Normal Dendritic Cell Mobilization to
Lymph Nodes under Conditions of Severe
Lymphatic Hypoplasia
This information is current as
of June 18, 2017.
Andrew M. Platt, Joseph M. Rutkowski, Catherine Martel,
Emma L. Kuan, Stoyan Ivanov, Melody A. Swartz and
Gwendalyn J. Randolph
Supplementary
Material
References
Subscription
Permissions
Email Alerts
http://www.jimmunol.org/content/suppl/2013/03/25/jimmunol.120260
0.DC1
This article cites 53 articles, 24 of which you can access for free at:
http://www.jimmunol.org/content/190/9/4608.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2013 by The American Association of
Immunologists, Inc. All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
J Immunol 2013; 190:4608-4620; Prepublished online 25
March 2013;
doi: 10.4049/jimmunol.1202600
http://www.jimmunol.org/content/190/9/4608
The Journal of Immunology
Normal Dendritic Cell Mobilization to Lymph Nodes under
Conditions of Severe Lymphatic Hypoplasia
Andrew M. Platt,*,1 Joseph M. Rutkowski,†,2 Catherine Martel,*,‡ Emma L. Kuan,*,3
Stoyan Ivanov,‡ Melody A. Swartz,† and Gwendalyn J. Randolph*,‡
L
ymphatic vessels mediate clearance of macromolecules
and immune cells, such as Ag-transporting dendritic cells
(DCs), from peripheral tissues (1–3). Absorptive initial
lymphatic capillaries, consisting of a single layer of endothelial
cells that form blind-ended termini, are present in most organs (2).
These capillaries transition into collecting lymphatic vessels (2)
characterized by valves and specialized muscle cells (3). Col-
*Immunology Institute, Mount Sinai School of Medicine, New York, NY 10029;
†
Institut Interfacultaire de Bioingénierie, École Polytechnique Fédérale de Lausanne,
Lausanne 1015, Switzerland; and ‡Department of Pathology and Immunology, Washington University, St. Louis, MO 63110
1
Current address: Institute of Infection, Immunity and Inflammation, University of
Glasgow, Glasgow, United Kingdom.
2
Current address: Touchstone Diabetes Center, University of Texas Southwestern
Medical Center, Dallas, TX.
3
Current address: Department of Immunology, Benaroya Research Institute, Seattle,
WA.
Received for publication September 14, 2012. Accepted for publication February 22,
2013.
This work was supported by National Institutes of Health Grant HL096539 (to
G.J.R. and M.A.S. [dual principal investigators]), National Institutes of Health
Grant AI046953, and American Heart Association Grant 0740052 (to G.J.R.). Confocal
laser scanning microscopy was performed at the Mount Sinai School of Medicine–
Microscopy Shared Resource Facility, supported by National Institutes of Health–National Cancer Institute Grant 5R24 CA095823-04, National Science Foundation
Grant DBI-9724504, and National Institutes of Health Grant 1 S10 RR0 9145-01.
Address correspondence and reprint requests to Dr. Gwendalyn J. Randolph, Department
of Pathology and Immunology, Washington University School of Medicine, 660 South
Euclid Avenue, Box 8118, St. Louis, MO 63110. E-mail address: grandolph@path.
wustl.edu
The online version of this article contains supplemental material.
Abbreviations used in this article: eGFP, enhanced GFP; DC, dendritic cell; HEV,
high endothelial venule; i.d., intradermal(ly); LN, lymph node; LYVE-1, lymphatic
vessel endothelial hyaluronan receptor-1; MHC-II, MHC class II; NP, neuropilin;
SMA, smooth muscle actin; VEGF, vascular endothelial growth factor; VEGFR3,
vascular endothelial growth factor receptor 3; WT, wild-type.
Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1202600
lecting vessels contact the subcapsular sinus of the regional
lymph node (LN) and subsequently drain into efferent vessels and
eventually to the thoracic duct, where lymph is returned to venous
blood.
Many questions remain unaddressed or unanswered in the nascent field of lymphatic biology. Lymphedematous diseases typically target skin and stem from impaired lymphatic transport (4).
However, relatively little analysis has investigated immunological
alterations in lymphedema patients, including whether immune
cell transport to LNs is severely decreased, as might be expected.
Addressing this issue will promote a better understanding of the
array of defects that occur in these diseases, including lymphedema
associated with breast cancer therapy or filariasis where maintaining
immune defense is critical.
DCs enter the lymphatic vasculature through lymphatic capillaries (5–9), gaining access through “button-like” junctions found in
initial lymphatic capillaries (10). DCs preferentially seek out areas
along lymphatic capillaries that have sparse basement membrane
(11, 12). However, the dependence on lymphatic capillaries and the
overall density at which lymphatic capillaries must be maintained
for DC mobilization to occur has not been formally evaluated.
It is widely assumed that impaired lymphatic transport of macromolecules would be paralleled by impaired immune cell trafficking (13). In this report, we studied DC migration from skin to
LNs in a mouse model (Chy mice) bearing an inactivating mutation
in the tyrosine kinase domain of vascular endothelial growth factor
receptor 3 (VEGFR3), which is mutated in the form of primary
lymphedema called Milroy’s disease (14). Chy mice have a loss of
lymphatic transport from skin, reportedly because of a devoid lymphatic capillary network in the skin (15). We illustrate in this study
that body extremities in Chy mice are indeed devoid of lymphatic
capillaries, but body trunk skin retains lymphatic capillaries at ∼10%
normal density. This residual density was insufficient to sustain
normal lymphatic transport of macromolecules, but it was remarkably sufficient to permit normal DC migration from skin to LNs.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
To address the requirement for lymphatic capillaries in dendritic cell (DC) mobilization from skin to lymph nodes (LNs), we used
mice bearing one inactivated allele of vascular endothelial growth factor receptor 3 (VEGFR3) where skin lymphatic capillaries are
reported absent. Unexpectedly, DC mobilization from the back skin to draining LNs was similar in magnitude, and kinetics to
control mice and humoral immunity appeared intact. By contrast, DC migration from body extremities, including ear and forepaws,
was ablated. An evaluation in different regions of skin revealed rare patches of lymphatic capillaries only in body trunk areas where
migration was intact. That is, whereas the ear skin was totally devoid of lymphatic capillaries, residual capillaries in the back skin
were present though retained only at ∼10% normal density. This reduction in density markedly reduced the clearance of soluble
tracers, indicating that normal cell migration was spared under conditions when lymphatic transport function was poor. Residual
lymphatic capillaries expressed slightly higher levels of CCL21 and migration of skin DCs to LNs remained dependent on CCR7 in
Chy mice. DC migration from the ear could be rescued by the introduction of a limited number of lymphatic capillaries through
skin transplantation. Thus, the development of lymphatic capillaries in the skin of body extremities was more severely impacted
by a mutant copy of VEGFR3 than trunk skin, but lymphatic transport function was markedly reduced throughout the skin,
demonstrating that even under conditions when a marked loss in lymphatic capillary density reduces lymph transport, DC
migration from skin to LNs remains normal. The Journal of Immunology, 2013, 190: 4608–4620.
The Journal of Immunology
Areas of skin without lymphatic capillaries supported no DC trafficking as expected. Thus, it appears that the lymphatic capillary
density needed to sustain normal DC migration to LNs is much lower
than the density needed to maintain normal molecular transport.
4609
the slope of area under the curve versus time. This was considered proportional to the actual rate of Cy5–dextran clearance. Left and right assessments
were made at each site (ear or back) in each mouse, and the two normalized
values were averaged to generate one mean value per mouse per site.
Flow cytometry
Materials and Methods
Mice
FITC painting assay
Epicutaneous application of FITC to study DC migration was performed on
the ears and on two areas of each side of the mouse back skin as described
previously (17). Briefly, FITC (8 mg/ml) was dissolved in acetone and
dibutyl phthalate (Sigma-Aldrich) and applied in 25-ml aliquots. Recovered LNs were teased and digested in 2.68 mg/ml collagenase D (Roche)
for 25 min at 37˚C. Then, 100 ml 100 mM EDTA was added for 5 min, and
cells were passed through a 100-mm cell strainer, washed, counted, and
stained for flow cytometry.
Adoptive transfer of DCs
WT or CCR72/2 bone marrow-derived DCs, generated by culture in GMCSF supernatant (18), were pulsed overnight with green or red fluorescent
polystyrene beads (1-mm diameter; Polysciences), and 4 3 10 ml (total
1 3 106 cells) was injected s.c. in the shaved back skin overlying the inguinal LN. Alternatively, spleens from WT or CCR72/2 mice were treated
with collagenase D as above. Following red cell lysis, splenic CD11b+
cells were purified by MACS purification (Miltenyi Biotec), labeled with
Cell Tracker Orange (Invitrogen), and transferred at a 1:1 ratio (4 3 10 ml;
total 1 3 106 cells) into shaved back skin overlying the brachial LN.
