Fascin Is Involved in the Antigen Presentation Activity ofMature Dendritic Cells1
Monther M. Al-Alwan,* Geoffrey Rowden, †‡ Timothy D. G. Lee,*§ and Kenneth A. West2‡*†
Maturation of dendritic cells (DC) is critical to their development into potent APCs. Upon maturation, DC up-regulate theexpression of MHC class II as well as costimulatory and adhesion molecules, all of which are important in Ag presentation. Inaddition, they undergo structural changes characterized by the expression of numerous long dendrites. Fascin is an actin-bundlingprotein that has been reported to be important for the development of dendrites. In this study, we evaluated fascin expression andfunction during DC maturation into potent APC. In vitro, treatment of bone marrow-derived DC (BM-DC) with GM-CSF resultedin increased levels of fascin expression. This increase correlated directly with an increase in MHC class II and B7-2 expression.Fascin expression was decreased by the addition of TGF-b and increased by the addition TNF-a to the culture. These cytokinessuppress or enhance DC maturation, respectively. Increased levels of fascin expression were found to correlate with increased APCactivity in a one-way MLR. Specific inhibition of fascin expression, using antisense oligonucleotides, markedly reduced this APCallostimulatory activity. These data demonstrate that fascin expression correlates with DC maturation into APC, and it plays asignificant role in the ability of DC to function as APC. This observation is the first evidence linking fascin-mediated dendriteformation with the APC activity of DC. The Journal of Immunology,2001, 166: 338–3450.
D endritic cells (DC)3 are now recognized as the most po-tent professional APC involved in initiating primary im-mune responses (1). Numerous studies have confirmed
that mature interdigitating dendritic cells (IDC) are the primary, ifnot the only, APC involved in the activation of naive Th cells (2,3). The importance of DC maturation for efficient T cell activationhas been extensively discussed in several previous reports (1, 4, 5).Immature DC that reside in the peripheral tissues (areas of high Agencounter) are equipped to capture and process Ag (6, 7). In re-sponse to activation signals, such as inflammatory cytokines, theseimmature DC undergo maturation, where they develop into potentT cell stimulators by up-regulating MHC class II as well as co-stimulatory and adhesion molecules. At the same time these DCmigrate into secondary lymphoid organs to stimulate Ag-specific Tcells (8–11). In the secondary lymphoid organs, DC express highlevels of the C-C chemokine, DC-CK1, which has been shown topreferentially attract naive (CD45RA1) T cells (12). In addition,DC maturation is associated with morphological changes, resultingin the expression of numerous delicate dendrites (13, 14). Thisdendrite formation has been associated, in other cells, with theexpression of fascin.
Fascin is a 55-kDa, actin-bundling protein that regulates therearrangement of cytoskeletal elements, and the interaction be-tween the cytoskeleton and the cell membrane in response to extra-
or intracellular signals (15). There is substantial evidence to linkfascin expression with dendrite formation. High levels of fascinexpression were found to be critical for dendrite development inneurons (16, 17). Recently, fascin expression was found to partic-ipate in the development of dendrites in mouse epidermal Lang-erhans cells (LC) (18). Moreover, when fascin was transfected intoa pig epithelial cell line, the cells became dendritic and their mo-tility was markedly increased (19).
Of interest is the fact that DC are the only leukocytes that havebeen demonstrated to express fascin (20). In the lymphoid organsonly IDC, which are involved in APC activity, show high levels offascin expression. This suggests a direct link between fascin ex-pression and the ability of DC to function as potent APC.
In this study, we evaluated fascin expression and function dur-ing DC maturation. The data presented here demonstrate that fas-cin expression is tightly regulated during maturation of DC. Theyalso demonstrate that fascin expression is critical for the Ag pre-sentation activity of mature DC.
Materials and MethodsAnimals and culture medium
Adult BALB/c and C57BL/6 mice were purchased from Charles RiverBreeding Laboratories (St. Constant, Quebec, Canada) and housed in theCarleton Animal Care Facility (Sir Charles Tupper Medical Building, Dal-housie University, Halifax, Canada). All animals were housed in compli-ance with the guidelines established by the Canadian Council on AnimalCare and were given standard rodent chow and water ad libitum. Themedium used for bone marrow-derived DC (BM-DC) culture was RPMI1640 (Sigma-Aldrich, Oakville, Ontario, Canada) supplemented with 5%heat-inactivated (30 min, 65°C) FCS (Life Technologies, Grand IslandNY), 100 U/ml penicillin, 100mg/ml streptomycin, and 5 mM 2-ME (Brit-ish Drug House, Toronto, Ontario, Canada). In this report, we referred tothis medium as complete RPMI (cRPMI).
Antibodies
A panel of mAbs was used in the immunocytochemistry and flow cytom-etry studies. All Abs were titered, and the optimal dilution was used. Theanti-fascin mAb (mouse IgG1) was a gift of Dr. Erik Langhoff, Pennsyl-vania State University (Hershey, PA). Anti-Ia (mouse IgG1), anti-Thy1.2(CD90; mouse IgG2b), anti-CD11c (N418; hamster IgG), purified mouse
Departments of *Microbiology and Immunology,†Pathology,‡Medicine, and§Sur-gery, Dalhousie University, Halifax, Nova Scotia, Canada
Received for publication July 6, 1999. Accepted for publication October 5, 2000.
The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by the Kidney Foundation of Canada, the Canadian Der-matology Foundation, and the Government of Saudi Arabia.2 Address correspondence and reprint requests to Dr. Kenneth A. West, Departmentof Medicine, Suite 5087, Dickson Building, 5820 University Avenue, Halifax, NovaScotia, B3H 2Y9 Canada. E-mail address: [email protected] Abbreviations used in this paper: DC, dendritic cell(s); BM-DC, bone marrow-derived DC; IDC, interdigitating DC; LC, Langerhans cell(s); cRPMI, completeRPMI; rm, recombinant mouse.
