Normal Lymphatic Development and Function in Mice Deficient for the Lymphatic Hyaluronan

MOLECULAR AND CELLULAR BIOLOGY, Jan. 2007, p. 595–604 Vol. 27, No. 20270-7306/07/$08.00�0 doi:10.1128/MCB.01503-06Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Normal Lymphatic Development and Function in Mice Deficient forthe Lymphatic Hyaluronan Receptor LYVE-1�†

Nicholas W. Gale,1‡ Remko Prevo,2‡ Jorge Espinosa,2 David J. Ferguson,3Melissa G. Dominguez,1 George D. Yancopoulos,1

Gavin Thurston,1 and David G. Jackson2*Regeneron Pharmaceuticals, Inc., 777 Old Saw Mill River Road, Tarrytown, New York 105911; MRC Human Immunology Unit,

Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DS, United Kingdom2; andNuffield Department of Pathology, John Radcliffe Hospital, Headington, Oxford OX3 9DS, United Kingdom3

Received 12 August 2006/Returned for modification 19 September 2006/Accepted 28 October 2006

The hyaluronan receptor LYVE-1 is expressed abundantly on the surfaces of lymphatic vessels and lymphnode sinus endothelial cells from early development, where it has been suggested to function both in celladhesion/transmigration and as a scavenger for hyaluronan turnover. To investigate the physiological role(s)of LYVE-1, we generated mice in which the gene for the receptor was inactivated by replacement with a�-galactosidase reporter. LYVE-1�/� mice displayed an apparently normal phenotype, with no obvious alter-ation in lymphatic vessel ultrastructure or function and no apparent change in secondary lymphoid tissuestructure or cellularity. In addition, the levels of hyaluronan in tissue and blood were unchanged. LYVE-1�/�

mice also displayed normal trafficking of cutaneous CD11c� dendritic cells to draining lymph nodes viaafferent lymphatics and normal resolution of oxazolone-induced skin inflammation. Finally, LYVE-1�/� micesupported normal growth of transplanted B16F10 melanomas and Lewis lung carcinomas. These resultsindicate that LYVE-1 is not obligatory for normal lymphatic development and function and suggest either theexistence of compensatory receptors or a role more specific than that previously envisaged.

The lymphatic vasculature represents a second circulatorysystem that plays an important role in the maintenance oftissue fluid homeostasis (16). In addition, the network of ves-sels and lymph nodes that constitute the lymphatic systemrepresents the main core of the immune system. For example,it is the afferent lymphatic vessels that act as vital conduits forantigen-presenting dendritic cells migrating from the tissues tothe draining lymph nodes, where they stimulate naı̈ve T cells tomount the primary immune response (30). These same vesselsalso provide routes for tumor metastasis; early colonization ofthe lymph nodes is a common finding for many common hu-man cancers (26). However, unlike the blood vasculature, themechanisms that regulate cellular recognition and traffickingwithin the lymphatic vasculature, including the complex sinusstructures of the lymph nodes, are poorly understood.

Many details about the development of the lymphatics havebecome known only within the past decade, largely as a resultof studies of gene-targeted mice (reviewed in reference 28).For mammals, it has been demonstrated that the lymphaticsdevelop centrifugally by budding from the thoracic vessels (32),confirming predictions made over a century ago that they havea venous origin (24; reviewed in reference 19). During mouse

embryogenesis, the first lymph sacs bud from the cardinal veinin response to the lymphangiogenic growth factor vascularendothelial growth factor C (VEGF-C) and its lymphatic en-dothelium-specific receptor VEGFR-3 around embryonic day10.5 (E10.5), and these subsequently give rise to the develop-ing vessel network in response to continuing cues from thehomeobox domain-related transcriptional factor PROX-1 (31,32). Patterning and interconnection of lymphatic vessels laterin development and during postnatal life are regulated by thegrowth factor Ang-2 and its TIE-2 receptor on lymphatic en-dothelium (5) and by Eph-EphrinB2 interactions (15), whilearterialization of larger lymphatic collecting vessels and valveformation is regulated by the transcription factor FoxC2 incombination with platelet-derived growth factor/platelet-de-rived growth factor receptor signaling (12, 21). Other pro-teins of lymphatic endothelial cells that contribute to lym-phatic vessel function include the class III semaphorinreceptor neuropilin 2 (33), the integrin �9�1 (6), and the38-kDa integral membrane mucoprotein podoplanin/T1alpha, deletion of which results variously in embryoniclethality and/or fluid leakage syndromes of chylothorax andchylous ascites (25). Finally, the lymphatic endothelial che-mokine receptor D6 appears to be important for clearanceof chemoattractants during the resolution phase of inflam-mation (11).

Another interesting and particularly abundant component oflymphatic endothelium is the transmembrane proteinLYVE-1, a member of the Link protein superfamily (1, 22)that binds the large extracellular matrix glycosaminoglycanhyaluronan [HA; (GlcNAc�1-4GlcUA)n] via a conserved C-type-lectin-like Link domain (3). Homologous to the leukocytehoming receptor CD44, which supports HA-mediated rolling

* Corresponding author. Mailing address: MRC Human Immunol-ogy Unit, Weatherall Institute of Molecular Medicine, John RadcliffeHospital, Headington, Oxford OX3 9DS, United Kingdom. Phone: 441865 222313. Fax: 44 1865 222502. E-mail: [email protected].

† Supplemental material for this article may be found at http://mcb.asm.org/.

‡ The contributions of both of these authors should be consideredequal.

� Published ahead of print on 13 November 2006.

595

on April 11, 2019 by guest

http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

on inflamed hemovascular endothelium (23), LYVE-1 beginsto express almost simultaneously with PROX-1 during devel-opment, and this expression thereafter remains almost ex-clusively confined to lymphatic vessel and lymph node sinusendothelium (19, 20). These latter properties have madeLYVE-1 a powerful and widely exploited molecular marker instudies of normal and pathological lymphangiogenesis (9).Nevertheless, the physiological function of LYVE-1 has re-mained unresolved (7). The known involvement of the lym-phatics and lymph nodes in particular in the clearance anddegradation of hyaluronan (4, 13), allied with the observedcapacity of LYVE-1-transfected cells to internalize hyaluronan(22), would tend to suggest a role in glycosaminoglycan ho-meostasis (8). Furthermore, its structural similarity to CD44and the observed capacity to form trimolecular complexes withCD44 and hyaluronan in vitro suggest a role for LYVE-1 inmediating the transmigration of CD44-expressing leukocytesacross lymphatic vessels (7). Definitive evidence in favor ofeither role, however, is lacking.

