In vivo expression profile of the antiviral restrictionfactor and tumor-targeting antigen CD317/BST-2/HM1.24/tetherin in humansElina Eriksona, Tarek Adama,1, Sarah Schmidta, Judith Lehmann-Kochb, Benjamin Overa, Christine Goffineta,2,Christoph Harterc, Isabelle Bekeredjian-Dinga, Serkan Serteld, Felix Lasitschkab,3, and Oliver T. Kepplera,3

aDepartment of Infectious Diseases, bInstitute of Pathology, cDepartment of Internal Medicine V, and dDepartment of Otolaryngology, Head and NeckSurgery, University of Heidelberg, 69120 Heidelberg, Germany

Edited by Malcolm A. Martin, National Institute of Allergy and Infectious Diseases, Bethesda, MD, and approved June 14, 2011 (received for reviewFebruary 1, 2011)

Human CD317 is an intrinsic immunity factor that restricts the re-lease of enveloped viruses, including the major pathogens HIV andLassa virus, from infected cells in culture. Its importance forinfection control in humans is unclear, due in part to its incom-pletely defined in vivo expression pattern. CD317 also has beenproposed as a selective target for immunotherapy of multiplemyeloma. To provide a framework for studies of the biological func-tions, regulation, and therapeutic potential of CD317, we performedmicroarray-based expression profiling in 468 tissue samples from25 healthy organs from more than 210 patients. We found thatCD317 protein was expressed to varying degrees in all organstested and detected in a number of specialized cell types, includinghepatocytes, pneumocytes, ducts of major salivary glands, pan-creas and kidney, Paneth cells, epithelia, Leydig cells, plasma cells,bone marrow stromal cells, monocytes, and vascular endothelium.Although many of these cell types are in vivo targets forpathogenic viruses, restriction by CD317 or virus-encoded antag-onists has been documented in only some of them. Limited celltype–dependent coexpression of CD317 with the IFN biomarkerMxA in vivo and lack of responsive stimulation in organ explantssuggest that interferons may only partially regulate CD317. Thisin vivo expression profiling sheds light on the biology and species-specificity of CD317, identifies multiple thus far unknown interac-tion sites of viruses with this restriction factor, and refutes theconcept of its restricted constitutive expression and primary IFNinducibility. CD317’s widespread expression calls into question itssuitability as a target for immunotherapy.

CD317, also referred to as BST-2, HM1.24, or tetherin, is alipid raft-associated type II transmembrane glycoprotein that

localizes to the cell surface and various intracellular membranes(1, 2). Expression of CD317 in cultured cells restricts the releaseof a diverse spectrum of enveloped viruses, including the humanpathogens HIV, Lassa virus, and Kaposi’s sarcoma–associatedherpesvirus (KSHV) (3–5). In cell lines, the inhibitory virologicaleffect induced by CD317 expression is rather striking, yet therelevance of CD317 for the control of virus infections in vivois unclear.CD317’s capacity to form virion-restraining “tethers” likely

underlies its ability to bridge virion and cellular membranes viaits N- and C-terminal membrane-anchoring domains and to or-ganize into parallel disulfide-bond homodimers (6–8). This cur-rent model of virus restriction implies that expression of CD317in target cells of productive infection is a prerequisite for itsability to interfere with virus replication. However, the patternand regulation of CD317 expression in the human host, partic-ularly on cells that are susceptible to productive infection, re-main poorly defined.Based on its apparently selective expression on the surface of

terminally differentiated B cells (9), CD317 has been proposedas a therapeutic target for multiple myeloma and related plasmacell–derived immunoproliferative disorders (9–13). Antibody-

based pharmacodelivery approaches are particularly interestingfor treatment of chemotherapy-resistant malignancies (14), andthe cytotoxicity of anti-CD317 mAbs has been demonstrated inmultiple myeloma xenotransplant models (10, 15). Based onthese results, the European Medicines Agency’s Committee forOrphan Medicinal Products issued a positive opinion for the useof a humanized anti-HM1.24/CD317 mAb for the treatment ofmultiple myeloma. More recently, antibody-based immunother-apy strategies have been proposed for various solid humantumors with high-level CD317 expression (16–20).In the present work, we characterized the in vivo expression

profile of CD317 in nontransformed tissues to provide a frame-work for studies of its biological functions in humans, with a fo-cus on its potential importance as an antiviral factor and ontherapeutic strategies proposing CD317 as a target for immu-notherapy of B-cell malignancies and solid tumors.