IgG1, PE-conjugated rat IgG-2a, FITC-labeled streptavidin, FITC- and PE-labeled rabbit anti-rat IgG2a, and PE-conjugated goat anti-hamster (IgG)Abs were all purchased from Cedarlane Laboratories (Hornby, Ontario,Canada). PE-conjugated anti-I-Ad/I-Ed mAb (rat IgG2a) was purchasedfrom PharMingen (San Diego, CA). Alexa 488 (green) goat anti-mouseIgG conjugate was purchased from Molecular Probes (Eugene, OR). Thehybridoma, GL1, producing the anti-B7-2 (rat IgG2a) mAb was purchasedfrom the American Type Culture Collection (ATCC; no. HB-253;Manassas, VA).
Epidermal skin sheet preparation
Epidermal skin sheets were prepared essentially as previously described byBaker et. al (21). Briefly, a mouse ear was dissected and rubbed with theback of a pair of forceps to separate the two halves. The separated skin wasthen cut into 3- to 4-mm2 squares and incubated in 3.8% NH4SCN (Sigma-Aldrich Canada) for 30 min at 37°C. The skin was then washed in PBS, andthe epidermis was separated from the dermis under a dissecting micro-scope. The epidermis was then fixed in cold (220°C) acetone (BDH) for5 min and stored in PBS at 4°C until stained. All staining of the epidermalskin sheet was performed in 1.5-ml microfuge vials (O.H. Johns, Missis-sauga, Ontario, Canada). The sheets were then mounted on slides, andpositive cells were counted under320 magnification power. The data ispresented as the number of positive cells per mm2 of the skin sheet.
Preparation of DC from bone marrow
DC were prepared from BALB/c bone marrow essentially as previouslydescribed (22, 23). Briefly, marrow was flushed out of mouse femurs andtibias using sterile cRPMI in a laminar flow hood. The marrow suspensionwas then passed through nylon mesh to remove bone marrow particles. Thecells were washed, and RBC were lysed with lysing buffer, containing 155mM NH4Cl in 10 mM Tris-HCl buffer (Sigma-Aldrich Canada) for 2 minat room temperature. FcR-positive cells were depleted by incubation (at37°C for 1 h) on a Petri dish coated with IgG. After washing, lymphocytesand Ia-positive cells were removed by incubation with a mixture of mAbsfor 60 min at 4°C followed by Low-Tox rabbit complement (CedarlaneLaboratories) for 60 min at 37°C. The mAbs used for depletion wereGK1.5 (anti-CD4), HO2.2 (anti-CD8), B21-2 (anti-Ia), and RA3-3A1/6.1(anti-B220/CD45R) (TIB 207, 150, 229, and 146, respectively; ATCC).Cells (53 106) were cultured in 50-ml flasks (Nunc, Naperville, IL) in 5ml cRPMI supplemented with either 50 U/ml recombinant mouse GM-CSF(rmGM-CSF; Cedarlane Laboratories)6 200 U/ml rmTNF-a (Life Tech-nologies), or 50 U/ml rmGM-CSF1 0.5 ng/ml TGF-b (CalBiochem-No-vaBiochem, La Jolla, CA). The cultures were fed every 3 days by aspirat-ing 85% of the medium and adding back fresh medium with growth factors.
Immunostaining
Immunostaining was performed on lymph node histological sections, epi-dermal skin sheets, or BM-DC cytospins. For cytospin preparation, 73 104
cells were cytocentrifuged onto poly-L-lysine (Sigma)-coated slides. Cellswere then fixed in cold (220°C) acetone for 2 min and stored at220°Cuntil use. For fascin staining, where mouse anti-fascin mAb was used,slides were fixed in 10% acetate-buffered formalin and incubated in a re-agent (Signet Kit; IDlabs, Ontario, Canada) designed to block nonspecificbinding of mouse Ab to mouse tissue. For all other staining, slides werefixed in cold (220°C) acetone (10 min), 4% paraformaldehyde (2 min),dextran-Tris buffer (15 min), and glycine-lysine buffer (15 min). Theseslides were then incubated (30 min) in 3% H2O2 to block endogenousperoxidase followed by 1 h in horse serum to block nonspecific binding.Titered primary Ab or isotype control (mouse IgG1) was then added to theslides for incubation overnight at room temperature followed by the ap-propriate biotinylated secondary Ab for 1 h at room temperature. The Abwas localized using streptavidin-HRP (Signet Kit; IDlabs) for 1 h at roomtemperature and 3-amino-9-ethyl-carbozole (AEC; Sigma) as a chromogen.
Flow cytometric analysis
Cultured DC were harvested on different days and suspended in 100ml ofPBS supplemented with 1% BSA. The cell suspension was then incubatedwith the appropriate primary Ab. All Abs were incubated for 30 min at4°C. Permeabilization was required to stain for fascin, which is a cyto-plasmic protein. Cells were permeabilized by incubation with 100% meth-anol (BDH) for 30 min at room temperature. For fascin-MHC class IIdouble staining, cells were stained with anti-fascin mAb followed by FITC-conjugated goat anti-mouse Ab. PE-conjugated anti-MHC class II mAbwas then added as a second Ab. For fascin-B7-2 double staining, cells werestained with anti-fascin mAb followed by FITC-conjugated goat anti-mouse Ab. Anti-B7-2 was then added followed by a PE-conjugated rabbit
anti-rat Ab. Cells were washed twice with 1% BSA-PBS after each stepand fixed with 1% paraformaldehyde-PBS. Fluorescence was analyzed ona total of 10,000 cells per sample using a flow cytometer (FACScan; Bec-ton Dickinson, Mountain View, CA).
T cell enrichment
T cells were enriched from C57BL/6 spleen by filtration through a warm(37°C for 1 h) nylon wool column as previously described (24). Briefly,spleens were homogenized to achieve a single cell suspension. After RBCwere lysed (as above), the cells washed with RPMI 1640, loaded into anylon wool column, and incubated at 37°C for 1 h. Selection of nonad-herent cells was followed by anti-B220 treatment followed by incubationwith rabbit Low-Tox complements (as above). T cell purity was routinelybetween 80 and 86%, as assessed by flow cytometry using Thy1.2 staining.