Injections were made using a 26-gauge Hamilton syringe.
Monocyte-derived DC migration
Green or red fluorescent polystyrene beads described above were diluted to
0.1% (w/v), and 5 ml was injected intradermally (i.d.) into both ears or
front footpads or the shaved back skin or cheek skin overlying the region
of the brachial, inguinal, or auricular LN of Chy or WT mice. Three days
later, left and right brachial, inguinal, or auricular LNs were respectively
pooled from each injected mouse and analyzed by flow cytometry (19).
Lymphatic clearance of dextran
A total of 5 ml 1% 70-kDa TRITC-conjugated, lysine-fixable dextran
(Molecular Probes) was injected i.d. into ear and back dermis. 5 min after
injection, auricular and brachial LNs were excised and snap-frozen in OCT
(Tissue-Tek). To quantify lymphatic transport of dextran, we developed
a simple clearance assay and performed it on mice treated to remove hair
from ears and back skin 1 d earlier. A total of 1 ml Cy5–dextran (5000 kDa;
Nanocs) at a concentration of 2 mg/ml in sterile PBS was injected via a
Hamilton syringe i.d. into the ear, or 5 ml of the same tracer at 0.2 mg/ml
was injected into back skin. Fluorescence was observed through skin using
a fluorescence stereomicroscope (M205FA Leica), and images of the skin
were acquired each minute for 15 min using constant exposure time. Fluorescence intensity and exposure times were adjusted to ensure that intensity
values were linearly proportional to the actual fluorescence. Images were
processed using ImageJ and FiJi software, and the rate of clearance was
determined by first calculating the area under the curve of fluorescence intensity in the injection region at each time point, normalized to the initial
value. The normalized rate of fluorescence decay was then calculated from
Immunohistochemistry
Ten-micron cross-sections of WT/Chy back skin and LNs were fixed in 4%
PFA for 10 min, stained with primary Abs for 30 min and secondary Ab for
30 min, or sections were stained with Hema-3 solution (Fisher). For wholemount staining, ears were dissected, hair was chemically removed, and the
ear was split into dorsal and ventral sheets. Excess fat and cartilage was
removed under a dissection microscope, and the tissue was incubated in
0.5 M ammonium thiocyanate in H2O for 20 min before the epidermis was
peeled off. The ears were then fixed in 100% acetone for 5 min, followed
by 80% methanol for 5 min before blocking. For the GFP transplant
experiments, ears were fixed in 1% PFA/20% sucrose (Fisher) for 1 h to
preserve GFP signal. Both ears and 60-mm en face frozen sections of back
skin were blocked overnight and incubated with primary Ab overnight, and
then, secondary Ab was added for 2 h. Primary Abs were anti-mouse
podoplanin (Angiobio), goat anti-CCL21 (R&D Systems), rabbit anti–
lymphatic vessel endothelial hyaluronan receptor-1 (LYVE-1) (Abcam),
anti-smooth muscle actin (Sigma-Aldrich), rabbit anti-collagen IV (Abcam),
rat anti–VE-cadherin (BD Biosciences), anti-langerin, anti-CD3, and antiB220 (BD Biosciences). These were detected with aminomethylcoumarin
acetate, Cy2-, Cy3-, or Cy5-conjugated secondary Abs (Jackson ImmunoResearch). Tissue was imaged by confocal microscopy using a Leica SP5
DM, and three-dimensional reconstructions and isosurface rendering were
performed using Volocity software.
Quantification of lymphatic capillary density and CCL21
intensity
Thick sections of frozen back skin overlying the brachial LN were stained
as above with anti–LYVE-1 and anti-CCL21. Using Volocity software,
CCL21+ tissue maps were generated based on signal intensity, and the total
area of CCL21+ vessel coverage was quantified. Ten regions were imaged
per animal, and the mean of these readings was taken as the value for one
animal and eight animals per group. For local density, only the mean of
regions of Chy mice that contained any lymphatic capillaries were included
and compared with WT mice where all regions contained vessels. CCL21
intensity was measured using ImageJ (National Institutes of Health, Bethesda,
MD) and reported as mean normalized fluorescence intensity of CCL21+
lymphatic endothelial cells. Measurements were taken from three mice
per group.
Inflammation and immunization
To induce inflammation, 1 ml emulsified CFA (diluted 1:1 in PBS; SigmaAldich) was injected into the right ear dermis of WT and Chy mice using
a Hamilton syringe. In the left ear of the same animals, 1 ml PBS was
injected as a control. Ear thickness was measured using Absolute Digimatic calipers (Mitutoyo). For immunization, 4 mg OVA mixed with 25
mg Ultrapure 0111:B4 LPS (InvivoGen, Nunningen, Switzerland) or 50 mg
OVA in emulsified CFA (diluted 1:1) was injected into the back skin
dermis of Chy and littermate WT mice. Fourteen to 21 d later, serum was
harvested and assessed for total anti-OVA IgG Ab by ELISA as described
previously (20). Fold increase in reactivity (OD) was normalized by dividing the OD value from various dilutions of plasma from the immunized
mouse by the OD reading in the nonimmunized state.
Skin digestion for flow cytometry
Four hours after injection of emulsified CFA into the ear or back skin
dermis, the skin was excised and cut into small segments using scissors. This
was then incubated in digestion solution (Liberase [1.75 mg/ml] and 2%
FCS in RPMI 1640 medium) for 25 min at 37˚C.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Male Chy (heterozygote) mice on a mixed C3H background, obtained from
the Medical Research Council Mammalian Genetics Unit Embryo Bank
(Harwell, U.K.), were crossed with wild-type (WT) littermates to obtain
heterozygote Chy offspring. Chy mice were also crossed 10 times with
C57BL/6J mates (The Jackson Laboratory). Experiments were performed in
6- to 12-wk-old Chy mice on both backgrounds, and no differences were
observed other than a reduced frequency (∼10% compared with ∼50%) of
mutant offspring in the C57BL/6 colony. Differences between sexes were also
not observed. The Chy mutation was identified by PCR using 59-GAAGACCTTGTATGCTAC-39/59-AGGCCAAAGTCGCAG-AA-39 primer
sequences. CCR72/2 mice were obtained from Jackson Laboratories, mice
expressing enhanced (e)GFP under the control of the b-actin promoter
were provided by M. Merad (Mount Sinai School of Medicine, New York,
NY), and K14-VEGFR3-Ig (16) were provided to M.A.S. by K. Alitalo
(University of Helsinki, Helsinki, Finland). Mice were housed in a specific
pathogen-free environment and were used in accordance with protocols
approved by animal welfare oversight committees at Mount Sinai, École
Polytechnique Fédérale de Lausanne, or Washington University.
Anti-CD11b, anti-CD11c, anti-CD8a, anti-Ly6G, and anti–I-A/I-E mAbs
were from BD Biosciences. Conjugated isotype-matched control mAbs
were obtained from eBioscience or BD Biosciences. Intracellular staining
for CD207 was performed after cells were fixed and permeabilized with
Cytofix/Cytoperm solution (BD Biosciences) for 20 min at 4˚C and then
washed in Perm/Wash buffer (BD Biosciences) and incubated for 20 min at
4˚C with control goat IgG (R&D Systems) or goat anti-langerin Ab (Santa
Cruz Biotechnology) diluted 1:400 in the Perm/Wash buffer. Cells were
washed twice in Perm/Wash buffer and incubated with 1:300 anti-goat
FITC Ab (Invitrogen) diluted in Perm/Wash buffer for 20 min at 4˚C.
4610
Skin transplantation model
Mice were anesthetized, and the ear skin was cleaned. In the donor animal,
a 0.25-cm incision was made in the dorsal side of the ear, and a small region of
skin was removed. After an identical area of ear skin was removed from the
recipient mice, the donor skin was sutured, using 9-0 polyamide monofilament sutures, onto the recipient ear. This was conducted on both ears of the
same animal. As a control for the procedure, a “sham surgery” group of mice
had an incision made on both ears. After a period of 14 d, 1-mm microspheres were injected into the ear dermis adjacent to the transplant, and the
mice were sacrificed 3 d later. To assess lymphatic capillary density, five
regions (one within the transplant site) of the ear were imaged per animal
and the mean of these readings was taken as the value for one animal.
Statistical analysis
Statistical comparisons were made by using the Student t test. One-way
ANOVAs were used for multiple comparisons, with a Tukey posttest. The
nonparametric Kruskal–Wallis test was used to analyze the aggregate data
in the humoral immune response, and a two-way ANOVA was used for
repeated measures in the footpad swelling experiments. All experiments
contained three or more replicate mice per experimental parameter.
Normal proportions of lymph-trafficking DCs in some LNs of
Chy mice
Skin-draining LNs, particularly the inguinal LN, of Chy mice exhibited reduced cellularity compared with WT littermates (Fig.