In a more direct approach to elucidating the true physiolog-ical function of LYVE-1, we have generated mice that lack theLYVE-1 gene by targeted replacement with a �-galactosidase(�-Gal) reporter. Here we report that LYVE-1 gene-targetedmice develop normally and establish a functional network oflymphatic vessels and lymph nodes that is indistinguishablefrom that of wild-type animals. We have found no evidence forany disruption in the metabolism of hyaluronan or in the de-velopment or compartmentalization of leukocyte subsets andno defect in dendritic cell trafficking or tumor growth. This lackof an obvious phenotype argues against a major structural,developmental, or regulatory role for LYVE-1 and suggestseither compensation by an as-yet-unidentified component or arole much more specialized than that previously envisaged.

MATERIALS AND METHODS

Generation of LYVE-1 gene-targeted mice. Gene targeting of the LYVE-1locus was accomplished using VelociGene technology and was initially reportedin a summary table (29). Briefly, a bacterial artificial chromosome (BAC)-basedtargeting vector was derived using short regions of homology flanking exons 2and 5 (termed 5� and 3� homology boxes, respectively) (Fig. 1A). These homol-ogy boxes were ligated to a reporter/selection cassette encoding transmembrane�-Gal (TM-LacZ) and a loxP-flanked neomycin resistance selection gene. Ho-mologous recombination in bacteria through these homology boxes was used toreplace the intervening sequences in a BAC containing the complete LYVE-1gene. The replacement leads to the deletion of exons 2 to 5 and interveningintrons and is designed, after splicing, to result in the fusion of the first 32 aminoacids of LYVE-1 (containing the signal peptide) to TM-LacZ. The BAC em-ployed in the construction was clone 343p22 (Incyte Genomics), which is ap-proximately 150 kb in size. The cassette replacement results in a targeting vectorwith homology arms of 95 kb and 7 kb and a deletion of approximately 7 kb (Fig.1A). This BAC-based targeting vector was used as described previously (29) totarget the LYVE-1 locus in F1H4 hybrid embryonic stem (ES) cells. SeveralTaqMan probes, termed “loss-of-allele” probes, were designed to the deletedregion and were used to establish that proper targeting had occurred in ES cellsand subsequently in mice as described previously (29) (Fig. 1A). Two indepen-dent ES cell clones were derived and tested, and both clones behaved identicallyin their abilities to generate chimeric F1 and F2 mice. Knockout mice werebackcrossed one additional generation onto the C57BL/6 background, resultingin an approximate 75% contribution of the C57BL/6 and 25% of the 129 back-grounds in the resulting N2F2 mice for studies reported here. Subsequent back-crossing onto the C57BL/6 background for more than five generations has pro-duced no apparent change in phenotype. All procedures were carried outaccording to relevant U.S. federal or United Kingdom Home Office guidelinesand institutional procedures as appropriate.

Whole-mount histochemistry and immunohistology. Whole-mount stainingfor �-Gal in LYVE-1�/� and LYVE-1�/� mice and tumor cryostat sections wasperformed as described previously (2). For whole-mount preparations of dermallymphatics, ear tissues were fixed overnight in 4% paraformaldehyde in phos-phate-buffered saline (PBS), and the dorsal segment was blocked for 3 h in 30mg/ml milk powder-10% fetal bovine serum (FBS) in PBS–0.1% (wt/vol) TritonX-100 prior to double staining overnight with rat anti-mouse CD31 (MEC 13.3)/Alexa594-conjugated goat anti-rat and hamster anti-mouse podoplanin (8.1.1)/Alexa488-conjugated goat anti-hamster antibodies. Preparation of intestinalwhole mounts and immunostaining for lymphatics was performed essentially asdescribed in reference 5 by use of antisera generated against mouse LYVE-1–human immunoglobulin G1 (IgG1)–Fc (1/10,000 dilution) or a directly biotinyl-ated goat anti-mouse VEGFR-3 antiserum (R&D Systems). Fluorescent anti-body-stained sections were viewed with a Bio-Rad Radiance 2000 laser scanningconfocal microscope equipped with argon and green helium/neon lasers andanalyzed using LaserSharp2000 software. Images were recorded in simultaneousscanning mode with lambda strobing at 4-�m intervals over a depth of 24 �musing a focusing motor and subsequently merged.

Immunohistology of cryostat sections. Tissues were embedded in OCT mount-ing medium (Tissue-Tek) and frozen in liquid N2 or in dry ice-ethanol and cutinto 10-�m sections for fixation in acetone. Sections were stained with hamsteranti-mouse podoplanin (clone 8.1.1; Developmental Studies Hybridoma Bank,IA), goat anti-mouse VEGFR-3 (R&D Systems), rat anti-peripheral lymph nodeaddressin (PNAd), clone MECA79 (Becton Dickinson), or biotinylated hyalu-ronan binding protein (bHABP) (Seikagaku). After being washed, slides wereincubated with the appropriate secondary antibodies: Alexa594-conjugated goatanti-hamster, Alexa594-conjugated donkey anti-goat, Alexa488-conjugatedstreptavidin (all from Molecular Probes), or fluorescein isothiocyanate (FITC)-conjugated goat anti-rat IgM (Southern Biotech). Then, viewing was performedwith a Zeiss Axiophot microscope with epifluorescent illumination.

Electron microscopy. Samples of formaldehyde-fixed skin and intestine wereembedded in Spurr’s epoxy resin, and thin sections were coated on Formvar-coated gold grids prior to staining with uranyl acetate and lead citrate. Prepa-rations were examined with a JEOL 1200EX electron microscope.

