Stabilin-1CLEVER-1, a type 2 macrophage marker, is an
Stabilin-1/CLEVER-1,a type 2macrophage marker,is an adhesion and scavenging molecule on human placental macrophages
Senthil Palani 1,Mikael Maksimow 1,Mari Miiluniemi 1,Kaisa Auvinen 1,Sirpa Jalkanen 1and Marko Salmi 1,2
1
MediCity Research Laboratory,University of Turku,and National Institute of Health and Welfare,Turku,Finland
2
Department of Medical Biochemistry and Genetics,University of Turku,Turku,Finland
Stabilin-1/common lymphatic endothelial and vascular endothelial receptor-1(CLEVER-1)is a multidomain protein present in lymphatic and vascular endothelial cells and type 2immunosuppressive macrophages.In adults,stabilin-1/CLEVER-1is a scavenging recep-tor and an adhesion molecule,but much less is known about its role during development.Here,we studied the expression and functions of macrophage stabilin-1/CLEVER-1in human placenta and during human https://www.wendangku.net/doc/8a16043976.html,ing newly generated mAbs,we found that stabilin-1/CLEVER-1is expressed on virtually all macrophages in term placenta,both in the decidua and in the placental villi.Placental stabilin-1/CLEVER-1was involved in the scavenging of Ac-LDL (acetylated low density lipoprotein)and in the uptake of ?uores-cently labeled model antigen OVA.siRNA-mediated suppression of stabilin-1/CLEVER-1altered the cytokine pro?le produced by placental macrophages.Stabilin-1/CLEVER-1on placental macrophages mediated their adhesion to placental vessels and supported their transmigration through vascular endothelium.Finally,we found that stabilin-1/CLEVER-1is induced very early in fetal macrophages,high endothelial venules,and lymphatic vessels in multiple lymphatic organs.Together,these data suggest that macrophage stabilin-1/CLEVER-1can potentially regulate leukocyte migration and scavenging during the development of the placenta and fetus.
Key words:Adhesion .Human placenta .Reproductive immunology .Type 2
macrophages
Supporting Information available online
Introduction
Macrophages can be divided into in?ammation-promoting type 1cells,and into anti-in?ammatory type 2cells [1–4].Although this classi?cation encompasses a spectrum of different activation
sta-
ges rather than just the two extreme poles,it is useful in understanding the versatility of macrophage phenotypes and functions.The type 2cells seem to play a key role in dampening immune responses during physiological (e.g.pregnancy)and pathological (e.g.tumor evasion)conditions.Therefore,great interest has arisen in ?nding good phenotypic markers for type 2cells in humans.Macrophage mannose receptor CD206and stabilin-1/common lymphatic endothelial and vascular endothe-lial receptor-1(CLEVER-1)are among the best markers for type 2macrophages in humans
[1–4].
Correspondence:Professor Marko Salmi e-mail:marko.salmi@utu.?
Stabilin-1/CLEVER-1,also known as FEEL-1,is a large,type I membrane protein expressed on certain endothelial cells and macrophages[5–8].On the endothelial cells,it is constitutively present on lymphatic endothelium.Blockade of stabilin-1/ CLEVER-1with mAbs inhibits lymphocyte–lymphatic endothelial interactions and leukocyte migration in the afferent arm of the lymphatic system both in vitro and in vivo[8–10].On the human vascular endothelium,stabilin-1/CLEVER-1is constitutively expressed only in high endothelial venules(HEVs)in secondary lymphatic tissues and in sinusoidal endothelium of liver,spleen, and bone marrow[6].However,it is also inducible on continuous ?at-walled endothelium under the conditions of in?ammation [7,9].On the vascular endothelium,stabilin-1/CLEVER-1 supports leukocyte adhesion and transmigration[9,10],and it has also been reported to modulate angiogenesis[11].Under normal conditions,blood monocytes are virtually negative for stabilin-1/CLEVER-1,and only a subpopulation of macrophages (alternatively activated cells)express stabilin-1/CLEVER-1in the skin and gut[7,12].However,blood monocytes can be induced to express stabilin-1/CLEVER-1by IL-4/dexamethasone stimula-tion in vitro,and they become stabilin-1/CLEVER-1positive in certain diseases in vivo[7,12].On transfectants and macro-phages,stabilin-1/CLEVER-1functions as a scavenging receptor by taking up oxidized and acetylated low density lipoprotein (Ac-LDL)or OVA,SPARC and placental lactogen,and possibly bacteria and apoptotic bodies[7,13–19].
During development,the polarization of placental and embryonic macrophages into type2-like cells has potentially an important role in regulating immune reactions and angiogenesis [20–23].Although stabilin-1/CLEVER-1is a well-established marker of type2macrophages,its possible role during pregnancy and fetal development is not well understood.Therefore,we analyzed the expression and immunological functions of human stabilin-1/CLEVER-1at the feto-maternal interface.
Results
Novel anti-human stabilin-1/CLEVER-1mAbs
Only three mAbs,MS-1,266,and372,against human stabilin-1/ CLEVER-1are available[6,8],and they have certain limitations for several immunological applications.Therefore,we immunoaf?nity puri?ed stabilin-1/CLEVER-1from human placenta using mAb372and generated new anti-stabilin-1/ CLEVER-1mAbs by immunizing rats with the puri?ed protein. Two of the new mAbs,9–11(rat IgG2a)and2–7(rat IgG1), stained stabilin-1/CLEVER-1transfectants,but not control cells. Both mAbs recognized the full-length stabilin-1/CLEVER-1 protein,as well as an N-terminal fragment encompassing amino acids1–1027of this large2570amino acid protein(Fig.1A). In immunoblottings,they also speci?cally recognized a molecule of270kDa from the full-length stabilin-1/CLEVER-1transfec-tants,and a110kDa molecule from the N-terminal fragment of stabilin-1/CLEVER-1(Fig.1B).Immunoaf?nity depletion of placental lysate with mAb372caused a concomitant loss of the 9–11signal,and,conversely,after9–11depletion no372signal was detectable(Fig.1C).Moreover,in immunohistochemistry mAbs9–11and2–7stained the same cells as mAb372both in placenta and in tonsil(Fig.1D,Supporting Information Fig.1, and data not shown).Together,these data show that9–11and 2–7are new mAbs speci?cally recognizing human stabilin-1/ CLEVER-1.