1A), and the popliteal LN, draining the footpad, which had the
most overt lymphedema (15), could not be found in most Chy
mice. CD11chi LN DC subsets can be broadly divided into “resident” and “lymph-migratory” DCs based on the expression of
higher levels of MHC class II (MHC-II) on the latter (21). LNresident DC subsets arrive to LNs through high endothelial venules (HEVs) (22) rather than lymphatics (23). Such resident DCs
were present in normal proportions in all Chy LNs examined,
including the auricular (which drains the ear), brachial (which
drains the upper back skin and front paws), and inguinal (which
drains the lower back and belly skin) LNs of Chy mice (data not
shown). Unexpectedly, the proportions and numbers of lymphmigratory MHC-IIhi DCs were also similar between WT and
Chy mice in the brachial and inguinal LNs draining the trunk skin
(Fig. 1B; data not shown), but their frequency was reduced in the
ear-draining auricular LN (Fig. 1B). As a distinct approach to
identifying lymph migratory DCs, we also analyzed LNs for the
frequency of CD207 (Langerin)+ DCs. CD207 is expressed only
on lymph-migratory DCs, including epidermal Langerhans cells
and CD207+ dermal DCs (24–26). CD207+ DCs were present in
all T cell zones of skin-draining Chy LNs (Fig. 1C). As for MHCIIhi DCs, proportions and numbers of CD207+ DCs were reduced
in the auricular LN of Chy mice relative to WT littermates but
were similar in the brachial and inguinal LNs (Fig. 1B, 1D, 1E;
data not shown). However, CD207 can be expressed by CD8a+
DCs (27), which may be LN-resident DCs, and thus, we assessed
CD8a expression by this population. In this study, we found that
all CD207+ DCs expressed CD8a in both genotypes (Fig. 1F);
however, the majority of CD8a+ DCs (60–80%) failed to express
CD207 (data not shown). Furthermore, because DCs constitute a
minor proportion of the LN cellularity, we examined whether the
reduced cellularity of Chy LNs reflected a reduction in numbers of
other cell types. Thus, we assessed the total number of B and
T lymphocytes in skin-draining LNs in Chy mice, shown in Table
I. In this study, we found that the numbers of all subsets of
lymphocytes, including B cells and CD4+ and CD8+ T cells,
were reduced in Chy mice. These results starkly contrasted with
findings that MHC-IIhi and CD207+ lymph-migratory DCs are
completely absent from all skin-draining LNs in K14-VEGFR3-Ig
mice (Fig. 1G) (20), which, like Chy mice, have been reported to
lack lymphatic capillaries in skin and exhibit reduced LN cellularity (16). Thus, against our expectations, lymph-migratory DCs
were not absent in any Chy LNs and were capable of normally
populating several Chy LNs.
No kinetic delay in DC migration to LNs in Chy mice
Although lymph migratory DC frequency in Chy LNs was normal in
resting LNs draining the back skin, it remained possible that migration of DCs to Chy LNs was slower than under normal conditions. Thus, we set out to quantify DC migration from skin to draining LNs and assess its kinetics in Chy mice. We used the established
“FITC painting” assay that allows for robust assessment of DC migration with known kinetics. Peak accumulation of FITC+ DCs in
LNs occurs at 18 h, with DC arrival just getting under way at 12 h
(17, 28). At 18 h after painting the back skin and ear, accumulation of
FITC+ DCs was nearly abrogated in the ear-draining auricular LNs of
Chy mice, but surprisingly, similar proportions of DCs appeared in
brachial and inguinal LNs of Chy and WT mice (Fig. 2A, 2B). The
total number of migrated DCs in the inguinal, but not brachial, LN
was reduced in Chy mice (data not shown), reflecting our findings
that several LNs were reduced in overall cellularity (Fig. 1A) and that
DC migration to LNs from lymphatics is typically maintained in
a manner proportional to LN cellularity (29).
Even at 12 h after FITC application, the time of earliest appearance of FITC+ DCs in LNs, no significant differences existed
between Chy and WT mice (Fig. 2C). In a second migration assay,
we quantified the mobilization of monocyte-derived DCs from
skin to LNs in Chy mice by assessing their transport of i.d.
injected, 1-mm fluorescent beads (19, 30). Intradermal 1-mm
beads do not freely flow to the LN but are carried there by
CD11cloCD11bhiMHC-IIhi DCs (19). Three days after bead injection i.d., bead+ DCs were absent in the auricular LN of Chy
mice, but similar proportions and numbers of bead+ DCs were
found in brachial and inguinal LNs of Chy and WT mice (Fig. 2D,
2E; data not shown), as seen after FITC skin painting. Thus, DC
mobilization to draining LNs appeared to be normal in Chy mice
in a highly region-dependent manner.
Surprised to detect migration of DCs to trunk skin-draining LNs,
we explored whether DC migration in Chy mice involved CCR7, the
master regulator of DC migration to and within WT LNs (21, 31, 32).
We thus competed CD45.1+ WT and CD45.2+CCR72/2 splenic
CD11b+ cells, labeled with Cell Tracker Orange, after they were
cotransferred i.d. into the back skin of CD45.2+ WT or CD45.2+ Chy
mice. Gating on Cell Tracker Orange+ transferred cells in the brachial LN revealed that donor cells were exclusively derived from WT
mice, demonstrating DC mobilization in Chy mice is CCR7 dependent (Fig. 2F, 2G). To confirm this result, we labeled WT and
CCR72/2 DCs with respectively different colored beads, injected
them s.c., and assessed their migration. Although migration from the
s.c. injection site up toward the dermis occurred independently of
CCR7 in both genotypes (Fig. 2H), only WT DCs progressed to the
LN (Fig. 2I). Thus, using three different assays that examined endogenous and adoptively transferred DCs, we conclude that DC migration is kinetically normal and dependent on CCR7 from the back
skin of Chy mice. From the ear, however, DC migration is abrogated.
Regionalized lymphatic defects dictate DC mobilization
Given that some lymph-migratory DCs still populated steady-state
auricular LNs of Chy mice (Fig. 1C–E), but migration of DCs
from the ear that drains to this LN was ablated (Fig. 2B, 2E), the
origin of these DCs in the auricular LN was unclear. Thus, we
addressed whether the auricular LN drained other sites that may
have intact lymphatic transit of DCs. To this end, we compared
endogenous DC migration to the auricular LN from the ear and
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Results
LYMPHATIC CAPILLARIES AND DC MIGRATION
The Journal of Immunology
4611
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 1. Normal proportions of lymph-trafficking DCs in some LNs of Chy mice. (A) Total cellularity was measured in the auricular, brachial, and
inguinal LNs of WT and Chy mice. (B) LN preparations from WT and Chy mice were analyzed for CD11c and MHC-II expression, and the proportion of
CD11c+MHC-IIhi cells is shown. n = 6–7 mice. (C) Immunofluorescence images of LNs stained for CD207 (green), CD3 (red), B220 (pink), and LYVE-1
(blue). Scale bar, 12 mm. Representative dot plots (D) and proportion (E) of CD207 expression by CD11c+ cells in skin-draining LNs. (F) CD8a expression
among CD11c+CD207+ cells in brachial LNs of Chy and WT mice. (G) Proportion of CD11c+MHC-IIhi DCs from pooled brachial, inguinal, and axillary
LNs from WT and K14-VEGFR3-Ig mice. s, WT mice; d, Chy or K14-VEGFR3-Ig mice. Each circle shows one mouse. n = 3–7 mice. Results are from at
least three independent experiments. *p , 0.05, **p , 0.01, ***p , 0.001.
cheek skin sites simultaneously using green and red beads, respectively, deposited i.d. in the ear or cheek. We found both ear
and cheek bead-transporting DCs in WT littermate auricular LNs
but only cheek-derived bead+ DCs in the auricular LNs of Chy
mice (Fig. 3A, 3B). We then carried out a similar approach to examine DC migration from the forepaw extremities. In this study,
we injected the two differently colored beads, respectively, in the
back skin or front footpad and assessed the brachial LN for migratory bead+ DCs. Bead+ DCs derived from both peripheral sites
were observed in WT LNs; however, only DCs derived from the
back skin and not the front footpad were observed in Chy mouse
brachial LNs (Fig. 3C, 3D). Thus, profound defects in DC migration in Chy mice are confined to regionalized areas, notably the
body’s extremities, limbs, and ears, but migration remains intact
when DCs originate from the body trunk and facial skin.
