ARTICLE
Received 23 Jun 2015 | Accepted 14 Jan 2016 | Published 12 Feb 2016
Zhen Xu1,2, Fumio Takizawa1, David Parra3, Daniela Gmez1, Louise von Gersdorff Jrgensen4, Scott E. LaPatra5 & J. Oriol Sunyer1
Gas-exchange structures are critical for acquiring oxygen, but they also represent portals for pathogen entry. Local mucosal immunoglobulin responses against pathogens in specialized respiratory organs have only been described in tetrapods. Since sh gills are considered a mucosal surface, we hypothesized that a dedicated mucosal immunoglobulin response would be generated within its mucosa on microbial exposure. Supporting this hypothesis, here we demonstrate that following pathogen exposure, IgT B cells proliferate and generate pathogen-specic IgT within the gills of sh, thus providing the rst example of locally induced immunoglobulin in the mucosa of a cold-blooded species. Moreover, we demonstrate that gill microbiota is predominantly coated with IgT, thus providing previously unappreciated evidence that the microbiota present at a respiratory surface of a vertebrate is recognized by a mucosal immunoglobulin. Our ndings indicate that respiratory surfaces and mucosal immunoglobulins are part of an ancient association that predates the emergence of tetrapods.
1 Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania, 413 Rosenthal building, 3800 Spruce Street, Philadelphia, Pennsylvania 19104, USA. 2 Department of Aquatic Animal Medicine, College of Fisheries, Huazhong Agricultural University, Wuhan, Hubei 430070, China.
3 Department of Cell Biology, Physiology and Immunology, Universitat Autnoma de Barcelona, Barcelona 08193, Spain. 4 Laboratory of Aquatic Pathobiology, Department of Veterinary Disease Biology, Faculty of Health and Medical Sciences, University of Copenhagen, Frederiskberg DK-1870, Denmark. 5 Research Division, Clear Springs Foods Inc., P O Box 712, Buhl, Idaho 83316, USA. Correspondence and requests for materials should be addressed to J.O.S. (email: mailto:[email protected]
Web End [email protected] ).
NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 1
DOI: 10.1038/ncomms10728 OPEN
Mucosal immunoglobulins at respiratory surfaces mark an ancient association that predates the emergence of tetrapods
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728
Individual vertebrate species have faced unique evolutionary pressures to continuously acquire oxygen from the environment, which has led to the development of diverse gas-
exchange structures (that is, respiratory organs)1. In most vertebrates, respiration occurs through specialized gas-exchange surfaces (that is, gill, lungs), although a small percentage of gas exchange may also occur through the skin (that is, cutaneous respiration)2. Design of specialized respiratory organs has evolved in response to several factors, most critically the environment where the organism lives, but also the size and phylogenetic position of the organism1. In general, vertebrates living in aqueous media have evolved evaginated gas-exchanges structures (that is, gills) while those acquiring oxygen from air have developed invaginated structures (that is, lungs) to prevent desiccation. An effective gas-exchange organ needs to maximize the efciency with which oxygen from the external media (water/ air) comes in contact with blood. In the case of teleost sh this is accomplished by four pairs of vascularized gill arches composed of hundreds of gill laments, which increase their contact surface by folding into the secondary lamella. In addition to respiration, gills perform other functions including osmoregulation, pH balance, ammonia excretion, hormone regulation, detoxication and immune defense1. Gills are in direct contact with the water and therefore are continuously exposed to environmental toxins and pathogens. Thus, there is an evident need for the sh to defend such a large and delicate surface from pathogenic attack. While little is known about how gill immunity operates, it has been reported that gill tissue contains a signicant number of innate and adaptive immune cells and that several innate and adaptive immune molecules and pathways operate in the gill36.
Teleost sh contain three immunoglobulin classes IgM, IgD and IgT/Z7. While IgM represents the most abundant class, strong IgM immune responses to infection or vaccination are mainly detected in plasma, whereas IgM titres in mucosal tissues such as gut, skin or gills remain low7. Moreover, IgM has been shown to coat a signicant portion of the trout skin, gut and nasal microbiota810. Secreted IgD has been identied in catsh and trout plasma11,12, although its function in teleosts still remains undetermined. As opposed to trout secretory IgD, the secreted form of catsh IgD lacks a V region and it has been suggested that this immunoglobulin may function as an innate pattern recognition molecule11. In addition to IgM and IgD, teleosts contain a third immunoglobulin class, IgT (also called IgZ), rst identied at the genome level in 2005 (refs 13,14). Except catsh and medaka, all other teleost species studied express this immunoglobulin15. We have previously reported that teleost IgT is an immunoglobulin specialized for gut and skin mucosal immunity8,9. More specically, we showed that IgT is the main immunoglobulin induced in the gut and skin mucosa on pathogenic challenge, and we also demonstrated that IgT plays a prevalent role in the coating of the microbiota found in these surfaces. Moreover, we discovered a novel B-cell lineage uniquely expressing surface IgT which represents the predominant B-cell subset in the trout gut-, skin- and nasal-associated lymphoid tissues (GALT, SALT and NALT)810. Overall, sh IgT represents the most ancient mucosal immunoglobulin found in a vertebrate species. In fact, the newly attributed role of IgT in mucosal immunity has challenged the previous paradigm that specialization of immunoglobulin isotypes into mucosal and systemic responses arose during tetrapod evolution8.
Three B-cell subsets have been identied in catsh11: IgM IgD , IgM IgD , IgM IgD , while generally two B-cell subsets can be identied in all other species that contain IgT/IgZ, including the IgM IgD IgT and the IgM IgD IgT (ref. 12). Moreover, one study has recently described a new population of trout IgD B cells devoid of surface IgM expression
(IgD IgM B cells), which was shown to represent a major B-cell subset in gill16. Since the same antibodies used in the aforementioned study had failed to detect these IgDIgM B cells in a previous study12, it was suggested that this B-cell subset might only be present in an European trout line16.
Gills are considered a mucosal surface, and through convergent evolution they have evolved structures and components similar to those of the mammalian respiratory mucosa1. In this regard, the epithelia of gill primary and secondary lamella is covered by one to four layers of cuboidal or squamous cell epithelia17, similar to what it is found in the upper respiratory tract of mammals. Another similarity between these surfaces is the existence of local mucus-secreting cells (that is, goblet cells), which are characteristic of type I mucosal epithelia18, and represent the rst line of defense and cell protection. In the lungs, which represent a type I mammalian mucosal surface, secretory IgA (sIgA) is produced locally and is a key immunoglobulin involved in mucosal respiratory immunological responses1921. A number of reports have described the induction of gill pathogen-specic IgM titres on pathogenic challenge or vaccination2225, while antigen-specic responses mediated by other teleost immunoglobulins are thus far unknown. Since IgT in sh plays a key role in mucosal immunity in the gut and skin, it is conceivable that IgT plays a pivotal role in gill immunity. However, it has recently been suggested that trout IgD may play an important function in gill immune responses12,16,24. Thus, the role of the different teleost immunoglobulins in the gill remains to be clearly delineated.
Thus far, local mucosal immunoglobulin responses in specialized respiratory surfaces have only been described in tetrapods. This study was therefore undertaken to examine whether a primordial association between gas-exchange structures and specialized mucosal immunoglobulin responses exists in species preceding the tetrapods. To this end, here we examined whether dedicated mucosal B-cell and immunoglobulin responses were present in the gills of the rainbow trout, a model species in the eld of evolutionary and comparative immunology26,27. The ndings here support our hypothesis that IgT is the main gill immunoglobulin involved in pathogen-specic immune responses while we found no evidence of a gill IgD IgM B-cell subset or pathogen-specic
IgD responses. For the rst time in a vertebrate species we show that the microbiota associated with a respiratory organ is recognized by a mucosal immunoglobulin, as evidenced here by the prevalent coating of gill microbiota with IgT. In addition, we demonstrate the local generation of IgT and proliferation of IgT B cells on pathogenic challenge, thus providing the rst demonstration that dedicated mucosal immunoglobulins are locally induced at the mucosal surface of a non-mammalian species.
ResultsImmunoglobulins and B cells in the gill of trout. Using an anti-IgT antibody, we detected by western blotting a protein in gill mucus consistent with the reported molecular mass of IgT8,9,14. To identify this immunoreactive band, the protein was puried using an anti-IgT afnity column and liquid chromatography-tandem mass spectrometry (LC/MSMS) conrmed that the puried protein corresponded to IgT (Supplementary Fig. 1). Gel-ltration analysis showed that a majority of gill mucus IgT are present in polymeric form, as it eluted at a fraction similar to that of trout mucosal IgM, a tetrameric immunoglobulin (Fig. 1a). Interestingly, a portion of gill mucus IgT consistently eluted in monomeric form. As opposed to polymeric IgT, IgD eluted as a monomer at the molecular weight range previously reported for serum IgD12. Gill mucus polymeric IgT migrated as a monomer by SDSPAGE under non-reducing conditions, indicating that polymeric IgT is associated by non-covalent interactions (Fig. 1b,
2 NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728 ARTICLE
a b
mAU
1,200
900
600
300
IgM
IgD
IgT
Elution volume (ml)
670 kDa
158 kDa
44 kDa17 kDa
Elution volume (ml)
Anti-IgM Anti-IgD Anti-IgT
8.5 8.5 8.5
11.5 11.5 11.5
A 280
kDa
0
5.0 10.0 15.0 20.0 Elution volume (ml)
700
400
180
260
140
100
70
50
kDa
6.5 7.5
20.06
8.5 9.5 10.5 11.5 12.5 13.5
c d e f
30
concentration (g ml1)
8,000
30
7,000
30 19.8
6,000
Gill mucus
concentration (g ml1)
20
20
20
Serum
80
IgT/IgM (100)
IgD/IgM (100)
5,000
4,000
10.1
10
60
35.60
10
10
1.55
1.22
20
40 9.52
5,534
0.16
0.89
0 IgT IgD
IgM
0 IgT
IgM
IgD
0 Gill mucus Serum
0 Gill mucus
Serum
g h
105
60 52.5 47.5
105
104
103
102
101
104
Anti-IgT
103
8.97 9.59
Anti-IgT
105
104
103
102
101
Anti-IgM
40
6.95
Cells
(% of total B cells in gill)
102
20
7.45
105
7.28
101
0
101 105
Anti-IgD
102
103
104
102
101 105
Anti-IgD
103
104
101
102
103
104
IgT+ IgM+IgD+
Anti-IgM
Figure 1 | Protein characterization of gill mucus immunoglobulins and identication of B cells in the gill of trout. (a) Fractionation of gill mucus (B0.5 ml) by gel ltration (upper) followed by immunoblot analysis of the fractions with anti-trout IgM, anti-trout IgD and anti-trout IgT-specic mAbs (lower). A280, absorbance at 280 nm. (b) SDSPAGE of gel-ltration fractions (415%) corresponding to elution volumes of 8.5 and 11.5 ml under non-reducing conditions followed by immunoblot analysis with mAbs to trout IgM, IgD or IgT. (c,d) Immunoblot and densitometric analysis of the concentration of IgT, IgM and IgD in gill mucus (c) and serum (d) (n 15 sh). (e,f) Ratio of IgT to IgM concentration (e) and IgD to IgM concentration (f) in gill mucus
and serum, calculated from the values shown in c,d. (g) Dot plots of leucocytes from gills. Numbers adjacent to outlined areas indicate percentage IgT B cells, IgM B cells or IgD B cells in the lymphocyte gate. (h) Frequency of IgT and IgM IgD B cells among total B cells. Results in ce and h are expressed as mean and s.e.m. obtained from 15 individual shes.