Mixed lymphocyte reaction
BM-DC were harvested on day 9 and treated with 25mg/ml mitomycin C(Sigma) for 30 min at 37°C. Treated DC were then applied in graded dosesto 23 105 T cell-enriched allogeneic spleen cells for 4 days. Cultures weremaintained in U-shaped 96-well plates (Nunc), in 200ml RPMI 1640 sup-plemented with 10% FCS, 50mM 2-ME, 100 U/ml penicillin, and 100mg/ml streptomycin. The cells were pulsed with 1mCi/ml of [3H]thymidine(ICN Pharmaceuticals, Costa Mesa, CA) in the last 18 h of incubation. Tcell proliferation was assessed by harvesting the cells on filtermats using acell harvester (Skatron, Sterling, VA) and measuring the thymidine uptakein a liquid scintillation counter (Beckman Coulter, Fullerton, CA).
Cultivation of bone marrow precursors with antisenseoligonucleotides
A fascin antisense oligonucleotide (oligo) and a matched sense controloligo of the same length (17 bp) were purchased from University CoreDNA Services, University of Calgary (Calgary, Alberta, Canada). The se-quence of the antisense oligo was provided by Dr. J. Bryan of BaylorCollege of Medicine (Houston, TX; J. Bryan, unpublished observation;Refs. 16 and 17). To stabilize the antisense and the control sense oligo,three base pairs at the 39and the 59end were phosphorothioated. Thesequence of the control oligo was CCGGCACCATGACCGCC and theantisense oligo was GGCGGTCATGGTGCCGG. Bone marrow precursorswere cultivated in cRPMI supplemented with GM-CSF alone, GM-CSFplus control oligo (2mM), or GM-CSF plus antisense oligo (2mM). Thecells were fed with new media supplemented with GM-CSF6 oligo every3 days. On day 9, DC were harvested and an MLR assay was performed toinvestigate the effect of inhibiting fascin expression on DC allostimulatoryactivity.
Image analysis
Five areas on a fascin-immunostained cytospin were randomly selected.Microscopic images on these areas were captured using a JVC model TK-1070U video camera (JVC, Scarborough, Ontario, Canada) and a Nuvista1
frame grabber board (Truevision, Indianapolis, IN) connected to a Macin-tosh computer. The captured images were analyzed using the softwarepackage NIH Image (website: rsb.info.nih.gov/NIH-IMAGE/) on a PowerMacintosh 6100/60 computer. The result is reported as percentage of fascinstaining in the total DC cell area.
ResultsFascin expression in DC in vivo
LC in the skin are the classical immature DC (25, 26). In contrast,lymph nodes contain mostly mature DC that have migrated fromdifferent interstitial sites in response to danger stimuli (1, 2). Wechose to examine fascin expression in lymph node and skin as afirst step to evaluate the correlation between fascin expression andDC maturation. Fascin binding was detected by immunohisto-chemistry using a biotinylated anti-mouse Ab as a secondary Aband streptavidin-HRP for visualization. The results (Fig. 1A) re-veal strong fascin staining of DC in the T cell-dependent areas ofthe lymph node (IDC). However, in the germinal center (B cell)areas of the lymph node, the follicular DC were essentially fascinnegative. This restriction of fascin expression to IDC is similar towhat has been reported previously in human lymph nodes (27).Fig. 1Bshows the staining of the epidermal skin sheet for fascin.
Very few fascin-positive cells could be seen. Image analysis re-vealed that only 106 17 per mm2 (n 5 7) were fascin positive inthis skin sheet. However, when these sheets were stained for MHCclass II (image not shown), significant numbers of positive cellswere observed (6886 158 cells per mm2).
These observations suggest that the expression of fascin in DCmay be regulated during maturation. Furthermore, they showedthat the IDC-restricted expression of this protein in the mouselymph node is similar to that in humans. This data is consistentwith a role for fascin in DC Ag presentation.
Fascin expression in in vitro-generated BM-DC
To confirm our in vivo observation regarding the correlation offascin expression with DC maturation and to investigate the func-tional role of fascin in DC, we generated DC in vitro. TheseBM-DC mature over 7–9 days in culture, thus allowing enoughtime to examine fascin expression during maturation (22). Bonemarrow precursor cells were allowed to grow for 9 days in thepresence of cRPMI supplemented with GM-CSF. This regimepushes DC precursors to mature into DC (22, 28, 29). On day 9,cells were examined for fascin expression by immunocytochem-istry. We observed a 29-fold increase in the percentage of cellsstaining for fascin at day 9 as compared with day 0. The stainingwas evenly distributed throughout the cytoplasm (Fig. 2A). More-over, cells that stained positively for fascin were large and dis-played long dendrites, features normally seen in mature DC. Wealso examined BM-DC for MHC class II expression, a moleculeknown to be up-regulated during DC maturation (8, 30). We ob-served that BM-DC stain strongly positive for MHC class II, re-
flecting their degree of maturity (Fig. 2B). These data demon-strated that our population of mature BM-DC expresses both fascinand MHC class II.
Fascin and MHC class II or B7-2 coexpression on BM-DCduring maturation
We used double color staining to compare the changes that occurin fascin expression and MHC class II expression during DC mat-uration. Day 9 BM-DC were double stained for fascin and MHCclass II and examined by flow cytometry. Nearly all (98.5%) of thefascin-positive cells were MHC class II positive (Fig. 3A). Fur-thermore, when we gated on the fascin-positive cells in Fig. 3A, wefound that 67.5% (square in Fig. 3A) of these cells express highlevels of MHC class II, reflecting their maturity. A similar patternof fascin and MHC class II double positivity was seen with day 3and day 6 BM-DC (data not shown). In comparison to day 3, weobserved 6.9- and 9.4-fold increases in fascin expression on day 6and day 9 BM-DC, respectively. Over the 9 days of examination,there was a direct correlation between the level of fascin expres-sion and MHC class IIhigh expression with a correlation coefficientyield of 0.98 (Fig. 3B). MHC class IIhigh expression has been usedpreviously as a marker of mature DC (8, 30). To confirm that thestrong correlation between fascin and MHC class II expressionwas indeed correlated with DC maturation, we examined a secondmarker of DC maturity, B7-2, during BM-DC maturation. B7-2 isa costimulatory molecule critical for T cell activation and is knownto be up-regulated during DC maturation (31). We used doublecolor staining to evaluate the correlation between fascin expressionand B7-2 expression in mature DC. Day 9 DM-DC were doublestained for fascin and B7-2 and examined by flow cytometry. Sim-ilar to the MHC class II double staining nearly all (96.0%) of the
FIGURE 1. A andB, Fascin expression in mouse lymph node histolog-ical section and epidermal skin sheet. Formalin-fixed lymph node sections(A) or epidermal skin sheet (B) were reacted with anti-fascin mAb followedby secondary goat anti-mouse biotinylated Ab. Staining was detected bythe HRP system. GC, Germinal center (B cell area); PZ, paracortical zone(T cell area).
