13 pages

Global Vascular Expression of Murine CD34, a Sialomucin-Like Endothelial Ligand for L-Selectin

Please download to get full document.

View again

of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Global Vascular Expression of Murine CD34, a Sialomucin-Like Endothelial Ligand for L-Selectin
  1994 84: 2554-2565 S Baumhueter, N Dybdal, C Kyle and LA Lasky  endothelial ligand for L-selectinGlobal vascular expression of murine CD34, a sialomucin-like Information about reproducing this article in parts or in its entirety may be found online at: Information about ordering reprints may be found online at: Information about subscriptions and ASH membership may be found online at: reserved.Copyright 2011 by The American Society of Hematology; all rights900, Washington DC 20036.weekly by the American Society of Hematology, 2021 L St, NW, Suite Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published  For personal use guest on February 7, 2014. bloodjournal.hematologylibrary.orgFrom For personal use guest on February 7, 2014. bloodjournal.hematologylibrary.orgFrom   Global Vascular Expression of Murine CD34, a Sialomucin-Like Endothelial Ligand for L-Selectin By Susanne Baumhueter, Noel Dybdal, Carrie Kyle, and Laurence A. Lasky Extravasation of leukocytes into organized lymphoid tissues and into sites of inflammation s critical to immune surveil- lance. Leukocyte migration to peripheral lymph nodes (PLN), mesenteric lymph nodes (MLN) and Peyer’s patches (PP) de- pends on L-selectin, which recognizes carbohydrate-bearing, sialomucin-like endothelial cell surface glycoproteins. Two of these ligands have been identified at the molecular level. One is the potentially soluble mucin, GlyCAM 1, which is almost exclusively produced by high endothelial venules (HEV) of PLN and MLN. The second HEV ligand for L-selectin is the membrane-bound sialomucin CD34. Historically, this molecule has been successfully used to purify human pluri- potent bone marrow stem cells, and limited data suggest that human CD34 is present on the vascular endothelium of several organs. Here we describe a comprehensive analysis of the vascular expression of CD34 in murine tissues using a highly specific antimurine CD34 polyclonal antibody. CD34 FFICIENT IMMUNE surveillance critically depends on the ability of the appropriate leukocyte subsets to ad- here to the vascular endothelium of lymphoid organs and adjacent to inflammatory sites. A number of adhesive sys- tems are brought into play during inflammation, including the leukocyte integrins, their immunoglobulin superfamily ligands, the selectins, and a newly described family of si- alomucin-like glycoproteins that appear to function as scaf- folds for the presentation of carbohydrate ligands to the se- lectins.I4 In conjunction with various cytokines, chemokines and their receptors, the combinatorial matrix formed by this diversity of adhesion and signaling molecules, directs the appropriate cell types to areas of inflammation in a tempo- rally correct manner, thus insuring both rapid and long-lived defensive responses to any of a variety of pathogenic in- sults.1.5 Leukocyte or L-selectin appears to be used for trafficking of lymphocytes to peripheral lymph nodes (PLN), mesenteric lymph nodes (MLN) and Peyer’s patches (PP), and together with P-selectin seems to mediate the low avidity rolling of neutrophils on venules adjacent to acute inflammatory sites in vivo.6-” Initial studies of neutrophil rolling with antibodies to L-selectin and with an L-selectidimmunoglobulin G (IgG) chimeric protein suggested that L-selectin recognized a car- bohydrate ligand that was expressed on mesenteric venules and potentially at all nonlymphoid vascular sites.’ Subse- quent experiments with soluble carbohydrates that inhibited E From the Departments of Immunology and Pharmacology Genen- Submitted March 31 1994; accepted June 22 1994. Address reprint requests to Laurence A. Lasky PhD Department of Immunology Genentech lnc 46 Pt. San Bruno Blvd San Fran- cisco CA 94080. The publication costs of this article were defrayed n part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. tech Inc San Francisco CA. 994 by The American Society of Hematology. 