ResultsCD317 Is Expressed on Specialized Cells in a Variety of Human Tissues.As a methodological basis for a comprehensive evaluation ofCD317 expression in vivo, we developed an immunohistochemicalstaining protocol for formalin-fixed, paraffin-embedded tissuesections, given that standard protocols yielded only low-levelstaining in tonsil (Fig. S1 and SI Results). The mouse anti-HM1.24mAb, specific for an epitope in the extracellular domain of CD317(15), was used as primary detection reagent. The specificity andsensitivity of this immunohistochemical staining protocol wasvalidated on patient-derived multiple myeloma (9), tonsil tissue,and human cell lines expressing high endogenous CD317 levels(21) or no CD317 (3) (SI Results and Figs. S2 and S3A).In a cross-sectional CD317 expression analysis, we used a tissue

microarray (TMA) containing 468 individual sections derived from25 paraffinized nontransformed tissue samples frommore than 210patients. CD317 expression was rated based on a combined pro-portion and intensity scoring system reported by Allred et al (22).Remarkably, CD317 was expressed in more than 40% of patientsamples in 21 of the tissues analyzed (Figs. 1 and 2A and Fig. S4).The expression scores varied considerably among tissues; the

Author contributions: E.E., F.L., and O.T.K. designed research; E.E., T.A., S. Schmidt, B.O.,C.G., and F.L. performed research; J.L.-K., C.H., S. Sertel, and F.L. contributed new re-agents/analytic tools; E.E., T.A., S. Schmidt, B.O., C.G., I.B.-D., F.L., and O.T.K. analyzeddata; and O.T.K. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.1Present address: Saint Louis University School of Medicine, St. Louis, MO.2Present address: Universitätsklinikum Ulm, Institut für Molekulare Virologie, Ulm,Germany.

3To whom correspondence may be addressed. E-mail: oliver.keppler@med.uni-heidelberg.de or felix.lasitschka@med.uni-heidelberg.de.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1101684108/-/DCSupplemental.

13688–13693 | PNAS | August 16, 2011 | vol. 108 | no. 33 www.pnas.org/cgi/doi/10.1073/pnas.1101684108

Dow

nloa

ded

by g

uest

on

Aug

ust 9

, 202

0

highest scores were seen in spleen, gallbladder, and stomach(>70% medium or high expression scores), whereas more moder-ate expression scores were seen in pancreas, adrenal gland, smallintestine, liver, and salivary gland (40–70% medium or high ex-pression scores) (Fig. 2B). More than half of the cases analyzed forheart, ovary, uterus, kidney, testis, and bladder tissues received lowormedium scores. CD317 expression in lung, appendix, skin, tonsil,fat, and thyroid samples of the TMA was low or negative (Figs. 1and 2B and Fig. S4). In a complementary approach, Western blotanalyses of snap-frozen samples of tissues with high (liver, gall-bladder, spleen) or low (thyroid) CD317 expression on immuno-histochemistry (Figs. 1 and 2) were confirmatory (Fig. S3B).Based on histological criteria, CD317 expression in organ

sections of the TMA was identified on type I and type II pneu-mocytes in the lung, intercalated and striated ducts of majorsalivary glands, squamous epithelium of the esophagus, gastricfundic epithelium and glands, Paneth cells of the small intestine,mononuclear cells in the lamina propria of the large intestine,acinar cells of the exocrine pancreas, hepatocytes, gallbladderepithelium, cells in the adrenal zona reticularis, collecting ductsof the kidney, Leydig cells in the testis, endometrial glands, and

vasculature (Fig. 1 and Fig. S4). Thus, in contrast to previouswork (9), we found CD317 to be widely expressed and exposedon a number of specialized cell types.

CD317 Is Ubiquitously Expressed on Blood Vessels. We next focusedon the CD317 expression on vasculature. Blood vessels in alltissues represented in the TMA demonstrated CD317 staining inthe lumen-oriented endothelial lining. In highly vascularizedorgans, such as spleen, the majority of CD317 expression was onblood vessels (Fig. 1 and Fig. S5A), but prominent vessel stainingwas observed in sections from gastrointestinal tract, musculartissue of the heart (Fig. 3A), liver, pancreas, and bladder as well(Fig. S5A). In line with these immunohistological findings,CD317 expression colocalized with expression of the blood vesselendothelial cell marker CD31 (platelet endothelial cell adhesionmolecule, PECAM-1) in the endothelium of larger arteries suchas coronary arteries and the aorta, as well as in smaller, organ-associated blood vessels such as those found in tonsil (Fig. 3Band Fig. S5B). We conclude that CD317 is highly expressed onblood vessels throughout the body.

ControlCD317

Sp

leen

ControlCD317

Uteru

sL

iver

Gallb

lad

der

Eso

ph

ag

us

Sto

mach

Sm

all In

testin

e

A

B

Pan

creas

Testis

Lung Liver Esophagus Gallbladder

Salivary G

lan

d

Lu

ng

Kid

ney

Fig. 1. CD317 is expressed in a variety of human tissues. (A) 20× magnification. (B) 40× magnification. TMA slides were stained with the anti-HM1.24/CD317mAb (Left) or an isotype control mAb (Right), followed by a biotinylated sheep α-mouse secondary antibody. Subsequently, sections were exposed to theavidin-containing ABC-AP Kit, followed by substrate development with New Fuchsin. Nuclei were counterstained with H&E.

