman, C. Morimoto.
1990
. Comitogenic effect of solid-phase immobilized anti-1F7 on human CD4 T cell activation via CD3 and CD2 pathways.
J. Immunol.
144
:
4092
29
Hegen, M., G. Nieddobitek, C. E. Klein, H. Stein, B. Fleischer.
1990
. The T-cell triggering molecule Tp103 is associated with dipeptidyl aminopeptidase IV activity.
J. Immunol.
144
:
2908
30
O’Connell, P. J., V. Gerkis, A. J. F. D’Apice.
1989
. CD13: a signal transduction molecule on myeloid cells?.
Transplant. Proc.
21
:
3826
31
Thompson, L. F., J. M. Ruedi, A. Glass, M. G. Low, A. H. Lucas.
1989
. Antibodies to 5′-nucleotidase (CD73), a glycosyl-phophatidylinositol-anchored protein, cause human peripheral blood T cells to proliferate.
J. Immunol.
140
:
1815
32
Dang, N. H., Y. Torimoto, K. Sugita, J. F. Daley, P. Schow, C. Prado, S. F. Schlossman, C. Morimoto.
1990
. Cell surface modulation of CD26 by 1F7 monoclonal antibody: analysis of surface expression and human T-cell activation.
J. Immunol.
145
:
3963
33
G. A., Piazza, H. M. Callanan, J. Mowery, D. Hixson.
1989
. Evidence for a type of dipeptidyl peptidase IV in fibronectin-mediated interactions of hepatocytes with extracellular matrix.
Biochem. J.
262
:
327
34
Dang, N. H., Y. Torimoto, S. F. Schlossman, C. Morimoto.
1990
. Human CD4 helper T cell activation: functional involvement of two distinct collagen receptors, 1F7 and VCA integrin family.
J. Exp. Med.
172
:
649
35
Aurivillius, M., O. C. Hansen, M. B. S. Lazrek, E. Bock, B. Orbink.
1990
. The cell adhesion molecule cell-CAM 105 is an ecto-ATPase and a member of the immunoglobulin family.
FEBS Lett.
264
:
267
36
Daikiri, K., S. Nakamura, S. Ikegami, M. Nakamura, T. Fujimori, N. Tamaoki, N. Tada.
1989
. Mouse Ly-31. 1 is an alloantigenic determinant D alkaline phosphatase predominantly expressed in the kidney and bone.
Immunogenetics
29
:
235
37
Marquez, C., A. De La Hera, E. Leonardo, L. Pezzi, A. Stasser, A. C. Martinez.
1990
. Identity of PB76 differentiation antigen and lymphocyte alkaline phosphatase.
Immunogenetics
20
:
947
38
Tiruppathi, C., Y. Miyamoto, V. Ganapathy, T. A. Roesel, G. M. Whitford, F. H. Leibach.
1990
. Hydrolysis and transport of proline-containing peptides in renal brush-border membrane vesicles from dipeptidyl peptidase IV-positive and negative rat strains.
J. Biol. Chem.
265
:
1476
39
Mouritsen, S., M. Meldal, O. Werdelin, A. Stryhn Hansen, S. Buus.
1992
. MHC molecules protect T cell epitopes against proteolytic destruction.
J. Immunol.
149
:
1987
40
Falk, K., O. Roetzschke, S. Stefanovic, G. Jung, H. G. Rammensee.
1994
. Pool sequencing of natural HLA-DR, -DQ and -DP ligands reveals detailed peptide motifs, constraints of processing and general rules.
Immunogenetics
39
:
230
41
Larsen, S., L. Ostergaard, S. Buus, A. Stryhn.
1996
. T cell responses affected by aminopeptidase N (CD13)-mediated trimming of major histocompatability complex class II-bound peptides.
J. Exp. Med.
184
:
183
42
Amoscato, A. A., J. W. Alexander, G. F. Babcock.
1989
. Surface aminopeptidase activity of human lymphocytes. I. Biochemcial and biologic properties.
J. Immunol.
142
:
1245
43
Amoscato, A. A., R. R. Spiess, A. M. Brumfield, R. B. Herberman, W. H. Chambers.
1993
. Surface aminopeptidase activity of rat NK cells.
Biochim. Biophys. Acta
1221
:
221
44
Berg, H. C..
1969
. Sulfanilic acid diazonium salt: a label for the outside of the human erythrocyte membrane.
Biochim. Biophys. Acta
183
:
65
45
Sallusto, F., M. Cella, C. Danieli, A. Lansavecchia.
1995
. Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatability complex class I compartment: downregulation by cytokines and bacterial products.
J. Exp. Med.
182
:
389
46
Steinman, R. M., J. Swanson.
1995
. The endocytic activity of dendritic cells.
J. Exp. Med.
182
:
283
47
Menrad, A., D. Speicher, J. Wacker, M. Herlyn.
1993
. Biochemical and functional characterization of aminopeptidase N expressed by human melanoma cells.
Cancer Res.
53
:
1450
48
Saiki, I., H. Fujii, J. Yoneda, F. Abe, M. Nakajima, T. Tsuruo, I. Azuma.
1993
. Role of aminopeptidase N (CD13) in tumor-cell invasion and extracellular matrix degradation.
Int. J. Cancer
54
:
137
49
Barrett, A. J..
1977
. Cathepsin D and other carboxyl proteinases. A. J. Barrett, ed.
Proteinases in Mammalian Cells and Tissues
209
-248. North-Holland Publishing Co., Amsterdam.
50
Kanayama, N., Y. Kajiwara, J. Goto, E. El Maradny, K. Maehara, K. Andou, T. Tereo.
1995
. Inactivation of interleukin-8 by aminopeptidase N (CD13).
J. Leukocyte Biol.
57
:
129
51
Delmas, B., J. Gelfi, R. L’Haridou, L. K. Vogel, H. Sjostrom, O. Noren, H. Lunde.
1992
. Aminopeptidase N is a major receptor for the enteropathogenic corona virus TGEV.
Nature
357
:
417
52
Yeager, C. L., R. A. Ashmun, R. K. Williams, C. B. Cardellichio, L. H. Shapiro, A. T. Look, K. V. Holmes.
1992
. Human aminopeptidase N is a major receptor for human corona virus 229E.
Nature
357
:
420
53
Giugni, T. D., C. Soderberg, D. J. Hamm, R. M. Bautista, K. O. Hedlund, E. Moller, J. A. Zaia.
1996
. Neutralization of human cytomegalovirus by human CD13-specific antibodies.
J. Infect. Dis.
173
:
106