Rare lymphatic capillaries exist in the back skin dermis of Chy
mice
We next analyzed distinct anatomical regions for lymphatic capillaries
by microscopy. Because macrophages can also express LYVE-1,
4612
LYMPHATIC CAPILLARIES AND DC MIGRATION
Table I. Numbers of different immune cell populations in skin-draining LNs of Chy mice
Auricular LN
WT
+
CD4 T cells
CD8+ T cells
B cells
Migratory CD207+ DCs
Migratory CD2072 DCs
LN-resident DCs
902
946
1512
4.3
10.4
63.8
6
6
6
6
6
6
191.8
126.0
391.4
0.1
3.0
6.0
Brachial LN
Chy
400
406
838
0.9
3.8
32.5
6
6
6
6
6
6
31.3
83.0
202.5
0.3
1.0
7.7
WT
538
595
711
4.7
8.9
75.3
6
6
6
6
6
6
78.5
125.2
152.2
1.1
2.3
7.0
Inguinal LN
Chy
318
271
580
2.4
5.4
46.0
6
6
6
6
6
6
177.3
114.4
295.8
1.2
2.5
7.3
WT
394
423
536
4.0
6.6
65.2
6
6
6
6
6
6
32.8
69.7
110.6
1.4
2.3
11.1
Chy
75
79
212
1.5
2.0
19.0
6
6
6
6
6
6
24.3
26.6
88.1
0.3
0.7
3.8
Skin-draining auricular, brachial, and inguinal LNs from Chy and WT mice were examined for the presence of CD4+ T cells (CD4+), CD8+ T cells (CD8+), B cells (B220+),
migratory CD11c+CD207+ DCs, migratory CD11c+CD2072 DCs (MHChi but CD2072) and LN-resident DCs (CD11c+MHClow-intermediate). Numbers (103) 6 SEM are shown.
Data are from two independent experiments.
Molecular transport from back skin lymphatic vessels of Chy
mice is impaired
The observation that DC migration was normal under conditions
of severe lymphatic hypoplasia led us to wonder whether the
rare lymphatic vessels in Chy mice had developed compensatory
mechanisms to restore transport function in general. However,
button-like junctions and portals appeared normal (Fig. 5A–D),
suggesting no major compensatory changes at this level. Qualitatively, soluble TRITC–dextran transport from the ear was ablated in Chy mice but still occurred from the back skin (Fig. 5E,
5F). To quantify transport of macromolecules from the skin, we
injected Cy5–dextran into ear or back skin and monitored its
clearance over time at the same sites from which we earlier
injected DCs. The median normalized rate of clearance was more
profoundly impaired from the ear than from the back skin (Fig.
5G), but transport from the back skin was nonetheless less than
half the rate observed in WT littermates (Fig. 5G). Thus, Chy
lymphatic vessels do not develop a fully sufficient compensatory
mechanism at the level of macromolecular transport to overcome
severe lymphatic hypoplasia.
Classical inflammation drives limited lymphangiogenesis but
fails to rescue DC mobilization in Chy mice
Because excess levels of vascular endothelial growth factor
(VEGF)-C can rescue the loss of lymphatic capillaries in the ear of
Chy mice (15), and VEGF-C levels are increased during inflammation (34, 35), we assessed whether a classical inflammatory
stimulus could induce lymphangiogenesis and DC migration in
mutant mice by injection of a small volume of CFA into the ear
dermis. CFA is upstream of lymphangiogenesis around LNs and
enhances DC migration (29, 36). We first assessed whether inflammatory cell recruitment to inflamed ear or back skin occurred
normally in Chy mice and found that, 4 h after CFA injection,
neutrophils could be found in similar proportions and numbers to
WT mice in both sites (Fig. 6A, 6B, Supplemental 2A, 2B).
Neutrophils can also enter afferent lymph and arrive in LNs rapidly (37), although other neutrophils may arrive through HEVs.
When we assessed the accumulation of neutrophils in the brachial
and auricular LNs at 4 h, the frequency of neutrophils in both skindraining LNs was similar to WT mice (Supplemental Fig. 2C–F),
suggesting that neutrophil migration in response to an inflammatory stimulus was intact.
Following CFA administration, we observed tissue edema in
the ear in both genotypes as expected. However, this inflammation
was notably increased and resolved more slowly in Chy mice (Fig.
6C). Moreover, the expected lymphadenopathy that occurs in WT
mice treated with CFA (29) failed to occur in draining auricular
LNs of Chy mice (Fig. 6D), and this was similar in the brachial
LN after CFA injection in the back skin (data not shown). Lymphatic density was increased in the ears of WT mice treated with
CFA, but this failed to reach significance (Fig. 6E). Despite the
strong inflammatory response, there was a significant but minor
increase in lymphatic density in Chy mutant mice, with some
isolated vessels present in the ears of some mice (Fig. 6E; data not
shown). However, DC mobilization from the ear remained abrogated in Chy mice 15 d after CFA administration (Fig. 6F, 6G).
Thus, induction of general inflammation by CFA injection is not
sufficient to promote restorative lymphangiogenesis in regions of
Chy mouse skin that are devoid of lymphatics.
Humoral immunity is diminished in K14-VEGFR3-Ig mice
following dermal immunization at sites where DC migration is
entirely abolished (20). When Chy and WT littermates were immunized with OVA mixed with LPS or CFA in the back skin
dermis or the footpad and assessed for anti-OVA IgG Abs in the
serum 14–21 d later, footpad immunizations where there are no
lymphatic vessels led to abrogated induction of anti-OVA IgG
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
tissues stained for LYVE-1 (Fig. 4A, 4B) were costained with
podoplanin, CCL21, and/or Prox-1 to identify lymphatic capillaries
(Fig. 4C, 4D; data not shown). Indeed, LYVE-1+ lymphatic capillaries were completely absent in the ear dermis (Fig. 4A), as reported previously (15). LYVE-1+ lymphatic capillaries were also
not seen in the back skin dermis of Chy mice by cross-section (Fig.
4B), where the subadipose layer was markedly thinned (Supplemental Fig. 1A). However, all Chy skin-draining LNs examined
contained LYVE-1+ terminal lymphatic capillaries (Supplemental
Fig. 1B) (1).
In thick en face sheets of back skin or whole-mount preparations
of ear dermis that allowed a larger region to be analyzed, uniform
lymphatic capillary networks were observed in WT mice (Fig. 4C,
left panel). Most regions of back skin in Chy mice were devoid of
lymphatic capillaries (Fig. 4C, middle panel) and in the ear entirely,
but sparse clusters were present in back skin (Fig. 4C, right panel).
These vessels still expressed the CCR7 ligand CCL21 (Fig. 4D) as
in WT mice (12). Quantification of lymphatic capillary density
revealed a ∼90% reduction in density in Chy dermis (Fig. 4E).
Even in scattered clusters of lymphatic vessels, the local density
was reduced by ∼60% compared with WT (Fig. 4F). CCL21 intensity was modestly elevated in Chy mice (Fig. 4G). Thus, regions
of skin in Chy mice that support migration of skin DCs contain very
rare networks of lymphatic capillaries, whereas the regions of skin
containing no lymphatic capillaries allow no transit of DCs from
skin. We also examined lymphatic precollecting/collecting vessels,
which are characterized by smooth muscle coverage and little or no
LYVE-1 expression (33). By whole-mount imaging, we observed
podoplanin+ LYVE-12 smooth muscle actin (SMA)+ collecting
vessels in the ear and back skin dermis of both WT and Chy mice
(Fig. 4H, 4I). These collectors were less frequent in the ear and
back skin dermis of Chy mice and most appeared blind-ended
without connection to LYVE-1+ capillaries (data not shown).
The Journal of Immunology
4613
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 2. No kinetic delay in DC migration to LNs in Chy mice. (A) Representative dot plots of FITC+CD11c+ cells in skin-draining LNs of WT
and Chy mice 18 h after FITC application to the dorsal ear and upper and lower back skin. (B) Percentage of CD11c+FITC+ cells in LNs of WT and Chy mice.
n = 5–7. (C) Percentage of CD11c+FITC+ cells in brachial LNs of WT and Chy mice 12 h after FITC application to upper back skin. (D) Representative dot
plots of MHC-II+bead+ cells in skin-draining LNs of WT and Chy mice 3 d after i.d. injection of 1 mm green fluorescent beads in the ear and back skin. (E)
Percentage of green bead+ cells in LNs of WT and Chy mice. s, WT mice; d, Chy mice. n = 4–7. (F) Spleen CD11b+ cells from CD45.1 WT and CD45.2
WT (left panel) or CD45.2 CCR72/2 (middle and right panels) mice were labeled with Cell Tracker Orange and mixed at 1:1 ratio and then injected into the
upper back skin dermis of WT/Chy CD45.2+ mice. At 18 h, CD45.1 and CD45.2 expression was examined in brachial LNs on total Cell Tracker Orange+
cells (F). (G) Results are shown as the mean percentage 6 SD of Cell Tracker Orange+ cells that were CD45.1+ versus CD45.2+ in WT and Chy brachial
LNs. (H) WT (red) and CCR72/2 (green) bone marrow-derived DCs were pulsed overnight with 1-mm beads, stimulated with (Figure legend continues)
4614
LYMPHATIC CAPILLARIES AND DC MIGRATION
FIGURE 3. Regionalized lymphatic defects dictate DC mobilization. (A) Representative dot plots of cells in the draining
auricular LN 3 d after injection of red
beads into the cheek skin dermis and green
beads into the ear dermis. (B) Percentage
of red bead+ cells (cheek skin) and green
bead+ cells (ear) in the auricular LN. (C)
Representative dot plots of red bead+ and
green bead+ cells in the draining brachial
LN 3 d after injection of red beads into the
upper back skin dermis and green beads
into the front footpad dermis. (D) Percentage of red bead+ cells (back skin) and
green bead+ cells (footpad) in the brachial
LN. Results depict three independent
experiments. s, WT mice; d, Chy mice.
n = 10–11; ***p , 0.001.