Stabilin-1/CLEVER-1is expressed on all macrophages, and in certain endothelial cells,in placenta
Stabilin-1/CLEVER-1is known to be present on placental macrophages[6,13],but it has not been determined whether it is uniquely expressed on a certain macrophage subpopulation and whether it is present on nonmacrophage cell types.There-fore,we used the new2–7stabilin-1/CLEVER-1mAb to stain formalin-?xed,paraf?n-embedded sections from term placentas obtained during normal births.Figure2A shows that stabilin-1/ CLEVER-1is strongly expressed on placental leukocytes,which morphologically resemble macrophages.In vessels,the express-ion of stabilin-1/CLEVER-1is variable.A few vessels have clearly positive endothelium(Fig.2A),in some vessels a punctate expression of stabilin-1/CLEVER-1was found on a few endothe-lial cells,and in most vessels all endothelial cells were negative for this receptor.All other cell types,including lymphocyte-like cells and trophoblasts,and other stromal components were stabilin-1/CLEVER-1negative.
To con?rm the identity of stabilin-1/CLEVER-1-positive leukocytes as macrophages,we used two-color immuno-?uorescent staining.These analyses showed that all CD681cells coexpressed stabilin-1/CLEVER-1(Fig.2B).All stabilin-1/ CLEVER-1-positive cells also expressed CD206,the macrophage mannose receptor,which is a well-established type2macrophage marker(Fig.2B).Nevertheless,stabilin-1/CLEVER-1did not stain all macrophages,since in in?amed tonsil(containing many type1 macrophages)most CD681cells were devoid of stabilin-1/ CLEVER-1staining(Supporting Information Fig.2).
We then isolated placental leukocytes by mechanical teasing and Ficoll-gradient centrifugation for multicolor-FACS analyses (Fig.2C).These analyses veri?ed that most macrophages co-expressed CD14,CD68,CD206,and stabilin-1/CLEVER-1.When analyzed from four placentas,9771%of the CD141macro-phages coexpressed CD68and,and9672%of the CD141 macrophages coexpressed CD206.In these samples,placental macrophages expressed stabilin-1/CLEVER-1at a relatively low level,and,in fact,no clear stabilin-1/CLEVER-1-negative sub-population was identi?able.Stabilin-1/CLEVER-1was expressed at approximately the same level on the surface of on CD141 macrophages(mean?uorescence intensity,MFI52172), CD681macrophages(MFI51972)and CD2061macrophages (MFI51772;n54for all).Notably,?ow cytometric staining was performed with nonpermeabilized cells,which demonstrates that stabilin-1/CLEVER-1is present on the cell surface in many
D
A
B
Control
Stabilin-11-1027
C
225150
10276
225Lysate:Blot:
15010276
rIgG mIgG
372
9-112-7
Stabilin-1fl
372S t a b i l i n -1 f l
M o c k S t a b i l i n -1 1-10279-112-7
M o c k S t a b i l i n -1 1-1027S t a b i l i n -1 f l S t a b i l i n -1 1-1027S t a b i l i n -1 f l Depletion:Blot:
3729-11
r I g G 9-11
m I g G
372
r I g G
9-11
EGFP
Figure 1.mAbs 2–7and 9–11are new stabilin-1/CLEVER-1-speci?c mAbs.(A)The stabilin-1/CLEVER-1constructs (an N-terminal fragment encoding for amino acids 1–1027and a full-length (?)molecule)were expressed as pIRES2-EGFP plasmids,and GFP ?uorescence from nonpermeabilized transfectants (and nontransfected controls)was analyzed by ?ow cytometry.After permeabilization,the anti-stabilin-1/CLEVER-1mAbs 372,9–11,and 2–7(and negative control mAbs)were used to stain the cells on the red channel.(B)Lysates from the stabilin-1/CLEVER-1transfectants and from EGFP control transfectants were subjected to immunoblotting with the indicated mAbs.Intact full-length stabilin-1/CLEVER-1migrates as a 270kDa molecule (black arrow),and the N-terminal fragment as a 110kDa molecule (white arrow).(C)Placental lysates were immunodepleted,and then immunoblotted with the indicated Abs.(D)T wo-color immuno?uorescent staining of placenta with the indicated mAbs.Scale bar,50m m.mIgG and rIgG,mouse and rat-negative control mAbs respectively.In (A–D)one representative experiment (out of three similar experiments)is shown.
placental macrophages.Moreover,stabilin-1/CLEVER-1was absent from small leukocytes,many of which were CD561NK cells,which constitute the second major leukocyte subpopulation in the placenta (Fig.2C,histogram,and data not shown).Thus,together the immunohistochemical and FACS analyses showed that all macrophages in term placentas in humans are
stabilin-1/
C
A
B
64%
26%
0%70%
17%
1%
rIgG
rIgG
9-11
9-11
C D 206
F L 4
C D 14
FL1FL1F L 2
2-7
rIgG
2-7
9-112-72-7
V
Figure 2.Stabilin-1/CLEVER-1is expressed on the surface of macrophages in human placenta.(A)Anti-stabilin-1/CLEVER-1mAbs 2–7(and a rat-negative control rIgG,9B5)were used to stain paraf?n sections of term placenta.V ,villus;D,decidua;and E,endothelial cells of a vessel.Note that
the vessel contains erythrocytes,indicating that it is a blood vessel.Scale bar,50m m.(B)Frozen sections of placenta were stained for two-color immuno?uorescence with the indicated mAbs.Scale bar,50m m.(C)Placental leukocytes were separated,and stained using multi-color staining against the indicated antigens.Representative ?ow cytometric dot blots from macrophages (red cells in the scatter plot),and histograms from small leukocytes (cyan cells in the scatter plot)are shown.All experiments (A–C)were repeated with at least three different donors.
CLEVER-1-positive type2cells,and the majority of them also express stabilin-1/CLEVER-1on the cell surface.