FITC-dextran and Evans blue dye lymphangiography. Fluorescence mi-crolymphangiography of the tail was carried out essentially as described inreference 27. Mice were anesthetized with ketamine (80 �g/g body weight) andmedetomidine (Domitol; 1 �g/g), placed on a heating pad, and injected with asolution of 10 mg/ml FITC-labeled dextran (2,000 kDa; Sigma) in PBS admin-istered over a period of 20 min through a 29-gauge 1/2-in. needle insertedintradermally in the tail tip. Dye uptake was monitored continuously using aZeiss inverted epifluorescence microscope.

For Evans blue dye lymphangiography, 40 �l of 1% (wt/vol) Evans blue wasinjected into hind-limb footpads of mice that were first anesthetized with isoflu-rane. After 20 min, mice were perfused with PBS and then photographed.

Measurement of tissue hyaluronan levels. Levels of hyaluronan extracted frommouse tissues were estimated using a competitive enzyme-linked immunosorbentassay (ELISA). For the extraction step, tissue samples were digested with papain(250 �g/ml) in 5 mM cysteine-5 mM EDTA, pH 7.5, at 60°C for 24 h, followedby inactivation of the enzyme (100°C, 10 min) and centrifugation (900 � g, 5 min)to remove particulate matter. For the ELISA, 96-well Nunc Maxisorp plates werecoated overnight (4°C) with 25 �g/ml rooster comb hyaluronan (Sigma) incoating buffer (15 mM sodium carbonate and 34 mM sodium bicarbonate, pH9.4) followed by blocking in bovine serum albumin (10 mg/ml in PBS, pH 7.5) for1 h at 37°C. Tissue samples (50 �l) or hyaluronan standards (12 to 1,600 ng/ml)were then mixed with biotinylated bHABP (50 �l, 1 mg/ml; Seikagaku) andincubated for 3 h at 37°C prior to washing and detection with peroxidase-conjugated streptavidin (1:500 DAKO) followed by O-phenylenediamine sub-strate (Sigma) and measurement of absorbance at 490 nm in a Bio-Rad platereader.

Determination of serum Ig levels. Levels of serum Ig classes were estimated byantigen capture ELISA. Briefly, ELISA plates (Maxisorp; Nunc) were coatedovernight (2 �g/ml) with either rat anti-mouse IgA, rat anti-mouse IgG1, ratanti-mouse IgG2a, rat anti-mouse IgG2b, rat anti-mouse IgG3 (all from BDPharmingen), rat anti-mouse IgM, or rat anti-mouse IgE (Southern Biotechnol-ogy) followed by rinsing and blocking (1 h) with PBS–0.1% Triton X-100 � 1%bovine serum albumin. Appropriately diluted sera were applied at room tem-perature for 1 h prior to detection with Ig-specific biotinylated secondary anti-bodies (0.25 �g/ml; Pharmingen) and poly-horseradish peroxidase-streptavidin(Endogen)/tetramethylbenzidine substrate (Pharmingen). Values were deter-mined from measurements of absorption at 430 nm in a Bio-Rad ELISA platereader.

596 GALE ET AL. MOL. CELL. BIOL.


http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

FIG. 1. Generation of LYVE-1 gene-targeted mice. (A) The native LYVE-1 locus (top) showing the intron-exon structure, location ofloss-of-allele probes (red circles), and 3� and 5� homology boxes (purple boxes) (see Materials and Methods). The targeting cassette (middle) showsthe reporter/selection replacement cassette comprised of a TM-LacZ gene and a neomycin selection gene. Synthetic 3� and 5� homology boxes(purple) ligated to the reporter/selection cassette direct its targeting to the appropriate location within the LYVE-1 gene by homologousrecombination in bacteria. The final targeted allele, achieved by homologous recombination in ES cells (bottom), in which exons 2 and 5 and theintervening sequences are precisely replaced with the targeting cassette, is shown. The normal LYVE-1 gene is ablated in the targeted allele, andsplicing of the targeted allele leads to the expression of mRNA containing a transmembrane �-Gal reporter gene. (B to E) X-Gal staining ofwhole-mount (B and D) sections of intestine whole mounts (viewed from the serosal surface) from heterozygous mice reveals �-Gal expressionin cells in a pattern identical to that for whole mounts stained with anti-LYVE-1 (C and E). Panels B and C show �-Gal versus immunohisto-chemical staining in similar regions of normal intestinal serosa; panels D and E show regions of intestine harboring Peyer’s patches. (F) Comparedto that with wild-type controls, whole-mount immunohistochemistry of intestine (viewed from the mucosal surface) with anti-LYVE-1 antiserareveals a complete absence of LYVE-1 expression in knockout (KO) mice.

VOL. 27, 2007 TARGETED DELETION OF LYVE-1 597


http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

Analysis of leukocyte populations by flow cytometry. For quantitative estima-tion of tissue leukocyte populations, lymphoid organs were dissected, mechani-cally disrupted, and put through a 70-�m nylon mesh prior to staining (30 min onice) with FITC-conjugated CD4 (GK1.5 clone), FITC-conjugated CD3, allophy-cocyanin-conjugated CD19, phycoerythrin (PE)-conjugated CD45R/B220-PE(clone RA3-6B2), NK1.1 (PK136) (all from BD Pharmingen), or allophycocya-nin-conjugated rat anti-mouse CD8� (clones 5H10 and 53-6.7) as appropriate.

Granulocytes, NK cells, and monocytes/macrophages were assayed using ratanti-mouse Ly6G (Gr-1) or Ly6C-FITC (RB6-8C5 clone), FITC-conjugated ratanti-mouse CD49b or DX5-PE (DX5 clone), and rat anti-mouse CD11b-APC(M1/70 clone) or F4/80 (BM8), respectively. Cells were analyzed on a BectonDickinson FACSCalibur instrument, and a total of 106 cells per sample wascollected.