FIGURE 2. A andB, Fascin and MHC class II expression in BM-DC.Day 9 BM-DC were harvested for cytocentrifugation. Cells were stainedwith anti-fascin (A) or anti-MHC class II (B) mAb followed by a secondarybiotinylated goat anti-mouse Ab and detected by the HRP system.
fascin-positive cells were B7-2 positive (Fig. 4A). B7-2 was alsoup-regulated on the BM-DC with time in culture in parallel withfascin expression (Fig. 4B). The highly significant correlation be-tween fascin expression and MHC class II and the up-regulation ofB7-2 expression in parallel with fascin strongly links fascin ex-pression with DC maturation.
Fascin expression correlates with BM-DC maturation
To demonstrate that the increase in fascin expression was due toBM-DC maturation rather than simply an effect of time in culture,we suppressed or enhanced BM-DC maturation with additionalgrowth factors and examined the expression of fascin and MHCclass II. BM-DC precursors were grown in cRPMI supplementedwith 1) GM-CSF alone; 2) GM-CSF plus TGF-b, which has beenreported to suppress DC maturation from bone marrow progenitors(23); or 3) GM-CSF plus TNF-a, which has been shown to en-hance DC maturation from bone marrow progenitors (23, 32).Cells were harvested on days 0, 3, 6, and 9 from each group,double stained for fascin and MHC class II, and analyzed by flowcytometry.
Regardless of the treatment group, there was an increase in fas-cin expression over time and, again, most of the fascin-positive
cells (946 6%) were MHC class II positive. However, the numberof fascin-positive DC was significantly (p , 0.001) reduced at day6 (by 80%) and at day 9 (68% reduction;p , 0.001) in the grouptreated with GM-CSF plus TGF-b compared with those treatedwith GM-CSF alone (Fig. 5). The cells in the group treated withTGF-b were small and had few mature DC-like cells when exam-ined in culture by light microscopy. In contrast, the number offascin-positive DC was significantly (p , 0.001) increased at day6 (by 12%) and at day 9 (26% increase;p , 0.001) in the grouptreated with GM-CSF plus TNF-a when compared with thosetreated with GM-CSF alone (Fig. 5). The number of CD11c-pos-itive cells in the TGF-b (32%), GM-CSF alone (33%), and TNF-a(34%) groups were not statistically different (ns;p . 0.05), indi-cating that the overall numbers of DC were similar. Light micro-scopy examination demonstrated more mature DC-like cells in thegroup treated with TNF-a. This clearly demonstrates that enhance-ment of BM-DC maturation with TNF-a increased fascin expres-sion. In contrast, suppression of BM-DC maturation with TGF-bmarkedly reduced fascin expression.
Similar patterns of MHC class II expression were observed onBM-DC when treated with these growth combinations (data notshown). These results provide further evidence linking fascin ex-pression with DC maturity.
FIGURE 3. A andB, Demonstration of fascin and MHC class II coex-pression on BM-DC by double staining. BM-DC were generated fromBM-DC progenitors in the presence of GM-CSF as described inMaterialsand Methods.A, On day 9, BM-DC were collected and permeabilized with100% methanol. The cells were then stained with the anti-fascin mAb or anisotype control followed by FITC-labeled goat anti-mouse Ab. Cells wereincubated with a PE-conjugated anti-MHC class II or PE-conjugated iso-typic control. For flow cytometric analysis, 104 cells were collected.B,BM-DC at days 0, 3, 6, and 9 were double stained for fascin and MHCclass II, as above. The figure displays the correlation between fascin-pos-itive cells and MHC class II high (correlation coefficient5 0.98). Cellswere called MHC class IIhigh if their expression was one log higher than thecontrol.
FIGURE 4. B7-2 expression on BM-DC during maturation. DC weregenerated from BM-DC precursors cultured in the presence of GM-CSF.A,Day 9 BM-DC were harvested and permeabilized with 100% methanol.The cells were stained with the anti-fascin mAb or an isotype controlfollowed by FITC-labeled goat anti-mouse Ab. Cells were then incubatedwith anti-B7-2 mAb or an isotype control followed by a secondary PE-labeled rabbit anti-rat Ab.B, Cells were also harvested at days 0, 3, 6, and9, and single stained with the anti-B7-2 mAb (solid line) or an isotypecontrol (dotted line) followed by a secondary FITC-labeled rabbit anti-ratAb. Fluorescence on a total of 104 cells was then analyzed by flowcytometry.
Increased levels of fascin correlates with enhanced Agpresentation activity by BM-DC
Upon maturation, DC increase their potency at stimulating naive Tcell proliferation in a MLR assay (1, 4). Because our data corre-lated fascin expression with DC maturation, we examined whetherthe increase in fascin expression would be associated with en-hanced allostimulatory activity.