DC mobilization from regions with no previous drainage can
be rescued by skin transplantation
Finally, we assessed whether we could rescue DC mobilization from
the most severely affected regions by directly introducing a limited
number of lymphatic capillaries. Specifically, we attempted to introduce lymphatic capillaries by transplantation of a small region of
WT (or Chy as a negative control) ear skin onto the ear of Chy mice
(Fig. 7A). We waited 14 d and then assessed lymphatic density and
DC mobilization. Lymphangiogenesis was induced in the ear even in
regions beyond the transplant area (regions 2–5 in Figure 7A, 7B)
and to a limited degree in a proportion of Chy recipients that received
Chy mutant donor skin (Fig. 7C; data not shown). However, the
density of lymphatic capillaries in transplant recipients of WT donor
skin was still less than half the density observed in unmanipulated
WT mice (Fig. 7C). The superior lymphangiogenesis in Chy mice
transplanted with WT skin indicated that the donor vessels expand
rather than the surgery driving endogenous lymphatic expansion.
However, to assess this directly, we transplanted dorsal ear skin from
b-actin–eGFP mice onto Chy recipients and assessed the lymphatic
network for host versus donor origin 2 wk later. In this study, we
found that the majority (61%) of the LYVE-1+ capillaries, both
within and outside the transplanted region, were of donor origin (Fig.
7D, 7G; data not shown). However, we could observe a lower
number (24%) of recipient eGFP2LYVE-1+ vessels (Fig. 7E, 7G)
and some vessels that appeared to be derived from both donor and
recipient endothelium (15%) (Fig. 7F, 7G). However, because the
donor is not a lymphatic-specific reporter, it remains possible that
the few GFP+ cells in Fig. 7F are not lymphatic endothelial cells.
By i.d. injecting beads adjacent to the transplanted region, we
assessed DC migration following skin transplantation or a sham
operation (Fig. 7H–J). WT skin transplanted onto Chy recipients
increased migration significantly over mice transplanted with
mutant skin and sham-operated mice (Fig. 7I, 7J), and indeed,
when we compared transplant or sham transplant data to the mi-
gration observed in Fig. 2, we found that transplant of WT onto
Chy mice restored migration significantly enough that the difference to unmanipulated WT mice was no longer observable. However, migration in the sham-operated mice or recipients of Chy
donor skin remained significantly reduced relative to that in WT
mice. Although even transplantation of Chy skin onto Chy mice
appeared to restore migration and induce lymphangiogenesis in
some mice, WT skin used in the transplant was more often effective. Indeed, if we defined restoration of migration as migration that
was quantitatively within 1 SD of the mean DC migration in unmanipulated WT mice, WT skin transplanted onto Chy mice rescued migration robustly in 64.7% of the recipients, whereas Chy
skin transplanted onto Chy mouse ears did so only 21.4% of the
time, and sham was minimal at 9.1% (data not shown). To ensure
that the VEGFR3 mutation had no effects other than on the formation of initial lymphatic capillaries, we transplanted Chy skin
onto WT recipients. In this study, we found that both lymphatic
density and DC mobilization were comparable to unmanipulated
WT mice (data not shown). Taken together, these data indicate that
transplantation effectively restores DC migration.
Discussion
As anticipated, our data suggest that the presence of lymphatic
capillaries in skin is essential for skin DC mobilization to LNs.
However, contrary to expectations, even rare, sparsely localized
lymphatic capillaries can be sufficient to support DC mobilization
without obvious kinetic delay or reduction in magnitude. Steadystate migration of lymph-trafficking MHC-IIhi or CD207+ DCs
from the back skin, but not the ear, in Chy mice was similar to WT
mice. However, CD207 can be expressed by CD8a+ DCs (27),
which may be LN-resident DCs seeding the LN through HEVs.
We found that all CD207+ DCs expressed CD8a in both genotypes
(Fig. 1F); however, the majority of CD8a+ DCs (60–80%) failed
to express CD207. These data suggest that CD8a+CD2072 DCs
are the LN-resident DCs that enter LNs as precursors through
HEVs (22), whereas CD207+CD8a+ DCs likely correspond with
lymph-migratory DCs. Indeed, CD8a expression by lymph-migratory
100 ng/ml LPS for 5 h, and injected s.c. in the back skin over the inguinal LN. At 16 h, inguinal skin and LNs were collected, and the migration of bead-labeled
cells toward CCL21+ (white) lymphatic capillary-rich dermis was examined (H). (I) Images of the draining inguinal LN in both WT and Chy mice 16 h after
injection of bead+ DCs. CCL21 staining (white); DCs that have been loaded with red flourescent beads (red); DCs that have been loaded with green
fluorescent beads (green). Results derive from two to four independent experiments. Scale bar in (H) and (I), 80 mm. n = 4. **p , 0.01, ***p , 0.001.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
(Fig. 6H). However, there was no significant difference in antiOVA IgG generated from immunizing Chy mice in the back skin
relative to WT littermates (Fig. 6I).
The Journal of Immunology
4615
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 4. Rare lymphatic capillaries exist in the back skin dermis of Chy mice. (A) Whole-mount staining of ear dermis with LYVE-1 (green) and
smooth muscle actin (SMA; red). Scale bar, 60 mm. (B) Cross-sections (10 mm) of back skin from WT and Chy mice stained for LYVE-1 (green) and SMA
(red). Scale bar, 60 mm. (C) Podoplanin (green), LYVE-1 (white), and SMA (red) staining of en face sections of back skin from WT (left panel) and Chy
(middle and right panels) mice. Scale bar, 150 mm. (D) CCL21 staining of lymphatic capillaries in back skin sections from WT and Chy mice. Scale bar,
150 mm. (E) Quantification of lymphatic capillary density. (F) Local density of lymphatic capillaries examining only the areas of Chy back skin that
contained vessels. Results are from eight mice per group. (G) Normalized mean fluorescence intensity of CCL21 staining. Measurements were taken from
three mice per group. (H and I) Whole-mount images of the ear dermis (H) and en face images of back skin dermis (I) of WT and Chy mice show podoplanin
(green), LYVE-1 (white), and SMA (red) staining. Scale bar, 32 mm (H) and 38 mm (I). Note that precollectors are podoplanin+LYVE-12 and exhibit partial
SMA coverage. *p , 0.05, ***p , 0.001.
DCs recently has been demonstrated in the mesenteric afferent lymphatics (38). Furthermore, migration of DCs after contact sensitization and migration of monocyte-derived DCs was also identical to
WT mice from the back skin.
The major difference between skin of the extremities versus that
of the body trunk was the presence of rare lymphatic capillaries in
the trunk skin compared with their complete absence in extremities
such as the ear. There were no obvious differences in button-like
4616
LYMPHATIC CAPILLARIES AND DC MIGRATION
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 5. Molecular transport from back skin lymphatic vessels of Chy mice is impaired. (A and B) Images show staining with LYVE-1 (green) and
VE-cadherin (red) on lymphatic capillaries in the back skin dermis of WT and Chy mice. Scale bar, 24 mm. (B) Enlargement of inset in (A) showing that the
structural pattern is highlighted more clearly after isosurface rendering of the image. Scale bar, 5.2 mm (left panel) and 5.08 mm (right panel). (C) Images
show staining of back skin lymphatic capillaries with collagen IV (white) and CCL21 (red) in WT and Chy mice. Scale bar, 24 mm. (D) Enlargement of
inset in (C). Yellow arrowheads indicate lymphatic portals. Scale bar, 11.1 mm (left panel) and 12 mm (right panel). (E and F) Sections of auricular (E) and
brachial (F) LNs 5 min after injection of 50 mg TRITC–dextran (red) into the ear and back i.d. Counterstained with DAPI (blue). Scale bar, 60 mm. (G)
Lymphatic transport measured in the ear and back skin. Each symbol represents pooled left and right data from a single mouse, and the bar shows the
median normalized rate of fluorescence decay. Measurements are from one to four mice per group in two independent experiments. **p , 0.01, assessed
using a one-tailed Student t test.
junctions (Fig. 5A, 5B) or portals (Fig. 5C, 5D) of residual lymphatic capillaries in Chy mice, but these vessels displayed slightly
more CCL21. Although this increase may augment DC migration
in the face of lymphatic hypoplasia, the increase may be too
modest to be considered a compensatory change. Furthermore, it
is unlikely that prelymphatic fluid channels (2) evolved to greater
efficiency in Chy mice, because such a modification would be
expected to translate into improved lymphatic transport of macromolecules. Instead, we found that clearance of soluble lymphatic tracers was markedly reduced in Chy mice, arguing against
compensatory improvements in processes associated with both
molecular and cellular transport. Thus, based on the present
findings, we propose that lymphatic capillaries in WT mouse skin
evolved to the density they did to allow for optimal molecular
transport and that the threshold of lymphatic density needed to
support the migration of actively motile DCs, and perhaps other
immune cells, is far lower. This conclusion, although unexpected,
now seems logical given that macromolecular transport through
lymphatic vessels occurs on a very different time scale (minutes)
to cellular transport (hours). That the rate of DC migration is
The Journal of Immunology
4617
regulated by other features of lymphatic vessels such as adhesion
of DCs within the lumen of lymphatic capillaries (12) is consistent
with the concept that DC migration is rate-limited by different
processes than molecular transport.