Silencing of placental stabilin-1/CLEVER-1reduces scavenging and cytokine secretion
To study the function of stabilin-1/CLEVER-1on placental macrophages,we isolated placental leukocytes and silenced stabilin-1/CLEVER-1by https://www.wendangku.net/doc/8a16043976.html,e of siRNAs with primary leukocytes is challenging,but we managed to knock down stabilin-1/CLEVER-1protein on average by60%with a speci?c siRNA in comparison with control siRNA-transfected cells (Fig.3A).The uptake and/or processing of a self-quenched model antigen BODIPY FL-labeled OVA(DQ-OVA)was similar in stabilin-1/CLEVER-1and control siRNA-treated placental macro-phages at a30min time point(Fig.3B).However,after4h,the stabilin-1/CLEVER-1siRNA-treated cells had taken-up/processed DQ-OVA signi?cantly less than control siRNA-transfected cells(Fig. 3B).Knockdown of stabilin-1/CLEVER-1also inhibited uptake of Ac-LDL(Fig.3C).Thus stabilin-1/CLEVER-1expressed on placental macrophages is a functionally intact scavenging receptor,and it may participate in the clearance of foreign substances and thereby it may also contribute to antigen presentation.
Stabilin-1/CLEVER-1might not be merely a marker of type2 macrophages,but it might have a functional role in macrophage polarization.Therefore,we silenced stabilin-1/CLEVER-1in placental leukocytes(mostly macrophages but also some mono-nuclear cells,see Fig.2C)and analyzed its effect on cytokine expression patterns using Multiplex assays(Fig.3D,and Supporting Information Table).The results showed that when stabilin-1/CLEVER-1was knocked down,there was a statistically signi?cant increase in the secretion of the proin?ammatory cytokine TNF-a into the culture medium.However,IL-10, important for type2polarization,was also increased when stabilin-1/CLEVER-1was knocked down.Moreover,silencing of stabilin-1/CLEVER-1did not affect the expression of a type2 macrophage marker MR(Fig.3E).Therefore,stabilin-1/ CLEVER-1does not seem to be selectively needed for the production of those cytokines which control type2macrophage polarization or for the maintenance of type2phenotype.
Stabilin-1/CLEVER-1on macrophages contributes to leukocyte–endothelial interactions
Stabilin-1/CLEVER-1on endothelial cells is known to mediate leukocyte adhesion and transmigration[9,10],but the adhesive role of stabilin-1/CLEVER-1expressed on macrophages has not been studied.Therefore,we determined the role of stabilin-1/CLEVER-1 in the binding of placental macrophages to placental blood vessels. To be able to dissect the contribution of macrophage stabilin-1/ CLEVER-1from vascular stabilin-1/CLEVER-1,we pretreated the leukocytes with anti-stabilin mAb9–11and washed away all unbound mAbs before adding the leukocytes on the placental sections.We found that placental leukocytes morphologically resembling macrophages(large,ruf?ed cells)speci?cally interacted with placental vessels(Fig.4A).These analyses showed that blocking of the macrophage stabilin-1/CLEVER-1by mAb9–11 reduced the binding by almost60%(Fig.4B).
To study the step of the adhesion cascade supported by macrophage stabilin-1/CLEVER-1,we performed in vitro?ow assays.In these assays,placental macrophages were pretreated with mAb372or control(both as whole mAbs),and then perfused over a con?uent human umbilical vein endothelial cell(HUVEC) monolayer under physiologically relevant shear force.Real-time imaging revealed that placental macrophages?rmly adhered to these endothelial cells,and some of them underwent trans-endothelial migration manifested by the typical transformation of phase-bright cells into phase-dark cells(Fig.4C,and Supporting Information Video).We found that blocking of macrophage stabilin-1/CLEVER-1did not signi?cantly alter?rm adhesion of placental leukocytes to HUVEC(Fig.4D).However,transmigra-tion of placental macrophages through the vascular endothelium was signi?cantly inhibited when macrophage stabilin-1/CLEVER-1 was blocked with mAb372(Fig.4D).These data show for the?rst time that stabilin-1/CLEVER-1expressed on leukocytes is an adhesion molecule involved in leukocyte traf?c.
Stabilin-1/CLEVER-1is expressed early in macrophages and lymphatic vessels in human fetuses
The new anti-stabilin mAb2–7allowed us to analyze the appearance of stabilin-1/CLEVER-1in lymphoid organs during fetal development using archived material.In primary lymphoid organs,stabilin-1/CLEVER-1was present in fetal thymus in macrophages,which mainly resided in septal areas,but some of which were also scattered within the cortex and medulla already in our earliest(18wk)thymus samples(Fig.5).In the secondary lymphatic organs,stabilin-1/CLEVER-1was found in the gut already at week11.During the ontogeny,it was present in the lymphatics within the villi and,when identi?able,in HEVs of lymphoid aggregates.In the spleen,the sinusoidal vessels were highly stabilin-1/CLEVER-1positive in all our samples (18–40wk).In our earliest LN samples(40wk),stabilin-1/ CLEVER-1was prominently expressed not only in the marginal sinus,and in the lymphatic vasculature,but also in HEV.In the liver,stabilin-1/CLEVER-1was found at the time of fetal hematopoesis in the vascular lining(18wk;Fig.5).Later on (23–40wk),its expression shifted more dominantly to macro-phage-like cells,which most likely are Kupffer cells.Thus, stabilin-1/CLEVER-1appears to be induced early on(before week11)during human development both in macrophages and in lymphatic and certain types of vascular endothelium.