0006-4971/94/8408-0009 3.00/0 2554 was detected on vessels in all organs examined and was expressed during pancreatic and skin nflammatory epi- sodes. A ubset of HEV-like vessels in he nflamed pancreas of nonobese diabetic (NOD) mice are positive for both CD34 and GlyCAM 1, and bind to an L-sdectin/imrnunoglobulin G (IgG) chimeric probe. Finally, we found that D34 is present on vessels of deafferentiated PLN, despite the fact hat these vessels are no longer able to nteract with L-selectin or sup- port lymphocyte binding in vitro or trafficking in vivo. Our data suggest that the egulation of posttranslational carbo- hydrate modifications of CD34 is critical in determining its capability to act as an L-selectin ligand. Based on ts ubiqui- tous expression, we propose that an appropriately glycosyl- ated form of vascular CD34 may act as a ligand for L-selec- tin-mediated leukocyte trafficking to both ymphoid and nonlymphoid sites. 994 by The American Society of Hematology L-selectin activity were consistent with this proposal, al- though these same carbohydrates also inhibited P-selectin binding in vitro.” More recent experiments with L-selectin transfected cells that lack the carbohydrate ligands for P- selectin have confirmed a role for L-selectin in neutrophil rolling.’2 The lectin domain of L-selectin recognizes sulfated, sialy- lated, fucosylated carbohydrates that are presented by three mucin-like glycoproteins expressed by PLN and MLN high endothelial venules (HEV).”-16 The first of these ligands to be described at the molecular level was GlyCAM 1, a protein rich in 0-linked carbohydrates whose expression is restricted to PLN and MLN HEV.I4 A soluble form of this mucin capable of binding to L-selectin was found in murine bl00d.l~ The second L-selectin ligand, srcinally termed Sgp90, has been recently characterized as the sialomucin-like glycopro- tein, CD34.l’ Unlike the apparently secreted GlyCAM 1, CD34 contains a transmembrane domain, and biochemical and transfection experiments have shown it to be stably asso- ciated with the cell surface. A mucin-like domain at its N- terminus appears to present a potentially large number of sialylated, sulfated 0-linked carbohydrate ligands to the lec- tin domain of L-~electin.’~.’~,’~ s would be expected for an endothelial ligand for L-selectin in PLN, CD34 was detected on he HEV, as well as on capillaries of these lymphoid organs. However, in contrast to GlyCAM 1, expression of CD34 is not confined to PLN. CD34 has long been used as a surface marker for the purification of pluripotent stem cells for bone marrow transplantations and data on a limited number of human tissues demonstrate vascular expression of CD34 in several nonlymphoid sites.20*21 he third mucin- like ligand for L-selectin is MadCAM 1.16 This mucosal lymph node HEV glycoprotein contains both immunoglobu- lin-like domains and a mucin-like domain that presents 0- linked carbohydrates to L-selectin. Purified MadCAM 1 was recently shown to support L-selectin mediated rolling in vitro.** A potential candidate for a scaffold that presents carbohy- drates to neutrophil L-selectin at nonlymphoid sites is vascu- lar CD34. The previous data demonstrating the ability of Blood, Vol 84, No 8 October 15). 1994 pp 2554-2565 For personal use guest on February 7, 2014. bloodjournal.hematologylibrary.orgFrom   GLOBAL VASCULAR EXPRESSION OF CD34 2555 PLN CD34 to bind to L-~electin,'~ ith the limited analysis of CD34 expression at nonlymphoid vascular sites in hu- mans,2o32' upport this notion. Here, we provide evidence that this sialomucin is expressed at vascular sites in all mu- rine organs and tissues examined, a result that further accen- tuates the possibility that vascular CD34 is involved in L- selectin mediated neutrophil rolling. In addition, we show that the vascular expression of CD34 is maintained at in- flammatory sites, consistent with a potential role in directing leukocyte traffic. Finally, our results support the hypothesis that, if vascular CD34 does not display the appropriate carbo- hydrate modifications, it does not appear to serve as an L- selectin ligand, and is insufficient to support lymphocyte trafficking to PLN. MATERIALS AND METHODS Antibody production and characterization. Polyclonal antibod- ies against murine CD34 were prepared as previously de~cribed.'