BM-DC were generated in the presence of different growth fac-tor combinations that we have shown above to alter fascin expres-sion and BM-DC maturation. At day 9, BM-DC were harvestedfrom the different groups and added in graded doses to a fixednumber of naive allogeneic T cells. T cell proliferation was thenassessed in a 4-day MLR assay. In all groups, there was an in-crease in T cell proliferation when DC numbers increased (Fig. 6).DC that were treated with GM-CSF were potent APC. However,the allostimulatory activity of BM-DC treated with GM-CSF plusTGF-b, which express a lower level of fascin, was markedly re-duced. This reduction become more evident at a higher DC/T cellratio, with a 91% (p , 0.001) reduction at 1:25 ratio. In contrast,the allostimulatory activity of BM-DC treated with GM-CSF plusTNF-a, which express a higher level of fascin, was increased by37% (p , 0.001) at the same DC/T cell ratio as compared withthose treated with GM-CSF alone. In addition, when we plotted thecorrelation between fascin-positive cells and the levels of T cellalloactivation in MLR, we found a strong correlation between fas-cin levels and T cell alloactivation with a correlation coefficientyield of 0.97. These data demonstrate a strong correlation betweenthe level of fascin expression and T cell allostimulation.
The effect of fascin inhibition in BM-DC on their allostimulatoryactivity
The correlation between fascin expression and allostimulatory ac-tivity does not prove a role for fascin in Ag presentation becauseDC maturation is also associated with the up-regulation of MHCclass II and costimulatory molecules. To isolate the effects of fas-cin on BM-DC allostimulatory activity, we used antisense oligo-nucleotides to inhibit fascin expression during maturation. Bonemarrow precursor cells were seeded in cRPMI supplemented withGM-CSF alone, GM-CSF plus antisense oligo, or GM-CSF pluscontrol oligo for 9 days, and then evaluated by MLR. BM-DCviability in the no oligo (97%), control oligo (95%), and antisenseoligo (95%) treatment groups was not significantly different (ns;p . 0.05). The change in DC morphology following antisenseoligo or control oligo treatment was assessed using double stainingwith anti-fascin and anti-CD11c Abs. Consistent with previous
FIGURE 5. Influence of different growth factor combinations on fascinexpression during BM-DC maturation. DC were generated from BM-DCprecursors cultured in the presence of GM-CSF alone (l), GM-CSF plusTNF-a (F), or GM-CSF plus TGF-b (Œ). The cells were harvested on days0 (without growth factors), 3, 6, and 9, permeabilized, and double stainedfor fascin and MHC class II. This figure was obtained from flow cytometricanalysis and represents the number of fascin-positive cells of 104 collected.The results are expressed as mean of positive cells6 SE and are repre-sentative of three independent experiments. Statistical significance was de-termined using one-way ANOVA wherepp indicate ap value of,0.01 andppp indicate ap value of,0.001.
FIGURE 6. A andB, Allostimulatory activity of day 9 BM-DC culturedunder different conditions and their correlation with fascin levels. DC weregenerated from BM-DC precursors cultured in the presence of GM-CSFalone (F), GM-CSF plus TNF-a (l), or GM-CSF plus TGF-b (Œ).BM-DC were harvested at day 9 for MLR or fascin staining.A, For MLR,BM-DC were treated with mitomycin C for 30 min at 37°C and added ingraded doses to 23 105 naive allogeneic T cells in a 96-well plate. Thecells were pulsed with [3H]thymidine in the last 18 h of the 4-day incu-bation. T cell proliferation was assessed by measuring the [3H]thymidineuptake in a liquid scintillation counter. The results are expressed as meanDPM 6 SE and are representative of three independent experiments. Sta-tistical significance was determined using one-way ANOVA wherepp in-dicates ap value of ,0.01, ppp indicates ap value of ,0.001, and nsindicates no significance.B, BM-DC from the different culture groups werestained for fascin as above. This figure represent the correlation betweenfascin levels from the different BM-DC culture group, and T cell allostimu-lation in MLR at DC/T cell ratio of 1:25. The correlation coefficient yieldis 0.97.
studies that demonstrated the role of fascin expression in dendriteformation, antisense oligo-treated BM-DC became smaller and hadfewer dendrites, whereas control-treated DC had a normal mor-phology (Fig. 7). In the control oligo-treated group, 376 1% ofthe CD11c1 cells were dendritic as compared with 66 0.1% in theantisense oligo-treated group (p , 0.001). Fascin expression in theantisense oligo-treated groups was suppressed by 70%, as judgedby image analysis, when compared with control oligo-treatedgroup (Fig. 8A). However, MHC class II and B7-2 expression onBM-DC were not significantly different (ns;p . 0.05) betweencontrol oligo- and antisense oligo-treated groups (Fig. 8,B andC).
In all groups, there was an increase in T cell proliferation withan increase in the number of DC (Fig. 9). We observed no signif-icant difference (ns;p . 0.05) in the allostimulatory activity be-tween DC that had been treated with GM-CSF alone and thosetreated with GM-CSF plus control oligo. However, the allostimu-latory activity of BM-DC that were treated with the GM-CSF plusantisense oligo was markedly suppressed (p , 0.001) comparedwith those treated with the GM-CSF plus control oligo or GM-CSFalone. Similar results were observed for the fascin antisense ex-periments when we used a population that contained 90% BM-DCgenerated by a different protocol (33). These data demonstrate forthe first time a direct role for fascin in the interaction between DCand T cells.
DiscussionIn this study, we have shown that fascin expression is critical in theAg presentation activity of mature DC. By phenotypical and func-tional examinations, we observed that fascin expression is stronglycorrelated with DC maturity. Furthermore, we found that fascinexpression plays a significant role in the function of DC as APC.The mechanisms by which fascin influences Ag presentation byDC remain to be determined.
Although DC have been described as the most potent profes-sional APC, their function in this respect is critically dependent ontheir degree of maturity (1, 4, 5). When DC-precursors leave thebone marrow they circulate in the blood and eventually reside in
nonlymphoid tissues as immature DC. At these sites, especially atinterfaces with the environment, immature DC are highly efficientin Ag uptake and processing but are poor in Ag presentation and
FIGURE 7. A–F, BM-DC morphology following antisense oligonucle-otides treatment. BM-DC precursors were treated with GM-CSF plus con-trol oligonucleotide or antisense oligonucleotide. On day 9, cells were cy-tocentrifuged and double stained with the anti-fascin mAb followed byAlexa 488 (green) goat anti-mouse IgG conjugate and anti-CD11c Ab fol-lowed by PE-conjugated goat anti-hamster (IgG). The slides were thenevaluated under an immunofluorescence microscope whereA andD dem-onstrate fascin expression,B andE demonstrate CD11c expression, andCand F demonstrate fascin and CD11c coexpression in control- and anti-sense-treated BM-DC, respectively.