Erikson et al. PNAS | August 16, 2011 | vol. 108 | no. 33 | 13689

IMMUNOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 9

, 202

0

Cell Type–Specific CD317 Expression in Hematopoietic Compartmentsand Intestinal Lamina Propria. We next characterized CD317 ex-pression in tonsil, peripheral blood mononuclear cells (PBMCs),

and bonemarrow. Based on documented CD317 expression on B-cell malignancies, including multiple myeloma, plasma cell leu-kemia, and Waldenström’s macroglobulinemia (9) (Fig. S2A), wefirst focused on cells of the B-cell lineage. In tonsil, little or noexpression of CD317 was found on naïve and immature B cells(soluble IgD+, CD21+), mature CD23+B cells, or cells expressingthe pan B-cell marker CD19 (Fig. 4A). A considerable fraction ofterminally differentiated CD138+ B cells, which localized pri-marily to the perifollicular region, stained positive for CD317(Fig. 4A, white arrows). In tonsil, no CD317 expression wasdetected on T cells, tissue-resident macrophages, dendritic cells(Fig. 4B), or CD66b+ granulocytes. Expression analysis onPBMCs demonstrated that CD3+ T cells did not express CD317,whereas CD19+ B cells expressed low levels on their surfaces (Fig.4C). CD14+monocytes were the only cell type in peripheral bloodthat expressed high levels of CD317 (Fig. 4C).An initial expression analysis in a previous study (23) indicated

preferential expression of CD317 on cell lines with characteristicsof bone marrow stroma; based on this, the surface protein wastermed bone marrow stromal antigen 2 (BST-2). In the presentstudy, we assessed CD317 expression in bone marrow aspiratesfrom healthy donors. A considerable fraction of nucleated bonemarrow cells stained positive for CD317 by immunohistochem-istry (Fig. 4D, Left). In Ficoll gradient–purified bone marrow, themajority of cells that expressed the stromal cell marker CD106(VCAM-1) (24) coexpressed CD317 on the surface (Fig. 4D).Furthermore, colabeling analyses of the lamina propria of thelarge intestine with cell type–specific markers identified CD138+

plasma cells, but not HIV target cells (i.e., CD4 T cells, macro-phages, dendritic cells), as the major CD317+ cell type (Fig. S6).These findings indicate that, along with terminally differentiatedB cells (9), monocytes and primary bone marrow stromal cellsexpress high levels of CD317 within the hematopoietic lineage.

Sple

en

Heart

Lung

Gallbla

dder

Liv

er

Ute

rus

Adrenal G

land

Testis

Sto

mach

Bla

dder

Skin

Sm

all Inte

stine

Prosta

te

Salivary G

land

Thym

us

Colo

n

Esophagus

Thyroid

Fat

Muscle

0

20

40

60

80

100

120

Sple

en

Pancreas

Sto

mach

Adrenal Gla

nd

Sm

all Inte

stine

Liv

er

Salivary G

land

Esophagus

Heart

Ute

rus

Kid

ney

Testis

Bla

dder

Thym

us

Colo

n

Prosta

te

Lung

Thyroid

0

20

40

60

80

100

Percen

tag

e o

f P

ositive C

ases

High

Medium

Low

A

B

Percen

tag

e o

f P

ositive C

ases

Appendix

Ovary

Kid

ney

Pancreas

Tonsil

Fat

Tonsil

Ovary

Appendix

Skin

Gallbla

dder

Muscle

Fig. 2. Rating of CD317 expression in human tissues. The expression of CD317 detected in tissues of the TMA by immunohistochemistry was rated based ona proportion and intensity scoring system (22). (A) Histogram bars depict the percentage of samples (n = 6–34 per tissue), in which the indicated tissue sectionscored positive. (B) Histogram bars depict the percentage of samples with high (solid), medium (hatched), or low (open) expression ratings. The percentage ofsamples with negative ratings is not shown.

Ao

rta

CD317/CD31CD317 CD31B

To

nsil

A Small Intestine Heart

Fig. 3. CD317 is expressed on CD31+ blood vessel endothelium. (A) Im-munohistochemical analysis of CD317 expression on blood vessels in smallintestine and heart. Black arrows indicate blood vessels with CD317+ endo-thelium. (B) Coexpression analysis of the vascular endothelial cell markerCD31 and CD317 in aorta and tonsil tissue. (Right) Merged three/two-colorimages for CD317 (red staining), CD31 (green staining), and nuclei (bluestaining). (Left and Middle) Individual fluorescence channels. Fluorescentimages were acquired at 10–60× magnification.