Analysis of dendritic cell trafficking by FITC skin painting. FITC skin paintingwas performed essentially as described previously (14). FITC (2.25 mg dissolvedin 150 �l acetone dibutylphthalate) was applied to the shaved abdominal skin ofwild-type, LYVE-1�/�, and LYVE-1�/� littermates. The following day, axillaryand inguinal lymph nodes were dissected, the two lymph nodes on the left andright side were pooled, digested with collagenase D (0.5 mg/ml 30 min at 37°C),put through a 70-�m nylon mesh, stained with PE-labeled hamster anti-mouseCD11c (HL3) (BD Pharmingen), and analyzed by flow cytometry to quantitatedendritic cells dually labeled with FITC and CD11c.

Oxazolone-mediated contact hypersensitivity response. Contact hypersensitiv-ity was induced in appropriate mice by initial sensitization with 100 �l 3%(wt/vol) oxazolone solution (Sigma) in 4:1 acetone-olive oil applied to the shavedabdominal skin followed by sequential elicitation 5 and 12 days later by appli-cation of 10 �l 1% (wt/vol) oxazolone in acetone-olive oil on both dorsal andventral aspects of the left ear. The control right ear was treated with acetone:olive oil only. Ear swelling was measured with a Mitutoyo loop handle dialthickness gauge (McMaster-Carr) immediately before and at various times afterelicitation.

Growth of subcutaneous tumors. B16F10 melanoma cells (kindly donated byKatja Simon, MRC Human Immunology Unit, Oxford) were resuspended inPBS, pH 7.5, and injected (0.5 � 106 in 200 �l) subcutaneously into wild-type(n � 9) or LYVE-1�/� (n � 11) mice. Tumor diameter was measured over aperiod of 5 weeks.

Lewis lung carcinoma cells (LL/2; American Type Culture Collection) wereinfected with retrovirus encoding VEGF-C in front of a green fluorescent pro-tein-internal ribosome entry site construct or with green fluorescent protein-internal ribosome entry site constructs alone for controls. Infected cell popula-tions were subjected to fluorescence-activated cell sorting twice on aDakoCytomation MoFlo high-speed cell sorter (Fort Collins, CO) prior to ex-pansion and subcutaneous implantation in appropriate mice (11 to 12 weeks ofage). After 14 days, tumors were harvested and processed for histological anal-ysis. Tumor sizes were recorded after sacrifice by use of calipers, and tumorvolumes were calculated using the formula length � width � height. Tumorswere cut into 80-�m sections, stained with X-Gal (5-bromo-4-chloro-3-indolyl-�-D-galactopyranoside) reaction, and photographed with a Zeiss Axiophot mi-croscope.

RESULTS

Generation of LYVE-1�/� mice. To generate mice lackingLYVE-1, we constructed a targeting vector that replaced theregion from codon 33 (in exon 2) through exon 5 with a TM-LacZ reporter gene (Fig. 1A). The targeted gene thus encodedthe N-terminal leader and the first eight amino acids from themature LYVE-1 protein fused to a transmembrane versionencoded by the TM-LacZ gene; when spliced, this gene en-codes a �-Gal protein which is targeted to the cell membrane.Mice carrying the mutated allele were screened by quantitativegenomic PCR using primers and probes as described for thewild-type or mutant allele ES cell screening in Materials andMethods (Fig. 1A). Mice heterozygous or homozygous for themutated LYVE-1 allele were born at normal Mendelian ratiosand displayed no obvious physical or behavioral abnormalities.

To confirm that expression of the LacZ gene was driven bythe LYVE-1 promoter, we stained heterozygous LYVE-1�/�

mouse tissues for �-Gal activity and compared this with the

pattern obtained using LYVE-1 antibody. As shown in Fig. 1Bto E, both the �-Gal and the LYVE-1 antibody revealed thesame extensive network of vessels characteristic of lymphaticson the surface of the intestine and associated Peyer’s patches.Finally, the total absence of LYVE-1 antigen from LYVE-1�/�

mice was confirmed for whole-mount tissue sections by immu-nohistochemistry with cognate antibody (for example, see re-sults for intestine shown in Fig. 1F).

Normal development, ultrastructure, and fluid drainagefunction of lymphatics in LYVE-1�/� mice. To investigatewhether LYVE-1 deficiency affects the growth or patterning ofthe lymphatic vessels, we stained whole-mount sections of in-testine with the lymphatic marker VEGFR-3. As shown in Fig.2A and B, the lymphatic vessels formed similar networks ofmostly small and often blind-ended capillaries in both wild-type and LYVE-1 gene-targeted mice. Similar results werefound for dermal lymphatics within whole-mount sections ofthe ear stained with the lymphatic marker podoplanin and thepanendothelial marker CD31, in which the morphology of themostly small and often blind-ended capillaries (diameter, 50 to100 �m) could be clearly seen (Fig. 2C and D). Moreover, thecomplex architecture of lymph node cortical and medullarysinuses—conduits for circulating lymphocytes and antigen-pre-senting cells—was also found to be conserved in LYVE-1gene-targeted mice, as visualized by immunofluorescencestaining for the lymphatic marker VEGFR-3 (Fig. 2F and G).Finally, transmission electron microscopy of initial lymphaticvessels in intestinal mucosa and skin (see Fig. S1 in the sup-plemental material) indicated a formation of overlapping in-terendothelial junctions in LYVE-1�/� mice similar to that inLYVE-1�/� mice. Consistent with this apparently normal pat-terning of the lymphatic vasculature in LYVE-1 gene-targetedmice, such animals showed no obvious signs of lymphedema.

In order to uncover more-subtle defects induced by genedisruption, we visualized the uptake and drainage of the lym-phatic tracking dye Evans blue after direct injection into thefootpad. As shown in Fig. 3A and B, the dye was distributedequally well between lymphatic vessels and draining lymphnodes in wild-type and LYVE-1 gene-targeted mice. To cor-roborate these findings using a second, independent method,we monitored the uptake and clearance of the fluorescenttracker dye FITC-dextran by the dense lymphatic networkpresent in the tail skin. To allow continuous monitoring ofuptake, the dye solution was injected at a constant pressure of45 cm H2O for 20 min, and images were recorded using afluorescence microscope. As shown in Fig. 3C and D, theFITC-dextran complex permeated the characteristic honey-comb network of dermal lymph vessels to similar depths (up to20 �m) and at similar rates in both wild-type and LYVE-1�/�

mice. Overall, these results indicate that LYVE-1 integrity isnot required for normal lymphatic fluid drainage function.