FIGURE 8. A–C, Fascin, MHC class II, and B7-2 levels on BM-DCfollowing antisense oligonucleotides treatment. BM-DC precursors weretreated with GM-CSF plus control oligonucleotide (bold) or antisense oli-gonucleotide (light). On day 9, cells were cytocentrifuged and stained forfascin (A). Fascin staining is expressed as a percentage of DC cell area6SE using the NIH Image Analysis software package as described inMa-terials and Methods. Day 9 DC were stained for MHC class II (B) or B7-2(C) and analyzed by flow cytometry. The data show MHC class II or B7-2expression as defined by mean fluorescence intensity (MFI). Statisticalsignificance was determined using one-way ANOVA wherepp indicates ap value of,0.01 and ns indicates no significance.
T cell activation (7, 34). Upon maturation, DC migrate via thelymphatics to the secondary lymphoid organs and become ex-tremely potent in Ag presentation, but poor in Ag uptake and pro-cessing (9–11). The increase in the Ag presentation ability of ma-ture DC is associated with up-regulation of a variety of molecules,including MHC class II and B7-2, the major triggers for the ini-tiation of T cell activation (35–39). DC express higher levels ofMHC class II and costimulatory molecules than other professionalAPC. However, this alone does not account for their greater po-tency in activating Ag-dependent immune responses (8, 31).
It has recently been recognized that other features unique to DCenhance their ability to present Ag to T cells. All of these functionsare dependent on the state of maturation of the DC. Mature DCretain MHC class II peptide complexes on their surface for pro-longed periods in culture, whereas other APC have a turnover mea-sured in hours (40, 41). Migration of DC from the periphery intothe lymph node is closely regulated during DC maturation throughchanges in chemokine receptor expression (42–44). In the lym-phoid organs only mature DC produce high levels of the DC-CK1chemokine, which preferentially attracts naive T cells (12). Thesefindings indicate that the changes that occur during DC maturationplay a significant role in their ability to act as potent APC.
DC also undergo a variety of morphological changes duringmaturation, including the development of numerous long dendrites(13, 14). Dendrites are a feature common to both neurons and DC,and both cell types express a cytoskeletal protein known as fascin(16, 20). Indeed, several studies have clearly demonstrated thatfascin is important for the development of dendrites (16, 18, 19).If dendrite formation is linked to APC activity then one wouldexpect a correlation between fascin expression, dendrite formation,and APC activity. We first confirmed that the DC in the IDC com-partment of the lymph node, where APC activity is highest, werefascin positive. This has been previously shown in human IDC(20), and we have here confirmed it in mice. We then confirmedthat LC in the skin, which are the classic immature DC, do notstain for fascin.
The in vivo demonstration of fascin expression in mature DC,and not in immature DC, suggests a link between maturation andfascin expression. To confirm this association we examined DCmaturation, in vitro, from bone marrow precursors. We found thatfascin was expressed in mature DC as evaluated by morphology.Fascin expression was also correlated with the up-regulation ofMHC class II and B7-2. These data establish an association be-tween fascin expression and maturation and a tentative link withAPC activity. To directly relate this to APC activity of mature DCwe performed MLR to assess APC activity under different condi-tions of reduced fascin expression. We found a strong correlationbetween the level of fascin and the ability of DC to activate T cellsin MLR. More specifically, fascin antisense oligonucleotides ef-fectively reduced fascin expression by 70%, and the cells treatedwith these oligonucleotides showed reduced alloactivation. Thisreduced allostimulatory effect was not due to changes in MHCclass II or B7-2 expression or reduced DC viability. Therefore,fascin is another DC protein regulated during maturation that iscritical for APC activity. These data provide the first evidencesuggesting that dendrite formation plays a functional role in theinteraction between DC and T cells.
Although we have shown that fascin expression is directly in-volved in enhancing the ability of DC to activate T cells, the exactmechanism underlying this process is not fully understood. Thereare a number of explanations that might account for this observa-tion. Fascin expression resulting in dendrite formation may in-crease the DC surface area, and this may favor interaction with agreater number of T cells. However, it may also be a much moreactive process. Fascin expression in epithelial cells results in activeextension of lamellipodia (19). Likewise, fascin might permit co-ordinated extension of dendrites maximizing the surface contactarea between DC and T cells (15).
Cell polarity may be another mechanism by which cytoskeletalproteins influence APC-T cell interactions. Recently, it has beendemonstrated that rearrangements of the T cell actin cytoskeletonresult in clustering of TCR molecules, thereby enhancing TCRcross-linking (45). These changes result in sustained T cell signal-ing, which is an important step in T cell activation (46). Throughits actin-bundling function fascin may induce a similar rearrange-ment of MHC molecules on the DC contributing to the immuno-logical synapse that develops between APC and T cells. Cell po-larity may also result in the directional secretion of cytokines byDC as has been demonstrated for T cells (47, 48). Finally, fascinis important in the motility and migration of cells (15, 19). Mi-gration of DC to lymph nodes during their maturation is critical forthe generation of the immune response, and this may be influencedby fascin expression.
In summary, this study clearly demonstrates that fascin is ex-pressed in DC upon maturation. More importantly, it indicates thatfascin expression in mature DC is critical for their generation ofdendrites and their ability to activate T cells. This observation isthe first evidence linking dendrite formation with the ability of DCto activate T cells. Although the exact nature of this interactionremains to be elucidated, further studies of the mechanisms thatcontrol fascin expression in DC may improve our understanding ofthe interaction between DC and T cells.
AcknowledgmentsWe thank Patricia Colp for her excellent technical assistance, and RobertDouglas and Lynn Thomas for their assistance on the use of the imageanalysis program.
References1. Steinman, R. M. 1991. The dendritic cell system and its role in immunogenicity.