13690 | www.pnas.org/cgi/doi/10.1073/pnas.1101684108 Erikson et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 9

, 202

0

Expression of CD317 on Human Plasmacytoid Dendritic Cells Is NotConstitutive. In mice, CD317 is a constitutively expressed surfaceantigen on plasmacytoid dendritic cells (pDCs) (25) and a fre-quently used marker for positive selection of pDCs from lym-phatic organs. In human tonsil, BDCA2+ pDCs were foundpredominantly in the T-cell–rich lymphocyte wall and to a lesserextent in the perifollicular region (Fig. S7A, Left). Coexpressionanalysis of CD317 and BDCA2 by in situ immunofluorescencemicroscopy (Fig. S7A, Right) and flow cytometry (Fig. S7B) foundno evidence of significant CD317 expression on pDCs from hu-man tonsil. Similarly, CD317 was not expressed on BDCA2+

pDCs from Ficoll gradient–purified PBMCs, in line with a recentreport (26). In isolated pDCs, stimulation with type I IFNs up-

regulated CD317 mRNA levels by twofold to fivefold (Fig. S7C)and CD317 surface levels by threefold to fivefold (Fig. S7D).Collectively, these results demonstrate a species-specific expres-sion pattern of CD317 on pDCs. In contrast to mice, CD317 is notconstitutively expressed on pDCs in human blood and tonsil, butcan be up-regulated by type I IFNs, at least on isolated cells.

Type I IFNs May Only Partially Regulate CD317 Expression In Vivo.Onvarious types of cultured cells, CD317 is up-regulated after ex-posure to type I IFNs (4, 27), consistent with its function as anantiviral restriction factor and the presence of IFN-response ele-ments in the CD317 promoter (28). We examined the degree towhich the prominent CD317 expression in certain cell types andtissues in vivo was induced by IFN. We began by performingcolabeling studies with the myxomavirus resistance protein A(MxA). The expression of MxA, which restricts the replication oforthomyxoviruses and other RNA viruses (29), is strictly de-pendent on stimulation by type I or type III IFNs, classifying MxAas a convenient marker for IFN bioactivity (30). Individual ex-pression analyses of each restriction factor showed that the ma-jority of MxA-expressing cells in tonsil were located withingerminal centers, with additional cells staining positive in thelymphocyte wall and perifollicular region (Fig. 5A). In contrast,the majority of CD317+ cells were found in the perifollicular re-gion, and few cells within germinal centers expressedCD317.Mostimportantly, double-immunofluorescence microscopy demon-strated that cells in tonsil typically expressed eitherMxAor CD317(Fig. 5A, Lower and Fig. S8A), and that coexpression was a rareevent (Fig. 5A, Lower Right, white arrow). Similarly, CD317+

pneumocytes did not express the IFN biomarker MxA, and mac-rophages found in the lung were MxA+CD317− (Fig. 5B, Left andFig. S8B). In contrast, frequent coexpression of both restriction

A

B

C

CD14-PECD19-PECD3-PE

CD

317- 660

D

CD

31

7C

D3

17

sIgD

CD19 CD23

CD21 CD138

CD138

DC-SIGN CD68 CD3

CD

31

7

CD3 CD19 CD14

CD106-PE

CD

317 - 660

R1

Forward Scatter

Sid

e S

catter

To

ns

il

To

ns

il

PB

MC

Bo

ne

M

arro

w

71%

29%

Fig. 4. Characterization of CD317 expression in tonsil, PBMCs, and bonemarrow. (A) Coexpression analysis of CD317 and several markers of the B-celllineage in human tonsil sections. (B) Coexpression analysis of CD317 (redstaining) on dendritic cells, macrophages, and T cells (green stainings) in tonsil.(C) CD317 coexpression analysis on T cells, B cells, and monocytes in PBMCs byimmunofluorescence microscopy of cells that had been seeded on coverslips,fixed, and permeabilized (Upper) or by flow cytometry of unfixed cells(Lower). FACS plots shown were gated on viable lymphocytes and monocytes.Formicroscopy, cells were stainedwith the anti-HM1.24mAb,Alexa Fluor 660–labeled secondary Ab, and PE-conjugated Abs to the respective lineagemarkers. (Scale bar: 10 μm.) White arrows indicate double-positive cells. (D)CD317 immunohistochemistry of bonemarrow. (Original magnification, 40×.)(Left) Coexpression analysis of the bone marrow stromal cell marker CD106and CD317 in a bone marrow aspirate. Shown is the forward scatterplot/sidescatterplot with the R1 live gate indicated and a FACS dot plot of CD106 andCD317. (Right) Relative percentages of CD317+ and CD317− cells amongCD106+ cells. Data are representative of the results from two donors.

A

B

CD317 MxA

To

nsil

GC

LW

PR

GC

LW

PR

CD317/MxA CD317/MxA

Lung Aorta

Fig. 5. Cell type–dependent coexpression of CD317 and the IFN biomarkerMxA. (A) (Upper) Immunohistochemical detection of CD317 and MxA intonsil. GC, germinal center; LW, lymphocyte wall; PR, perifollicular region.(Original magnification, 10×.) (Lower) Coexpression analysis of CD317 (redstaining) and MxA (green staining) in tonsil by immunofluoresence micros-copy. (Scale bars: Left, 100 μm; Right, 10 μm.) The white arrow indicatesa coexpressing cell. (B) Coexpression analysis in lung tissue and aorta.Merged three-color images for CD317 (red staining), MxA (green staining),and nuclei (blue staining) are shown. Images from additional donors [tonsil(n = 2), aorta (n = 3), and lung tissue (n = 2)] are shown in Fig. S8.