Chy LNs draining the skin exhibited a reduced cellularity in
the face of lymphatic hypoplasia. That the popliteal LN, which
drains only the rear footpad and not the body trunk where lymphatics are at least partly intact (39), is actually absent in Chy
mice suggests that functional lymphatics are needed for a LN to
form properly or be sustained. Although LN anlagen may be Prox1 and therefore lymphatic vessel independent (40), maintenance
of LN integrity may require lymph flow. Indeed, this possibility is
consistent with reduced LN size when afferent lymphatics are
surgically severed (41), which also results in altered HEV maturation and T–B cell compartmentalization. Furthermore, DCs,
including those from the lymph, coordinate LN T cell homeostasis
at least in part through enhancing HEV formation (42–44). However, critical elements derived from lymph likely include more than
migratory DCs because skin-draining LNs still form in K14VEGFR3-Ig mice that lack lymph-migrating DCs (Fig. 1G) (20).
Indeed, we observed that the typical lymphadenopathy associated with immunization failed in Chy mice. Although DCs have been
linked to lymphadenopathy, our data suggest that intact DC migration may not be sufficient to coordinate lymphadenopathy in Chy
mice. Earlier work from our laboratory revealed that lymphangiogenesis was associated with and required for lymphadenopathy
(29). Thus, the failure of Chy mice to exhibit lymphadenopathy may
be due to attenuated signaling of VEGFR3 by lymphatic endothelial
cells. In the face of smaller baseline LNs in the Chy mouse, and
their inability to exhibit lymphadenopathy in response to immunization, we find normal humoral immune responses at sites where
DC migration remains proportional to overall LN cellularity, a result that is distinct from the impaired humoral responses observed in
mice devoid of lymphatic capillaries in skin and devoid of lymphtrafficking DCs. This finding, in our view, highlights the greater
importance of LN cellular composition, such as balanced ratios
between lymph-homing and resident DCs and lymphocytes, than
overall cellularity in driving an effective immune response.
Chy mice carry a dominant-negative mutation in one allele. They
serve as a useful tool to examine the VEGFR3 axis in lymphatic
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 6. Classical inflammation drives limited lymphangiogenesis but fails to rescue DC mobilization in Chy mice. (A and B) Four hours after PBS
(N) or CFA (n) injection in the ear (A) or back skin (B), skin was digested, and the total number of live CD45+CD11b+Ly6G+ neutrophils is shown. (C)
Graph depicts ear thickness of WT and Chy mice after injection of PBS or CFA into the ear dermis. (D) Plot shows the total cellularity of the draining
auricular LN 18 d after CFA treatment. (E) Plot depicts the area of the ear dermis covered by LYVE-1+ vessels 18 d after injection of PBS or CFA. (F and G)
Results shown are the proportion (F) and total number (G) of green bead+ cells present in the auricular LN of WT and Chy mice injected with green
fluorescent beads on day 15 postinjection with PBS or CFA. LNs were harvested 3 d after injection of green beads. s, WT mice; d, Chy mutant mice. Each
circle represents one mouse. (H and I) Four micrograms of OVA mixed with 25 mg Ultrapure 0111:B4 LPS (InvivoGen) or 50 mg OVA in emulsified CFA
(diluted 1:1) was injected into the footpad (H) or back skin dermis (I) of Chy and littermate WT mice. Fourteen to twenty-one days later, serum was
harvested and assessed for total anti-OVA IgG Ab by ELISA. Results show the OD value of plasma at various dilutions. Results are representative of two
independent experiments. n = 7–10. *p , 0.05, **p , 0.01, ***p = 0.001.
4618
LYMPHATIC CAPILLARIES AND DC MIGRATION
biology, because mutations in Milroy patients with primary lymphedema are found in the vegfr3 gene (15). The regionalized
lymphatic defects in Chy mice are reminiscent of primary lymphedema patients, where body extremities, particularly lower extremities, are typically most affected (45, 46). The regional differences in lymphatic capillaries of the Chy mouse also adds a
layer of complexity to the role of VEGFR3 in lymphatic development and highlights differences in anatomic regions even within
the skin. Chy-3 mice, which are haploinsufficient for vegfc, also
exhibit a regionalized form of lymphedema primarily affecting the
lower limbs (47). VEGFR3 signaling is thought to be modulated
through the coreceptor neuropilin (NP)-2 in the control of anatomic differences in lymphatic hypoplasia (15). NP-2, which binds
VEGF-C and VEGF-D and is internalized with VEGFR3 upon ligand stimulation (48), is expressed by intestinal but not cutaneous
lymphatic vessels (15). NP-1 was recently shown to regulate the
recycling of VEGFR2 (49). Because haploinsufficient Chy mice
are not completely devoid of VEGFR3 signaling, regional differences in receptor recycling and other pathways that affect signaling
output could be crucial to the development of residual, functional
lymphatic capillaries.
Through skin transplantation, we demonstrated that the introduction of a limited number of lymphatic capillaries restores DC
mobilization, suggesting this approach may prove to have therapeutic value. Indeed, some microsurgery techniques have shown
success in lymphedema patients (50, 51) as well as in experimental
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 7. DC mobilization from regions with no previous drainage can be rescued by skin transplantation. (A) Representative photomicrograph after
WT (or Chy) dorsal ear skin was transplanted onto the dorsal ear dermis of Chy recipient mice. Five regions (as indicated) were examined for the presence
of LYVE+ lymphatic capillaries. (B) Representative LYVE-1 (green) and SMA (red) staining of ear dermis within (left panel, region 1) and outside (right
panel, region 4) the transplanted region from Chy mice transplanted with WT donor skin. Scale bar, 170 mm. (C) Lymphangiogenesis was assessed 17
d after surgery using Volocity software, and the area of ear dermis covered by LYVE-1+ vessels was measured in unmanipulated WT and Chy mice, Chy
recipients of WT and Chy donor skin, and mice undergoing sham surgery. n = 3–19. (D–F) Representative images of ear dermis of Chy recipients of
b-actin–eGFP donor skin 2 wk posttransplant. Images show LYVE-1 (red, left panel), eGFP signal (green, middle panel), and the colors merged (right
panel). Note the presence of LYVE-1+ vessels, which are GFP+ (D), GFP2 (E) and a mix of the two (F). Scale bar, 28 mm (D), 80 mm (E), 37 mm (F). (G)
LYVE-1+ vessels present in the recipient ear dermis were assessed for GFP signal and quantified, and the percentage of vessels in each group is shown. (H)
Representative dot plot of MHC-II+green bead+ cells in the auricular LN of Chy recipients of WT donor skin 3 d after i.d. injection of 1-mm green
fluorescent beads into the ear. Beads were injected 14 d posttransplant. (I and J) Results shown are the proportion (I) and total number (J) of green bead+
cells present in the auricular LN of transplant recipients or mice undergoing sham surgery. Each circle represents one individual transplant and two
transplants per mouse were performed. Results are representative of four independent experiments. n = 11–17. *p , 0.05, **p , 0.01.
The Journal of Immunology
Acknowledgments
We thank Kari Alitalo for providing K14-VEGFR3-Ig mice, Hwee Ying
Lim and Veronique Angeli for technical advice, and David Zawieja for helpful discussion. We thank the Microvascular Surgery Core at the Mount Sinai
School of Medicine for expert assistance. We also thank Drs. M. Ingersoll
and E. Gautier for helpful discussion and critical reading of the manuscript.
Disclosures
The authors have no financial conflicts of interest.
References
1. Randolph, G. J., V. Angeli, and M. A. Swartz. 2005. Dendritic-cell trafficking to
lymph nodes through lymphatic vessels. Nat. Rev. Immunol. 5: 617–628.