Macrophages in the villi of placenta were stabilin-1/CLEVER-1 negative at week7(our earliest sample),but clearly positive from week11onward(Fig.5).In one embryonic umbilical cord that we had available from week23,stabilin-1/CLEVER-1was present in the vascular endothelium both in the umbilical artery and in
R
e
l
a
t
i
v
e
e
x
p
r
e
s
s
i
o
n
D
Q
-
O
V
A
u
p
t
a
k
e
p
e
r
c
e
n
t
a
g
e
A
B
Neg co Stabilin-1
30min
siRNA:
Time:
4hours
3-372
mIgG
Neg co Stabilin-1Neg co Stabilin-1
Stabilin-1
Neg co
siRNA
I
L
-
2
I
-
6
L
I
L
-
3
1
I
F
N
-
γ
G
M-
C
S
F
T
N
F
-
α
N
e
c
o
g I
-
4
L
I
L
-
5
I
L
-
1
C
y
t
o
k
i
n
e
c
o
n
c
(
R
e
l
a
t
i
v
e
e
x
p
r
e
s
s
i
o
n
)
D
E
Stabilin-1
b
Neg co
siRNA
30min
siRNA:
Time:
4hours
Neg co Stabilin-1Neg co Stabilin-1
A
c
L
D
L
u
p
t
a
k
e
p
e
r
c
e
n
t
a
g
e
3-155
mIgG
R
e
l
a
t
i
v
e
e
x
p
r
e
s
s
i
o
n
siRNA:Neg co Stabilin-1
Figure3.Silencing of placental stabilin-1/CLEVER-1alters antigen processing,scavenging,and cytokine secretion.(A)Placental macrophages were isolated,transfected with stabilin-1/CLEVER-1,or control siRNA,cultured for2days and analyzed for stabilin-1/CLEVER-1expression.(B and C)The silenced cells were used for the(B)DQ-OVA and(C)Ac-LDL uptake assays.(D)The culture supernatants from the silenced cells were collected and the levels of secreted cytokines were determined using Milliplex Map assays.(E)Expression of mannose receptor on the silenced cells.In all experiments,representative histograms and quantitation of the?uorescence(geometric MFI)are shown,and the data from the different donors are normalized by assigning value100%to control treated cells.All data are mean7SEM,n53–4.?p o0.05,??p o0.01(Student’s t-test;with Bonferoni correction in(D)).
the vein (Fig.5).Also in the other organs (e.g.heart,lung,kidney,adipose tissue,connective tissue,adrenal,and thyroid),we found numerous stabilin-1/CLEVER-1-positive,macrophage-like cells in all available samples (18–40wk,data not shown).In a notable contrast,the neuronal tissue of the brain was devoid of stabilin-1/CLEVER-1positive cells.Thus,stabilin-1/CLEVER-1-positive macrophages are present in multiple organs from the early stages of human ontogeny,and can thus exert their immunological functions in the fetus already during the ?rst trimester of pregnancy.
Discussion
We report here that human stabilin-1/CLEVER-1is expressed in all placental macrophages,both at the fetal and at the maternal side,at term.It can contribute to the scavenging functions and antigen presentation at this location,but it is not needed for type 2polarization as such.Macrophage stabilin-1/CLEVER-1also mediates leukocyte–endothelial contacts.During fetal development in man,stabilin-1/CLEVER-1appeared before week 11both in the macrophages and in the vessels.Thus,stabilin-1/CLEVER-1has the potential to contribute to the immunological responses during the pregnancy both at the feto-maternal interface and in the fetus itself.
The expression of stabilin-1/CLEVER-1in placental macro-phages has been reported already when the molecule was iden-ti?ed 20years ago [6].However,it has not been known on what subpopulation of macrophages stabilin-1/CLEVER-1is present and when it is induced during human ontogeny.We saw that virtually all placental macrophages were positive for stabilin-1/CLEVER-1,and most of them expressed stabilin-1/CLEVER-1on the cell surface.This is notable,since stabilin-1/CLEVER-1is expressed only in a small subpopulation of type 2macrophages in different settings.For instance,in cancer tissue only about one-third of all type 2macrophages are stabilin-1/CLEVER-1positive (our unpublished observations and [19,24,25]).In addition,stabilin-1/CLEVER-1has been reported to be a late marker of the type 2polarization program [26].Moreover,the surface expression of stabilin-1/CLEVER-1in placental macrophages is intriguing.In normal blood monocytes,stabilin-1/CLEVER-1is virtually indistinguishable on the cell surface although it is
present
A
B
mIgG
PV-1
20406080100120
R e l a t i v e a d h e s i o n
C rIgG mIgG Adhesion Transmigration
Leukocyte pretreat
9-11
20406080100
372mIgG 372% o f c o n t r o l
D
Figure 4.Macrophage stabilin-1/CLEVER-1supports leukocyte–endothelial contacts.(A and B)Freshly isolated placental leukocytes were treated with anti-stabilin-1/CLEVER-1(9–11F(ab)2)or control (MEL-14F(ab)2)mAbs,and their binding to placental vessels was studied in the in vitro frozen sections.(A)Representative images from PV1-positive blood vessels (mIgG,a mouse negative control mAb 3G6)are shown.Binding of round leukocytes (some pointed out by arrows)to small and larger blood vessels (outlined by a dashed line)are shown.Scale bar,50m m.(B)The quantitative adhesion data are mean 7SEM,n 53.(C)Placental leukocytes were isolated,treated with control (AK-1),and anti-stabilin-1/CLEVER-1(372)mAbs and subjected to in vitro laminar ?ow assays (shear stress,0.75dyn/cm 2)on cultured HUVEC.A representative still image demonstrates the adhesion (phase bright cells,some pointed out by open arrows)and transmigration (phase-dark cells,two pointed out by white arrows).(D)The number of ?rmly adherent and transmigrated cells (mean 7SEM,n 56)was quantitated.The number of adherent/transmigrated cells in the control mAb-treated samples was normalized to 100%in each independent experiment (three different placentas and four different HUVEC).(B,D)?p o 0.05,??p o 0.01(Student’s t -test).
intracellularly.However,in familial hypercholesterolinemia patients,stabilin-1/CLEVER-1is found on the surface of circulat-ing monocytes[19].Therefore,placenta appears to be one of the few tissues,if not the only one,in which stabilin-1/CLEVER-1is expressed in all macrophages.Moreover,the subcellular locali-zation(cytoplasmic versus surface)is clearly different in distinct monocyte/macrophage populations,and this may be critical in the regulation of the functions of stabilin-1/CLEVER-1.Finally,since macrophages are the only stabilin-1/CLEVER-1-positive leukocyte type in the placenta,we can conclude that in all our functional assays,although we have not separated macrophages from other leukocyte types,we have always measured the function of macrophage stabilin-1/CLEVER-1.