Erikson et al. PNAS | August 16, 2011 | vol. 108 | no. 33 | 13691

IMMUNOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 9

, 202

0

factors was found on vessel endothelium, as exemplified in aortasamples from several patients (Fig. 5B, Right and Fig. S8C).To explore whether exogenous IFN stimulation can trigger

CD317 expression in tonsil explants, we exposed dispersed hu-man lymphoid aggregate cultures (tonsil-HLAC; ref. 31) to IFN-α for 24 h. This ex vivo stimulation enhanced CD317 expressionon tonsillar lymphocytes only marginally, whereas a strong andconcentration-dependent increase of MxA expression was seen(Fig. S9A). As a reference, IFN-α also readily triggered surfaceexpression of CD317 on Jurkat-TAg cells (Fig. S9B).Collectively, these results suggest a cell type– and/or tissue-

related IFN dependence of CD317 expression. Whereas markedCD317–MxA coexpression was found in vessel endothelium,their nearly mutually exclusive expression pattern in lymphoidand pulmonary tissue and insensitivity to exogenous IFN stim-ulation suggest that type I IFNs may only partially regulateCD317 levels in vivo.

DiscussionOur in vivo expression profiling of the antiviral restriction factorand potential tumor-targeting antigen CD317 in nontransformedhuman organs documented widespread tissue expression ona number of specialized cell types. This stands in contrast toa previous report suggesting selective expression of CD317 onthe surface of nontransformed, terminally differentiated B cells(9). Explicitly, that earlier study described a lack of CD317 ex-pression on PBMCs, lymph nodes, liver, spleen, kidney, andheart, applying the identical anti-HM1.24 mAb. Although we canconfirm CD317 expression on plasma cells, our study identifiedmany additional CD317+ cell types and tissues throughout thebody, greatly expanding the expression profile of this protein. Wesuspect that differences in the sensitivity of immunodetectionmight underlie this discrepancy. Our findings refute the widelyheld belief that the constitutive expression of CD317 in humansis highly restricted, a misperception that has spurred antibody-based immunotherapy strategies for multiple myeloma and cer-tain solid human tumors (9–13, 16–20). In line with our findings,a previous Northern blot analysis documented CD317 mRNA inseveral tissues, including pancreas, liver, lung, and heart (23).The role ofCD317 in the control of virus replication in vivo is not

completely understood; however, several lines of evidence point toan important antiviral capacity of CD317 in mammals. First, var-ious completely unrelated viruses (i.e., lentiviruses, herpes viruses,filoviruses, orthomyxoviruses) encode CD317 antagonists withdistinct modes of action to overcome the restriction imposed byCD317 (5). Second, anti-CD317 activities have evolved in threedifferent lentiviral genes (nef, env, and vpu) (32), implying a criticalrole of CD317 in the lentivirus–host adaption and cross-speciestransmission. In support of this notion, a nef-deleted simian im-munodeficiency virus rapidly acquired compensatory changes ingp41 in infected rhesus macaques that restored resistance toCD317/tetherin (33). Third,CD317 proteins fromdifferent speciesshow evidence of positive selection, consistent with adaptations inresponse to invading viruses (32). The present study establishesthat CD317 is expressed on a variety of specialized cell types knownto be in vivo targets for enveloped viruses. For example, vascularendothelial cells, identified as high CD317 expressors, are majorreplication sites of hemorrhagic fever viruses (e.g., Lassa virus,Ebola virus, Hantaan virus), as well as KSHV, vesicular stomatitisvirus, and herpes simplex virus (34). In the lung, CD317+ type I andII alveolar pneumocytes are targeted by various respiratorypathogens, including influenza viruses (35), severe acute re-spiratory syndrome coronavirus (36), and respiratory syncytial vi-rus (37). Remarkably, cell culture experiments have alreadydemonstrated that CD317 can exert antiviral activity against someof these pathogens, including Lassa virus, vesicular stomatitis virus,influenza A virus, Ebola virus, and KSHV (5, 38). Intriguingly, thelatter three viruses encode CD317 antagonists, indicating evolu-

tionary adaptation of these pathogens. In addition, hepatocytes,monocytes, epithelial cells, terminally differentiated B cells, andbone marrow stromal cells, which express CD317 in vivo, are sitesof replication for diverse viruses. The insights provided by thecurrent expression profiling underscore CD317’s in vivo relevanceas an antiviral factor and suggest the possibility of future studies toinvestigate whether some of these other viruses are affected by thisrestriction factor and/or have evolved antagonists.AlthoughHIV encodes potent CD317 antagonists, namely Vpu