right panel). As expected, mucosal IgD and IgM migrated as a monomer and a polymer, respectively, under the same SDSPAGE conditions (Fig. 1b, left and middle panels). Next, we found that although the protein concentration of IgT was B13- and B581-fold lower than that of IgM in gill mucus and serum, respectively, (Fig. 1c,d), the ratio IgT/IgM was B124-fold higher in the gill mucus when compared with that of serum (Fig. 1e). A similar situation occurs with IgD, which displayed the same concentration as IgT in mucus, while its serum concentration was B4 times more than that of IgT and B155 times less than that of IgM (Fig. 1c,d). The ratio IgD/IgM was
B10-fold greater in gill mucus than in serum (Fig. 1f). The abundance of B cells in the gills was analysed by ow cytometry. Two main B-cell subsets were detected in the gills (Fig. 1g), one expressing only surface IgT (IgT B cells), and the second one expressing both surface IgD and IgM (IgD IgM B cells). From these two populations, IgT B cells were more abundant in gills(52.5% of all B cells) (Fig. 1h), suggesting a role of this subset in gill immunity. For comparison, the same B-cell subsets were found in the trout NALT and spleen (Supplementary Fig. 2), where IgT and IgM B cells are the predominant NALT and spleen B-cell subset, respectively, as previously reported10.
IgT constitutes the main Ig isotype coating the gill microbiota. Similar to the situation in mammals for sIgA28,29, we have previously reported that a large fraction of the gut and skin bacterial microbiota are coated with IgT, and to a lesser extent with IgM8,9. Here we adapted the methodology, which we had used to assess immunoglobulin coating of trout gut and skin bacteria8,9, to the gill mucosa, and we extended it to the analysis of IgD coating. By ow cytometry, we detected that a large fraction of gill microbiota were coated with IgT (B51%), whereas a smaller proportion stained for IgM (B31%) or IgD (B10%) (Fig. 2a,b). Interestingly, we found that B4.23% of gill microbiota were double-coated with IgT and IgM and to a much lesser extent were double-coated with IgT and IgD (B1.09%), IgM and IgD (B0.91%) or triple-coated with the three immunoglobulins (B0.19%) (Fig. 2c). Similar results were obtained by immunouorescence microscopy (Fig. 2d; isotype-matched control antibodies, Supplementary Fig. 3). Western blot analysis further demonstrated the presence of the three immunoglobulins on gill bacteria (Fig. 2e). Notably, we found that more than 19% of total IgT present in gill mucus was used for bacterial coating, while only B1.5% of gill mucosal IgM and IgD was being utilized for that purpose (Fig. 2f).
NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 3
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728
a b c
P < 0.0001 P < 0.0001
100
Bacteria coated (%)
15
P = 0.0069
P = 0.0053
1,200
900
P < 0.0001 P < 0.0001
Isotype control Anti-IgT Anti-IgM Anti-IgD
Bacteria coated (%)
80
10
60
Cells
600
300
0 101 102 103 104 105
40
5
20
0 IgT IgM IgD
0 IgTIgM
IgTIgD
IgMIgD
IgTIgMIgD
d Nuclei Anti-IgT
Anti-IgM Anti-IgD Merge
IgT IgM IgD nuclei
e
f
1 2 3 4 5 6 7
80 P < 0.0001
IgT
IgM
IgD 400
kDa
700
Ig coated to bacteria
(% of total)
180
60 P < 0.0001
40
20
0 IgT IgM IgD
Figure 2 | Trout gill bacteria are predominantly coated with IgT. (a) Representative ow cytometry histograms showing the staining of gill bacteria with IgT, IgM and IgD. Bacteria were stained with anti-trout IgT (green line), anti-trout IgM (red line) or anti-trout IgD (magenta line) mAbs or isotype controls (shaded histograms). (b,c) Percentage of gill bacteria coated with IgT, IgM or IgD (b) or coated with IgTand IgM, IgTand IgD, IgM and IgD, IgTand IgM and IgD (n 16) (c). The median percentage is shown by a red line. Statistical differences were evaluated by the nonparametric MannWhitney test. (d)
Differential interference contrast (DIC) images of gill bacteria stained with a DAPI-Hoeschst solution (blue), anti-IgT (green), anti-IgM (red) or anti-IgD (magenta), and merging IgT, IgM and IgD stainings (Merge). (Isotype-matched control antibody staining is shown in Supplementary Fig. 3 online). (Scale bars, 5 mm). (e) Immunoblot analysis of IgT, IgM and IgD on gill bacteria. Lane 1, 0.1 mg of puried IgT, IgM or IgD; lanes 27, gill bacteria (n 6 sh). (f)
Percentage of total gill mucus IgT, IgM or IgD coating gill bacteria (n 13). The median is shown by a red line. Statistical differences were evaluated by one
way ANOVA with Bonferroni correction. Data are representative of at least three independent experiments. ANOVA, analysis of variance.
Pathogen-specic immunoglobulin responses in the gill. The prevalence of IgT B cells in the gill of trout and the predominant coating of gill bacteria by IgT led us to hypothesize an important role of IgT in gill mucosal immunity. However, a role for IgD and a putative subset of IgD/IgM B cells have been suggested in the gills of teleosts12,16,24. To evaluate which immunoglobulin played a dominant role in gill immunity, we assessed immunoglobulins and B-cell responses in sh exposed to a parasite, Ichthyophthirius multiliis (Ich), that has a tropism for the skin and gills of salmonids30. After infection, we could detect the presence of Ich trophonts in the gills of infected sh by immunouorescence microscopy (Fig. 3) using an anti-Ich antibody9. Strikingly, 25 days post infection most gill parasites were overwhelmingly recognized by IgT, while only some parasites were slightly coated with IgM (Fig. 3ad; isotype-matched control antibodies, Supplementary Fig. 4a). However, we could never detect parasites coated by IgD (Supplementary Fig. 5). Immunouorescent micrographs of gill samples showed very few IgT and IgM B cells in the gill laments of control sh (Fig. 4a). However, 25 days post infection, a moderate
(B3.1-fold) but signicant increase of IgT B cells was observed in the gills of infected sh (Fig. 4b). Notably, survivor sh (3 months post infection) contained large accumulations of IgT
B cells in the gill laments when compared with those of control sh (Fig. 4c; isotype-matched control antibodies, Supplementary Fig. 4b). It is worth mentioning that IgT B cells were noticeably accumulated in the primary lamella, while a few were present in the secondary lamella of survivor sh (Fig. 4c). In contrast, the number of IgM B cells did not change in the infected and survivor sh when compared with the controls (Fig. 4ac;
isotype-matched control antibodies, Supplementary Fig. 4b). Interestingly, we observed that some IgT B cells appeared to be secreting IgT (Fig. 4c, white arrows) and we noted an increase in cell size within the IgT population (Supplementary Fig. 6), which suggested the presence of IgT-producing plasmablasts- or plasma-like cells. Results from ow cytometry of B-cell populations in gills from control, infected and survivor sh mirrored those obtained by immunouorescence microscopy. Thus, it was observed that B6.92.4% of cells from the lymphocyte gate were
IgT B cells in the gills of control sh whereas a signicant
4 NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728 ARTICLE
Anti-Ich DIC merge
Anti-IgM Anti-IgT
Ich IgM IgT nuclei
a
b
c
d
Figure 3 | Parasites are mainly coated by IgT in the gill of infected trout. Four different microscope images (ad) of slides showing immunouorescence staining of Ich parasites in gill cryosections from trout infected with Ich after 25 days (n 6). From left to right: Ich (magenta),
IgM (red) and IgT (green) with nuclei stained with DAPI (blue); DIC images showing merged staining (isotype-matched control antibody staining, Supplementary Fig. 4a online). Scale bars, 20 mm.
increase of IgT B cells was found in infected (B10.52.5%) and survivor sh (B15.23.7%) (Fig. 4d). In contrast, the percentage of IgM B cells did not change in any sh group (Fig. 4d). Along with this increment of IgT B cells observed in gills of infected and survivor sh, the IgT protein concentration in the gill mucus of these sh groups increased by B7- and B10-fold, respectively, when compared with control sh, while the levels of IgM and IgD remained unchanged (Fig. 4e). In serum, only approximately two and threefold increase of IgT concentration was detected in infected and survivor sh, respectively. The concentration of serum IgM increased approximately three and twofold in infected and survivor sh, respectively, when compared with control sh (Fig. 4f). In contrast, the concentration of IgD did not change signicantly either in the gill mucus or serum of the same sh groups (Fig. 4e,f).
The large increases in IgT B cells and IgT protein levels in the gills of infected and survivor sh suggested the generation of
IgT-specic immune responses. Supporting this prediction, we detected signicant increases in parasite-specic IgT binding in up to 1/40 and 1/100 diluted gill mucus of infected and survivor sh, respectively, in which we found B5.2- and B3.9-fold binding increases, respectively, when compared with that of control gill mucus (Fig. 5ac). In serum, however, parasite-specic IgT binding was detected only at the 1/10 serum dilution of the survivor group. In contrast, parasite-specic IgM binding were observed in up to 1/1,000 and 1/4,000 serum dilutions from infected and survivor sh, respectively (Fig. 5df). We could not detect any parasite-specic IgD binding either in the gill mucus or serum from infected and survivor sh (Fig. 5af).