Annu. Rev. Immunol. 9:271.
FIGURE 9. Allostimulatory activity of BM-DC after fascin inhibition.BM-DC precursors were treated with GM-CSF alone (F), GM-CSF pluscontrol oligonucleotide (f), or antisense oligonucleotide (l). On day 9,cells were harvested and treated with mitomycin C and added in gradeddoses to naive allogeneic 23 105 T cells in a 96-well plate. The cells werethen pulsed with [3H]thymidine in the last 18 h of the 4-day incubation. Tcell proliferation was assessed by measuring the [3H]thymidine in a liquidscintillation counter. The results are expressed as mean DPM6 SE and arerepresentative of three independent experiments. Statistical significancewas determined using one-way ANOVA wherepp indicates ap value of,0.01,ppp indicates ap value of,0.001, and ns indicates no significance.
2. Kripke, M. L., C. G. Munn, A. Jeevan, J. M. Tang, and C. Bucana. 1990. Evi-dence that cutaneous antigen-presenting cells migrate to regional lymph nodesduring contact sensitization.J. Immunol. 145:2833.
3. Moll, H., H. Fuchs, C. Blank, and M. Rollinghoff. 1993. Langerhans cells trans-port Leishmania majorfrom the infected skin to the draining lymph node forpresentation to antigen-specific T cells.Eur. J. Immunol. 23:1595.
4. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control ofimmunity. Nature 392:245.
5. Rescigno, M., C. Winzler, D. Delia, C. Mutini, M. Lutz, andP. Ricciardi-Castagnoli. 1997. Dendritic cell maturation is required for initiationof the immune response.J. Leukocyte Biol. 61:415.
6. Buelens, C., V. Verhasselt, D. De Groote, K. Thielemans, M. Goldman, andF. Willems. 1997. Human dendritic cell responses to lipopolysaccharide andCD40 ligation are differentially regulated by interleukin-10.Eur. J. Immunol.27:1848.
7. Hart, D. N. 1997. Dendritic cells: unique leukocyte populations which control theprimary immune response.Blood 90:3245.
8. Inaba, K., M. Pack, M. Inaba, H. Sakuta, F. Isdell, and R. M. Steinman. 1997.High levels of a major histocompatibility complex II-self peptide complex ondendritic cells from the T cell areas of lymph nodes.J. Exp. Med. 186:665.
9. Austyn, J. M. 1996. New insights into the mobilization and phagocytic activity ofdendritic cells.J. Exp. Med. 183:1287.
10. McWilliam, A. S., D. Nelson, J. A. Thomas, and P. G. Holt. 1994. Rapid den-dritic cell recruitment is a hallmark of the acute inflammatory response at mu-cosal surfaces.J. Exp. Med. 179:1331.
11. Roake, J. A., A. S. Rao, P. J. Morris, C. P. Larsen, D. F. Hankins, andJ. M. Austyn. 1995. Dendritic cell loss from nonlymphoid tissues after systemicadministration of lipopolysaccharide, tumor necrosis factor, and interleukin 1.J. Exp. Med. 181:2237.
12. Adema, G. J., F. Hartgers, R. Verstraten, E. de Vries, G. Marland, S. Menon,J. Foster, Y. Xu, P. Nooyen, T. McClanahan, et al. 1997. A dendritic-cell-derivedC-C chemokine that preferentially attracts naive T cells.Nature 387:713.
13. Teunissen, M. B., J. Wormmeester, S. R. Krieg, P. J. Peters, I. M. Vogels,M. L. Kapsenberg, and J. D. Bos. 1990. Human epidermal Langerhans cellsundergo profound morphologic and phenotypical changes during in vitro culture.J. Invest. Dermatol. 94:166.
14. Larsen, C. P., S. C. Ritchie, T. C. Pearson, P. S. Linsley, and R. P. Lowry. 1992.Functional expression of the costimulatory molecule, B7/BB1, on murine den-dritic cell populations.J. Exp. Med. 176:1215.
15. Stossel, T. P. 1993. On the crawling of animal cells.Science 260:1086.16. Edwards, R. A., and J. Bryan. 1995. Fascins, a family of actin bundling proteins.
Cell Motil. Cytoskeleton 32:1.17. Edwards, R. A., H. Herrera-Sosa, J. Otto, and J. Bryan. 1995. Cloning and ex-
pression of a murine fascin homolog from mouse brain.J. Biol. Chem. 270:10764.
18. Ross, R., X. L. Ross, J. Schwing, T. Langin, and A. B. Reske-Kunz. 1998. Theactin-bundling protein fascin is involved in the formation of dendritic processesin maturing epidermal Langerhans cells.J. Immunol. 160:3776.
19. Yamashiro, S., Y. Yamakita, S. Ono, and F. Matsumura. 1998. Fascin, an actin-bundling protein, induces membrane protrusions and increases cell motility ofepithelial cells.Mol. Biol. Cell 9:993.
20. Mosialos, G., M. Birkenbach, S. Ayehunie, F. Matsumura, G. S. Pinkus, E. Kieff,and E. Langhoff. 1996. Circulating human dendritic cells differentially expresshigh levels of a 55-kD actin-bundling protein.Am. J. Pathol. 148:593.
21. Baker, K. W., S. McKee-Protopapas, and J. E. Habowsky. 1983. The Langerhanscell in hairless mouse epidermis.Scan. Electron Microsc.457.
22. Inaba, K., M. Inaba, N. Romani, H. Aya, M. Deguchi, S. Ikehara, S. Muramatsu,and R. M. Steinman. 1992. Generation of large numbers of dendritic cells frommouse bone marrow cultures supplemented with granulocyte/macrophage colo-ny-stimulating factor.J. Exp. Med. 176:1693.
23. Yamaguchi, Y., H. Tsumura, M. Miwa, and K. Inaba. 1997. Contrasting effectsof TGF-b 1 and TNF-a on the development of dendritic cells from progenitorsin mouse bone marrow.Stem Cells 15:144.
24. Julius, M. H., E. Simpson, and L. A. Herzenberg. 1973. A rapid method for theisolation of functional thymus-derived murine lymphocytes.Eur. J. Immunol.3:645.