and Env, where the virus actually encounters CD317 in vivoremains unclear. In secondary lymphoid organs and gut epithe-lium, major cell types for productive infection (i.e., CD4 T cells,macrophages, dendritic cells, and pDCs) did not express signifi-cant levels of CD317. Although the subset of CD16+ monocytesmight serve as an HIV-1 reservoir (39), it seems unlikely that thisminor population of infected, CD317+ cells exerts sufficient se-lection pressure for HIV to preserve CD317 antagonistic activi-ties. It is conceivable that systemic responses in HIV-infectedindividuals enhance CD317 expression in CD4 T cells, or thatsusceptible cells in other anatomic compartments, such as genitialmucosa, show a different profile of constitutive expression andregulation. In support of the former notion, a previous studyfound that CD4 T cells in blood and lymph nodes from Africangreen monkeys and rhesus macaques rapidly up-regulated CD317mRNA after simian immunodeficiency virus infection (40).Secretion of type I IFNs from virus-infected cells is a hallmark

of antiviral immunity. Potential target cells of infection that re-ceive these signals increase expression of IFN-stimulated genes,many of which are linked to antiviral functions (41–43). The IFNresponsiveness of CD317 in cultured primary cells is subtle;monocyte-derived macrophages up-regulated CD317 mRNAlevels by less than twofold (44), and protein levels were elevatedonly slightly (27). Moreover, in the present study, type I IFNstimulation increased CD317 levels on isolated pDCs onlymoderately, and CD317 expression on lymphocytes in tonsil waslargely unresponsive to IFN. In line with the latter observation,virtually no coexpression of CD317 and the strictly IFN-inducedrestriction factor MxA was observed in tonsillar and pulmonarytissue. In contrast, the abundant MxA–CD317 coexpression invascular endothelial cells is consistent with an IFN-mediatedcostimulation. Taken together, these findings suggest that type IIFNs may be a key regulator of CD317 expression only in certaincell types and tissues. Conceivably, other cytokines (e.g., TNF-α,IL-1, IL-6) secreted by the host in response to a viral insult maypossibly trigger CD317 expression in vivo (28).Along with antiviral activity, other functions of CD317 are

starting to emerge for which knowledge of CD317’s in vivo ex-pression profile is of interest, including its proposed capacity asa negative trans regulator of pDC responses (26) and as an or-ganizer of the subapical actin cytoskeleton in polarized epithelialcells (45). In support of the relevance of the latter function,gallbladder epithelial cells, which express high levels of CD317 insitu (Figs. 1 and 2), are characterized by a particularly pro-nounced structural polarization (46). From a therapeutic per-spective, the widespread expression of human CD317, inparticular its presence on a number of vital cell types, calls intoquestion the proposed development of tumor-selective, CD317-based targeting strategies.

Materials and MethodsParaffinized Tissues. All paraffinized tissue specimens were provided by thetissue bank of the National Center for Tumor Diseases (Heidelberg, Germany)and approved by Heidelberg University’s Ethics Committee (approval no. 206/2005). For the systematic expression analysis of CD317, an in-house TMAcontaining a total of 468 paraffinized tissue sections from 25 differenthealthy organs was used. After immunohistochemical staining for CD317,TMA slides were scanned for image acquisition using the Aperio ScanScopeCS, analyzed with Aperio ImageScope software, and rated in principle based

13692 | www.pnas.org/cgi/doi/10.1073/pnas.1101684108 Erikson et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 9

, 202

0

on the immunoreactive scoring system of Allred et al. (22). Each section wasassigned two scores, one score for the portion of cells that stained positive (0,<1% positive cells; 1, 1–10% positive cells; 2, 11–33% positive cells; 3, 34–66%positive cells; 4, >66% positive cells), and another score for the intensity ofstaining (0, no staining; 1, weak staining; 2, intermediate staining; 3, strongstaining) The overall amount of CD317 was then calculated as the sum of thescores for the portion of positive, tissue-specific cells and staining intensity. Atotal score of 7 was categorized as high staining, scores of 5 and 6 as mediumstaining, and scores of 2–4 as low staining (Fig. 2B).

Immunoenzyme Staining. The primary and secondary antibodies used arelisted in Tables S1 and S2. Isotype- and concentration-matched mouse con-trol mAbs and/or preimmune rabbit Ig served as negative controls. Immu-noenzyme stainings of CD317 were performed on 2-μm paraffin sections offormalin-fixed tissues using standard avidin-biotin anti–alkaline phospha-tase techniques (Vectastain; Vector Laboratories). Antigen retrieval wasachieved by steam-cooking the slides in 10 mM citrate buffer (pH 6.1; Dako)for 30 min. A solution of 10% Earle’s balanced salt solution (EBSS, Sigma-Aldrich) supplemented with 1% Hepes, 0.2% BSA, and 0.1% saponin (allfrom Sigma-Aldrich), pH 7.4, was used as a washing and permeabilization

buffer. Primary Ab dilutions also were prepared in this buffer with 4%γ-venin (Behring) added and incubated overnight at 4 °C. Biotinylated sheepanti-mouse IgG was applied as a secondary reagent for 30 min at roomtemperature. Naphthol AS-biphosphate (Sigma-Aldrich) with New Fuchsin(Merck) was used as the substrate for alkaline phosphatase. Slides wereviewed with an Olympus BX45 microscope.