To further support that IgT is the predominant immunoglobulin induced in the gill mucosa on pathogenic challenge, we also infected sh with Flavobacterium columnare, a bacterial pathogen that infects the gills of salmonids. We detected approximately vefold increase in bacteria-specic IgT binding in 1/10 diluted gill mucus of infected and survivor sh, when compared with the same dilution of control gill mucus (Supplementary Fig. 7ac). In serum, bacteria-specic IgT binding was detected at the 1/10
serum dilution of the infected and survivor group. However, bacteria-specic IgM binding was observed in up to 1/100 serum dilution from infected and survivor sh (Supplementary Fig. 7df). Similar to what we found in Ich, we could not detect any bacteria-specic IgD binding in the gill mucus and serum from infected and survivor sh (Supplementary Fig. 7af).
Local proliferative and immunoglobulin responses. The substantial increase of IgT B cells observed in the gill of survivor sh could be due to an inltration of these cells from systemic lymphoid organs or to a process of local IgT B proliferation. To clarify this point, we assessed the in vivo proliferative responses of
B cells on treatment of control and survivor sh with uorescent EdU (5-ethynyl-20-deoxyuridine, Alexa Fluor 488), a thymidine analogue that incorporates into DNA during cell division31. By ow cytometry, we detected an overall increase in proliferation (B2.6-fold) within the whole gill leucocyte population of survivor sh when compared with that of control animals (Supplementary Fig. 8). Interestingly the largest increase in proliferation occurred within the B-cell population (B8.8-fold), followed by the large cells (B2.6-fold) and the non-B-lymphocyte (B2.4-fold) subsets. Within the B-lymphocyte population, we detected a signicant increase in the percentage of proliferating IgT B cells (EdU IgT B cells) in the gills of survivor sh (B8.11.8%
of all IgT B cells) when compared with that of control animals (B2.50.77% of all IgT B cells) (Fig. 6a,c). However, no differences were detected in the percentage of EdU IgM B cells between control and survivor sh (Fig. 6b,c). Immunouorescent microscopic analysis of gills from EdU-injected sh yielded results very similar to those obtained by ow cytometry. Thus, we found an approximately fourfold increase in the number of EdU IgT B cells in survivor sh when compared with that of control sh, while no differences in proliferating IgM B cells were detected between these two groups (Fig. 6df). Systemic lymphoid tissues of survivor sh, however, did not show increased IgT B-cell proliferative responses when compared with control animals, while a rise of EdU IgM B cells was detected in head kidney of survivors (Supplementary Fig. 9).
These results suggested that the parasite-specic IgT responses shown in Fig. 5 were locally generated in the gills rather than being produced and transported from systemic lymphoid organs. To test this hypothesis, we analysed parasite-specic immunoglobulin titres from medium derived of cultured gill, head kidney and spleen explants obtained from control and survivor sh. Supporting our prediction, we found that the highest parasite-specic IgT responses were detected in the survivor gill explants, whereas the dominant IgM responses were localized in the head kidney and spleen explants (Fig. 7). Conversely, low or negligible parasite-specic IgT responses were detected in the survivor spleen and head kidney explants while very low IgM responses were measured in gill explants (Fig. 7).
pIgR in the gill of trout. One important feature of sIgA and sIgM in mammals is their transport to mucosal surfaces by the polymeric immunoglobulin receptor (pIgR)32. We have previously found that the secretory component of trout pIgR (tSC) is associated to IgT and IgM in the gut and skin mucus, respectively8,9. In this study, the tSC was detected in gill mucus but not in the serum (Fig. 8a). To assess whether tSC was associated to gill mucus IgT, IgM and IgD, we carried out coimmunoprecipitation assays in gill mucus using antibodies against tSC (tpIgR) and immunoglobulins. Our results showed that antibodies against trout IgT, IgM and IgD were able to coimmunoprecipitate tSC in gill mucus (Fig. 8b for IgT, Supplementary Fig. 10a,b for IgM and IgD). Moreover, IgT, IgM and
NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 5
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728
a
Control gill
Infected gill
Survivor gill
SL
SL SL SL
SL
SL
SL PL
PL
SL SL
SL
SL
PL PL
PL PL PL
PL
b
SL
SL
SL
PL
SL
SL
SL
SL
PL
PL
PL
c
SL SL SL SL
SL
PL PL PL
PL PL
SL
IgT IgM nuclei
d e f
30
Control Infected Survivor
(% lymphocytes in gill)
Serum Ig concentration
(g ml1)
35,000
Control Infected Survivor
40 Control
Infected Survivor
30,000
**
30
25,000
**
20
20,000
B cells
20
***
***
15,000
10
10,000
10
5,000
60
***
***
**
30 0 IgT IgM IgD
0 IgT+ IgM+IgD+
Gill mucus Ig concentration
(g ml1)
0 IgT IgM IgD
Figure 4 | Increases in IgT B cells in the gill of trout infected with Ich. (ac) DIC images of immunouorescence staining on trout gill cryosections from uninfected sh (a), 25 days infected sh (b) and survivor sh (c, upper), stained for IgT (green) and IgM (red); nuclei are stained with DAPI (blue)
(isotype-matched control antibody staining, Supplementary Fig. 4b online). Enlarged images of the areas outlined in c are showing some IgT B cells possibly secreting IgT (white arrowhead, c, lower). Primary lamellae (PL) and secondary lamellae (SL) are shown. Scale bars, 20 mm. Data are representative of at least three different independent experiments (n 89 per group). (d) Percentage of gill IgT and IgM IgD B cells in the
lymphocyte gate of uninfected control sh, infected sh and survivor sh measured by ow cytometric analysis. (e,f) Concentration of IgT, IgM and IgD in gill mucus (e) and serum (f) of control, infected and survivor sh (n 12 per group). **Po0.01 and ***Po0.001 (one-way ANOVA with Bonferroni
correction). Data in df are representative of at least three independent experiments (mean and s.e.m.). ANOVA, analysis of variance.
IgD in the gill mucus also could be immunoprecipitated by the anti-tpIgR antibody (Fig. 8c for IgT, Supplementary Fig. 10a,b for IgM and IgD). Using immunouorescence microscopy, most of the pIgR-containing cells were observed in the primary gill lamella, while a few pIgR-positive cells were located in the secondary gill lamella. We also noticed that some of those pIgR-containing cells were associated with IgT staining (Fig. 8d,e, isotype-matched control antibody in Supplementary Fig. 4c).
DiscussionSpecialized gas-exchange structures are constantly exposed to the environment and the pathogens therein. Both invaginated (that is,
lungs) and exvaginated (that is, gills) respiratory organs are separated from their environment by a mucosal epithelium, which is likely to have been subjected to comparable selective forces throughout evolutionary time. Thus, we hypothesized that respiratory mucosal tissues in both primeval and modern bony vertebrates evolved similar molecular mechanisms for mucosal adaptive immunity. This study was therefore undertaken to examine which B-cell and immunoglobulin responses play a key role in controlling pathogens and the microbiota in the gill of rainbow trout, a teleost sh model representing one of the oldest living bony vertebrate species.
In this study, we provide the rst structural and functional characterization of all teleost immunoglobulins, including
6 NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728 ARTICLE
a b c
Gill mucus Ig binding to
parasite (1:2 dilution)
Serum Ig binding to
parasite (1:10 dilution)
kDa
700
Control
Parasite-specific Ig binding (relative to control)
180
400
700
IgT IgM
25 IgT IgM IgD
Parasite-specific Ig binding
(relative to control)
20
25 IgT IgM IgD
Survivor gill mucus dilution
20
***
*** **
700
IgD
IgD
IgD
IgD
15
15
Control
Infected
Survivor
10
**
10
** *
*
180
400
5
** *
5
IgT
IgM
*
0
1:2 1:10 1:40 1:100
0 1:2 1:10 1:40 1:100
Infected gill mucus dilution
d e f
kDa
Control
180
400
700
180
400
IgT
IgM
20 IgT IgM IgD
Infected serum dilution
Parasite-specific Ig binding
(relative to control)
20 IgT IgM IgD
Survivor serum dilution
Parasite-specific Ig binding
(relative to control)
***
15
15
Control
Infected
Survivor
10
**
10
**
IgT
IgM
*
5
5
* *
** *
0
0
1:10 1:100 1:1000 1:4000
1:10 1:100 1:1,000 1:4,000
Figure 5 | Immunoglobulin responses in the gill mucus and serum from infected and survivor sh. (a) Western blot analysis of IgT-, IgM- and IgD-specic binding to Ich in gill mucus (dilution 1:2) from infected and survivor sh. (b,c) IgT-, IgM- and IgD-specic binding to Ich in dilutions of gill mucus from infected (b) and survivor (c) sh, evaluated by densitometric analysis of immunoblots and presented as relative values to those of control sh (n 9
per group). (d) Western blot analysis of IgT-, IgM- and IgD-specic binding to Ich in serum (dilution 1:10) from infected and survivor sh. (e,f) IgT-, IgM-and IgD-specic binding to Ich in dilutions of serum from infected (e) and survivor (f) sh, evaluated by densitometric analysis of immunoblots and presented as relative values to those of control sh (n 9 per group). *Po0.05, **Po0.01 and ***Po0.001 (unpaired Students t-test). Data are
representative of at least three independent experiments (mean and s.e.m.).
0
a b c
Control
Control
Survivor
Control Survivor
104
104
20
0.09% 1.08% 0.13% 0.18%
EdU+ Ig+ cells (%)
Control Survivor
103
103
15
102
EdU
102
EdU
10
101
101
5
100
100 104
103
100
100 104
103
101
102
103
104
100
101
102
101
102
103
104
100
101
102
IgT+
IgM+
IgM+
Anti-lgT
Anti-lgM
d e f
EdU IgT/IgM/EdU EdU IgT/IgM/EdU
20
Control Survivor
***
EdU+ Ig+ cells (%)
SL
15
***
10
SL
PL PL
PL SL SL SL
SL
5
PL
0
IgT+
Survivor
EdU IgT IgM nuclei
Figure 6 | Proliferative responses of IgT and IgM B cells in the gill of sh that survived Ich infection. (a,b) Representative ow cytometry dot plot showing proliferation of IgT B cells (a) and IgM B cells (b) in gill leucocytes of control and survivor sh. The percentage of lymphocytes representing proliferative B cells (EdU ) is shown in each dot plot. (c) Percentage of EdU cells from the total gill IgT or IgM B-cell populations in control or survivor sh (n 12). (d,e) Immnouorescence analysis of EdU incorporation by IgT or IgM B cells in gill of control (d) and survivor sh (e). Gill cryosections were
stained for EdU (magenta), trout IgT (green), trout IgM (red) and nuclei (blue) detection (n 9). Primary lamellae (PL) and Secondary lamellae (SL) are
shown. White arrowheads point to cells double stained for EdU and IgT. Scale bars, 20 mm. Data are representative of at least three different independent experiments. (f) Percentage of EdU cells from the total gill IgT or IgM B-cell populations in control or survivor sh counted from d,e. Data in c and f are representative of at least three independent experiments (mean and s.e.m.). Statistical analysis was performed by unpaired Students t-test. ***Po0.001.
NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 7
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728
a b c
Gill medium Ig binding to parasite (1:2 dilution)
Head kidney medium Ig binding to parasite (1:2 dilution)
Spleen medium Ig binding to parasite (1:2 dilution)
kDa
Control
Survivor
kDa
kDa
IgT
IgM
Survivor
Control Survivor
Control
IgT
IgM
180 IgT
700 IgM
IgD
180
700
400
180
700
400
IgD
IgD
20 IgT IgM IgD
10
10
**
IgT IgM IgD
Parasite-specific Ig binding
(relative to control)
IgT IgM IgD
10
Parasite-specific Ig binding
(relative to control)
Parasite-specific Ig binding
(relative to control)
***
**
*
8
8
15
**
6
6
4
4
** *
*
*
5
*
2
2
0
0
0
1:2 1:10 1:40 1:100
1:2 1:10 1:40 1:100
1:2 1:10 1:40 1:100
Survivor gill medium dilution
Survivor head kidney medium dilution
Survivor spleen medium dilution
Figure 7 | Local IgT-, IgM- and IgD-specic responses in gill explants of survivor sh. Gill, head kidney and spleen explants (B50 mg each) from control and survivor sh were cultured for 7 days. (ac) Western blot analysis of IgT-, IgM- and IgD-specic binding to Ich in the culture medium of gill (a, upper), head kidney (b, upper) and spleen (c, upper) (dilution 1:2) from control and survivor sh. (ac) IgT-, IgM- and IgD-specic binding to Ich in dilutions of culture medium from gill (a, lower), head kidney (b, lower) and spleen (c, lower) from control and survivor sh, measured by densitometric analysisof immunoblots and presented as relative values to those of control sh (n 912 per group). *Po0.05, **Po0.01 and ***Po0.001 (unpaired
Students t-test). Data are representative of at least three independent experiments (mean and s.e.m.).
secreted IgD (sIgD), IgM (sIgM) and IgT (sIgT) at a mucosal surface of a teleost sh. We found that sIgT is mainly polymeric in the gill mucus, as previously shown for gut and skin sIgT8,9. In contrast, sIgD was in monomeric form as previously reported for serum IgD12. Overall, the amount of sIgT in gill mucus was found to be approximately vefold higher and approximately fourfold lower than that reported in the skin and gut mucus, respectively8,9. Thus, as in mammals3335, mucosal immunoglobulin concentrations vary at individual mucosal surfaces in sh. In addition, we found that the ratios of IgD/IgM and IgT/ IgM in gill mucus were 10-fold and 100-fold higher, respectively, than those found in trout serum. The signicantly higher IgT/ IgM ratio in the gill mucus suggested a potential role for IgT in gill immunity.
The presence of all three trout immunoglobulins in the gill mucus implied a transport mechanism from the epithelium into the gill luminal area. Similar to sIgA and sIgM in the mammalian respiratory tract36, we observed that the putative tSC was associated with sIgT, sIgM and sIgD in gill mucus. Furthermore, we found that tpIgR was mainly expressed in the epithelial cells at the wall of the gill primary lamella, thus implying a role for gill tpIgR in the transport of all three immunoglobulins into the gill mucus. Notably, this is the rst description of a pIgR implicated in IgD secretion in a vertebrate respiratory tract, since in mammals the receptor responsible for IgD secretion in the lung lumen is still unknown37. While the lack of a J chain in mammalian IgD suggested that it would not be transported by the pIgR, all trout immunoglobulin classes as well as amphibian IgX (a dedicated mucosal immunoglobulin that is secreted into the gut lumen) also do not associate with J chain38. In light of this data, it is tempting to hypothesize that mammalian IgD may use pIgR for its transport into the lung lumen. Although the mechanisms of interaction of trout mucosal IgM, IgD or IgT with tpIgR in the absence of J chain are currently unclear, a recent theoretical model predicts that interaction of IgT to tpIgR can occur in the absence of a J chain and would be mediated by the binding of
domains 1 and 5 of tpIgR with the constant domain 3 of the IgT heavy chain39.
In addition to being required for the transport of sIgA and sIgM into the lung lumen, the SC is also implicated in anchoring mucosal antibodies in mucus, ensuring immune exclusion at mucosal surfaces4042. Indeed, sIgA plays a key role in immune exclusion, as sIgA-coated bacteria are prevented from trans-locating from the luminal mucus into the epithelium, thereby avoiding potential systemic infections. To determine if mucosal immunoglobulins mediated immune exclusion in the teleost gills, we analysed the capacity of each gill immunoglobulin to associate with microbiota. Notably, a large population of bacteria in the gill mucus was predominantly coated by sIgT and, to a much lesser degree also by sIgM and sIgD. These data parallel that reported for trout gut, skin and nose microbiota810, thus further highlighting that sIgT is the main immunoglobulin interacting with microbiota at all teleost mucosal surfaces. While at a signicantly lower level than that of sIgT, we also demonstrate that teleost sIgD can coat microbiota. Whether this capacity is conserved in sIgD from mammals or other vertebrates, remains to be investigated. Together, our ndings represent the rst identication of microbiota recognition by a specialized mucosal immunoglobulin in the gas-exchange surface of a vertebrate species. Although the lungs of mammals were thought to be sterile and devoid of microbiota43, the presence of a diverse microbiota has recently been conrmed in the lungs of human44 and mice45. Based on our ndings suggesting a key role for sIgT in microbiota homeostasis in the gills of teleosts, it seems reasonable to predict that sIgA may also play a key role in the homeostasis of the lung microbiota. In fact, we propose that since the gill is an evaginated respiratory surface, it offers an excellent model to study the role of mucosal immunoglobulins with regard to respiratory microbiota homeostasis, since it allows for easy and clean evaluations of both the microbiota and IgT. In contrast, in mammals, studies of the lung microbiome are often confounded by potential contamination/carryover of the upper airway microbiome into the lower airways44,46.
8 NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728 ARTICLE
a b
Serum
Gill mucus
(kDa)
193.8
120.3
88.9
36.231.0
16.3
ColP kDa
kDa
Prebleed -IgT
Prebleed -pIgR
-pIgR
-pIgR
45
45
IB
IB
-IgT
-IgT
180
c
ColP
180
6.6
d
Anti-pIgR Anti-IgT Merge
SL SL PL
SL SL PL
SL SL PL
PL
PL PL
pIgR IgT nuclei
e
Figure 8 | Trout pIgR associates with gill sIgs. (a) SDSPAGE under reducing conditions of trout serum and gill mucus (B5 mg each), followed by immunoblot analysis using anti-trout pIgR antibody. (b) Co-immunoprecipitation (CoIP) of gill mucus with anti-trout IgT antibody, followed by immunoblot analysis (IB) under reducing conditions (pIgR detection, upper panels) or non-reducing conditions (IgT detection, lower panels). (c) CoIP of gill mucus with rabbit anti-trout pIgR followed by IB under non-reducing conditions (IgT detection, upper panels) and reducing conditions (pIgR detection, lower panels). IgG puried from rabbits serum before immunization (Pre-bleed) served as negative control for rabbit anti-trout pIgR and rabbit anti- trout IgT, respectively (left lane on each panel for b,c). (d) Immunouorescence staining for pIgR with IgT in gill cryosections of rainbow trout. Differential interference contrast images of gill cryosections were stained with anti-trout pIgR (magenta), anti-trout IgT (green) and DAPI for nuclei (blue) (n 9). (isotype-matched control
antibodies for anti-pIgR in Supplementary Fig. 4c online) (e) Enlarged sections of the areas outlined in d without DIC showing some pIgR/IgTcolocalization (white arrowhead). Scale bars, 20 mm. Primary lamellae (PL) and secondary lamellae (SL) are shown. Data are representative of at least three independent experiments.
To date, gill immunoglobulin and B-cell responses have been poorly studied. Thus far, the only reported pathogen- or antigen-specic antibody responses in the gill pertain to the IgM class. In the catsh, Ich-specic IgM responses have clearly been shown30,47. Interestingly catsh is a teleost species that appears to lack IgT. In rainbow trout, IgT- and IgM-coated Ich trophonts have been observed in the gills of immune sh, and IgT and IgM transcript levels were moderately increased in the gills of immune sh after challenge with Ich48,49. This suggested that IgT and IgM responses against the parasite may occur in the gills. Based on these data, we chose this parasite to evaluate the specic contribution of sh immunoglobulin isotypes in mediating mucosal responses against the parasite in the gills. Importantly, our study has addressed for the rst time the detection of pathogen-specic titres of all three existing teleost immunoglobulins in a mucosal tissue. Overall, we found that strong
specic IgT immune responses against the parasite were specically detected in the gill mucus, while parasite-specic IgM responses were almost exclusively detected in the serum. Moreover, studies performed with a bacterial pathogen (F. colmunare) supported the results obtained with the parasite, thus establishing further the prevalent induction of pathogen-specic IgT and IgM responses in the gill mucus and serum, respectively. Thus, similar to what we have previously reported for the skin and gut mucosa8,9, these ndings demonstrate that IgT responses are conned to mucosal surfaces while IgM responses are systemic (that is, serum responses). Importantly, it is worth noting that this represents the rst study in which a bacterial pathogen is shown to induce dominant IgT responses in a sh mucosal site, thus suggesting that IgT is probably induced by a variety of pathogens in addition to parasites. In contrast to recent reports hypothesizing a role for IgD in the gills12,16,24, our
NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 9
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728
ndings suggest that this antibody is not involved in pathogen-specic immunity in the gill mucosa. Moreover, we were unable to detect a recently reported B-cell subset expressing surface IgD, but devoid of surface IgM16 in either naive, infected or survivor sh. Thus, it would appear that the only role of IgD in the gill mucosa is the recognition and coating of a small percentage of the microbiota. However, at this point we cannot exclude the possibility that different pathogens or stimulation routes from the ones used here may induce relevant IgD responses in the gill or other mucosal or systemic compartment. Overall, the gill IgT responses to pathogens and microbiota described here recapitulate those found in the skin and gut mucosa. However, we detect both the highest pathogen-specic IgT titres and the greatest percentages of microbiota coated by IgT in the gut8 and gills. In the skin, the lower mucosal IgT titres and lower percentage of coated microbiota correlate with the lower percentages of IgT B cells and lower concentrations of skin
IgT in naive sh when compared with what was found in gill and gut8,9. This may indicate that potency of the IgT responses in sh mucosal areas is related to the relative abundance of IgT B cells in those sites. Moreover, the steady-state levels of IgT B cells in these mucosal tissues may depend on the composition and/or abundance of their microbiota since in mammals it is well-known that the production of IgA and number of IgA B cells in the gut is dependent on the abundance of microbiota50. Indeed, important differences exist in the diversity and composition of the microbiota of trout mucosal surfaces, including the gut, gills, nose and skin51. Hence, it is conceivable that differences in the microbiota in the different mucosal surfaces may drive the steady-state abundance of IgT B cells, which in turn may modulate the potency and repertoire of IgT responses in that mucosal area. Further work is clearly warranted to conrm this hypothesis.