Hyaluronan homeostasis. Aside from its role in tissue fluidhomeostasis, the lymphatic system plays a significant role in theuptake and turnover of extracellular matrix components. In-deed, hyaluronan, the primary ligand for LYVE-1, is degradedin lymph nodes after release from its proteoglycan bindingpartners in the interstitial tissue matrix (4). Furthermore,LYVE-1 itself can function in vitro as an endocytic receptor forhyaluronan and is expressed abundantly in vivo within lymphnode sinus endothelium (22). We therefore investigated the



http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

FIG. 2. Lymphatic patterning and lymph node architecture in LYVE-1 gene-targeted mice. (A and B) Whole-mount VEGFR-3 staining ofintestine (viewed from the mucosal surface) from wild-type and LYVE-1 gene-targeted mice reveal that lymphatic vessels appear in the normaldensity and in a normal morphological pattern in control and knockout (ko) mice. (C and D) Whole-mount dorsal ear skin sections of wild-typeand LYVE-1 gene-targeted mice were stained with hamster anti-mouse podoplanin (in green) and rat anti-mouse CD31 (in red). Frozen sectionsfrom wild-type (F) and LYVE-1 gene-targeted (G) mice were stained with goat anti-mouse VEGFR-3 and rabbit anti-LYVE-1.



http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

effect of LYVE-1 gene deletion on hyaluronan homeostasis bymeasuring the steady-state levels of the glycosaminoglycan indifferent tissues. The results (Fig. 4A) showed no significantdifference in the concentrations of hyaluronan in skin, muscle,intestine, and lung between the two groups of animals, andonly trace levels within liver. This was also confirmed by prob-ing tissue sections from wild-type and knockout mice with abiotinylated bHABP, which revealed similar distributions ofhyaluronan within, e.g., the dermis of the ear (Fig. 4B and C)as well as in liver, lymph node, and intestine (not shown).Finally, a comparison of the serum hyaluronan levels of normal(0.783 0.081 �g/ml; n � 12) and knockout (0.762 0.048�g/ml; n � 12) mice indicated no difference due to LYVE-1gene deletion.

Leukocyte populations are unaltered in LYVE-1�/� mice.Given the abundant expression of LYVE-1 in normal lymphnode sinus endothelium and the role of secondary lymphoidtissue in regulating resident and circulating leukocyte popula-tions, we considered the possibility that LYVE-1 deficiencymight have an impact on leukocyte development and/or com-partmentalization. Hence, we carried out a comprehensiveanalysis of the number and composition of major leukocytepopulations, including the CD4� and CD8� (helper and cyto-toxic) T-cell subsets, CD19� B cells, NK1.1/Ly49b� NK cells,Ly6G/Gr1� neutrophils, and F4/80/CD11b� monocytes/mac-rophages in peripheral blood, spleen, thymus, and lymph nodesby use of flow cytometry. The results (Fig. 5A and B; also, seeFig. S2 in the supplemental material) revealed no significant

FIG. 3. Direct visualization of fluid drainage in LYVE-1 gene-targeted mice. (A and B) Distribution of Evans blue dye after injection into thehind footpads of wild-type (A) and LYVE-1 gene-targeted (B) mice. The dye labels large lymphatic vessels in the leg (insets) and draining lymphnodes within the abdomen (main images in panels A and B; iliac and renal nodes are shown by arrowheads) to similar extents in both control andknockout (KO) mice. (C and D) Levels of FITC-dextran uptake after injection into the dermis are similar in lymphatic vessels of the tails ofwild-type (C) and LYVE-1 gene-targeted (D) mice as observed by fluorescence microscopy. Images represent tail lymphatics at successive distances(as shown) from the original injection site.



http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

differences between any of these leukocyte populations withineither wild-type mice or homozygous/heterozygous LYVE-1gene-targeted mice. In addition, ELISA determination of in-dividual serum immunoglobulin classes (IgG, IgM, and IgA)and subtypes (IgG1, IgG2a, IgG2b, and IgG3) indicated nosignificant differences between wild-type and knockout mice(see Fig. S3 in the supplemental material). Together, theseresults indicate that LYVE-1 function is not required for thedevelopment or compartmentalization of immune cells withinprimary or secondary lymphoid tissue.

Migration of dendritic cells in afferent lymph. Based onhomology with the leukocyte-homing receptor CD44, whichsupports rolling on blood vascular endothelium, we had spec-ulated that LYVE-1 might mediate the adhesion or transmi-gration of cells entering the lymphatic vasculature (7, 10). Aprominent example of such trafficking is the migration of an-tigen-presenting cells from tissue to draining lymph nodes dur-ing normal immune surveillance. To compare the process innormal mice with that in knockout mice, we measured thetrafficking of epidermal dendritic cells to lymph node via af-ferent lymphatics using the technique of FITC skin painting.Dissection of the draining lymph nodes 18 h after dye appli-cation, followed by staining with the dendritic cell markerCD11c and flow cytometry, revealed that both the total CD11c-positive dendritic cell populations and the numbers of skin-derived CD11c/FITC double-positive cells were comparable in

wild-type mice and both heterozygous LYVE-1�/� and ho-mozygous LYVE-1�/� mice (Fig. 6A). Hence we conclude thatLYVE-1 is not required for either entry or migration of den-dritic cells through the afferent lymphatics.

Development and resolution of skin inflammation. Each ofthe parameters described above was used to evaluate the roleof LYVE-1 in normal tissue homeostasis. To determinewhether LYVE-1 function is manifest under conditions of tis-sue stress, we subjected both wild-type and LYVE-1�/� miceto allergen-induced contact hypersensitivity, a condition thatpromotes T-cell-mediated skin inflammation characterized bylocal tissue edema and the mobilization of epidermal Langer-hans cell migration to draining lymph nodes via afferent lym-phatics. To provoke inflammation, mice were sensitized by theconsecutive topical application of oxazolone on the shavedabdominal skin and challenged 5 days later by the applicationof oxazolone on the ear. Measurement of edema as deter-mined by ear thickness and microscopic examination of eartissue after treatment showed no difference in the responses ofthe two groups of mice in either the magnitude or the rate ofresolution of inflammation (Fig. 6B). These results suggest thatthe presence of LYVE-1 on lymphatic endothelial cells is re-quired neither for T-cell-mediated hypersensitivity nor forclearance of the attendant inflammation/edema.