Additional information, including procedures for double-immunofluo-rescence staining, purification of pDCs and PBMCs, HLAC from tonsil, IFN-αstimulation, immunoblotting, CD317 mRNA analysis, flow cytometry, con-focal microscopy, and details on the TMA, is provided in SI Materialsand Methods.

ACKNOWLEDGMENTS. We thank Drs. Christine Falk, Teunis Geijtenbeek,Otto Haller, and Georg Kochs; Chugai Pharmaceuticals for the gift ofreagents; John Moyers, Jutta Scheurer, Antje Heidtmann, and Ina Ambielfor expert technical assistance; and Oliver T. Fackler for discussion andcomments on the manuscript. This work was funded in part by DeutscheForschungsgemeinschaft KE742/4-1, by the European HIV-ACE researchnetwork (HEALTH-F3-2008-201095) (to O.T.K.), and SFB 938/Z2 (to F.L.).O.T.K. is a member of the CellNetworks Cluster of Excellence EXC81.

1. Kupzig S, et al. (2003) Bst-2/HM1.24 is a raft-associated apical membrane protein withan unusual topology. Traffic 4:694–709.

2. Habermann A, et al. (2010) CD317/tetherin is enriched in the HIV-1 envelope anddown-regulated from the plasma membrane upon virus infection. J Virol 84:4646–4658.

3. Neil SJ, Zang T, Bieniasz PD (2008) Tetherin inhibits retrovirus release and is antag-onized by HIV-1 Vpu. Nature 451:425–430.

4. Van Damme N, et al. (2008) The interferon-induced protein BST-2 restricts HIV-1 re-lease and is down-regulated from the cell surface by the viral Vpu protein. Cell HostMicrobe 3:245–252.

5. Evans DT, Serra-Moreno R, Singh RK, Guatelli JC (2010) BST-2/tetherin: A new com-ponent of the innate immune response to enveloped viruses. Trends Microbiol 18:388–396.

6. Perez-Caballero D, et al. (2009) Tetherin inhibits HIV-1 release by directly tetheringvirions to cells. Cell 139:499–511.

7. Schubert HL, et al. (2010) Structural and functional studies on the extracellular do-main of BST2/tetherin in reduced and oxidized conformations. Proc Natl Acad Sci USA107:17951–17956.

8. Yang H, et al. (2010) Structural insight into the mechanisms of enveloped virustethering by tetherin. Proc Natl Acad Sci USA 107:18428–18432.

9. Goto T, et al. (1994) A novel membrane antigen selectively expressed on terminallydifferentiated human B cells. Blood 84:1922–1930.

10. Ozaki S, et al. (1999) Humanized anti-HM1.24 antibody mediates myeloma cell cy-totoxicity that is enhanced by cytokine stimulation of effector cells. Blood 93:3922–3930.

11. Chiriva-Internati M, et al. (2003) Testing recombinant adeno-associated virus-geneloading of dendritic cells for generating potent cytotoxic T lymphocytes againsta prototype self-antigen, multiple myeloma HM1.24. Blood 102:3100–3107.

12. Jalili A, et al. (2005) Induction of HM1.24 peptide-specific cytotoxic T lymphocytes byusing peripheral-blood stem-cell harvests in patients with multiple myeloma. Blood106:3538–3545.

13. Kawai S, et al. (2006) Construction of a conventional non-radioisotope method toquantify HM1.24 antigens: Correlation of HM1.24 levels and ADCC activity of thehumanized antibody against HM1.24. Leuk Res 30:949–956.

14. Cheson BD, Leonard JP (2008) Monoclonal antibody therapy for B-cell non-Hodgkin’slymphoma. N Engl J Med 359:613–626.

15. Kawai S, et al. (2006) Antitumor activity of humanized monoclonal antibody againstHM1.24 antigen in human myeloma xenograft models. Oncol Rep 15:361–367.

16. Becker M, et al. (2005) Distinct gene expression patterns in a tamoxifen-sensitivehuman mammary carcinoma xenograft and its tamoxifen-resistant subline MaCa3366/TAM. Mol Cancer Ther 4:151–168.

17. Cai D, et al. (2009) Up-regulation of bone marrow stromal protein 2 (BST2) in breastcancer with bone metastasis. BMC Cancer 9:102.

18. Kawai S, et al. (2008) Interferon-α enhances CD317 expression and the antitumoractivity of anti-CD317 monoclonal antibody in renal cell carcinoma xenograft models.Cancer Sci 99:2461–2466.

19. Silveira NJ, et al.; Head and Neck Genome Project GENCAPO (2008) Searching formolecular markers in head and neck squamous cell carcinomas (HNSCC) by statisticaland bioinformatic analysis of larynx-derived SAGE libraries. BMC Med Genomics 1:56.

20. Wang W, et al. (2009) HM1.24 (CD317) is a novel target against lung cancer for im-munotherapy using anti-HM1.24 antibody. Cancer Immunol Immunother 58:967–976.

21. Goffinet C, et al. (2009) HIV-1 antagonism of CD317 is species-specific and involvesVpu-mediated proteasomal degradation of the restriction factor. Cell Host Microbe 5:285–297.