Immunostaining with monocytes and macrophages is prone to artifacts due to abundant expression of Fc receptors on these cell types.All earlier stainings in the literature have been done with whole Ig molecule anti-stabilin-1/CLEVER-1Abs.Therefore,the derivation of new mAbs and the generation of F(ab)2-fragments in the current study is an advance for obtaining reliable expres-sion data.Moreover,the generation of new Abs,which work well
Gut Liver PLN
Spleen Placenta 111840
4040
182340
1823401,5 yr
1,5 yr 7112340
Thymus Umbilical cord
182340
23
23
A
WP
Figure5.Stabilin-1/CLEVER-1is induced early on during fetal development.The expression of stabilin-1/CLEVER-1in the indicated fetal organs at the indicated time points(numbers in the upper right corners are gestational ages in weeks,except1.5year,which indicates a1.5-year-old child) was studied by immunohistochemistry with the anti-stabilin-1/CLEVER-1mAb2–7on paraf?n sections.Thymus:C,cortex;M,medulla;Peripheral LN(PLN):black arrow,marginal sinus;white arrow,HEV;Spleen:WP,white pulp;Umbilical cord:A,artery,V,vein.Scale bar,100m m.
in formalin-?xed paraf?n-embedded material will greatly facil-itate future analyses of stabilin-1/CLEVER-1using archived pathological materials.
Stabilin-1/CLEVER-1has been established as a good marker for type2macrophages both in humans and in mice.However,its functional role,if any,in macrophage polarization has not been analyzed.Here,we showed that siRNA-mediated knockdown of stabilin-1/CLEVER-1does not selectively alter the cytokine secretion into a direction that would be compatible with dim-inished type2differentiation.Thus,although loss of stabilin-1/ CLEVER-1caused an increase in the production of pro-in?ammatory TNF-a,it also induced the secretion of type2 differentiating cytokine IL-10.Hence,although stabilin-1/ CLEVER-1is a good marker for type2macrophages,it does not appear to support their polarization at least by modulating cytokine expression.
We found that placental macrophages use stabilin-1/ CLEVER-1to scavenge foreign substances.In stabilin-1/ CLEVER-1siRNA-treated cells,the proteolytic processing of DQ-OVA was modestly,but signi?cantly diminished.We believe that the diminished processing of DQ-OVA seen in our experiments is primarily due to defective uptake of OVA that carries foreign modi?cations rendering it susceptible to detection by scavenging receptors.This is in line with our?nding that stabilin-1/ CLEVER-1-silenced placental leukocytes showed decreased scavenging of Ac-LDL,an established ligand for stabilin-1/ CLEVER-1[27].Since the transfection of primary monocytes is challenging,we were able to reduce stabilin-1/CLEVER-1expres-sion only by about60%in our experiments.This strongly suggests that we may have underestimated the true contribution of stabilin-1/CLEVER-1in these processes.Placental macrophages also use stabilin-1/CLEVER-1to internalize placental lactogen[13].Thus, it appears that placental stabilin-1/CLEVER-1is an ef?cient multifunctional scavenging receptor already during pregnancy.
Endothelial stabilin-1/CLEVER-1mediates transmigration of lymphocytes through blood and lymphatic vessels both in vitro and in vivo[9,10],but the adhesive role of stabilin-1/CLEVER-1 on leukocytes had not been studied earlier.We found that inhi-bition of macrophage stabilin-1/CLEVER-1reduced their inter-action with blood vessels in placenta.Moreover,in vitro?ow assays indicated that macrophage stabilin-1/CLEVER-1contri-butes to the transmigration,but not the adhesion,step during the adhesion cascade.Although placental macrophages are mostly terminally differentiated resident cells,these?ndings have at least two implications.First,it is known that in familial hypercholesterolinemia blood monocytes express stabilin-1/ CLEVER-1on the cell surface[12].Our data therefore suggest that stabilin-1/CLEVER-1can be used by these leukocytes to transmigrate through the blood endothelium.Second,it is possible that stabilin-1/CLEVER-1could also be induced on the monocyte surface in the blood of pregnant women,and then it could help in the traf?cking of these cells to the placenta.
We found that stabilin-1/CLEVER-1is expressed in tissue macrophages and in both blood and lymphatic vessels from early on during the fetal development.It was present in the earliest samples that we had(week11)in the fetal organs,and already at week11also in the placental villi.The expression pattern of fetal stabilin-1/CLEVER-1was similar to that seen in the adults.This is clearly different from other adhesion molecules,such as mucosal and peripheral LN addressins,vascular adhesion protein-1,and HECA-452(cutaneous lymphocyte antigen CLA),which show clear time-dependent changes in their expression patterns [28–30].Together,our current data suggest that stabilin-1/ CLEVER-1,and in more general,type2macrophages are dispersed throughout the body very early on,and can therefore regulate the immune balance already before any contacts with outside antigens.
Materials and methods
Abs and tissues
A mouse antihuman stabilin-1/CLEVER-1mAb372(mouse IgG1) has been described previously[8].It was directly conjugated to Alexa-488by Molecular Probes.The new antistabilin-1/ CLEVER-1mAbs were produced against immunoaf?nity puri?ed protein from placental lysates using a modi?cation of a previously described protocol[8].Brie?y,Sprague–Dawley rats were immunized with the puri?ed antigen in incomplete Freund’s adjuvant s.c.four times.Thereafter,the draining LNs were isolated,and the lymphocytes fused with SP2/0cells using ClonaCell-HY technology(Stem Cell Technologies).The hybrid-oma supernatants were tested using immunohistochemistry,and positive cells cloned by limiting dilution.The immunization protocol was approved by the local ethical committee on the use of experimental animals.The isotype of the mAbs was deter-mined using Rat monoclonal Ab isotyping test kit(AbD Serotec). F(ab)2fragments of9–11were commercially generated by GenScript,and MEL-14was digested in house using a F(ab)2 preparation kit from Pierce.
The other mAbs used were CD206Alexa-647and CD206 Alexa-488(Biolegend,mouse IgG1),CD68Alexa-647and CD68-FITC(Santa Cruz biotechnology and Dako respectively,mouse IgG1),CD56Alexa-647(BD Pharmingen,mouse IgG1),CD14-FITC(Southern Biotech,mouse IgG2a),CD14-PE(BD Pharmin-gen,mouse IgG2a).The anti-PV-1mAb(a vascular endothelium marker)has been described previously[31].As isotype controls IgG1Alexa-488,IgG1Alexa-647,IgG2a-PE,IgG1-FITC,all from BD Pharmingen,and MEL-14(rat IgG2a),9B5(rat IgG),3G6and AK-1(both mouse IgG1)[10]were used.
The second stage reagents were FITC-conjugated anti-mouse IgG(whole molecule)F(ab)2fragment(Sigma),goat anti-mouse IgG1Alexa-546(Molecular Probes),goat anti-rat IgG Alexa-546 (Molecular Probes),and FITC-conjugated goat polyclonal F(ab)2 against anti-rat F(ab)2(Abcam).
Placentas were obtained from births at the Turku University Hospital.In?amed tonsil tissue was obtained from tonsillec-tomies.Fetal tissues were collected from spontaneous abortions
and abortions induced for medical indications through the Department of Obstetrics at the Turku University Hospital as described in detail earlier[28].In brief,they included samples from11,18,23,and40wk fetuses,and from a1.5-year-old child. For the present study,thymus,spleen,LN,gut,liver,lungs, kidneys,brain,heart,placenta,and umbilical cord were analyzed.Not all tissues were available from all age groups,and in general,organs from two donors were studied.All human samples were collected under appropriate ethical permissions.
Immunohistochemistry
For paraf?n staining,small pieces of tissues were collected into 10%formalin and embedded in paraf?n.Frozen blocks of placenta were made in OCT and stored atà701C.
Paraf?n-embedded sections were stained with anti-stabilin-1/ CLEVER-1mAbs2–7(1:10dilution of a culture supernatant)and a negative control9B5(1m g/mL)after a Proteinase K treatment for antigen retrieval.After washings,Vectastain kit for rat Igs was used to develop the reaction.3,3-Diaminobenzidine in PBS containing 0.03%hydrogen peroxide was used as the chromogen.Finally,the sections were counterstained with hematoxylin,dehydrated, cleared in xylene,and permanently mounted in DePex.
Acetone-?xed,6-m m frozen sections from placenta were stained with the stabilin-1/CLEVER-1Ab372and negative control for30min in a humid chamber.Alexa-546-conjugated goat anti-mouse Ig was used as the second stage.Thereafter, CD68-FITC,CD206Alexa-488,or an isotype control mouse IgG1 Alexa-488was added.In other experiments,anti-stabilin mAbs 9–11and2–7were detected with an Alexa546-conjugated second stage reagent,and then372-Alexa488was added. Finally,the slides were mounted with antifade Prolong gold (Invitrogen)and examined with Olympus BX60microscope. Frozen sections were also stained using the immunoperoxidase technique exactly as described previously[28].
Isolation of macrophage from human placenta
Placental tissue was cut into small pieces and the fragments were grinded mechanically using steel meshworks and syringe plunger. After?ltering through silk(100m m mesh size),the cell suspen-sions were washed with HBSS.The pellets were resuspended in HBSS and laid over Ficoll gradients.The leukocyte-containing interfaces were collected,washed,and used for analyses.
Flow cytometry staining
The cells isolated from placenta were preblocked with human Igs (100m g/mL)for10min at41C.The cells were then incubated with9–11F(ab)2primary Ab for stabilin-1/CLEVER-1and negative control MEL-14F(ab)2(20m g/mL)for30min at41C. After washings,FITC-conjugated goat anti-rat F(ab)2–F(ab)2was added for30min at41C.After washings,the other phenotypic marker Abs(CD14PE1CD68Alexa-647,CD14-PE1CD206 Alexa-647,or CD14-PE1CD56Alexa-647)were added for 30min at41C.Finally,the cells were washed and?xed,analyzed using FACSCalibur and CellQuest software from BD Biosciences.
Immunoblottings
The speci?city of the new stabilin-1/CLEVER-1mAbs was tested with lysates from HEK cells and HEK stable transfectant cell lines [10]expressing the whole length and a3kb fragment(N-terminal amino acids1–1027)of stabilin-1/CLEVER-1.In total, 100m g protein/well was separated using SDS-PAGE,and transferred onto a nitrocellulose membrane.After blocking(5% milk powder and0.1%Tween-20in PBS),the primary Abs were added in the blocking solution.After washings with0.1%Tween in PBS,the membrane was incubated with appropriate perox-idase-conjugated secondary anti-mouse and anti-rat Igs,and developed using ECL(Amersham).
Placental tissue was chopped into small pieces and homo-genized with1%NP-40containing lysis buffer.After clari?cation by centrifugation,CNBr-Sepharose4B beads armed with3–372, 9–11,3G6,or MEL-14were used to immunodeplete stabilin-1/ CLEVER-1(or control antigens)from the lysates.The af?nity depletion was repeated two times with fresh beads.The lysates were then subjected to immunoblotting as described above.
Cell culturing and siRNA transfections
Cells isolated from placenta were cultured in6well plates (2?106cells/well)in IMDM(containing10%FCS1L-gluta-mine).At day2,the cells were harvested,washed,permeabilized with a short(15s)acetone treatment.After washings and blocking,the cells were stained with anti-stabilin-1/CLEVER-1 Ab(372)followed by FITC-conjugated secondary anti-mouse Ab.
Transfections with stabilin-1/CLEVER-1(UCAA-GUCGCUGCCUGCAUA)and negative control siRNA were done using commercial transfection kits(Human Monocyte Nucleo-fector kit)from Lonza.Brie?y,2?106cells were washed with PBS,resuspended in100m L of transfection buffer with0.5m g of siRNA and nucleofected with the program Y-001.After the transfections,the cells were cultured in the growth medium.
Uptake/processing of DQ-OVA and scavenging of
Ac-LDL
Control and stabilin-1/CLEVER-1siRNA-transfected cells were harvested at day2.The cells were treated with DQ-OVA BODIPY FL?uorophore(Molecular Probes,at a concentration of10m g/ mL)in RPMI1640medium containing10%FCS.The cells were incubated at5%CO2incubator at371C for30min or4h,washed with ice-cold PBS,?xed,and analyzed using a FACSCalibur.
DQ-OVA is a self-quenched conjugate of OVA,which emits green light after it has been proteolytically degraded.The uptake of Alexa488-labeled Ac-LDL(Invitrogen,10m g/mL in RPMI1640 containing10%FCS)was analyzed in a similar manner.
Cytokine analysis
Cell-culture supernatant from the mock and stabilin-1/CLEVER-1 siRNA-transfected cells were cleared from the cellular debris by centrifugation at18500?g for20min at41C.The supernatants (50m L/reaction)were then used for cytokine analysis by Milliplex Map kit(high-sensitivity human cytokines)from Millipore according to the manufacturer’s instructions.The standards were prepared from the kit according to the manu-facturer’s instructions.
Frozen section adhesion assays
A modi?ed Stamper-Woodruff frozen section assay was used
[32].In brief,the isolated placental cells were treated with9–11 F(ab)2and control Ab MEL-14F(ab)2for30min at41C in RPMI 1640medium containing10%FCS(the binding medium).After washings,they were applied onto frozen sections cut from placenta(within wax-pen circles),and incubated in100m L of the binding medium on an orbital shaker(60rpm)for30min at41C. The cell suspension was then gently tilted away from the slides, and the adherent cells were?xed in2%glutaraldehyde.The sections were then read under dark-?eld microscope,and the number of cells adherent to vessels was counted.
In vitro?ow assays
The?ow assays were performed as described previously[9,33] with certain modi?cations.HUVECs were isolated from umbilical cords and used at passages2–4.They were cultured into con?uence in plastic?ow chamber channels(Ibidi integrated BioDiagnostics;1m-Slide VI ibiTreat),and induced with500U/ mL TNF-a for4h.The placental macrophages were isolated, pretreated with anti-stabilin-1/CLEVER-1mAb372(2m g/1?106 cells)or isotype-matched control mAb AK-1for30min,washed, and resuspended in the binding medium(HBSS with calcium and magnesium11%HSA).The cells were then perfused through the channels at0.75dyn/cm2shear for5min at room temperature. Thereafter,binding medium was allowed to?ow through the capillaries for20min.The interaction of leukocytes with endothelial cells was recorded with a phase-contrast videomicro-scope at the time points of15and25min from ten prede?ned microscopic?elds.
The number of adherent(bound phase bright cells)and transmigrated(phase-dark cells)cells was counted from the videos of?ine as described previously[9,33].The transmigration percentage is obtained by dividing the number of transmigrated cells with the all interacting (adhesive plus transmigrated)cells.We scored4279 (mean7SEM,n56)transmigrating,and801746(mean7SEM, n56)adherent cells,corresponding to1573,and282716 transmigrated and adherent cells/mm2,respectively,in each control capillary.Thus,in the control capillaries on average5.2%of the bound leukocytes had transmigrated at the 20min time point.
Statistical analyses
Comparisons between two groups were performed using two-sided paired Student’s t-test.Bonferoni correction was used for multiple testing(cytokine levels).p o0.05was considered
signi?cant.
Acknowledgements:The authors thank Riikka Sjo¨roos,Teija Kanasuo,Etta-Liisa Va¨a¨na¨nen and Sari Ma¨ki for technical help, the obstetricians in the Turku University Hospital for the placentas,and Anne Sovikoski-Georgieva for secretarial help. This work was supported by the Finnish Academy,and by the Sigrid Juselius Foundation(to S.J.and M.S.).
Con?ict of interest:The authors declare no?nancial or commercial con?ict of interest.
References
1Mosser, D.M.and Edwards,J.P.,Exploring the full spectrum of macrophage activation.Nat.Rev.Immunol.2008.8:958–969.
2Gordon,S.and Martinez,F.O.,Alternative activation of macrophages: mechanism and functions.Immunity2010.32:593–604.
3Biswas,S.K.and Mantovani,A.,Macrophage plasticity and interaction with lymphocyte subsets:cancer as a paradigm.Nat.Immunol.2010.11: 889–896.
4Auffray, C.,Sieweke,M.H.and Geissmann,F.,Blood monocytes: development,heterogeneity,and relationship with dendritic cells.Annu.
Rev.Immunol.2009.27:669–692.
5Politz,O.,Gratchev, A.,McCourt,P. A.,Schledzewski,K.,Guillot,P., Johansson,S.,Svineng,G.et al.,Stabilin-1and-2constitute a novel family of fasciclin-like hyaluronan receptor homologues.Biochem.J.2002.
362:155–164.
6Goerdt,S.,Walsh,L.J.,Murphy,G.F.and Pober,J.S.,Identi?cation of a novel high molecular weight protein preferentially expressed by sinusoi-dal endothelial cells in normal human tissues.J.Cell Biol.1991.113: 1425–1437.
7Goerdt,S.,Bhardwaj,R.and Sorg,C.,Inducible expression of MS-1high-molecular-weight protein by endothelial cells of continuous origin and by dendritic cells/macrophages in vivo and in vitro.Am.J.Pathol.1993.142: 1409–1422.
8Irjala,H.,Elima,K.,Johansson,E.-L.,Merinen,M.,Kontula,K.,Alanen,K., Grenman,R.et al.,The same endothelial receptor controls lymphocyte traf?c both in vascular and lymphatic vessels.Eur.J.Immunol.2003.33: 815–824.
9Salmi,M.,Koskinen,K.,Henttinen,T.,Elima,K.and Jalkanen,S., CLEVER-1mediates lymphocyte transmigration through vascular and lymphatic endothelium.Blood2004.104:3849–3857.
10Karikoski,M.,Irjala,H.,Maksimow,M.,Miiluniemi,M.,Granfors,K., Hernesniemi,S.,Elima,K.et al.,Clever-1/Stabilin-1regulates lymphocyte migration within lymphatics and leukocyte entrance to sites of in?am-mation.Eur.J.Immunol.2009.39:3477–3487.
11Adachi,H.and Tsujimoto,M.,FEEL-1,a novel scavengin receptor with bacteria-binding and angiogenesis-modulating activities.J.Biol.Chem.
2002.277:34264–34270.
12Mosig,S.,Rennert,K.,Krause,S.,Kzhyshkowska,J.,Neunubel,K.,Heller, R.and Funke,H.,Different functions of monocyte subsets in familial hypercholesterolemia:potential function of CD141CD161monocytes in detoxi?cation of oxidized LDL.FASEB J.2009.23:866–874.
13Kzhyshkowska,J.,Gratchev,A.,Schmuttermaier,C.,Brundiers,H.,Krusell, L.,Mamidi,S.,Zhang,J.et al.,Alternatively activated macrophages regulate extracellular levels of the hormone placental lactogen via receptor-mediated uptake and transcytosis.J.Immunol.2008.180:3028–3037.
14Kzhyshkowska,J.,Workman,G.,Cardo-Vila,M.,Arap,W.,Pasqualini,R., Gratchev,A.,Krusell,L.et al.,Novel function of alternatively activated macrophages:stabilin-1-mediated clearance of SPARC.J.Immunol.2006.
176:5825–5832.