25. Larsen, C. P., R. M. Steinman, M. Witmer-Pack, D. F. Hankins, P. J. Morris, andJ. M. Austyn. 1990. Migration and maturation of Langerhans cells in skin trans-plants and explants.J. Exp. Med. 172:1483.
26. Schuler, G., and R. M. Steinman. 1985. Murine epidermal Langerhans cells ma-ture into potent immunostimulatory dendritic cells in vitro.J. Exp. Med. 161:526.
27. Sonderbye, L., R. Magerstadt, R. N. Blatman, F. I. Preffer, and E. Langhoff. 1997.Selective expression of human fascin (p55) by dendritic leukocytes.Adv. Exp.Med. Biol. 417:41.
28. Heufler, C., F. Koch, and G. Schuler. 1988. Granulocyte/macrophage colony-stimulating factor and interleukin 1 mediate the maturation of murine epidermalLangerhans cells into potent immunostimulatory dendritic cells.J. Exp. Med.167:700.
29. Witmer-Pack, M. D., W. Olivier, J. Valinsky, G. Schuler, and R. M. Steinman.1987. Granulocyte/macrophage colony-stimulating factor is essential for the vi-ability and function of cultured murine epidermal Langerhans cells.J. Exp. Med.166:1484.
30. Witmer-Pack, M. D., J. Valinsky, W. Olivier, and R. M. Steinman. 1988. Quan-titation of surface antigens on cultured murine epidermal Langerhans cells: rapidand selective increase in the level of surface MHC products.J. Invest. Dermatol.90:387.
31. Inaba, K., M. Witmer-Pack, M. Inaba, K. S. Hathcock, H. Sakuta, M. Azuma,H. Yagita, K. Okumura, P. S. Linsley, and S. Ikehara. 1994. The tissue distri-bution of the B7-2 costimulator in mice: abundant expression on dendritic cellsin situ and during maturation in vitro.J. Exp. Med. 180:1849.
32. Caux, C., C. Dezutter-Dambuyant, D. Schmitt, and J. Banchereau. 1992. GM-CSF and TNF-a cooperate in the generation of dendritic Langerhans cells.Nature360:258.
33. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, andG. Schuler. 1999. An advanced culture method for generating large quantities ofhighly pure dendritic cells from mouse bone marrow.J. Immunol. Methods 223:77.
34. Caux, C., Y. J. Liu, and J. Banchereau. 1995. Recent advances in the study ofdendritic cells and follicular dendritic cells.Immunol. Today 16:2.
35. Sakane, T., and I. Green. 1979. Specificity and suppressor function of human Tcells responsive to autologous non-T cells.J. Immunol. 123:584.
36. Hausman, P. B., D. P. Stites, and J. D. Stobo. 1981. Antigen-reactive T cells canbe activated buy autologous macrophages in the absence of added antigen.J. Exp.Med. 153:476.
37. Glimcher, L. H., and E. M. Shevach. 1982. Production of autoreactive I region-restricted T cell hybridomas.J. Exp. Med. 156:640.
38. Romain, P. L., S. F. Schlossman, and E. L. Reinherz. 1984. Surface moleculesinvolved in self-recognition and T cell activation in the autologous mixed lym-phocyte reaction.J. Immunol. 133:1093.
39. Karsh, J., J. B. Harley, R. Goldstein, and A. I. Lazarovits. 1993. Ro/SSA inhibitsthe autologous mixed lymphocyte reaction.Clin. Exp. Immunol. 91:103.
40. Harding, C. V., R. W. Roof, and E. R. Unanue. 1989. Turnover of Ia-peptidecomplexes is facilitated in viable antigen-presenting cells: biosynthetic turnoverof Ia vs. peptide exchange.Proc. Natl. Acad. Sci. USA 86:4230.
41. Pure, E., K. Inaba, M. T. Crowley, L. Tardelli, M. D. Witmer-Pack, G. Ruberti,G. Fathman, and R. M. Steinman. 1990. Antigen processing by epidermal Lang-erhans cells correlates with the level of biosynthesis of major histocompatibilitycomplex class II molecules and expression of invariant chain.J. Exp. Med. 172:1459.
42. Yanagihara, S., E. Komura, J. Nagafune, H. Watarai, and Y. Yamaguchi. 1998.EBI1/CCR7 is a new member of dendritic cell chemokine receptor that is up-regulated upon maturation.J. Immunol. 161:3096.
43. Dieu, M. C., B. Vanbervliet, A. Vicari, J. M. Bridon, E. Oldham, S. Ait-Yahia,F. Briere, A. Zlotnik, S. Lebecque, and C. Caux. 1998. Selective recruitment ofimmature and mature dendritic cells by distinct chemokines expressed in differentanatomic sites.J. Exp. Med. 188:373.
44. Sallusto, F., P. Schaerli, P. Loetscher, C. Schaniel, D. Lenig, C. R. Mackay,S. Qin, and A. Lanzavecchia. 1998. Rapid and coordinated switch in chemokinereceptor expression during dendritic cell maturation.Eur. J. Immunol. 28:2760.
45. Valitutti, S., M. Dessing, K. Aktories, H. Gallati, and A. Lanzavecchia. 1995.Sustained signaling leading to T cell activation results from prolonged T cellreceptor occupancy: role of T cell actin cytoskeleton.J. Exp. Med. 181:577.
46. Iezzi, G., K. Karjalainen, and A. Lanzavecchia. 1998. The duration of antigenicstimulation determines the fate of naive and effector T cells.Immunity 8:89.
47. Dustin, M. L., M. W. Olszowy, A. D. Holdorf, J. Li, S. Bromley, N. Desai,P. Widder, F. Rosenberger, P. A. van der Merwe, P. M. Allen, and A. S. Shaw.1998. A novel adaptor protein orchestrates receptor patterning and cytoskeletalpolarity in T-cell contacts.Cell 94:667.
48. Janeway, C. A., Jr., and K. Bottomly. 1994. Signals and signs for lymphocyteresponses.Cell 76:275.