The overwhelming prevalence of IgT versus IgM titres detected in the gill mucus was reected in the degree of IgT coating on the gill parasites, which was much higher than that seen for IgM. In contrast, previous studies in the skin and gut mucosa showed that pathogens were selectively coated by IgT with negligible IgM8,9. The minor but consistent staining of many parasites by IgM is probably due to the low but detectable local production of parasite-specic IgM in the gill of survivor sh. Alternatively, it is possible that microlesions produced by the parasite may result in leakage of plasma from damaged microvessels, thus enabling specic IgM from plasma to bind the parasite. Interestingly in mammals, both IgA and IgG play an important role in the lung mucosal immune responses, whereas in the upper respiratory tract and all other inductive sites, IgA is the main player52.
The high titres of pathogen-specic IgT detected in gill correlated with large accumulations of IgT B cells in the same animals. We have previously shown similar increases in IgT B cells in the skin and gut of sh that have survived Ich and
Ceratonova Shasta infections, respectively8,9. Whether such IgT B-cell accumulations are the result of local B-cell proliferation or migration from other lymphoid organs remains a critical question. To determine if the Ich-specic IgT B-cell response occurred locally, we developed innovative in vivo proliferation assays and gill explant cultures. Here we show for the rst time potent local proliferative IgT B-cell responses and pathogen-specic IgT production in the mucosal-associated lymphoid tissues (MALT) of a sh species. Such responses strongly suggest that accumulation of IgT B cells in the gills is due to their local proliferation, rather than migration from other organs. In further support of localized proliferation, IgT B-cell proliferative responses were not detected in the head kidney or the spleen of the same sh. Since mucosal B cells proliferate at the inductive sites in mammals, our results imply that the gills are inductive
sites for the generation of IgT mucosal immune responses. In addition, the production of high titres of Ich-specic IgT in gill explant cultures, conrmed the local production of Ich-specic IgT in gill tissue. In which specic area of the gill the immune response develops, remains to be determined. In recent years a novel gill tissue, the interbranchial lymphoid tissue (ILT) has been described and shown to be a rich T-cell area, although low transcripts levels of IgM and IgT have also been detected within this tissue53. Due to the high likelihood that the pathogen-specic gill IgT responses observed in this study are T-cell dependent, it is conceivable that at least a portion of local BT-cell interactions involved in this response may have occurred within or nearby the ILT. However, these BT-cell interactions may also occur in other areas where IgT B cells proliferate, as shown in this study.
Future studies are warranted to determine the importance of the ILT in pathogen-specic IgT responses in the gill.
Together, our data support a model in which on parasite infection, antigen-presentation, as well as B-cell activation and proliferation occurs within the gills, leading to the production of parasite-specic IgT (Fig. 9). At this point, we cannot exclude the possibility that on gill infection, parasite antigens are taken up by gill antigen-presenting cells (APCs) which then migrate into systemic lymphoid tissues (that is, head kidney and/or spleen). At that point, activated IgT B cells generated systemically would be imprinted to migrate into the gill where they would proliferate and turn into plasmablast or plasma cells to produce parasite-specic IgT. Alternatively, since gills are highly vascularized, antigens from parasite debris could travel from the gills into the lymphoid tissues through the vascular system, where they could be taken up by systemic APCs to initiate an immune response. Future work is warranted to address further the detailed mechanisms underlying the local production of IgT-specic responses in the gill. Finally, it is worth noting that proliferating IgT B cells in the gills appear to be randomly distributed either alone or in small groups. This would suggest the absence of organized lymphoid structures in this tissue (that is, lymph nodes), which is in agreement with the already well-known lack of organized lymphoid structures in teleost sh7,54.
In conclusion, we have identied a previously unrecognized key role of the IgT system in mucosal immune responses to pathogens and microbiota in the gill of a teleost sh. More critically, we show the generation of local IgT B-cell proliferative and pathogen-specic IgT responses in the gills, thus providing the rst demonstration of locally induced B-cell and secretory immunoglobulin responses in the mucosa of a non-tetrapod species. Moreover, we show for the rst time that the microbiota present within the main respiratory organ of a vertebrate is recognized by a dedicated mucosal antibody class. Thus, our ndings indicate that the recognition of pathogens and microbiota by specic mucosal immunoglobulins in specialized respiratory surfaces forms part of a conserved immune strategy that predates the emergence of tetrapod species. In addition, our results reinforce the formerly stated view26 that specialized mucosal immunoglobulins (that is, IgA, IgT) of all vertebrate MALTs function under the guidance of primordially conserved principles. Since many sh pathogens enter the host through the gill mucosa, our ndings also have special relevance from an applied perspective as they may lead to the development of sh vaccines that have the capacity to induce gill IgT mucosal immune responses.
Methods
Fish maintenance. Rainbow trout obtained from Limestone Springs Fish Farm were maintained in the laboratory of J.O.S. as described8. Fish were acclimatized for 14 days at 15 C in an aerated aquarium with internal biolters and feddaily with dry pellets at 1% biomass. All animal procedures were approved by the Institutional Animal Care and Use Committees of the University of Pennsylvania.
10 NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728 ARTICLE
Ich infection
Migration to spleen/head kidney ?
b
Gill
Spleen/head kidney
Ag
Blood presentation
vessel
a
SL PL SL
Ag presentation
T
T
T
B
B-cell activation
B-cell proliferation
c
B
T
B-cell activation
IgT-coated microbiota
Mucus layer
Differentiation to plasma cells
B-cell proliferation
Migration to infection sites ?
?
B
Cell surface IgT
Ich
Ich Ag
B
T
IgT+ B-cells
CD4+ T cells
Epithelial cells
Ag-presenting cells
Plasma cells
Polymeric IgT
Polymeric IgT with tSC
MHC/TCR/Ag complex
Cytokines
Unknown
Commensal bacteria
pIgR
?
Figure 9 | Proposed model of local IgT and IgT B-cell induction in the gills. (a) Scheme of a typical teleost gill lament displaying the primary (PL) and secondary (SL) lamellae. (b) Induction of local IgT responses in the gill based on our ndings. On Ich infection, Ich antigen (Ag) are taken up by antigen-presenting cells (APC) and presented to naive CD4-T cells. B cells are then activated by Ag-specic CD4-T cells and start proliferating. Activated B cells may differentiate further into plasmablasts or plasma cells to produce large amounts of parasite-specic IgT that recognize Ich within the gill PLs or the Ich present in the mucus. We cannot rule out the possibility that on antigen uptake, loaded gill APCs may migrate into central secondary lymphoid organs (that is, spleen or head kidney) where the resulting activated IgT B cells may then home into the gill. Alternatively, since gills are highly vascularized, antigens from parasite debris may travel from the gills into the lymphoid tissues through the vascular system, where they could be taken up by systemic APCs to initiate an immune response. In the potential scenarios (as indicated by a ?) where immune responses develop in the spleen or head kidney, we propose that IgT B-cell proliferation does not occur in the spleen or head kidney because in Supplementary Fig. 9 we do not detect differences in IgT B-cell proliferative responses between control and survivor sh. This would suggest that if activated IgT B cells are generated in the spleen or head kidney, then they are probably imprinted to home into the gills where they may then proliferate. (c) IgT produced by IgT-secreting B cells is transported from the epithelium into the mucus layer (grey colour) by the tpIgR. The microbiota contained in the mucus is predominantly coated by sIgT.
Ich parasite isolation and infection. For Ich parasite isolation, heavily infected rainbow trout were killed by an overdose of MS-222 and placed in a container with water to allow trophonts and tomonts exit the host. After 4 h, the sh were removed, and the trophonts were left in the water at 15 C for 12 days to let tomocyst formation and subsequent theront release. For parasite infection, two different sets of experiments were performed. On one hand, a group of sh was infected with a single dose of B5,000 theronts per sh added directly into the aquarium, and tissue samples and uids (serum and gill mucus) were taken after 25 days (infected sh). On the other hand, a group of sh was monthly exposed over a 3-month period with B10,000 theronts per sh and tissue samples and uids (serum and gill mucus) were taken 1 month after the last challenge (survival sh). Both sets of experiments were performed at least three times. For each one, the same numbers of control sh were maintained in a similar tank but without parasites.
F. columnare infection. Two types of challenges with F. columnare were performed. In the rst one (infected group), 20 sh (B40 g per sh) were exposed by bath with a dose of 10 ml F. columnare suspension (OD 0.75 at 540 nm) in B4 l
water for 1 h. Fish were sampled (serum and gill mucus) 3 weeks after exposure. For the second type of challenge (survivor group), 20 sh (B40 g per sh) were
exposed by bath with a dose of 10 ml F. columnare suspension (OD 0.75 at
540 nm) in B4 l water for 4 h. These sh were let to recover from infection (survivor sh). To this end, survivor sh were obtained by allowing the infected sh to recover from the disease during a period of 3 months at which point sh were sampled (serum and gill mucus). Both sets of infections were performed at least three times. Control sh (mock infected) were maintained in similar tanks, and were exposed to the same culture medium without the bacteria. Throughout this time, the sh were maintained in a ow through aquaria at 15 C, and fed daily with dry pellets at 1% biomass.