15Kzhyshkowska,J.,Mamidi,S.,Gratchev,A.,Kremmer,E.,Schmutterma-ier, C.,Krusell,L.,Haus,G.et al.,Novel stabilin-1interacting chitinase-like protein(SI-CLP)is up-regulated in alternatively activated macrophages and secreted via lysosomal pathway.Blood2006.107: 3221–3228.
16Kzhyshkowska,J.,Gratchev,A.,Martens,J.H.,Pervushina,O.,Mamidi,S., Johansson,S.,Schledzewski,K.et al.,Stabilin-1localizes to endosomes and the trans-Golgi network in human macrophages and interacts with GGA adaptors.J.Leukoc.Biol.2004.76:1151–1161.
17Cupurdija,K.,Azzola,D.,Hainz,U.,Gratchev,A.,Heitger,A.,Takikawa, O.,Goerdt,S.,et al.,Macrophages of human?rst trimester decidua express markers associated to alternative activation.Am.J.Reprod.
Immunol.2004.51:117–122.
18Tamura,Y.,Adachi,H.,Osuga,J.,Yahagi,N.,Sekiya,M.,Okazaki,H., Tomita,S.et al.,FEEL-1and FEEL-2are endocytotic receptors for advanced glycation end products.J.Biol.Chem.2002.278:12613–12617.
19Schledzewski,K.,Falkowski,M.,Moldenhauer,G.,Metharom,P., Kzhyshkowska,J.,Ganss,R.,Demory, A.et al.,Lymphatic endothelium-speci?c hyaluronan receptor LYVE-1is expressed by stabilin-11,F4/801,CD11b1macrophages in malignant tumours and wound healing tissue in vivo and in bone marrow cultures in vitro: implications for the assessment of lymphangiogenesis.J.Pathol.2006.
209:67–77.
20Gustafsson,C.,Mjosberg,J.,Matussek,A.,Geffers,R.,Matthiesen,L., Berg,G.,Sharma,S.et al.,Gene expression pro?ling of human decidual macrophages:evidence for immunosuppressive phenotype.PLoS One 2008.3:e2078.
21Nagamatsu,T.and Schust,D.J.,The contribution of macrophages to normal and pathological pregnancies.Am.J.Reprod.Immunol.2010.63: 460–471.
22Houser,B.L.,Tilburgs,T.,Hill,J.,Nicotra,M.L.and Strominger,J.L.,T wo unique human decidual macrophage populations.J.Immunol.2011.186: 2633–2642.23Pucci,F.,Venneri,M.A.,Biziato,D.,Nonis,A.,Moi,D.,Sica,A.,Di Serio,C.
et al.,A distinguishing gene signature shared by tumor-in?ltrating Tie2-expressing monocytes,blood‘‘resident’’monocytes,and embryonic macrophages suggests common functions and developmental relation-ships.Blood2009.114:901–914.
24Marttila-Ichihara,F.,Auvinen,K.,Elima,K.,Jalkanen,S.and Salmi,M., Vascular adhesion protein-1enhances tumor growth by supporting recruitment of Gr-11CD11b1myeloid cells into tumors.Cancer Res.2009.
69:7875–7883.
25Marttila-Ichihara,F.,Turja,R.,Miiluniemi,M.,Karikoski,M.,Maksimow, M.,Niemela,J.,Martinez-Pomares,L.et al.,Macrophage mannose receptor on lymphatics controls cell traf?cking.Blood2008.112:64–72.
26Movahedi,K.,Laoui,D.,Gysemans,C.,Baeten,M.,Stange,G.,Van den Bossche,J.,Mack,M.et al.,Different tumor microenvironments contain functionally distinct subsets of macrophages derived from Ly6C(high) monocytes.Cancer Res.2010.70:5728–5739.
27Kzhyshkowska,J.,Gratchev,A.,Brundiers,H.,Mamidi,S.,Krusell,L.and Goerdt,S.,Phosphatidylinositide3-kinase activity is required for stabilin-1-mediated endosomal transport of acLDL.Immunobiology2005.210: 161–173.
28Salmi,M.,Alanen,K.,Grenman,S.,Briskin,M.,Butcher, E. C.and Jalkanen,S.,Immune cell traf?cking in uterus and early life is dominated by the mucosal addressin madcam-1in humans.Gastroenterology2001.
121:853–864.
29Salmi,M.and Jalkanen,S.,Developmental regulation of the adhesive and enzymatic activity of vascular adhesion protein-1(VAP-1)in humans.
Blood2006.108:1555–1561.
30Horst,E.,Meijer,C.J.L.M.,Duijvestijn,A.M.,Hartwig,N.,Van der Harten,H.J.and Pals,S.T.,The ontogeny of human lymphocyte recirculation:high endothelial cell antigen(HECA-452)and CD44homing receptor expression in the development of the immune system.Eur.J.
Immunol.1990.20:1483–1489.
31Keuschnigg,J.,Henttinen,T.,Auvinen,K.,Karikoski,M.,Salmi,M.and Jalkanen,S.,The prototype endothelial marker PAL-E is a leukocyte traf?cking molecule.Blood2009.114:478–484.
32Stamper H.B.,Jr.and Woodruff,J.J.,Lymphocyte homing into lymph nodes:In vitro demonstration of the selective af?nity of recirculating lymphocytes for high-endothelial venules.J.Exp.Med.1976.144:828–833.
33Koskinen,K.,Vainio,P.J.,Smith,D.J.,Pihlavisto,M.,Yla-Herttuala,S., Jalkanen,S.and Salmi,M.,Granulocyte transmigration through endothelium is regulated by the oxidase activity of vascular adhesion protein-1(VAP-1).Blood2004.103:3388–3395.
Abbreviations:Ac-LDL:acetylated low density lipoproteináCLEVER-1: common lymphatic endothelial and vascular endothelial receptor-1áDQ-OVA:BODIPY FL-labeled OVAáHEV:high endothelial venuleáHUVEC:human umbilical vein endothelial cell
Full correspondence:Professor Marko Salmi,MediCity Research Laboratory,University of T urku,T ykisto¨katu6A,FIN-20520T urku, Finland
Fax:1358-2-333-7000
e-mail:marko.salmi@utu.?
Received:22/12/2010
Revised:21/3/2011
Accepted:4/4/2011
Accepted article online:7/4/2011