Growth and metastasis of subcutaneous tumors. Finally, weconsidered the requirement for LYVE-1 integrity on tumorgrowth. The development of a lymphatic supply by either coop-tion or lymphangiogenesis is known to influence tumor fluiddrainage and, consequently, tumor hydrostatic pressure.Hence, any defects in lymphatic vessel formation might beexpected to limit tumor growth. To investigate these phenom-ena, we transplanted syngeneic (C57BL/6) B16F10 melanomacells subcutaneously into wild-type and LYVE-1�/� mice andmonitored tumor diameter over a 5-week period. The resultsshowed there was no obvious difference in tumor growth ratesbetween the two groups of animals (Fig. 7A). We also inves-

FIG. 5. Leukocyte populations in LYVE-1 gene-targeted mice. (Aand B) Leukocyte populations prepared from peripheral blood(A) and spleen (B) (see Materials and Methods) of wild-type andLYVE-1�/� (KO) mice were stained with directly conjugated antibod-ies as indicated and analyzed by flow cytometry. Values are the meanpercentages of positive cells in the leukocyte gate standard errors(n � 3). Mono/Mac, monocytes/macrophages.

FIG. 4. Hyaluronan levels in LYVE-1 gene-targeted mice. (A) Thelevels of hyaluronan in tissues from wild-type and LYVE-1�/� (KO)mice were determined by a competitive ELISA-like assay using bio-tinylated bHABP as described in Materials and Methods. (B) Frozensections of ear tissue stained with bHABP/Alexa488 streptavidin(green) and podoplanin-Alexa594-conjugated goat anti-hamster 594(red) to label hyaluronan and lymphatic vessels, respectively, show nosignificant differences between wild-type and LYVE-1 gene-targetedmice.



http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

tigated growth and lymph vessel development in a Lewis lungcarcinoma that had been engineered to overexpress the lym-phangiogenic growth factor VEGF-C by retroviral gene trans-fer. Again, these tumors showed similar growth rates in bothwild-type and LYVE-1 gene-targeted mice (Fig. 7B), and therewere no obvious differences in either the number of lymphat-ics, their patterns, or the depths of their penetration into tumortissue, as assessed by �-Gal histochemistry (Fig. 7C). Together,these results indicate that LYVE-1 is not obligatory for eitherthe growth or the apparent function of tumor-associated lym-phatics.

DISCUSSION

The near-ubiquitous expression and relative abundance ofLYVE-1 in lymphatics in the adult and its appearance as oneof the earliest lymphatic-specific genes during embryonic de-velopment had lead to the expectation that the receptor mustplay some fundamental role in maintaining either lymphatic

architecture or normal lymphatic function (7, 19). Surprisingly,however, we found no developmental defects in LYVE-1-de-leted mice. Rather, they displayed normal vessel morphologyand patterning, as assessed by a comprehensive survey of dif-ferent tissues by whole-mount histochemistry and immunohis-tology. Moreover, there was no evidence of edema, a conditionassociated with defective fluid drainage function that manifestsas chylothorax or chylous ascites, respectively, in mice withtargeted deletion of other genes expressed in lymphatic endo-thelium, such as those encoding the �9�1 integrin (6) and themembrane sialomucin podoplanin (25). This was confirmed bylymphangiography with FITC-dextran and Evans blue dyes,which showed uptake and clearance both by superficial dermallymphatics and by deeper collectors supplying the draininglymph nodes.

Recent studies in mouse embryogenesis have shown thatLYVE-1 is present in cardinal vein endothelial cells before theinitial separation of the hemovascular and lymphatic lineagesand just after the first expression of VEGFR-3, whose integrity

FIG. 6. Trafficking of dendritic cells and resolution of skin inflammation in LYVE-1 gene-targeted mice. (A) Analysis of dendritic cell antigenuptake and migration by FITC skin painting. FITC dissolved in acetone dibutylphthalate was applied to the shaved abdominal skin of wild-type,LYVE-1�/�, and LYVE-1�/� littermates (n � 3). The following day, axillary and inguinal lymph nodes were dissected, pooled together as left andright nodes, stained with PE-labeled CD11c, and analyzed by flow cytometry. Values are mean percentages standard errors for positive cells fromthe left and right draining lymph nodes from three mice (n � 3 � 2). (B) Delayed-type hypersensitivity response in LYVE-1 gene-targeted mice.Wild-type and LYVE-1�/� (Lyve-1KO) littermates were sensitized by a topical application of oxazolone (Ox) on the abdomen and challenged 5days later by an application of oxazolone to the left ear while the right ear was treated with vehicle only (ctrl). A second elicitation was performed7 days after the first elicitation. Ear swelling was measured with a dial thickness gauge immediately before and at various times after elicitation.Values shown are mean absolute increases in ear thickness in 10�2 mm for four mice.



http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

is essential not only for lymphatic but also for cardiac devel-opment (32). LYVE-1 is also present just prior to the expres-sion of PROX-1, whose polarization in the cardinal vein signalsthe onset of budding and formation of the nascent lymph sacsat E9 to E10.5 (31, 32). The lack of any profound embryolog-ical phenotype in gene-deleted mice is therefore intriguing andsuggests either that the expression of LYVE-1 at these earlytime points is not required for differentiation or lymphangio-genesis or that its loss can be compensated by other function-ally similar molecules (also see below). Although such com-pensatory molecules have yet to be identified, they are unlikelyto include CD44, as this molecule is not normally expressed bylymphatic endothelium and does not appear to be induced inthe lymphatics of LYVE-1 knockout tissues (unpublished ob-servation).