2. Schmid-Schönbein, G. W. 1990. Microlymphatics and lymph flow. Physiol. Rev.
70: 987–1028.
3. Muthuchamy, M., and D. Zawieja. 2008. Molecular regulation of lymphatic
contractility. Ann. N.Y. Acad. Sci. 1131: 89–99.
4. Rockson, S. G. 2010. Causes and consequences of lymphatic disease. Ann. N. Y.
Acad. Sci. 1207(Suppl 1): E2–E6.
5. Smith, J. B., G. H. McIntosh, and B. Morris. 1970. The traffic of cells through
tissues: a study of peripheral lymph in sheep. J. Anat. 107: 87–100.
6. Silberberg-Sinakin, I., G. J. Thorbecke, R. L. Baer, S. A. Rosenthal, and
V. Berezowsky. 1976. Antigen-bearing langerhans cells in skin, dermal lymphatics and in lymph nodes. Cell. Immunol. 25: 137–151.
7. Larsen, C. P., R. M. Steinman, M. Witmer-Pack, D. F. Hankins, P. J. Morris, and
J. M. Austyn. 1990. Migration and maturation of Langerhans cells in skin
transplants and explants. J. Exp. Med. 172: 1483–1493.
8. Lukas, M., H. Stössel, L. Hefel, S. Imamura, P. Fritsch, N. T. Sepp, G. Schuler, and
N. Romani. 1996. Human cutaneous dendritic cells migrate through dermal lymphatic vessels in a skin organ culture model. J. Invest. Dermatol. 106: 1293–1299.
9. Stoitzner, P., S. Holzmann, A. D. McLellan, L. Ivarsson, H. Stössel, M. Kapp,
U. Kämmerer, P. Douillard, E. Kämpgen, F. Koch, et al. 2003. Visualization and
characterization of migratory Langerhans cells in murine skin and lymph nodes
by antibodies against Langerin/CD207. J. Invest. Dermatol. 120: 266–274.
10. Baluk, P., J. Fuxe, H. Hashizume, T. Romano, E. Lashnits, S. Butz,
D. Vestweber, M. Corada, C. Molendini, E. Dejana, and D. M. McDonald. 2007.
Functionally specialized junctions between endothelial cells of lymphatic vessels. J. Exp. Med. 204: 2349–2362.
11. Pflicke, H., and M. Sixt. 2009. Preformed portals facilitate dendritic cell entry
into afferent lymphatic vessels. J. Exp. Med. 206: 2925–2935.
12. Tal, O., H. Y. Lim, I. Gurevich, I. Milo, Z. Shipony, L. G. Ng, V. Angeli, and
G. Shakhar. 2011. DC mobilization from the skin requires docking to immobilized CCL21 on lymphatic endothelium and intralymphatic crawling. J. Exp.
Med. 208: 2141–2153.
13. Miteva, D. O., J. M. Rutkowski, J. B. Dixon, W. Kilarski, J. D. Shields, and
M. A. Swartz. 2010. Transmural flow modulates cell and fluid transport functions
of lymphatic endothelium. Circ. Res. 106: 920–931.
14. Brice, G., A. H. Child, A. Evans, R. Bell, S. Mansour, K. Burnand, M. Sarfarazi,
S. Jeffery, and P. Mortimer. 2005. Milroy disease and the VEGFR-3 mutation
phenotype. J. Med. Genet. 42: 98–102.
15. Karkkainen, M. J., A. Saaristo, L. Jussila, K. A. Karila, E. C. Lawrence,
K. Pajusola, H. Bueler, A. Eichmann, R. Kauppinen, M. I. Kettunen, et al. 2001.
A model for gene therapy of human hereditary lymphedema. Proc. Natl. Acad.
Sci. USA 98: 12677–12682.
16. Mäkinen, T., L. Jussila, T. Veikkola, T. Karpanen, M. I. Kettunen,
K. J. Pulkkanen, R. Kauppinen, D. G. Jackson, H. Kubo, S. Nishikawa, et al.
2001. Inhibition of lymphangiogenesis with resulting lymphedema in
transgenic mice expressing soluble VEGF receptor-3. Nat. Med. 7: 199–205.
17. Robbiani, D. F., R. A. Finch, D. Jäger, W. A. Muller, A. C. Sartorelli, and
G. J. Randolph. 2000. The leukotriene C4 transporter MRP1 regulates CCL19
(MIP-3b, ELC)-dependent mobilization of dendritic cells to lymph nodes. Cell
103: 757–768.
18. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rössner, F. Koch, N. Romani, and
G. Schuler. 1999. An advanced culture method for generating large quantities of highly
pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223: 77–92.
19. Randolph, G. J., K. Inaba, D. F. Robbiani, R. M. Steinman, and W. A. Muller.
1999. Differentiation of phagocytic monocytes into lymph node dendritic cells
in vivo. Immunity 11: 753–761.
20. Thomas, S. N., J. M. Rutkowski, M. Pasquier, E. L. Kuan, K. Alitalo,
G. J. Randolph, and M. A. Swartz. 2012. Impaired humoral immunity and tolerance in K14-VEGFR-3-Ig mice that lack dermal lymphatic drainage. J.
Immunol. 189: 2181–2190.
21. Ohl, L., M. Mohaupt, N. Czeloth, G. Hintzen, Z. Kiafard, J. Zwirner,
T. Blankenstein, G. Henning, and R. Förster. 2004. CCR7 governs skin dendritic cell migration under inflammatory and steady-state conditions. Immunity 21: 279–288.
22. Liu, K., G. D. Victora, T. A. Schwickert, P. Guermonprez, M. M. Meredith,
K. Yao, F. F. Chu, G. J. Randolph, A. Y. Rudensky, and M. Nussenzweig. 2009.
In vivo analysis of dendritic cell development and homeostasis. Science 324:
392–397.
23. Jakubzick, C., M. Bogunovic, A. J. Bonito, E. L. Kuan, M. Merad, and
G. J. Randolph. 2008. Lymph-migrating, tissue-derived dendritic cells are minor
constituents within steady-state lymph nodes. J. Exp. Med. 205: 2839–2850.
24. Poulin, L. F., S. Henri, B. de Bovis, E. Devilard, A. Kissenpfennig, and
B. Malissen. 2007. The dermis contains langerin+ dendritic cells that develop
and function independently of epidermal Langerhans cells. J. Exp. Med. 204:
3119–3131.
25. Ginhoux, F., M. P. Collin, M. Bogunovic, M. Abel, M. Leboeuf, J. Helft,
J. Ochando, A. Kissenpfennig, B. Malissen, M. Grisotto, et al. 2007. Bloodderived dermal langerin+ dendritic cells survey the skin in the steady state. J.
Exp. Med. 204: 3133–3146.
26. Bursch, L. S., L. Wang, B. Igyarto, A. Kissenpfennig, B. Malissen, D. H. Kaplan,
and K. A. Hogquist. 2007. Identification of a novel population of Langerin+
dendritic cells. J. Exp. Med. 204: 3147–3156.
27. Flacher, V., P. Douillard, S. Aı̈t-Yahia, P. Stoitzner, V. Clair-Moninot, N. Romani,
and S. Saeland. 2008. Expression of langerin/CD207 reveals dendritic cell heterogeneity between inbred mouse strains. Immunology 123: 339–347.
28. Allan, R. S., J. Waithman, S. Bedoui, C. M. Jones, J. A. Villadangos, Y. Zhan,
A. M. Lew, K. Shortman, W. R. Heath, and F. R. Carbone. 2006. Migratory
dendritic cells transfer antigen to a lymph node-resident dendritic cell population
for efficient CTL priming. Immunity 25: 153–162.
29. Angeli, V., F. Ginhoux, J. Llodrà, L. Quemeneur, P. S. Frenette, M. Skobe,
R. Jessberger, M. Merad, and G. J. Randolph. 2006. B cell‑driven lymphangiogenesis in inflamed lymph nodes enhances dendritic cell mobilization.
Immunity 24: 203–215.
30. Qu, C., E. W. Edwards, F. Tacke, V. Angeli, J. Llodrá, G. Sanchez-Schmitz,
A. Garin, N. S. Haque, W. Peters, N. van Rooijen, et al. 2004. Role of CCR8 and
other chemokine pathways in the migration of monocyte-derived dendritic cells
to lymph nodes. J. Exp. Med. 200: 1231–1241.
31. Förster, R., A. Schubel, D. Breitfeld, E. Kremmer, I. Renner-Müller, E. Wolf, and
M. Lipp. 1999. CCR7 coordinates the primary immune response by establishing
functional microenvironments in secondary lymphoid organs. Cell 99: 23–33.