Collection of gill mucus and bacteria. For sampling, sh were anaesthetized with MS-222 and serum was collected and stored as described9. To obtain sh gill mucus, we modied the method described by Lumsden et al.55 To obtain the mucus, blood in the gills was rst removed by perfusion with PBSheparin through the heart until the gills were completely blanched. Gill arches were excised and rinsed with PBS three times to remove the remaining blood. Thereafter, gills were incubated for 12 h at 4 C, with occasional shaking in protease inhibitor buffer (1 PBS, containing 1 protease inhibitor cocktail (Roche), 1 mM
phenylmethylsulfonyl uoride (Sigma); pH 7.2) at a ratio of 1 g of gill tissue per ml of buffer. The suspension (gill mucus) was collected into an Eppendorf tube,
NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 11
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728
vigorously vortexed and centrifuged at 400g for 10 min at 4 C to remove trout cells. To separate gill bacteria from mucus, the cell-free supernatant was thereafter centrifuged at 10,000g for 10 min. The resulting supernatant (containing gill mucus) was harvested, ltered with a 0.45 mm syringe lter (Millipore) and stored at 4 C prior to use the same day, whereas the pellet (containing gill bacteria) was washed three times with PBS (pH 7.2) and resuspended for further analysis.
Purication and LC-MS/MS analysis of gill mucus IgT. Gill mucus IgT was partially puried using an anti-IgT afnity column as previously described for gut and skin mucus IgT8,9. Briey, a gill mucus sample pooled from several trout individuals was applied to the anti-trout IgT column equilibrated in PBSEDTA (1 PBS, containing 20 mM EDTA; pH 7.2). After several washes of the afnity
column with PBSEDTA, the bound IgT were eluted with 0.1 M glycine (pH 2.5) and immediately neutralized with 1 M Tris (pH 9.0). The eluted IgT fractions were thereafter concentrated. Approximately 2 mg of afnity puried trout IgT was resolved by SDSPAGE on a 415% Tris-HCl Gradient Ready Gel (Bio-Rad) under non-reducing conditions and stained with Bio-Safe Coomassie (Bio-Rad). LC-MS/ MS analysis of the IgT band was performed by the Proteomics Facility at the School of Medicine, the University of Pennsylvania. The resulting masses and MS/MS spectra were searched against the non-redundant NCBI database using the TurboSEQUEST Browser.
Isolation of leucocytes from the gill and nasal cavities. The leucocytes from trout gill were obtained using a modied methodology from that used to isolate gill leucocytes in dab (Limanda limanda)56. Briey, rainbow trout were anaesthetized with MS-222 and blood was collected from the caudal vein. Gills were extensively perfused and washed with cold PBS as described above to avoid blood contamination. Thereafter, gills were incubated in PBSDTT (containing0.37 mg ml 1 EDTA and 0.14 mg ml 1 dithiothreitol DTT) for 30 min with continuous shaking. Thereafter, the gill pieces were mechanically disaggregated on a 100-mm cell shredder and the cell fraction was collected. The remaining nondisaggregated gill tissue was then transferred into a 50 ml tube containing 10 ml collagenase buffer (Invitrogen, 0.15 mg ml 1 in Hanks with 5% foetal bovine serum FBS and 100 U ml 1 penicillin and 100 mg ml 1 streptomycin) for 2 h at room temperature with continuous shaking. Thereafter, the remaining gill pieces were mechanically disaggregated on a 100-mm cell shredder and the cell fraction was collected. The cell fractions from the above gill treatments were pooled, washed with Dulbeccos modied eagle medium (DMEM) and passed through a 100-mm nylon mesh. The resulting cell suspension was washed three times in fresh DMEM and layered over a 51/34% discontinuous Percoll gradient. After 35 min of centrifugation at 400g, leucocytes lying at the interface of the gradient were collected and washed with DMEM medium. The isolation of leucocytes from trout NALT was performed as described previously10. Briey, leucocytes from the trout olfactory organ were isolated by mechanical agitation of both olfactory rosettes in DMEM medium (supplemented with 5% FBS, 100 U ml 1 penicillin and100 mg ml 1 streptomycin) at 4 C for 30 min. Leucocytes were subsequently collected, and the aforementioned procedure was repeated four times. Thereafter, the olfactory organ pieces were treated with PBS (containing 0.37 mg ml 1 EDTA and 0.14 mg ml 1 dithiothreitol DTT) for 30 min followed by enzymatic digestion with collagenase (Invitrogen, 0.15 mg ml 1) for 2 h at 20 C. All cell fractions obtained from the olfactory organ after mechanical and enzymatic treatments were pooled and washed with modied DMEM. Cells were thereafter layered over a 51/34% discontinuous Percoll gradient and after 35 min of centrifugation at 400g, leucocytes lying at the interface of the gradient were collected and washed with DMEM medium.
SDSPAGE and western blot. Serum and gill mucus samples were resolved on 415% SDSPAGE Ready Gel (Bio-Rad) under non-reducing and/or reducing conditions. The gels were either stained with Bio-Safe Coomassie or transferred onto Sequi-Blot PVDF membranes (Bio-Rad). For western blot analysis, the membranes were blocked with 8% skim milk and incubated with anti-trout IgT (rabbit pAb), anti-trout IgM (mouse monoclonal antibody (mAb)) or biotinylated anti-trout IgD (mouse mAb) antibodies followed by incubation with peroxidaseconjugated anti-rabbit, anti-mouse IgG (GE Healthcare) or streptavidin (Thermo Scientic Pierce). Immunoreactive bands were visualized using the HyGLO Chemiluminescent HRP Antibody Detection Reagent (Denville Scientic Products). For quantitative analyses of IgM, IgD and IgT in serum and gill mucus, western blot lms were scanned and the signal strength of each band was determined by using ImageQuant TL software (GE Healthcare). Thereafter, the concentration of IgM, IgD and IgT were determined by plotting the obtained signal strength values on a standard curve generated for each blot using known amounts of puried trout IgM, IgD or IgT. Original images of the western blot analyses are shown in Supplementary Fig. 11.
Flow cytometry. For ow cytometry studies of B cells in the gill, spleen and NALT, mouse anti-trout IgT and mouse anti-trout IgM mAbs were labelled with allophycocyanin (APC) and phycoerythrin (PE) using the Mix-n-Stain Antibody labelling kit (Biotium). Trout leucocytes (5 105) were rst stained with biotiny
lated mouse anti-trout IgD (1 mg ml 1) on ice for 40 min. After washing three
times, Brilliant Violet 421 Streptavidin (1 mg ml 1, Biolegend), APC-labelled mouse anti-trout IgM and PE-labelled mouse anti-trout IgT (1 mg ml 1, each)
were added to detect IgD , IgT and IgM B cells, respectively. Samples were thereafter incubated on ice for 40 min and washed three times prior to ow cytometric analysis. Gill bacteria were stained with biotinylated mouse anti-trout IgD (1 mg ml 1) or with biotinylated mouse IgG1 (as isotype control) at 4 C for 1 h with continuous agitation. After washing three times, Brilliant Violet 421
Streptavidin (1 mg ml 1), APC-labelled mouse anti-trout IgM and PE-labelled mouse anti-trout IgT (1 mg ml 1, each), or APC-labelled mouse IgG1 and PE-labelled mouse IgG2b (Biolegend, 1 mg ml 1, each) as isotype controls were added and incubated for 45 min at 4 C. To discriminate bacteria from debris, gill bacteria were labelled with BacLight Green bacterial stain (Invitrogen), following the manufacturers instructions. After three washes, analysis of stained leucocytes or bacteria was performed with a FACSCanto II (BD Biosciences) and FlowJo software (Tree Star).
Gel ltration. Gel ltration to analyse the monomeric or polymeric state of Igs in trout gill mucus was performed as described previously for gut mucus8. Briey, fractions containing the IgM, IgT or IgD were separated by gel ltration using a Pharmacia Superdex 200 column (GE Healthcare). The column was previously equilibrated with cold PBS, and protein fractions were eluted at 0.5 ml min 1 with
PBS using a fast protein LC instrument withKTApurier systems (GE Healthcare). Protein elution was monitored by absorbance at 280 nm. Identication of IgM, IgD and IgT in the eluted fractions was performed by western blot analysis using anti-IgM, anti-IgD and anti-IgT antibodies, respectively, as described above. A standard curve was generated by plotting the elution volume of the standard proteins in a Gel Filtration Standard (Bio-Rad) against their known molecular weight, which was then used to determine the molecular weight of the eluted IgT, IgM and IgD by their elution volume. Original images of the western blot analyses are shown in Supplementary Fig. 11.
Immunouorescence microscopy studies. Cryoblocks of gill were taken using Optimal Cutting Temperature medium (Tissue-Tek). Cryoblocks were thereafter used to generate cryosections for immunohistochemistry assays. Gill cryosections, 6-mm thick, were used to perform immunouorescence studies. For the detection of IgT , IgM and IgD B cells in the gill, cryosections of tissue samples were xed for 3 min in 10% (vol/vol) neutral buffered formalin. Background auto-uorescence was eliminated by treatment of cryosections for 10 min with 0.1 M glycine, pH 2.3. For the detection of Ich parasite at the same time of IgT and
IgM B cells, antibodies were labelled using the Mix-n-Stain Antibody labelling kit. We used CF555-labelled rabbit anti-trout IgT (0.6 mg ml 1), CF640R-labelled mouse anti-trout IgM (2 mg ml 1) and CF488A-labelled rabbit anti-Ich(1 mg ml 1) to incubate the cryosections overnight at 4 C. For the detection of Ich parasite at the same time of IgT and IgD B cells, we used CF555-labelled rabbit anti-trout IgT (0.6 mg ml 1), CF640R-labelled mouse anti-trout IgD (2 mg ml 1)
and CF488A-labelled rabbit anti-Ich (1 mg ml 1) to incubate the cryosections overnight at 4 C. For detection of gill tpIgR, we used the same methodology described to stain skin tpIgR in cryosections using our rabbit anti-tpIgR9. As controls, the rabbit IgG pre-bleed and the mouse-IgG1 isotype antibodies were labelled with the same antibody labelling kits and used at the same concentrations. All slides were stained with DAPI (40,6-diamidino-2-phenylindole; 1 mg ml 1;
Invitrogen) before mounting with uorescent microscopy mounting solution. Gill bacteria were stained as previously described in the above section and stained bacteria were cytospinned on glass slides and mounted with uorescent microscopy mounting solution. Images were acquired and analysed using a Nikon Eclipse E600 uorescence microscope and the iVision-Mac (v4.0.14) scientic imaging processing software (Bio Vision Technologies).