22. Allred DC, et al. (1993) Association of p53 protein expression with tumor cell pro-liferation rate and clinical outcome in node-negative breast cancer. J Natl Cancer Inst85:200–206.

23. Ishikawa J, et al. (1995) Molecular cloning and chromosomal mapping of a bonemarrow stromal cell surface gene, BST2, that may be involved in pre–B-cell growth.Genomics 26:527–534.

24. Schaumann DH, Tuischer J, Ebell W, Manz RA, Lauster R (2007) VCAM-1–positivestromal cells from human bone marrow producing cytokines for B lineage progeni-tors and for plasma cells: SDF-1, flt3L, and BAFF. Mol Immunol 44:1606–1612.

25. Swiecki M, Colonna M (2010) Unraveling the functions of plasmacytoid dendritic cellsduring viral infections, autoimmunity, and tolerance. Immunol Rev 234:142–162.

26. Cao W, et al. (2009) Regulation of TLR7/9 responses in plasmacytoid dendritic cells byBST2 and ILT7 receptor interaction. J Exp Med 206:1603–1614.

27. Miyagi E, Andrew AJ, Kao S, Strebel K (2009) Vpu enhances HIV-1 virus release in theabsence of Bst-2 cell surface down-modulation and intracellular depletion. Proc NatlAcad Sci USA 106:2868–2873.

28. Ohtomo T, et al. (1999) Molecular cloning and characterization of a surface antigenpreferentially overexpressed on multiple myeloma cells. Biochem Biophys Res Com-mun 258:583–591.

29. Haller O, Staeheli P, Kochs G (2007) Interferon-induced Mx proteins in antiviral hostdefense. Biochimie 89:812–818.

30. Wenzel J, et al. (2005) Type I interferon-associated recruitment of cytotoxic lym-phocytes: A common mechanism in regressive melanocytic lesions. Am J Clin Pathol124:37–48.

31. Jekle A, et al. (2003) In vivo evolution of human immunodeficiency virus type 1 to-ward increased pathogenicity through CXCR4-mediated killing of uninfected CD4 Tcells. J Virol 77:5846–5854.

32. Kirchhoff F (2010) Immune evasion and counteraction of restriction factors by HIV-1and other primate lentiviruses. Cell Host Microbe 8:55–67.

33. Serra-Moreno R, Jia B, Breed M, Alvarez X, Evans DT (2011) Compensatory changes inthe cytoplasmic tail of gp41 confer resistance to tetherin/BST-2 in a pathogenic nef-deleted SIV. Cell Host Microbe 9:46–57.

34. Valbuena G, Walker DH (2006) The endothelium as a target for infections. Annu RevPathol 1:171–198.

35. Guarner J, Falcón-Escobedo R (2009) Comparison of the pathology caused by H1N1,H5N1, and H3N2 influenza viruses. Arch Med Res 40:655–661.

36. Guo Y, Korteweg C, McNutt MA, Gu J (2008) Pathogenetic mechanisms of severeacute respiratory syndrome. Virus Res 133:4–12.

37. Johnson JE, Gonzales RA, Olson SJ, Wright PF, Graham BS (2007) The histopathologyof fatal untreated human respiratory syncytial virus infection. Mod Pathol 20:108–119.

38. Yondola MA, et al. (2011) Budding capability of the influenza virus neuraminidasecan be modulated by tetherin. J Virol 85:2480–2491.

39. Coleman CM, Wu L (2009) HIV interactions with monocytes and dendritic cells: Virallatency and reservoirs. Retrovirology 6:51.

40. Jacquelin B, et al. (2009) Nonpathogenic SIV infection of African green monkeys in-duces a strong but rapidly controlled type I IFN response. J Clin Invest 119:3544–3555.

41. Towers GJ (2007) The control of viral infection by tripartite motif proteins and cy-clophilin A. Retrovirology 4:40.

42. Goila-Gaur R, Strebel K (2008) HIV-1 Vif, APOBEC, and intrinsic immunity. Retro-virology 5:51.

43. Jeon YJ, Yoo HM, Chung CH (2010) ISG15 and immune diseases. Biochim Biophys Acta1802:485–496.

44. Greenwell-Wild T, et al. (2009) Interleukin-27 inhibition of HIV-1 involves an in-termediate induction of type I interferon. Blood 114:1864–1874.

45. Rollason R, Korolchuk V, Hamilton C, Jepson M, Banting G (2009) A CD317/tetherin-RICH2 complex plays a critical role in the organization of the subapical actin cyto-skeleton in polarized epithelial cells. J Cell Biol 184:721–736.

46. Oldham-Ott CK, Gilloteaux J (1997) Comparative morphology of the gallbladder andbiliary tract in vertebrates: Variation in structure, homology in function and gall-stones. Microsc Res Tech 38:571–597.

Erikson et al. PNAS | August 16, 2011 | vol. 108 | no. 33 | 13693

IMMUNOLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 9

, 202

0