The lack of any defect in constitutive dendritic cell traffick-ing in LYVE-1�/� gene-deficient mice and their ability toresolve oxazolone skin contact-induced hypersensitivity andattendant edema formation argues against the role we origi-

nally proposed for the receptor as a gatekeeper for leukocyteadhesion/transmigration (10). In light of evidence that den-dritic cells have the capacity to synthesize and present HA ontheir surfaces (17), we had anticipated that the migration ofthese cells would be at least partially disrupted in LYVE-1knockout mice. However, one very recent finding from ourlaboratory relevant to this issue is that hyaluronan binding inlymphatic vessels in the context of normal tissues may be re-versibly inactivated (18). Although we have not fully definedthe molecular basis for inactivation, the emerging picture is ofa reversible modification to LYVE-1 sugar chains that preventsinteraction of the receptor with glycosaminoglycan while re-taining expression at the cell surface. How this intriguing pro-cess is regulated is currently unknown. Perhaps the uniquephysiological function of LYVE-1, much like that of its homo-logue CD44, will be apparent only under specific conditions oftissue damage or disease.

A final possibility is that the loss of LYVE-1 in gene-deletedmice can be compensated by other HA receptors. The appar-ent maintenance of normal HA levels could be achievedthrough CD44 expressed on fibroblasts and tissue macro-phages or the endocytic HARE molecule (hyaluronan receptoron endothelium) present in lymph node, spleen, and liver sinusendothelium, for example (34, 35). Further studies usingLYVE-1/CD44 knockout mice and other double-knockoutmice will be required to resolve this issue.

ACKNOWLEDGMENTS

This work was supported by Unit funding from the United KingdomMedical Research Council and by a project grant from Cancer Re-search UK (A399/C581).

We thank Anthony Dore, Nicholas Papadopoulos, Li Pan, VirginiaHughes, and Mary Simmons for assistance with flow cytometry andimmunochemical staining; Ella Joffe for assistance with delayed-hy-persensitivity experiments; and Jocelyn Holash and Irene Noguera forassistance with Lewis carcinoma experiments. We gratefully acknowl-edge David Valenzuela, Andrew Murphy and all the other members ofthe VelociGene team at Regeneron for the generation of the LYVE-1knockout mice.

REFERENCES

1. Banerji, S., J. Ni, S. X. Wang, S. Clasper, J. Su, R. Tammi, M. Jones, andD. G. Jackson. 1999. LYVE-1, a new homologue of the CD44 glycoprotein,is a lymph-specific receptor for hyaluronan. J. Cell Biol. 144:789–801.

2. Bianchi, L. M., K. Dinsio, K. Davoli, and N. W. Gale. 2002. Lac z histo-chemistry and immunohistochemistry reveal ephrin-B ligand expression inthe inner ear. J. Histochem. Cytochem. 50:1641–1645.

3. Day, A. J., and G. D. Prestwich. 2002. Hyaluronan-binding proteins: tying upthe giant. J. Biol. Chem. 277:4585–4588.

4. Fraser, J. R., W. G. Kimpton, T. C. Laurent, R. N. Cahill, and N. Vakakis.1988. Uptake and degradation of hyaluronan in lymphatic tissue. Biochem.J. 256:153–158.

5. Gale, N. W., G. Thurston, S. F. Hackett, R. Renard, Q. Wang, J. McClain, C.Martin, C. Witte, M. H. Witte, D. G. Jackson, T. Sato, C. Suri, P. A.Campochiaro, S. J. Wiegand, and G. D. Yancopoulos. 2002. Angiopoietin-2is required for postnatal angiogenesis and lymphatic patterning, and only thelatter role is rescued by angiopoietin-1. Dev. Cell 3:411–423.

6. Huang, X. Z., J. F. Wu, R. Ferrando, J. H. Lee, Y. L. Wang, R. V. Farese, Jr.,and D. Sheppard. 2000. Fatal bilateral chylothorax in mice lacking theintegrin �9�1. Mol. Cell. Biol. 20:5208–5215.

7. Jackson, D. G. 2004. Biology of the lymphatic marker LYVE-1 and appli-cations in research into lymphatic trafficking and lymphangiogenesis. APMIS112:526–538.

8. Jackson, D. G. 2004. The lymphatic endothelial hyaluronan receptor LYVE-1.Glycoforum www.glycoforum.gr.jp/science/hyaluronan/HA28/HA28E.html.

9. Jackson, D. G. 2001. New molecular markers for the study of tumor lym-phangiogenesis. Anticancer Res. 21:4279–4283.

10. Jackson, D. G., R. Prevo, S. Clasper, and S. Banerji. 2001. LYVE-1, thelymphatic system and tumor lymphangiogenesis. Trends Immunol. 22:317–321.

FIG. 7. Growth and lymphangiogenesis of subcutaneous tumors inLYVE-1�/� mice. (A) B16F10 melanoma cells (0.5 � 106) were in-jected subcutaneously into wild-type (n � 9), LYVE-1�/� (het) (n �10), and LYVE-1�/� (ko) (n � 8) mice. Values represent mean tumordiameters measured after 5 weeks. (B) Lewis lung carcinoma cellsretrovirally transfected with VEGF-C were injected (1 � 106) subcu-taneously into LYVE-1 knockout, heterozygous, and wild-type litter-mates (three mice per group), and tumor volume was determinedsubsequently (T, �14 days) as described in Materials and Methods.(C) Fourteen-day Lewis lung carcinoma tumors from heterozygousand LYVE-1 knockout mice stained with X-Gal to visualize intratu-moral lymphatics.



http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

11. Jamieson, T., D. N. Cook, R. J. B. Nibbs, A. Rot, C. Nixon, P. McLean, A.Alcami, S. A. Lira, M. Wiekowski, and G. J. Graham. 2005. The chemokinereceptor D6 limits the inflammatory response in vivo. Nat. Immunol. 6:403–411.

12. Kriederman, B. M., T. L. Myloyde, M. H. Witte, S. L. Dagenais, C. L. Witte,M. Rennels, M. J. Bernas, M. T. Lynch, R. P. Erickson, M. S. Caulder, N.Miura, D. Jackson, B. Brooks, and T. W. Glover. 2003. FOXC2 haploinsuf-ficient mice are a model for human autosomal dominant lymphedema-distichiasis syndrome. Hum. Mol. Genet. 12:1179–1185.

13. Laurent, T. C., and J. R. Fraser. 1992. Hyaluronan. FASEB J. 6:2397–2404.14. Macatonia, S. E., S. C. Knight, A. J. Edwards, S. Griffiths, and P. Fryer.

1987. Localization of antigen on lymph node dendritic cells after exposure tothe contact sensitizer fluorescein isothiocyanate. Functional and morpholog-ical studies. J. Exp. Med. 166:1654–1667.

15. Makinen, T., R. H. Adams, J. Bailey, Q. Lu, A. Ziemiecki, K. Alitalo, R.Klein, and G. A. Wilkinson. 2005. PDZ interaction site in ephrinB2 isrequired for the remodeling of lymphatic vasculature. Genes Dev. 19:397–410.

16. Mortimer, P. S. 1997. Lymphatics. Recent Adv. Dermatol. 15:175–192.17. Mummert, M. E., D. Mummert, D. Edelbaum, F. Hui, H. Matsue, and A.

Takashima. 2002. Synthesis and surface expression of hyaluronan by den-dritic cells and its potential role in antigen processing. J. Immunol. 169:4322–4331.

18. Nightingale, T., S. Banerji, and D. G. Jackson. 2005. Functional regulationof the lymphatic endothelial hyaluronan receptor LYVE-1: the role of N-glycosylation, p. 615–618. In E. A. Balazs and V. C. Hascall (ed.), Hyaluro-nan, vol. 2. Matrix Biology Institute, Edgewater, NJ.

19. Oliver, G. 2004. Lymphatic vasculature development. Nat. Rev. Immunol.4:35–45.

20. Oliver, G., and M. Detmar. 2002. The rediscovery of the lymphatic system:old and new insights into the development and biological function of thelymphatic vasculature. Genes Dev. 16:773–783.

21. Petrova, T. V., T. Karpanen, C. Norrmen, R. Mellor, T. Tamakoshi, D. N.Finegold, R. Ferrell, D. Kerjaschki, P. Mortimer, S. Yla-Herttuala, N.Miura, and A. Alitalo. 2004. Defective valves and abnormal mural cellrecruitment underlie lymphatic vascular failure in lymphedema distichiasis.Nat. Med. 10:974–981.

22. Prevo, R., S. Banerji, D. J. Ferguson, S. Clasper, and D. G. Jackson. 2001.Mouse LYVE-1 is an endocytic receptor for hyaluronan in lymphatic endo-thelium. J. Biol. Chem. 276:19420–19430.

23. Pure, E., and C. A. Cuff. 2001. A crucial role for CD44 in inflammation.Trends Mol. Med. 7:213–221.

24. Sabin, F. R. 1902. The lymphatic system in human embryos, with consider-ation of the morphology of the system as a whole. Am. J. Pathol. 1:367–389.

25. Schacht, V., M. I. Ramirez, Y. K. Hong, S. Hirakawa, D. Feng, N. Harvey, M.Williams, A. M. Dvorak, H. F. Dvorak, G. Oliver, and M. Detmar. 2003. T1alpha/podoplanin deficiency disrupts normal lymphatic vasculature forma-tion and causes lymphedema. EMBO J. 22:3546–3556.

26. Stacker, S. A., M. G. Achen, L. Jussila, M. E. Baldwin, and K. Alitalo. 2002.Lymphangiogenesis and cancer metastasis. Nat. Rev. Cancer 2:573–583.

27. Swartz, M. A., D. A. Berk, and R. K. Jain. 1996. Transport in lymphaticcapillaries. I. Macroscopic measurements using residence time distributiontheory. Am. J. Physiol. 270:324–329.

28. Tammela, T., T. V. Petrova, and K. Alitalo. 2005. Molecular lymphangio-genesis: new players. Trends Cell Biol. 15:434–441.

29. Valenzuela, D. M., A. J. Murphy, D. Frendewey, N. W. Gale, A. N. Econo-mides, W. Auerbach, W. T. Poueymirou, N. C. Adams, J. Rojas, J. Yasen-chak, R. Chernomorsky, M. Boucher, A. L. Elsasser, L. Esau, J. Zheng, J. A.Griffiths, X. Wang, H. Su, Y. Xue, M. G. Dominguez, I. Noguera, R. Torres,L. E. MacDonald, A. F. Stewart, T. M. DeChiara, and G. D. Yancopoulos.2003. High-throughput engineering of the mouse genome coupled with high-resolution expression analysis. Nat. Biotechnol. 21:652–659.

30. von Andrian, U. H., and T. R. Mempel. 2003. Homing and cellular traffic inlymph nodes. Nat. Rev. Immunol. 3:867–878.

31. Wigle, J. T., N. Harvey, M. Detmar, I. Lagutina, G. Grosveld, M. D. Gunn,D. G. Jackson, and G. Oliver. 2002. An essential role for Prox1 in theinduction of the lymphatic endothelial cell phenotype. EMBO J. 21:1505–1513.

32. Wigle, J. T., and G. Oliver. 1999. Prox-1 function is required for the devel-opment of the murine lymphatic system. Cell 98:769–778.

33. Yuan, L., D. Moyon, L. Pardanaud, C. Breant, M. J. Karkkainen, K. Alitalo,and A. Eichmann. 2002. Abnormal lymphatic vessel development in neuro-pilin 2 mutant mice. Development 129:4797–4806.

34. Zhou, B., J. A. Weigel, L. Fauss, and P. H. Weigel. 2000. Identification of thehyaluronan receptor for endocytosis (HARE). J. Biol. Chem. 275:37733–37741.

35. Zhou, B., J. A. Weigel, A. Saxena, and P. H. Weigel. 2002. Molecular cloningand functional expression of the rat 175-kDa hyaluronan receptor for endo-cytosis. Mol. Biol. Cell 13:2853–2868.



http://mcb.asm

.org/D

ownloaded from

http://mcb.asm.org/

Normal Lymphatic Development and Function in Mice Deficient for the Lymphatic Hyaluronan

Documents

Transcript of Normal Lymphatic Development and Function in Mice Deficient for the Lymphatic Hyaluronan