Intracellular staining. Gill leucocytes were isolated as described above. Cells were xed with xation buffer (Biolegend) and washed twice with permeabilization wash buffer (Biolegend) following the manufacturers instructions. Permeabilized cells were stained with monoclonal mouse anti-trout IgT (1 mg ml 1), or mouse IgG2b as isotype control (Biolegend; 1 mg ml 1) in permeabilization wash buffer at 4 C for 30 min. Secondary antibody APC-goat anti-mouse IgG2b (Jackson ImmunoResearch Laboratories) was added after washing and incubated for 30 min at4 C. Analyses of stained leucocytes were performed by ow cytometry using a FACSCanto II and FlowJo software.
Proliferation of B cells in the gill of trout. Survivor and control sh (B50 g) were anaesthetized with MS-222 and i.v. injected with 200 mg EdU (Invitrogen) in 100 ml of PBS. After 24 h, leucocytes from gill, spleen or head kidney were obtained as described above, and 5 106 cells were incubated with mAb mouse anti-trout
IgM and anti-trout IgT (1 mg ml 1 each) on ice for 40 min. After washing three times, APC-goat anti-mouse IgG (Jackson ImmunoResearch Laboratories Inc.) was used as secondary antibody to detect IgM or IgT B cells. After 40 min incubation on ice, cells were washed three times and xed. EdU cell detection was performed according to the manufacturers instructions (Click-iT EdU Alexa Fluor 488 Flow Cytometry Assay Kit, Invitrogen). Cells were thereafter analysed in a FACSCanto II with Diva Software. For immunouorescence analysis, gill
12 NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728 ARTICLE
cryosections were incubated overnight at 4 C with rabbit anti-trout IgT(0.2 mg ml 1) and mouse anti-trout IgM (IgG1 isotype; 1 mg ml 1). After washing with PBS, cryosections were incubated for 2 h at room temperature with Cy3-conjugated AfniPure Donkey anti-rabbit IgG or Cy5-conjugated-AfniPure Goat anti-mouse IgG1 (Jackson ImmunoResearch Laboratories Inc.) at 2.5 mg ml 1 each. As primary control Abs, the rabbit pre-bleed and the mouse-IgG1 isotype controls were labelled with the same antibody labelling kits and used at the same concentrations. Stained cells were xed and EdU cell detection was performed according to the manufacturers instructions (Click-iT EdU Alexa Fluor 488
Imaging Kit, Invitrogen). Cell nuclei were stained with DAPI (1 mg ml 1) before mounting with uorescent microscopy mounting solution. Images were acquired and analysed using a Nikon Eclipse E600 uorescence microscope and the iVision-Mac scientic imaging processing software.
Tissue explants culture. Control and survivor sh were killed with an overdose of MS-222, and blood was removed through the caudal vein to minimize the blood content in the collected organs. Thereafter, head kidney, spleen and gills were collected. Approximately 50 mg of each tissue was submerged in 70% ethanol for 1 min to eliminate possible bacteria on their surface and then washed twice with PBS. Thereafter, tissues were placed in a 24-well plate and cultured with 400 ml
DMEM medium (Gibco), supplemented with 10% FBS, 100 U ml 1 penicillin and 100 mg ml 1 streptomycin, with 5% CO2 at 17 C. After 7 days culture, super-natants were harvested, centrifuged and stored at 80 C until further analysis.
Binding of trout immunoglobulins to Ich. To assess whether infected and survivor sh had generated pathogen-specic immunoglobulins, we measured the capacity of IgT, IgM and IgD from serum, gill mucus or tissue (gill, spleen and head kidney) explant supernatants to bind to Ich using a pull-down assay previously described by us9. In short, parasites (B100 tomonts) were preincubated with a solution of 0.5% BSA in PBS (pH 7.2) for 2 h at 4 C. Thereafter, parasites were incubated with diluted gill mucus or serum or tissue (gill, head kidney and spleen) explant supernatants from infected, survivor or control sh for 2 h at 4 C in a 300 ml volume. Dilutions were made with PBS containing 0.5% BSA (pH 7.2).
After incubation, the tomonts were washed with PBS, and bound proteins were eluted with Laemmli Sample Buffer (Bio-Rad) and boiled for 5 min at 95 C. The eluted material was resolved on 415% SDSPAGE Ready Gel under non-reducing conditions, and the presence of IgT, IgM or IgD was detected by western blotting using the anti-trout IgT, IgM or IgD antibodies as described above. Original images of the western blot analyses are shown in Supplementary Fig. 11.
Co-immunoprecipitation studies. We followed the same methodology previously reported by us to detect the association of tpIgR to IgT and IgM in gut mucus9. To detect whether polymeric trout IgT, IgM and IgD present in the gill mucus were associated to a secretory component-like molecule derived from tSC, we performed co-immunoprecipitating analysis using anti-trout IgT (pAb), anti-trout IgM (mAb) or anti-trout IgD (mAb) antibodies with the goal to potentially co-immunoprecipitate the tSC. To this end, 10 mg of anti-IgT, anti-IgM or anti-IgD were incubated with 200 ml of trout gill mucus. As control for these studies, the same amount of rabbit IgG (puried from the pre-bleed serum of the rabbit) were used as negative controls for anti-IgT. Similarly, a mouse-IgG1 isotype control (BioLegend) was used as negative controls for anti-IgM and anti-IgD. After overnight incubation at 4 C, 20 ml of protein G Agarose (Invitrogen) was added into each reaction mixture and incubated for 1 h at 4 C. Thereafter, the beads were washed ve times with PBS, and subsequently bound proteins were eluted in Laemmli Sample Buffer (Bio-Rad). The eluted material was resolved by SDSPAGE on 415% Tris-HCl Gradient ReadyGels (Bio-Rad) under reducing (for tSC detection) or non-reducing (for IgT, IgM or IgD detection) conditions. Western blot was performed with anti-pIgR, anti-IgT, anti-IgM or anti-IgD antibodies as described above. Original images of the western blot analyses are shown in Supplementary Fig. 11.
Statistical analysis. An unpaired Students t-test (Excel version 11.0; Microsoft) and one-way analysis of variance with Bonferroni correction (Prism version 6.0; GraphPad) were used for analysis of differences between groups. P values of r0.05 were considered statistically signicant.
References
1. Maina, J. N. Structure, function and evolution of the gas exchangers: comparative perspectives. J. Anat. 201, 281304 (2002).
2. Glover, C. N., Bucking, C. & Wood, C. M. The skin of sh as a transport epithelium: a review. J. Comp. Physiol. B 183, 877891 (2013).
3. Aquilino, C., Castro, R., Fischer, U. & Tafalla, C. Transcriptomic responses in rainbow trout gills upon infection with viral hemorrhagic septicemia virus (VHSV). Dev. Comp. Immunol. 44, 1220 (2014).
4. Pennacchi, Y., Leef, M. J., Crosbie, P. B., Nowak, B. F. & Bridle, A. R. Evidence of immune and inammatory processes in the gills of AGD-affected Atlantic salmon, Salmo salar L. Fish Shellsh Immunol. 36, 563570 (2014).
5. Valenzuela-Miranda, D., Boltana, S., Cabrejos, M. E., Yanez, J. M. & Gallardo-Escarate, C. High-throughput transcriptome analysis of ISAV-infected Atlantic salmon Salmo salar unravels divergent immune responses associated to head-kidney, liver and gills tissues. Fish Shellsh Immunol. 45, 367377 (2015).
6. Shi, M. et al. RNA-seq proles from grass carp tissues after reovirus (GCRV) infection based on singular and modular enrichment analyses. Mol. Immunol. 61, 4453 (2014).
7. Salinas, I., Zhang, Y. A. & Sunyer, J. O. Mucosal immunoglobulins and B cells of teleost sh. Dev. Comp. Immunol. 35, 13461365 (2011).
8. Zhang, Y. A. et al. IgT, a primitive immunoglobulin class specialized in mucosal immunity. Nat. Immunol. 11, 827835 (2010).
9. Xu, Z. et al. Teleost skin, an ancient mucosal surface that elicits gut-like immune responses. Proc. Natl Acad. Sci. USA 110, 1309713102 (2013).
10. Tacchi, L. et al. Nasal immunity is an ancient arm of the mucosal immune system of vertebrates. Nat. Commun. 5, 5205 (2014).
11. Edholm, E. S. et al. Identication of two IgD B-cell populations in channel
catsh, Ictalurus punctatus. J. Immunol. 185, 40824094 (2010).12. Ramirez-Gomez, F. et al. Discovery and characterization of secretory IgD in rainbow trout: secretory IgD is produced through a novel splicing mechanism.J. Immunol. 188, 13411349 (2012).13. Danilova, N., Bussmann, J., Jekosch, K. & Steiner, L. A. The immunoglobulin heavy-chain locus in zebrash: identication and expression of a previously unknown isotype, immunoglobulin Z. Nat. Immunol. 6, 295302 (2005).
14. Hansen, J. D., Landis, E. D. & Phillips, R. B. Discovery of a unique Ig heavy-chain isotype (IgT) in rainbow trout: Implications for a distinctive B cell developmental pathway in teleost sh. Proc. Natl Acad. Sci. USA 102, 69196924 (2005).
15. Fillatreau, S. et al. The astonishing diversity of Ig classes and B cell repertoires in teleost sh. Front. Immunol. 4, 28 (2013).
16. Castro, R. et al. CCR7 is mainly expressed in teleost gills, where it denes an IgD IgM- B lymphocyte subset. J. Immunol. 192, 12571266 (2014).
17. Wilson, J. M. & Laurent, P. Fish gill morphology: inside out. J. Exp. Zool. 293, 192213 (2002).
18. Iwasaki, A. Mucosal dendritic cells. Annu. Rev. Immunol. 25, 381418 (2007).19. Salvi, S. & Holgate, S. T. Could the airway epithelium play an important role in mucosal immunoglobulin A production? Clin. Exp. Allergy 29, 15971605 (1999).
20. Chiu, C., Ellebedy, A. H., Wrammert, J. & Ahmed, R. B cell responses to inuenza infection and vaccination. Curr. Top. Microbiol. Immunol. 386, 381398 (2015).