32. Braun, A., T. Worbs, G. L. Moschovakis, S. Halle, K. Hoffmann, J. Bölter,
A. Münk, and R. Förster. 2011. Afferent lymph-derived T cells and DCs use
different chemokine receptor CCR7-dependent routes for entry into the lymph
node and intranodal migration. Nat. Immunol. 12: 879–887.
33. Mäkinen, T., R. H. Adams, J. Bailey, Q. Lu, A. Ziemiecki, K. Alitalo, R. Klein,
and G. A. Wilkinson. 2005. PDZ interaction site in ephrinB2 is required for the
remodeling of lymphatic vasculature. Genes Dev. 19: 397–410.
34. Alitalo, K., T. Tammela, and T. V. Petrova. 2005. Lymphangiogenesis in development and human disease. Nature 438: 946–953.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
models of skin transplantation to areas of secondary lymphedema
(52). In addition to restoration of DC migration, we also observed
lymphatic capillaries in and adjacent to the transplant in the WT–
Chy group and, to a lesser extent, the Chy–Chy group. The lymphangiogenesis was higher in the WT–Chy group compared with
the Chy–Chy or sham surgery groups, suggesting that, at least to
some extent, donor LECs may proliferate and migrate to form new
vessels. Indeed, our GFP donor experiments lend credence to this
possibility, and the presence of collecting vessels in the ear of Chy
mice (Fig. 4H) suggest donor capillaries may anastomose with
endogenous collectors. Consistent with the heightened lymphangiogenesis in WT–Chy transplants, most lymphangiogenesis in
adults occurs through sprouting from existing vessels (53). The
more limited restoration, which occurred in the groups lacking
pre-existing lymphatic vessels (Chy transplanted tissue or Chy
sham-operated mice), may be due to local increases in VEGF-C as
part of the surgery-associated inflammation. It is possible that the
appearance of a low number of LYVE-1+ lymphatic capillaries
observed after Chy–Chy transplant (Fig. 7C) either results from
sprouting of, or anastomosis with, the collecting vessels already
present (Fig. 4H), thereby facilitating DC migration in a small
number of these recipients.
In summary, we show that the presence of very rare lymphatic
capillaries in Chy mouse skin retards molecular transport through
lymphatics but allows DCs to mobilize from the skin to LNs
through the lymphatic vasculature normally and without delay.
Furthermore, these results reveal regional differences in lymphatic
capillaries in the Chy mouse, with residual capillaries found on the
body trunk while they are absent in the extremities. These data
importantly add to our understanding of the requirements for DC
transit through the lymphatic vasculature, aid in defining how the
immune system is affected by lymphatic hypoplasia, and may
have a translational aspect and lend further credence to surgical
treatment of patients with peripheral lymphedema.
4619
4620
45. Bollinger, A., G. Isenring, U. K. Franzeck, and U. Brunner. 1983. Aplasia of
superficial lymphatic capillaries in hereditary and connatal lymphedema (Milroy’s
disease). Lymphology 16: 27–30.
46. Mellor, R. H., C. E. Hubert, A. W. Stanton, N. Tate, V. Akhras, A. Smith,
K. G. Burnand, S. Jeffery, T. Mäkinen, J. R. Levick, and P. S. Mortimer. 2010.
Lymphatic dysfunction, not aplasia, underlies Milroy disease. Microcirculation
17: 281–296.
47. Dellinger, M. T., R. J. Hunter, M. J. Bernas, M. H. Witte, and R. P. Erickson.
2007. Chy-3 mice are Vegfc haploinsufficient and exhibit defective dermal superficial to deep lymphatic transition and dermal lymphatic hypoplasia. Dev.
Dyn. 236: 2346–2355.
48. Kärpänen, T., C. A. Heckman, S. Keskitalo, M. Jeltsch, H. Ollila,
G. Neufeld, L. Tamagnone, and K. Alitalo. 2006. Functional interaction
of VEGF-C and VEGF-D with neuropilin receptors. FASEB J. 20: 1462–
1472.
49. Ballmer-Hofer, K., A. E. Andersson, L. E. Ratcliffe, and P. Berger. 2011.
Neuropilin-1 promotes VEGFR-2 trafficking through Rab11 vesicles thereby
specifying signal output. Blood 118: 816–826.
50. Lee, B. B., Y. W. Kim, D. I. Kim, J. H. Hwang, J. Laredo, and R. Neville. 2008.
Supplemental surgical treatment to end stage (stage IV-V) of chronic lymphedema. Int. Angiol. 27: 389–395.
51. Campisi, C., C. Bellini, C. Campisi, S. Accogli, E. Bonioli, and F. Boccardo.
2010. Microsurgery for lymphedema: clinical research and long-term results.
Microsurgery 30: 256–260.
52. Slavin, S. A., A. D. Van den Abbeele, A. Losken, M. A. Swartz, and R. K. Jain.
1999. Return of lymphatic function after flap transfer for acute lymphedema.
Ann. Surg. 229: 421–427.
53. He, Y., I. Rajantie, M. Ilmonen, T. Makinen, M. J. Karkkainen, P. Haiko,
P. Salven, and K. Alitalo. 2004. Preexisting lymphatic endothelium but not endothelial progenitor cells are essential for tumor lymphangiogenesis and lymphatic metastasis. Cancer Res. 64: 3737–3740.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
35. Baluk, P., T. Tammela, E. Ator, N. Lyubynska, M. G. Achen, D. J. Hicklin,
M. Jeltsch, T. V. Petrova, B. Pytowski, S. A. Stacker, et al. 2005. Pathogenesis of
persistent lymphatic vessel hyperplasia in chronic airway inflammation. J. Clin.
Invest. 115: 247–257.
36. Vigl, B., D. Aebischer, M. Nitschké, M. Iolyeva, T. Röthlin, O. Antsiferova, and
C. Halin. 2011. Tissue inflammation modulates gene expression of lymphatic
endothelial cells and dendritic cell migration in a stimulus-dependent manner.
Blood 118: 205–215.
37. Yang, C. W., B. S. Strong, M. J. Miller, and E. R. Unanue. 2010. Neutrophils
influence the level of antigen presentation during the immune response to protein
antigens in adjuvants. J. Immunol. 185: 2927–2934.
38. Cerovic, V., S.A. Houston, C.L. Scott, A. Aumeunier, U. Yrlid, A.M. Mowat, and
S.W. Milling. 2013. Intestinal CD103‑ dendritic cells migrate in lymph and
prime effector T cells. Mucosal Immunol. 6: 104‑113.
39. Tilney, N. L. 1971. Patterns of lymphatic drainage in the adult laboratory rat. J.
Anat. 109: 369–383.
40. Vondenhoff, M. F., S. A. van de Pavert, M. E. Dillard, M. Greuter, G. Goverse,
G. Oliver, and R. E. Mebius. 2009. Lymph sacs are not required for the initiation
of lymph node formation. Development 136: 29–34.
41. Mebius, R. E., P. R. Streeter, J. Brevé, A. M. Duijvestijn, and G. Kraal. 1991.
The influence of afferent lymphatic vessel interruption on vascular addressin
expression. J. Cell Biol. 115: 85–95.
42. Wendland, M., S. Willenzon, J. Kocks, A. C. Davalos-Misslitz, S. I. Hammerschmidt,
K. Schumann, E. Kremmer, M. Sixt, A. Hoffmeyer, O. Pabst, and R. Förster. 2011.
Lymph node T cell homeostasis relies on steady state homing of dendritic cells.
Immunity 35: 945–957.
43. Moussion, C., and J. P. Girard. 2011. Dendritic cells control lymphocyte entry to
lymph nodes through high endothelial venules. Nature 479: 542–546.
44. Kumamoto, Y., L. M. Mattei, S. Sellers, G. W. Payne, and A. Iwasaki. 2011.
CD4+ T cells support cytotoxic T lymphocyte priming by controlling lymph
node input. Proc. Natl. Acad. Sci. USA 108: 8749–8754.
LYMPHATIC CAPILLARIES AND DC MIGRATION
Supplemental Material
Supplementary Figure 1. Back skin histology and intranodal lymphatics in Chy mice. (A)
Representative histologic staining using Hema-3 of cross-sections of the back skin of WT and Chy
mice. (B) LYVE-1 staining (green) of auricular LNs from WT and Chy mice. Bar=60μm. Results are
representative of 2 independent experiments.
Supplementary Figure 2. Neutrophil accumulation in the skin and draining lymph nodes during
inflammation. (A-B) Proportions of live CD45+ cells that were CD11b+Ly6G+ in the ear (A) and back
skin (B) 4 hours after injection of PBS (white bars) or CFA (black bars). (C-E) Proportions (C+E) and
total numbers (D+F) of CD11b+Ly6G+ cells in the auricular (C+D) and brachial (E+F) LNs 4 hours
after injection of PBS (white bars) or CFA (black bars) into the ear or back skin dermis. ns=not
significant. Results are representative of 2 independent experiments.