~ Briefly, a recombinant mCD34hgG chimera was purified from transfected cell supernatants using protein G affinity chromatogra- phy. The material was then cleaved with immobilized papain (Pierce, Rockford, IL) s specified by the manufacturers instructions (3 hours at 37°C) and passed over a second protein G column to remove the human IgG Fc portion.15 One hundred micrograms of the flow through containing the extracellular domain of CD34 was used to inoculate rabbits together with complete Freund's adjuvant. Anti- body titers were determined in an enzyme-linked immunosorbent assay (ELISA) using recombinant CD34hgG as immobilized anti- gen. Rabbit serum was first depleted of antihuman IgG antibodies by passage over a human IgG column and anti-CD34 antibodies were purified from the flow through on a CD34hgG affinity column. Bound antibody was eluted with 0.1 mol/L acetic acid/O.l5 mom NaCl (pH 3.0), immediately neutralized with 1 molL Tris (pH 8.8 . dialyzed against three changes of phosphate-buffered saline (PBS), and stored at -70°C in 10 GlyceroVPBS. Immunohistochemistry. Immunohistochemistry was performed as previously described either on fresh frozen or periodate-lysine- paraformaldehyde (PLP) fixed tissue ~ecti0ns.l~ riefly, 5 to 8-pm paraffin sections were deparaffinized in xylene for 10 minutes, rinsed in water, and endogenous peroxidase was quenched with 1 H,O, for 30 minutes. For staining with the CD34 antibody, sections were immersed in 0.1 mom citrate buffer, pH 6.0, and microwaved twice on high for 3 minutes followed by 20 minutes at room temperature. For staining with the anti-GlyCAM 1 antibody, which has been previously described,14 he citrate antigen retrieval step was replaced by a 3-minute pepsin digestion at 37°C. Nonspecific binding was blocked by a 20 to 30-minute preincubation with 10% normal goat serum followed by a 30-minute room temperature incubation with the appropriate dilution of antibody. The sections were rinsed in PBS, incubated with biotinylated goat antirabbit IgG (Vector Labora- tories, Burlingame, CA) for 30 minutes, rinsed in PBS, and incubated with the Vector Elite ABC reagents (streptavidin horseradish peroxi- dase conjugate), as specified by the manufacturer's instructions. Sub- strate (DAB, Dako, Glastrup, Denmark) was added for 5 minutes. The sections were counterstained with Mayers' hematoxylin, dehy- drated, and mounted in synthetic mounting medium. Modifications that were made for frozen tissue sections included the addition of an acetone fixation step and omission of the deparaffinization and antigen retrieval (citrate/pepsin) steps. Evaluation of staining was done using a scale from 0 to 3 where 3 represents strongest reactivity. In all cases, no staining was observed with preimmune sera from either the mCD34 or GlyCAM 1 innoculated animals. In addition, in the limited number of cases that we examined, the staining of the tissues could be specifically blocked by the inclusion of recombinant mCD34 or the GlyCAM 1 peptide. Finally, we also found a lack of staining due to antibodies directed against the residual human IgG 1 contained in the purified mCD34 preparation. This was particularly true in the case of the lack of staining of B cells in either the inflamed or noninflamed peripheral lymph nodes. Staining of sections with L-selectidIgG was done using a proce- dure modified from previous method^.^^^^^ L-selectidIgG was conju- gated to gold particles as previously de~cribed.'~ ive to eight-mi- crometer cryostat sections of PLP fixed frozen tissue were placed onto Vectabond-coated slides (Vector Labs, Burlingame, CA), dried, fixed with cold acetone and dried. One hundred microliters of 200 pglmL CD4 IgG containing 1 bovine serum albumin (BSA) in PBS was added to the slide and incubated for 15 minutes at 4°C on a rotating platform set at 70 rpm. Sections were then incubated with 100 pL of 50 pg/mL L-selectidIgG chimera in the presence or absence of 10 mmoVL EGTA or with control CD4hgG alone for 45 minutes at 4°C and 70 rpm. The sections were gently decanted, fixed with cold 2.5% glutaraldehyde in PBS for 30 minutes, and washed in three changes of distilled water. L-selectidIgG gold staining was developed using the IntenSE silver enhancement kit from Amers- ham, Arlington Heights, L ections were rinsed, counterstained with Mayers' hematoxylin and mounted in CrystaVMount (Bio- media, Foster City, CA). Induction of injammatory responses. Mice were primed by spot- ting 25 pL of a solution containing 1 mg/mL oxazolone (Sigma, St Louis, MO) in 80% acetone/20% olive oil (voVvol) on both hindlegs. Draining lymph nodes, inflamed by the criteria that they were grossly enlarged with newly immigrated lymphocytes, were harvested after 5 days and processed for immunohistochemistry as described above. An immediate type hypersensitivity RH) esponse was induced by spotting 10 pL of the same oxazolone solution onto the right ears of primed mice on day 5. The left ears were treated with solvent only and served as control. Ears were harvested 3 hours after induction of the IT response and processed for immunohistochemistry. Non- obese diabetic (NOD) mice were obtained from Taconic, German- town, New York, and pancreases of male and female animals were harvested at 20 weeks after birth. The incidence of development of diabetes is approximately 80 in female and 30% in male mice at the age of 24 weeks, and the noninflamed male mouse pancreas was used as a control for CD34 staining in normal tissue. Deufferentiation of Pm. Deafferentiation was done as re- viously de~cribed.'~ riefly, popliteal lymph nodes were exteriorized, and afferent lymphatics were severed, leaving the efferent lymphat- ics and blood vessels intact. Mice were euthanized 1 week after deafferentiation. The completeness of deafferentiation was deter- mined by injection of 50 pL of a 10% India ink solution into the ipsilateral footpad. Lymph nodes with intact afferent lymphatics, as determined by uptake of India ink, were discarded. The contralateral popliteal lymph node served as the unoperated control. RESULTS Production of a speciJic polyclonal antibody directed against murine CD34. High titer antibody production against murine CD34 was achieved using a recombinant form of this protein purified from the supernatant of stably transfected mammalian ~ells.'~~*~ he hinge, CH2, and CH3 domains of human IgG 1 were attached to the extracellular domain of murine CD34 (mCD34ngG) (Fig lA),23 and this construct was transfected into human embryonic kidney (293) cells. Purification of the mCD34AgG on a protein G sepharose column enabled the isolation of milligram quanti- ties of the fusion protein with a molecular weight of -90 For personal use guest on February 7, 2014. bloodjournal.hematologylibrary.orgFrom   2556 A Murine CD34 BAUMHUETER ET AL II II I Signal sequence Cys rich TM Cytoplasmic region domain Murine CD34 Human gG Chimera 32 I I1 II I 12 Signal sequence Cys rich Human IgG1 Fc region Fig 1. (A) Schematic representation of the murine CD34 cDNA (upper) and the mCD34 human IgG chimera (lower), which was stably transfected into 293 cells. s B) Analysis of the affinity purified CD34/lgG fusion protein by reducing SDS polyacrylamide gel electrophoresis; Coomassie blue stained gel is shown with molecular weight markers n lane 1 and purified mCD34/lgG in lane 2. Molecular weights are shown as kD. kD Fig IB, lane 2 . To enhance the antibody response to the murine CD34, the bulk of the human IgG 1 Fc was removed by digestion with immobilized papain followed by protein G sepharose chromatography. The resultant flow through from this column contained approximately 80% pure extracellular domain of CD34, which was used for immuni- zation of rabbits. The anti-CD34 antibody was immunoaffinity purified from rabbit serum using immobilized recombinant mCD34/ IgG and tested for recognition of native CD34 in a number of systems. Figure 2 illustrates that the antibody specifically recognizes cell surface CD34 on normal rat kidney NRK) cells that were transfected with a full-length CD34 expres- sion construct Fig 2A) and also recognizes NIH3T3 cells, which were shown to express high levels of CD34 mRNA Fig 2B .26 Immunoprecipitation analysis of the CD34 transfectants as well as of NIH3T3 cells showed a band at -100 kD that was not seen using preimmune serum data not shown). In addition, previous immunohistochemical analysis demon- strated that this polyclonal antiserum specifically recognized capillaries and HEV in murine PLN.I5 Finally, this antiserum has been used for the isolation of hematopoietic stem cells from murine bone marrow and fetal liver C. Jordan, S Baumhueter, L. Lasky and W. Matthews, unpublished data). Taken together, these data suggest that the affinity purified anti-CD34 antiserum is a highly specific reagent for the de- tection of murine CD34 in biochemical and histologic exper- iments. Distribution of CD34 in normal mouse tissues The dis- tribution of CD34 in murine tissues was investigated by immunohistochemistry. The results of this survey are sum- marized in Table 1. In all organs examined, prominent stain- ing of capillary and postcapillary venules was observed and Fig 3 illustrates the staining of vascular endothelium in brain Fig 3A), kidney Fig 3B), and thymus Fig 3C) and the staining of megakaryoblasts and a small percentage of blast- like cells, most likely hematopoietic progenitors, in bone marrow Fig 3D). Staining was always lumenally oriented with the exception of presumptive hematopoietic progenitors and megakaryoblasts in the bone marrow where t was appar- ently cytoplasmic, as well as on the cell surface. In some tissues, eg, lymph nodes and thymus, capsules were reactive with the CD34 antibody, which most likely ecognizes endo- thelial cells in these structures see also comments, Table Q) > 8 Q) U 60 40 20 Log fluorescence intensity Fig 2. Fluorescence-activated cell sorter analysis of the surface expression of CD34 on (A) NRK cells ransfected with the ull-length murine CD34 cDNA and B) NIH3T3 cells that were shown to express high levels of mRNA.= Profiles hown are cells stained with the sec- ondary antibody only 0). ith preimmune serum (shaded) and with anti-CD34 antiserum W). For personal use guest on February 7, 2014. bloodjournal.hematologylibrary.orgFrom   GLOBAL VASCULAR EXPRESSION OF CD34 Table 1. Summary of CD34 Vascular Staining Pattern 2557 Tissue Result: Vascular Staining omments Lymph node, axillary, inguinal 3+ HEV and capillaries within node and mandibular Lymph node, mesenteric Thymus Spleen Bone, sternum, vertebrae Brain Pituitary gland Eye and lacrimal gland Salivary gland Esophagus Stomach Small and large intestine Liver Pancreas Trachea Lung Kidney Ovary Uterus Cervix and vagina Mammary gland As with axillary node. Mesenteric vessels as with other large caliber vessels examined both endothelial and adventitial staining present 3+ without trafficking) and vessels of adipose tissue 3+ of sparse vascular structures in red and white pulp, concentrated in marginal zone of white pulp and in adjacent red 3+ endothelium within thymus (with nd Pulp 3+ endothelium in section and 2+ adventitia in surrounding large caliber vessels 3+ endothelium throughout meninges, neuropil and choroid plexus. Luminal orientation primarily evident 2+ endothelium of vessels but minimal or no staining of sinusoids of pars distalis 3+ endothelium including neovascularization of bilateral corneal lesions 3+ endothelium throughout mixed and serous glands 3+ endothelium in periesophageal loose connective and adipose tissue 3+ endothelium in muscularis and submucosa. Lamina propria 3+ endothelium but staining may also be more diffuse in loose connective tissue of lamina propria prominent in lymphoid nodules and PP. Endothelium of villous capillaries clearly stained. In larger vessels of submucosa and adjacent mesentery adventitia surrounding vessels also stains 3+ endothelium of vessels and lymphatics, portal triads. No staining in sinusoids or central veins 3+ endothelium 3+ endothelium in all areas. HEV 3+ endothelium in peritracheal loose Alveolar capillaries-3+. Inconsistent 2-31 connective tissue and adipose large caliber pulmonary vessels and peribronchiole lymphatics 3+ glomerular capillaries, remaining vessels and capillaries also 3+ but do not stand out as prominently as the glomerular capillaries 3+ endothelium, capillaries throughout parenchyma and surrounding follicles clearly stained 3+ capillary endothelium within lamina propria of endometrium. 3+ remaining endothelium 3+ endothelium 3+ endothelium 3+ staining also present on narrow outer layer of node capsule As with axillary node 2+ Narrow external band of capsule In adjacent bone marrow 3+ staining of random blast type cells, presumptive hematopoietic progenitor cells. l+ cytoplasmic staining in large cells which are most consistent with megakaryblasts 1 perinuclear cytoplasmic and possible nuclear staining multifocally throughout neuropil, varies; glial and some neuronal staining Staining most prominent n pars nervosa due to increased number of vessels Connective tissue of choroid layer also stained Serous glands to 1 staining of acinar epithelium, no 1 nonspecific staining in adipose and loose connective tissue staining in mucous portion of mixed glands 1 nonspecific staining in adipose and loose connective tissue 2+ lumenal surface of gall bladder epithelium 3+ capsule and periductal band. 3+ narrow band surrounds l+ nonspecific staining in adipose tissue. many islets; nonspecific acinar cytoplasmic staining to l+ staining of basal area of proximal convoluted tubular eDithelium l+ adventitia and connective tissue surrounding follicles and in adjacent periuterine tube 2+ lamina propria of endometrium. Endometrial epithelium negative l submucosa and muscularis; epithelium negative 2-3+ in connective tissue surrounding mammary glands For personal use guest on February 7, 2014. bloodjournal.hematologylibrary.orgFrom 
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!