21. Chiu, C. & Openshaw, P. J. Antiviral B cell and T cell immunity in the lungs. Nat. Immunol. 16, 1826 (2015).
22. dos Santos, N. M. et al. The gill is a major organ for antibody secreting cell production following direct immersion of sea bass (Dicentrarchus labrax, L.) in a Photobacterium damselae ssp. piscicida bacterin: an ontogenetic study. Fish Shellsh Immunol. 11, 6574 (2001).
23. Esteve-Gassent, M. D., Nielsen, M. E. & Amaro, C. The kinetics of antibody production in mucus and serum of European eel (Anguilla anguilla L.) after vaccination against Vibrio vulnicus: development of a new method for antibody quantication in skin mucus. Fish Shellsh Immunol. 15, 5161 (2003).
24. Makesh, M., Sudheesh, P. S. & Cain, K. D. Systemic and mucosal immune response of rainbow trout to immunization with an attenuated Flavobacterium psychrophilum vaccine strain by different routes. Fish Shellsh Immunol. 44, 156163 (2015).
25. Valdenegro-Vega, V. A. et al. Effects of single and repeated infections with Neoparamoeba perurans on antibody levels and immune gene expression in Atlantic salmon (Salmo salar). Fish Shellsh Immunol. 42, 522529 (2015).
26. Sunyer, J. O. Fishing for mammalian paradigms in the teleost immune system. Nat. Immunol. 14, 320326 (2013).
27. Berthelot, C. et al. The rainbow trout genome provides novel insights into evolution after whole-genome duplication in vertebrates. Nat. Commun. 5, 3657 (2014).
28. Pabst, O. New concepts in the generation and functions of IgA. Nat. Rev. Immunol. 12, 821832 (2012).
29. Gutzeit, C., Magri, G. & Cerutti, A. Intestinal IgA production and its role in host-microbe interaction. Immunol. Rev. 260, 7685 (2014).
30. Dickerson, H. W. & Findly, R. C. Immunity to Ichthyophthirius infections in sh: a synopsis. Dev. Comp. Immunol. 43, 290299 (2014).
31. Salic, A. & Mitchison, T. J. A chemical method for fast and sensitive detection of DNA synthesis in vivo. Proc. Natl Acad. Sci. USA 105, 24152420 (2008).
32. Mostov, K. E. Transepithelial transport of immunoglobulins. Annu. Rev. Immunol. 12, 6384 (1994).
33. Powell, K. R., Shorr, R., Cherry, J. D. & Hendley, J. O. Improved method for collection of nasal mucus. J. Infect. Dis. 136, 109111 (1977).
34. Okada, T., Konishi, H., Ito, M., Nagura, H. & Asai, J. Identication of secretory immunoglobulin A in human sweat and sweat glands. J. Invest. Dermatol. 90, 648651 (1988).
NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications 13
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10728
35. Aufricht, C. et al. Salivary IgA concentration is inuenced by the saliva collection method. Eur. J. Clin. Chem. Clin. Biochem. 30, 8183 (1992).
36. Mostov, K. E. & Blobel, G. A transmembrane precursor of secretory component. The receptor for transcellular transport of polymeric immunoglobulins. J. Biol. Chem. 257, 1181611821 (1982).
37. Baker, K., Lencer, W. I. & Blumberg, R. S. Beyond IgA: the mucosal immunoglobulin alphabet. Mucosal Immunol. 3, 324325 (2010).
38. Castro, C. D. & Flajnik, M. F. Putting J chain back on the map: how might its expression dene plasma cell development? J. Immunol. 193, 32483255 (2014).
39. Kaetzel, C. S. Coevolution of mucosal immunoglobulins and the polymeric immunoglobulin receptor: evidence that the commensal microbiota provided the driving force. ISRN Immunol. 2014, 541537 (2014).
40. Phalipon, A. et al. Secretory component: a new role in secretory IgA-mediated immune exclusion in vivo. Immunity 17, 107115 (2002).
41. Cerutti, A., Chen, K. & Chorny, A. Immunoglobulin responses at the mucosal interface. Annu. Rev. Immunol. 29, 273293 (2011).
42. Mantis, N. J., Rol, N. & Corthesy, B. Secretory IgAs complex roles in immunity and mucosal homeostasis in the gut. Mucosal Immunol. 4, 603611 (2011).
43. Marsland, B. J. & Gollwitzer, E. S. Host-microorganism interactions in lung diseases. Nat. Rev. Immunol. 14, 827835 (2014).
44. Beck, J. M., Young, V. B. & Huffnagle, G. B. The microbiome of the lung. Transl. Res. 160, 258266 (2012).
45. Gollwitzer, E. S. et al. Lung microbiota promotes tolerance to allergens in neonates via PD-L1. Nat. Med. 20, 642647 (2014).
46. Segal, L. N., Rom, W. N. & Weiden, M. D. Lung microbiome for clinicians. New discoveries about bugs in healthy and diseased lungs. Ann. Am. Thorac. Soc. 11, 108116 (2014).
47. Findly, R. C., Zhao, X., Noe, J., Camus, A. C. & Dickerson, H. W. B cell memory following infection and challenge of channel catsh with Ichthyophthirius multiliis. Dev. Comp. Immunol. 39, 302311 (2013).
48. Olsen, M. M. et al. Cellular and humoral factors involved in the response of rainbow trout gills to Ichthyophthirius multiliis infections: molecular and immunohistochemical studies. Fish Shellsh Immunol. 30, 859869 (2011).
49. von Gersdorff Jorgensen, L., Heinecke, R. D., Skjodt, K., Rasmussen, K. J. & Buchmann, K. Experimental evidence for direct in situ binding of IgM and IgT to early trophonts of Ichthyophthirius multiliis (Fouquet) in the gills of rainbow trout, Oncorhynchus mykiss (Walbaum). J. Fish Dis. 34, 749755 (2011).
50. Round, J. L. & Mazmanian, S. K. The gut microbiota shapes intestinal immune responses during health and disease. Nat. Rev. Immunol. 9, 313323 (2009).
51. Lowrey, L., Woodhams, D. C., Tacchi, L. & Salinas, I. Topographical mapping of the rainbow trout (Oncorhynchus mykiss) microbiome reveals a diverse bacterial community with antifungal properties in the skin. Appl. Environ. Microbiol. 81, 69156925 (2015).
52. Spiekermann, G. M. et al. Receptor-mediated immunoglobulin G transport across mucosal barriers in adult life: functional expression of FcRn in the mammalian lung. J. Exp. Med. 196, 303310 (2002).
53. Austbo, L. et al. Transcriptional response of immune genes in gills and the interbranchial lymphoid tissue of Atlantic salmon challenged with infectious salmon anaemia virus. Dev. Comp. Immunol. 45, 107114 (2014).
54. Salinas, I. The mucosal immune system of teleost sh. Biology 4, 525539 (2015).
55. Lumsden, J. S., Ferguson, H. W., Ostland, V. E. & Byrne, P. J. The mucous coat on gill lamellae of rainbow-trout (oncorhynchus-mykiss). Cell Tissue Res. 275, 187193 (1994).
56. Davidson, G. A., Lin, S. H., Secombes, C. J. & Ellis, A. E. Detection of specic and constitutive antibody secreting cells in the gills, head kidney and peripheral blood leucocytes of dab (Limanda limanda). Vet. Immunol. Immunopathol. 58, 363374 (1997).
Acknowledgements
We would like to thank the Morphology Core of the Center for Molecular Studies in Digestive and Liver Diseases (the University of Pennsylvania), especially Adam Bedenbaugh, for the help and advice provided for the immunohistochemistry work. We would also like to thank Dr Gordon Ruthel (Penn Vet Imaging Facility, the University of Pennsylvania). The anti-rainbow trout IgD mAb was provided by the U.S. Veterinary Immune Reagent Network, which was funded by the USDA NIFA #2010-65121-20649 award. This work was supported by the US Department of Agriculture Grant USDANRI-2013-01107 to J.O.S., the National Science Foundation Grant NSF-IOS-1457282 to J.O.S., the National Institutes of Health Grant 2R01GM085207-05 to J.O.S., a JSPS Postdoctoral Fellowships for Research Abroad to F.T., and by the European Commission under the Work Programme 2012 of the 7th Framework Programme for Research and Technological Development of the European Union (Grant Agreement 311993 TARGETFISH).
Author contributions
Z.X. performed most of the experimental work including those from Figs 18. F.T. contributed to data from Fig. 1, Fig. 9 and Supplementary Fig. 2. D.P. contributed to data shown in Fig. 2. D.G. contributed to data from Fig. 3. L.v.G.J. provided the anti-Ich antibody and contributed to setting up the Ich infection model. S.E.L. contributed to addressing the role of gill IgT to bacterial trout pathogens. Z.X., F.T., D.P. and D.G. contributed to data analysis. J.O.S. designed the overall study, contributed to data analysis and wrote the main body of the paper.
Additional information
Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications
Web End =http://www.nature.com/ http://www.nature.com/naturecommunications
Web End =naturecommunications
Competing nancial interests: The authors declare no competing nancial interests.
Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/
Web End =http://npg.nature.com/ http://npg.nature.com/reprintsandpermissions/
Web End =reprintsandpermissions/
How to cite this article: Xu, Z. et al. Mucosal immunoglobulins at respiratory surfaces mark an ancient association that predates the emergence of tetrapods. Nat. Commun. 7:10728 doi: 10.1038/ncomms10728 (2016).
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
Web End =http://creativecommons.org/licenses/by/4.0/
14 NATURE COMMUNICATIONS | 7:10728 | DOI: 10.1038/ncomms10728 | http://www.nature.com/naturecommunications
Web End =www.nature.com/naturecommunications
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Copyright Nature Publishing Group Feb 2016
Abstract
Gas-exchange structures are critical for acquiring oxygen, but they also represent portals for pathogen entry. Local mucosal immunoglobulin responses against pathogens in specialized respiratory organs have only been described in tetrapods. Since fish gills are considered a mucosal surface, we hypothesized that a dedicated mucosal immunoglobulin response would be generated within its mucosa on microbial exposure. Supporting this hypothesis, here we demonstrate that following pathogen exposure, IgT+ B cells proliferate and generate pathogen-specific IgT within the gills of fish, thus providing the first example of locally induced immunoglobulin in the mucosa of a cold-blooded species. Moreover, we demonstrate that gill microbiota is predominantly coated with IgT, thus providing previously unappreciated evidence that the microbiota present at a respiratory surface of a vertebrate is recognized by a mucosal immunoglobulin. Our findings indicate that respiratory surfaces and mucosal immunoglobulins are part of an ancient association that predates the emergence of tetrapods.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer