- Email: [email protected]
What could be the mechanisms of immunological memory in fish?
Accepted Manuscript What could be the mechanisms of immunological memory in fish? Takuya Yamaguchi, Edwige Quillet, Pierre Boudinot, Uwe Fischer PII:
S1050-4648(18)30035-4
DOI:
10.1016/j.fsi.2018.01.035
Reference:
YFSIM 5089
To appear in:
Fish and Shellfish Immunology
Received Date: 26 October 2017 Revised Date:
18 January 2018
Accepted Date: 21 January 2018
Please cite this article as: Yamaguchi T, Quillet E, Boudinot P, Fischer U, What could be the mechanisms of immunological memory in fish?, Fish and Shellfish Immunology (2018), doi: 10.1016/ j.fsi.2018.01.035. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
2 x bath vaccine 10 + 2 weeks
AquavacERM
RI PT
non-vaccinated
M AN U
SC
adoptive transfer of PBL
mock inf.
Y.r. low
-
-
EP
TE D
bath challenge
Y.r. high
Y.r. low
mock inf.
+
-
-
-
CD8+, IgM+ and IgT+ splenocytes
Low
High
Antigen-specific IgM secretion
Low
High
Mortality
AC C
Y.r. high
Graphical abstract
ACCEPTED MANUSCRIPT
1 2 3 4
What could be the mechanisms of immunological memory in fish?
RI PT
5 6 7
Takuya Yamaguchi1, Edwige Quillet3, Pierre Boudinot2 and Uwe Fischer1*
SC
8 9
M AN U
10 11 12 13
1
14
10, Greifswald-Insel Riems 17493, Germany; E-Mails: [email protected] (T.Y.);
15
[email protected] (U.F.)
16
2
17
France
18
3
19
78350 Jouy-en-Josas, France
Laboratory of Fish Immunology, Institute of Infectology, Friedrich-Loeffler-Institut, Südufer
TE D
Virologie et Immunologie Moléculaires, INRA, Université Paris-Saclay, 78350 Jouy-en-Josas,
Génétique animale et biologie intégrative, INRA, AgroParisTech, Université Paris-Saclay,
EP
20 21 22
24
* Corresponding author
AC C
23
25
Key words
26
Rainbow trout; Vaccine; Yersinia ruckeri; Adaptive immunity; Adoptive transfer; Targetfish
ACCEPTED MANUSCRIPT
27
Abstract (256 words) Vaccination is the most effective strategy to control infectious diseases in species with
29
adaptive immunity. In human and in mouse, vaccination typically induces specific memory cells,
30
which can mediate a fast anamnestic response upon infection by the targeted pathogen. In these
31
species, successful vaccination induces a long-lasting protection, long after the titres of specific
32
antibodies and the frequency of specific T cells have returned to steady state. Vaccination is also
33
an important challenge in aquaculture, since alternative treatments are either too costly, or, in the
34
case of antibiotics, are harmful for the environment or may result in dangerous resistances.
35
However, the mechanisms of the long-term protection elicited by vaccines in fish remain poorly
36
understood. Although fish possess typical B- and T-cells expressing diverse repertoires of
37
immunoglobulins and T-cell receptors, many features of antigen specific responses are different
38
from what is known in mouse and in human. Memory is one of the most elusive properties of
39
fish adaptive immunity, and its basis is widely unknown. In this opinion article, we discuss the
40
concept of immune memory in the context of the fish immunity. We illustrate the complexity of
41
this question by discussing the results of experiments showing that protection can be passed
42
through adoptive transfer of leukocytes from vaccinated donor fish to naive histocompatible
43
recipients. Combined with tools developed in Targetfish and in previous projects, such as
44
monoclonal antibodies against B- and T-cell markers, we propose that such models of protection
45
transfer provide excellent systems to dissect the mechanisms of B- and T-cell memory in the
46
future.
SC
M AN U
TE D
EP AC C
47
RI PT
28
ACCEPTED MANUSCRIPT
48
1. How to define immune memory?
49
Immune memory is a fundamental characteristic of the adaptive immune system in
51
vertebrates. However, the definition of memory is still changing, under the influence of new
52
mechanisms of long-term protection. A recent article [1] put in perspective different points of
53
view about the nature of immunological memory. According to Donna Farber, memory is
54
defined by three conditions: (1) memory immune cells are long-lived and maintained
55
independently of stimulation by the antigen, (2) memory immune cells are epitope specific and
56
(3) memory immune cells are intrinsically changed by the previous encounter with the antigen.
57
Andreas Radbruch and Klaus Rajewsky consider that immunological memory is defined by "a
58
stimulus-specific change of immune reactivity that persists in the absence of the stimulus". Rolf
59
Zinkernagel reminds the classical definition of a memory response as a faster and stronger
60
response against a previously encountered antigen, compared with the primary response, and
61
emphasizes that the key question is whether or not protective immunity is maintained in an
62
antigen-dependent or -independent manner. Finally, Mihail Netea proposes a wider definition of
63
memory as the capacity of the immune system to keep and recall information on encountered
64
antigens or pathogens, which also includes the concept of innate immune memory or trained
65
immunity [2]. While adaptive immune memory involves permanent genetic changes such as
66
mutations and recombinations, innate immune memory is characterized by epigenetic
67
reprogramming involving sustained changes in gene expression and cell physiology. These
68
multiple definitions and concepts underscore the complexity of the question, while these authors
69
essentially refer to the immunity of well-known species of mammals. Overall, however, it is
70
accepted in mammals that vaccination against an infectious pathogen is based on the
71
establishment of populations of memory cells after an intentional immunization with a
72
homologous attenuated/inactivated strain of the corresponding pathogen or with parts of it.
73
Beyond individual protection, mammalian immunological memory is certainly also important for
74
the survival of offsprings in utero and during the neonatal life. This is mainly due to antibodies,
75
since protecting antibodies can be transferred from the mother to offspring during the foetal life
76
in many species, and also via colostrum (reviewed in [3,4]).
AC C
EP
TE D
M AN U
SC
RI PT
50
77
Importantly, the modalities of secondary responses, which allow assessment of
78
immunological memory, vary considerably across vertebrates. Differences between primary and
ACCEPTED MANUSCRIPT
secondary responses have been observed from fish to mammals. Typical secondary responses,
80
similar to those described in mouse and human, are fast and lead to higher titres of specific
81
antibodies when compared to primary responses; they are found in mammals and birds, where Ig
82
gene hypermutation occurs in germinal centres [5-7]. Early works in frogs proved the existence
83
of secondary responses in "lower" vertebrates, but showed they were not accompanied with a
84
substantial increase in antibody affinity, though Ig genes get hypermutated even in tadpoles [8].
85
In fish, the modalities of secondary responses and long-term protection are even less similar to
86
the typical situations described in mammals.
88
SC
87
RI PT
79
2. Secondary response and long-term protection in teleost fish The existence of a true secondary response with typical characteristics in fish was initially
90
debated [9]: the primary B cell response is generally much slower and develops over several
91
weeks or even months especially in cold water. For example, the antibody response against the
92
hapten dinitrophenyl keyhole limpet haemocyanin (DNP-KLH) in rainbow trout peaked 50 days
93
post immunization, and antibody titres remained high for a long time, at least until 150 days post
94
immunization [10]. Thus, the maintenance of protection might be partly due to a long-term
95
primary response and long-lasting increase of specific antibody titres. However, significant
96
differences between primary and secondary responses were demonstrated in several fish species.
97
A faster secondary response along with elevated antibody titres (about doubled titres) were
98
reported after immunization with bovine serum albumin [11] and with sheep red blood cells
99
(SRBC) [12] in carp, as well as against the pathogenic bacteria Y. ruckeri [13] and Aeromonas
100
salmonicida [14] in rainbow trout. The dose dependency of the secondary antibody response was
101
studied in detail in carp, further supporting the presence of a cell based immune memory: fish
102
were immunized with 105, 107 and 109 SRBC by intramuscular injection and challenged 1, 3, 6
103
or 10 months later with the same dose; with 109 SRBC, a secondary response was already
104
observed at 3 months and was maintained until later time points, and even with the lowest dose,
105
titres reached comparable or higher levels. However, the high-dose, 109 SRBC, did not lead to a
106
clear secondary response with enhanced titres, while a high primary response was induced [12].
107
These data showed that the secondary response was at least quantitatively higher than the
108
primary response, supporting the existence of a certain level of immunological memory.
AC C
EP
TE D
M AN U
89
ACCEPTED MANUSCRIPT
The existence of efficient “memory” in fish is also advocated by the (relatively) long-term
110
protection afforded by vaccination. For example, in rainbow trout, bath vaccination with a
111
bacterin of Y. ruckeri, the causative agent of enteric red mouth disease (ERM), was shown to
112
provide a significant protection of vaccinated fish 4 and 8 weeks post-vaccination [15]. DNA
113
vaccination of rainbow trout against infectious hematopoietic necrosis (IHN) has been shown to
114
confer protection over a period of two years, which would cover the life cycle of a commercial
115
fish [16]. It remains unclear whether a continuous stimulation of the immune system through
116
long-term expression of the viral G protein expression from the plasmid is required for this
117
protection. Importantly, reporter protein expression from injected plasmids can still be detected
118
over long periods [17, 18]. Another example is DNA vaccination against viral haemorrhagic
119
septicaemia (VHS), which was also shown to induce long-lasting protection post vaccination
120
[19]. While protection established shortly after vaccination was cross-protective in terms of a
121
heterologous viral pathogen this cross-protection was absent two months after vaccination, while
122
the specific response to the homologous virus persisted. In most cases, this protection seems to
123
be based on anamnestic secondary responses, which provide a better defence against the
124
pathogens compared to a naive immune system. Analysis of gene expression analysis showed
125
that such a protection was somehow associated with up- or down-regulation events of the
126
expression of immune related molecules such as cytokines and transcriptomic factors [15, 20,
127
21]. However, the detailed cellular basis of protection including T- and B-cell dynamics after
128
fish vaccination is often poorly understood and, most importantly, the persistence of the antigen
129
between the initial immunization and the challenge is most often not directly evaluated.
EP
TE D
M AN U
SC
RI PT
109
Other key features associated with immune B cell memory formation in mammals are class
131
switch and affinity maturation. While an Ig class switch seems to be absent in teleosts, at least in
132
all species studied in sufficient detail, it is generally believed that most lower vertebrates can
133
hypermutate Ig genes but lack the microenvironment required for the selection of B cells with
134
hypermutated Ig receptors of high affinity for the antigen [22, 23]. The enzymatic machinery
135
needed for somatic hypermutation of Ig genes is present in teleosts [24-26], and indeed
136
hypermutation was observed in zebrafish mature Ig repertoires [27]. Interestingly, long-term
137
studies of the antibody responses in rainbow trout showed that high affinity antibodies appear
138
late and become dominant, more than 25 weeks after (primary) immunization [28, 29]. The
139
kinetics of antibody affinity in trout immunized with TNP-KLH revealed that a relatively low
AC C
130
ACCEPTED MANUSCRIPT
affinity antibody subpopulation appears early, does not achieve high titres and is transient; an
141
intermediate affinity subpopulation appears later (five weeks after immunization), achieves
142
relatively high titres and persists longer than the low affinity antibodies; and the highest affinity
143
subpopulation emerges only after 15 weeks at still high titres [30, 31]. These studies suggest, that
144
a slow affinity maturation may actually occur in fish through alternative mechanisms.
RI PT
140
The vast diversity of fish (and their immune systems) should also be considered in
146
investigations about putative immune memory in this group. Thus, species related to cod
147
(Gadidae) and pipefish lack MHC class II and CD4 [32, 33], which implies that B cell responses
148
cannot rely on a T cell help as in mammals. In these species, B cell memory, if it exists, must be
149
based on particular pathways and this assumption should be considered with regard to the
150
successful vaccination protocols in these fish species.
M AN U
SC
145
In this general context, the definition and the putative location of fish memory cells also
152
remain quite elusive. B cells from rainbow trout head kidney still produced specific antibodies
153
later than 20 weeks post immunization with TNP-KLH [34]. In contrast, spleen or blood
154
leukocytes contained very few or no TNP-specific antibody secreting cells 10 - 15 weeks after
155
immunization. Thus, rainbow trout anterior kidney antibody secreting cells may resemble
156
mammalian long-lived plasma cells, and they are likely responsible for sustained serum antibody
157
titres [34]. In the channel catfish, it has been proposed that hypermutation of Ig genes may occur
158
in melanomacrophage centres of the spleen, since activation-induced cytidine deaminase (an
159
enzyme causing somatic hypermutation, gene conversion, and class-switch recombination of
160
immunoglobulin genes in mammalian B cells) is expressed in these structures [35], but the
161
cellular context remains to be further dissected in fish (discussed in [26]). Overall, fish memory
162
cells are very poorly defined and their existence remains debatable.
EP
AC C
163
TE D
151
164
3. A system to further define the basis of immune memory: transfer of
165
protection through cell transplant in rainbow trout.
166
If fish possess memory cells, their phenotypes, frequency, lifespan and functions remain
167
essentially unknown. The lack of antibodies against key surface markers has certainly hampered
168
progress in this field, making difficult the definition and isolation of relevant cell subsets.
169
Furthermore, the lack of syngeneic or congenic animals has hindered cell transfer experiments,
170
which would help defining a “memory” cell compartment. Taking advantage of the availability
ACCEPTED MANUSCRIPT
171
of clonal, isogenic rainbow trout, we could recently transfer protection by transplantation of
172
peripheral blood leukocytes (PBL) from vaccinated fish to histocompatible individuals. The
173
potential interest of this experimental set up for the study of immunological memory will be
174
discussed below. Isogenic fish were vaccinated by immersion using a commercial vaccine against ERM, and
176
boosted ten weeks later, also by immersion (material and methods in supplements). Control fish
177
of the same isogenic progeny were mock-vaccinated and -boosted. Two weeks after the boost,
178
PBL were transferred to histocompatible, naive recipients from the same fish clone, i.e. with the
179
same genetic background. Recipients were challenged by bath with different dosages (mock, low
180
and high) of Y. ruckeri twenty-four hours later and mortalities were recorded (Table 1). The high
181
dose challenge led to high mortality (63%) of recipient fish that had received PBL from non-
182
vaccinated donors (group3). While mortalities were observed between day 5 and 12 after
183
challenge, most fish died until day 9. In contrast, fish which had received PBL from vaccinated
184
donors (group 6), were fully protected, indicating that protection against Y. ruckeri was conferred
185
to the recipients by adoptive transfer of PBL from vaccinated donors.
M AN U
SC
RI PT
175
Following PBL transfer, dynamics of CD8α+, IgM+ and IgT+ lymphocytes in the blood, gills
187
and spleens of recipients that had survived the infection were analysed. We are aware that bath
188
challenge did not ensure that all fish had received the same amount of bacteria, and that the
189
survival might have been affected by this factor; however, these observations allow a first
190
comparison between immune cell populations in protected fish and in the other groups on 22
191
(Fig. 1a) and 29 days (Fig. 1b; Fig 1c) after transfer, which are corresponding to 21 and 28 days
192
post challenge, respectively.
EP
TE D
186
Twenty two days post transfer, the total number of CD8α+ T cells in non-challenged
194
recipients which had received PBL from non-vaccinated fish (group 1), was about 2 × 104 in
195
PBL and approx. ten times more in spleen and gills (Fig. 1a). In comparison, non-challenged
196
recipients, which were injected with PBL from vaccinated donors (group 4), had much more
197
CD8α+ T cells in the gills (about 2.5 times more than those of group 1), while this trend was not
198
observed in spleen and PBL. This suggests that T cells might have accumulated (or proliferated)
199
in gills three weeks after the transfer. In challenged recipients, twenty-one days after bath
200
infection, the number of CD8α+ T cells was higher in PBL of recipients, which had received
201
cells from non-vaccinated donors (group 2 and 3), but not in those transferred with cells from
AC C
193
ACCEPTED MANUSCRIPT
vaccinated donors (group 5 and 6). This trend was also observed in the spleen of recipients
203
challenged at low dose (compare group 2 to group 5), but not in other cases. This suggests that
204
although recipient groups challenged with low dose did not show mortality, different responses
205
and recruitments of T cells occurred between fish transferred with cells from vaccinated and
206
control fish. These patterns were not observed 28 days post challenge, the most obvious pattern
207
being then an increase (approx. 3.5 times) of CD8α+ T cells in the spleen of high dose
208
challenged recipient fish, which had received PBL from vaccinated donors (and were protected,
209
group 6). In the same recipient group, there was an increase (approx. 2.5 times) of CD8α+
210
splenocytes between day 21 and 28 (Fig. 1c) post challenge, while such an increase was not
211
observed in recipient fish which had received PBL from control donors (group 3, in which high
212
mortalities were recorded). To what extent this protection relates to a CD8α+ T cell response
213
remains to be analysed. Of note, no differences in CD8α+ T cell numbers were observed 28 days
214
post challenge, between non-challenged fish, which had received PBL from vaccinated (group 4)
215
and control donors (group 1). Additional studies will be necessary to establish firmly that this
216
kinetics is reproducible and to further elucidate the role of CD8α+ T cells in the transferred
217
protection to ERM.
M AN U
SC
RI PT
202
Similar to CD8α+T cells, twenty two days post transfer, IgM+ and IgT+ B cell numbers were
219
about three times increased in gills of non-challenged recipients, which had received PBL from
220
vaccinated donors (group 4) compared to those which had received PBL from non-vaccinated
221
control donors (group 1). Again, this pattern was not found twenty nine days post transfer. The
222
only remarkable trend regarding B cells was observed at the second time point of sampling. 28
223
days after high dose infection an accumulation (or proliferation) of both IgM+ and IgT+ B cells in
224
the spleen of challenged fish was observed, especially in recipients which had received PBL
225
from vaccinated donors (group 6). Combining the most obvious increase in B cell numbers in
226
high dose challenged recipients that have received immune cells from vaccinated donors with the
227
fact they are fully protected (Table 1) points on a role of B cells in the immune response against
228
ERM.
AC C
EP
TE D
218
229
An interesting observation was made in the spleen, where the numbers of CD8α+, IgM+ and
230
IgT+ cells of high dose challenged recipient fish, which had received PBL from vaccinated
231
donors (group 6) at 28 d p.i. were higher than in those at 21 d p.i. Such an increase was not
232
observed in low dose challenged (group 1 and 4) or mock challenged (group 2 and 5) recipients
ACCEPTED MANUSCRIPT
233
(Fig. 1C). This suggests that B- and T-cell responses might continue developing at this stage in
234
the spleen of group 6 recipients at 28 d p.i. Importantly, such trends should be taken with great caution, since they come from a pool of 3
236
- 6 survivors from a single campaign of experiments. Although a single individual with a
237
remarkable phenotype may induce a bias, we believe the strongest trends are interesting.
238
Although the modest numbers of fish used in this first set of experiment do not fully demonstrate
239
trends, they provide a first frame of analysis to be confirmed by further studies. The adoptive
240
transfer model appeared to be excellent to pursue the characterization of mechanisms involved in
241
the transfer of “memory” and/or protection.
SC
RI PT
235
In order to check the titres of secreted IgM antibodies against Y. ruckeri, sera as well as muci
243
from gills and skin were collected from surviving recipients 28 days after challenge and were
244
analysed by ELISA. Among all examined groups, the titres of anti-Y. ruckeri IgM in muci and
245
sera were clearly elevated only in recipients of group 6 that had received PBL from vaccinated
246
donors and that thereafter had been challenged with high dose of Y. ruckeri (Fig. 2) further
247
suggesting that a secondary response is dose dependent. Nevertheless, these results indicate that
248
transfer of PBL from vaccinated donors successfully transferred the capacity to produce high
249
titres of anti-Y. ruckeri antibodies upon infection in the recipients, hence to mount a "memory"
250
response. This was correlated with high levels of protection. In contrast, recipient fish that had
251
received PBL from non-immunized donors (group 3) showed high mortality after high dose
252
challenge, and survivors did not develop anti-Y. ruckeri IgM antibodies. Interestingly, not all
253
recipients of group 6 were seroconverted although full protection and high levels of IgM+ and
254
IgT+ cells were recorded. Group 6 can be divided into non- and high IgM responders (which is
255
partially also true for vaccinated donors). This implies that several mechanisms may explain
256
survival and that, although genetically identical fish were used in these experiments, immune
257
responses to a given pathogen may show considerable interindividual variability in fish. These
258
observations underline the importance of following these parameters when tracking immune
259
memory.
AC C
EP
TE D
M AN U
242
260 261
4. Perspectives and discussion
262
The available literature illustrates and the case study described above exemplifies that much
263
remains to be understood and defined regarding the protection against pathogens - including the
ACCEPTED MANUSCRIPT
protection induced by vaccination - and a putative immunological memory in fish. An important
265
issue would be to identify the cellular compartment responsible for the long-term protection.
266
Whether the mechanisms triggered by such cells may be named "memory" would be the next
267
question. Several mechanisms can be proposed on how protection has been transferred from
268
vaccinated donor fish to recipient fish. One possibility is that effector cells were transferred in an
269
activated stage induced by vaccination plus boost, and that thus have directly contributed to
270
rescue recipients from infection. Such activated effector cells could have acted specifically for
271
example via specific antibody production, neutralising the pathogen. It is also reasonable to
272
consider that transferred donor effector cells mounted a more efficient immune response in
273
concert with the hitherto naïve immune system of the recipient. Thus, innate immunity of the
274
recipients complemented by the action of transferred cells could be efficient enough to promote
275
survival. Last but not least, bona fide memory cells could have been transferred to recipients
276
allowing a fast and effective response to the infection and leading to high survival. These
277
different hypotheses could be tested by performing transfer of particular sorted cell subsets, and
278
detailed characterization of the response to different challenges (homologous versus
279
heterologous pathogens, multiples routes). Using a similar experimental set-up in isogenic
280
ginbuna crucian carp [36] where Yamazaki et al. have shown that adoptive transfer of CD4+ and
281
CD8α+ cells from vaccinated donors protects recipients against a homologous Edwardsiella
282
tarda challenge. The carriers of protection were CD8α+ cells and, to a lesser extent, CD4+ cells.
283
However in this study, ginbuna lymphocytes were isolated from donor fish only 8 days after
284
immunization, which is probably enough to mount a specific immune responses in this species,
285
but clearly too short to study long-term, persistent "memory" cells.
EP
TE D
M AN U
SC
RI PT
264
To get more insights in the mechanisms of long-term protection and memory, transfer
287
experiments of sorted cell subsets should be performed in models similar to the one described
288
above in trout. When the size of the fish and the time line allows, it would be very powerful tool
289
to follow immune parameters such as antigen specific antibody titres in individual fish, using
290
non-lethal sampling, before and after challenge.
AC C
286
291
In conclusion, models such as the cell transfer between histocompatible fish presented in this
292
article will be instrumental to understand the mechanisms and impact of anamnestic memory
293
responses in fish. These future analyses will benefit from antibodies specific for lymphocytes
294
markers. A better characterization of markers such as CD27, CD38, CD45RO, CD45RA and
ACCEPTED MANUSCRIPT
CD62L (and production of the corresponding reagents) in fish would be also extremely useful.
296
Besides, the development of new high-throughput transcriptomic approaches will likely be
297
instrumental for the definition of memory cell subsets in fish. Modifications of lymphocyte
298
populations in lymphoid organs after PBL transfer and or challenge suggests that transferred
299
cells might be prone to recruitment into defined territories, or to re-localisation of immune cells.
RI PT
295
AC C
EP
TE D
M AN U
SC
300
ACCEPTED MANUSCRIPT
301
Acknowledgements We are grateful to Prof. Oriol Sunyer, University of Pennsylvania, USA, for providing us the
303
monoclonal antibody against IgT, to Prof. Inger Dalsgaard, Danish Technical University for
304
providing us with the Yersinia ruckeri O1 strain and to Prof. Kurt Buchmann, University of
305
Copenhagen, Denmark for technical advice regarding ELISA. We also thank Mrs. Susann
306
Schares for excellent technical assistance. This work was supported by the EU FP7 grant 311993
307
(TargetFish) for Pierre Boudinot, Takuya Yamaguchi and Uwe Fischer.
RI PT
302
AC C
EP
TE D
M AN U
SC
308
ACCEPTED MANUSCRIPT
References
310
[1] D.L. Farber, M.G. Netea, A. Radbruch, K. Rajewsky, R.M. Zinkernagel, Immunological
311
memory: lessons from the past and a look to the future, Nat. Rev. Immunol. 16 (2016) 124-8,
312
https://doi.org/10.1038/nri.2016.13.
RI PT
309
313
[2] M.G. Netea, L.A. Joosten, E. Latz, K.H. Mills, G. Natoli, H.G. Stunnenberg, L.A. O’Neill,
314
R.J. Xavier, Trained immunity: a program of innate immune memory in health and disease,
315
Science 352 (2016) aaf1098, 1-9, https://doi.org/10.1126/science.aaf1098
[3] A. Marchant, M. Sadarangani, M. Garand, N. Dauby, V. Verhasselt, L. Pereira, G. Bjornson,
317
C.E. Jones, S.A. Halperin, K.M. Edwards et al., Maternal immunisation: collaborating with
318
mother nature. Lancet Infect. Dis. 17 (2017) e197-e208. https://doi.org/10.1016/S1473-
319
3099(17)30229-3.
M AN U
SC
316
320
[4] M.F. Jennewein, B. Abu-Raya, Y. Jiang, G. Alter, A. Marchant, Transfer of maternal
321
immunity and programming of the newborn immune system, Semin. Immunopathol. 39
322
(2017) 605-13, https://doi.org/10.1007/s00281-017-0653-x
324 325 326
[5] L. Du Pasquier, Meeting the demand for innate and adaptive immunities during evolution, Scand. J. Immunol. 62 (2005) 39-48, https://doi.org/10.1111/j.1365-3083.2005.01608.x.
TE D
323
[6] M. Flajnik, L. du Pasquier, Evolution of the immune system in Fundamental immunology 7th Edn. (Paul, W. E., ed.) Lippincott-Williams & Wilkins, Philadelphia, (2012) 67-128. [7] R. Mehr, H. Edelman, D. Sehgal, R. Mage, Analysis of mutational lineage trees from sites of
328
primary and secondary Ig gene diversification in rabbits and chickens, J. Immunol. 172
329
(2004) 4790-6, https://doi.org/10.4049/jimmunol.172.8.4790.
AC C
EP
327
330
[8] M. Wilson, A. Marcuz, L. du Pasquier, Somatic mutations during an immune response in
331
Xenopus tadpoles, Dev. Immunol. 4 (1995) 227-34, http://dx.doi.org/10.1155/1995/38639.
332
[9] F. Rothe, H. Ambrosius, Contributions on the immunobiology of poikilothermic vertebrates.
333
V. The proliferation of antibody-producing cells in tortoises, Acta. Biol. Med. Ger. 21 (1968)
334
525-36.
335
[10] M. Cossarini-Dunier, F.X. Desvaux, M. Dorson, Variability in humoral responses to DNP-
336
KLH of rainbow trout (Salmo gairdneri). Comparison of antibody kinetics and
ACCEPTED MANUSCRIPT
337
immunoglobulins spectrotypes between normal trouts and trouts obtained by gynogenesis or
338
self-fertilization, Dev. Comp. Immunol. 10 (1986) 207-17, https://doi.org/10.1016/0145-
339
305X(86)90004-2. [11] R.R. Avtalion, Temperature effect on antibody production and immunological memory, in
341
carp (Cyprinus carpio) immunized against bovine serum albumin (BSA), Immunology 17
342
(1969) 927-31.
RI PT
340
[12] G.T. Rijkers, E. M. H. Frederix-Wolters, W. B. Van Muiswinkel, The immune system of
344
cyprinid fish. The effect of antigen dose and route of administration on the development of
345
immunological memory in carp (Cyprinus carpio), M.J. Manning (ed.), Phylogeny of
346
Immunological Memory, Elsevier/North-Holland, Amsterdam, (1980) 93-102.
M AN U
347
SC
343
[13] M. Cossarini-Dunier, Secondary response of rainbow trout (Salmo gairdneri Richardson) to
348
DNP-haemocyanin
and
Yersinia
ruckeri,
349
https://doi.org/10.1016/0044-8486(86)90027-X.
Aquaculture
52
(1986)
81-6,
[14] M.F. Tatner, A. Adams, W. Leschen, An analysis of the primary and secondary antibody
351
response in intact and thymectomized rainbow trout, Salmo gairdneri richardson, to human
352
gamma globulin and Aeromonas salmonicida. J. Fish. Biol. 31 (1987) 177-95,
353
https://doi.org/10.1111/j.1095-8649.1987.tb05224.x.
TE D
350
[15] M.K. Raida, K. Buchmann, Bath vaccination of rainbow trout (Oncorhynchus mykiss
355
Walbaum) against Yersinia ruckeri: effects of temperature on protection and gene expression,
356
Vaccine 26 (2008) 1050-62, https://doi.org/10.1016/j.vaccine.2007.12.029.
EP
354
[16] G. Kurath, K.A. Garver, S. Corbeil, D.G. Elliott, E.D. Anderson, S.E. LaPatra, Protective
358
immunity and lack of histopathological damage two years after DNA vaccination against
359
infectious
360
https://doi.org/10.1016/j.vaccine.2005.07.068.
AC C
357
hematopoietic
necrosis
virus
in
trout,
Vaccine
24
(2006)
345-54,
361
[17] J.M. Dijkstra, H. Okamoto, M. Ototake, T. Nakanishi, Luciferase expression 2 years after
362
DNA injection in glass catfish (Kryptopterus bicirrhus), Fish Shellfish Immun. 11 (2001)
363
199-202, https://doi.org/10.1006/fsim.2000.0303.
ACCEPTED MANUSCRIPT
364
[18] T.C. Tonheim, J. Leirvik, M. Løvoll, A.I. Myhr, J. Bøgwald, R.A. Dalmo, Detection of
365
supercoiled plasmid DNA and luciferase expression in Atlantic salmon (Salmo salar L.) 535
366
days
367
https://doi.org/10.1016/j.fsi.2007.03.015
injection,
Fish
Shellfish
Immun.
23
(2007)
867-76,
RI PT
after
368
[19] N. Lorenzen, E. Lorenzen, K. Einer-Jensen, S.E. LaPatra, Immunity induced shortly after
369
DNA vaccination of rainbow trout against rhabdoviruses protects against heterologous virus
370
but not against bacterial pathogens, Dev. Comp. Immunol.
371
https://doi.org/10.1016/S0145-305X(01)00059-3.
SC
26 (2002) 173-9,
[20] M.K. Purcell, G. Kurath, K.A. Garver, R.P. Herwig, J.R. Winton, Quantitative expression
373
profiling of immune response genes in rainbow trout following infectious haematopoietic
374
necrosis virus (IHNV) infection or DNA vaccination, Fish Shellfish Immun. 17 (2004) 447-
375
62, https://doi.org/10.1016/j.fsi.2004.04.017.
M AN U
372
[21] N.O. Harun, T. Wang, C.J. Secombes, Gene expression profiling in naïve and vaccinated
377
rainbow trout after Yersinia ruckeri infection: insights into the mechanisms of protection seen
378
in vaccinated fish, Vaccine 29 (2011) 4388-99, https://doi.org/10.1016/j.vaccine.2011.04.003.
379
[22] M. Wilson, E. Hsu, A. Marcuz, M. Courtet, L. Du Pasquier, C. Steinberg, What limits
380
affinity maturation of antibodies in Xenopus--the rate of somatic mutation or the ability to
381
select mutants?, EMBO J. 11 (1992) 4337-47
TE D
376
[23] E. Bengtén, M. Wilson, Antibody repertoires in fish. In Pathogen-Host Interactions:
383
Antigenic Variation v. Somatic Adaptations, Springer International Publishing (2015) 193-
384
234, https://doi.org/10.1007/978-3-319-20819-0.
AC C
EP
382
385
[24] V.M. Barreto, Q. Pan-Hammarstrom, Y. Zhao, L. Hammarstrom, Z. Misulovin, M.C.
386
Nussenzweig, AID from bony fish catalyzes class switch recombination, J. Exp. Med. 202
387
(2005) 733-8, https://doi.org/10.1084/jem.20051378.
388
[25] L. Kato, A. Stanlie, N. A. Begum, M. Kobayashi, M. Aida, T. Honjo, An evolutionary view
389
of the mechanism for immune and genome diversity, J. Immunol. 188 (2012) 3559-66,
390
https://doi.org/10.4049/jimmunol.1102397.
ACCEPTED MANUSCRIPT
391 392
[26] B.G. Magor, Antibody Affinity Maturation in Fishes—Our Current Understanding, Biology, 4 (2015) 512-24, https://doi.org/10.3390/biology4030512. [27] N. Jiang, J.A. Weinstein, L. Penland, R.A. White, D.S. Fisher, S.R. Quake, Determinism
394
and stochasticity during maturation of the zebrafish antibody repertoire. P. Natl. Acad. Sci.
395
USA, 108 (2011) 5348-53. https://doi.org/10.1073/pnas.1014277108.
RI PT
393
[28] S.L. Kaattari, H.L. Zhang, W. Khor, I.M. Kaattari, D.A. Shapiro, Affinity maturation in
397
trout: clonal dominance of high affinity antibodies late in the immune response, Dev. Comp.
398
Immunol. 26 (2002) 191-200, https://doi.org/10.1016/S0145-305X(01)00064-7.
SC
396
[29] K.D. Cain, D.R. Jones, R.L. Raison, Antibody–antigen kinetics following immunization of
400
rainbow trout (Oncorhynchus mykiss) with a T-cell dependent antigen. Dev. Comp. Immunol.
401
26 (2002) 181-90, https://doi.org/10.1016/S0145-305X(01)00063-5.
M AN U
399
402
[30] J. Ye, E.S. Bromage, S.L. Kaattari, The strength of B cell interaction with antigen
403
determines the degree of IgM polymerization, J. Immunol. 184 (2010) 844-50,
404
https://doi.org/10.4049/jimmunol.0902364.
[31] J. Ye, I.M. Kaattari, S.L. Kaattari, The differential dynamics of antibody subpopulation
406
expression during affinity maturation in a teleost, Fish Shellfish Immun. 30 (2011) 372-7,
407
https://doi.org/10.1016/j.fsi.2010.11.013.
TE D
405
[32] B. Star, A.J. Nederbragt, S. Jentoft, U. Grimholt, M. Malmstrom, T.F. Gregers, T.B.
409
Rounge, J. Paulsen, M.H. Solbakken, A. Sharma et al. The genome sequence of Atlantic cod
410
reveals
411
https://doi.org/10.1038/nature10342.
unique
immune
system,
Nature
477
(2011)
207–10,
AC C
a
EP
408
412
[33] D. Haase, O. Roth, M. Kalbe, G. Schmiedeskamp, J.P. Scharsack, P. Rosenstiel, T.B.
413
Reusch, Absence of major histocompatibility complex class II mediated immunity in pipefish,
414
Syngnathus typhle: evidence from deep transcriptome sequencing, Biol. Lett. 9 (2013),
415
20130044, https://doi.org/10.1098/rsbl.2013.0044.
416
[34] E.S. Bromage, I.M. Kaattari, P. Zwollo, S.L. Kaattari, Plasmablast and plasma cell
417
production and distribution in trout immune tissues, J. Immunol. 173 (2004) 7317-23,
418
https://doi.org/10.4049/jimmunol.173.12.7317.
ACCEPTED MANUSCRIPT
419
[35] H.L. Saunders, A.L. Oko, A.N. Scott, C.W. Fan, B.G. Magor, The cellular context of AID
420
expressing cells in fish lymphoid tissues, Dev. Comp. Immunol. 34 (2010) 669-76,
421
https://doi.org/10.1016/j.dci.2010.01.013. [36] M. Yamasaki, K. Araki, T. Nakanishi, C. Nakayasu, A. Yamamoto, Role of CD4+ and
423
CD8α+ T cells in protective immunity against Edwardsiella tarda infection of ginbuna crucian
424
carp, Carassius auratus langsdorfii. Fish shellfish Immun. 36 (2014) 299-304,
425
https://doi.org/10.1016/j.fsi.2013.11.016.
AC C
EP
TE D
M AN U
SC
426
RI PT
422
ACCEPTED MANUSCRIPT
427
Legends
428
Table 1
429 430
Summary of the experimental design, including groups, recorded mortalities and numbers of sampled fish.
RI PT
431 432 433
Figure 1
Numbers of CD8α+, IgM+ and IgT+ leukocytes in six recipient groups. 21 (A) and 28 (B) days
435
after bath (or mock) challenge, percentages of three leukocyte subpopulations in PBL, gill
436
leukocytes and splenocytes were analysed by flow cytometry, and the total numbers of leukocyte
437
subpopulations per fish were calculated. Pooled samples from each group (three to six
438
fish/group; shown in Table 1) were analysed. Data from PBL of group 1 are not available for 28
439
d p.i. (N.A.) due to a technical failure (blood clotting). In these figures, the results are shown as
440
follows; left vertical panels depict PBL, centre panels gill and right panels spleen; upper
441
horizontal panels depict CD8α+, middle panels IgM+ and lower panels IgT+ cells. (C) Here
442
percentages of the three leukocyte subpopulations in splenocytes from group 1, 3, 4 and 6 were
443
compared on 21 and 28 d p.i. (note the increase of CD8α+ cells in group 6 at day 28). Black bars
444
show the data from day 28 p.i., white bars show the data from 21 d p.i., respectively. The X-axis
445
corresponds to the total numbers of the respective leukocyte subpopulations per fish (× 104).
446 447 448
Figure 2.
EP
TE D
M AN U
SC
434
Titres of anti-Y. ruckeri IgM antibodies in the vaccinated donor fish (left dot column) and in
450
the six recipient groups 1 to 6. 28 days after challenge, muci from gill and skin as well as sera
451
were collected from survived fish (three to six fish per group as shown in Table 1) and were
452
individually analysed by ELISA except skin- and gill- muci from vaccinated donor fish. For
453
skin- and gill- muci from vaccinated donor fish, samples from 2 fish were pooled. The Y-axis
454
shows OD values. Dots (●) represent individual OD values. Hyphens (-) represent mean OD
455
values of each group. To distinguish between positive and negative samples, cut-off levels were
456
calculated (threefold standard deviations of negative control ODs plus their corresponding mean
457
OD values) and are shown as red lines.
AC C
449
Total No. of fish
Infection status
22 0 (mock)
1 nonvaccinated
3
0
0%
6
6
6
14
63.6 %
4
4
4
0
0%
3
3
3
Low
0
0%
6
6
6
High
0
0%
6
6
6
4
22 0 (mock)
EP
AC C
22
TE D
High
6
28 dpi
3
22 22
28 dpi
3
3 AquavacERM vaccinated
21 dpi
0%
Low
5
No. of FCM-analyzed fish
Mortality
No. of ELISAanalyzed fish
0
22
2
No. of dead fish
SC
Donor
M AN U
Group
RI PT
ACCEPTED MANUSCRIPT
Table 1
ACCEPTED MANUSCRIPT
22 days post transfer (= 21 d p i) Gill
PBL
0
6 3
IgM
High
5 Low 2 4 0 1
3
IgT
5
High Low
2 4 1
0
RI PT 4
6
+ + + 0
6
2
200
+ + + 0
80
20
SC
0
1
M AN U
4
40
600
10
20
30
40
20
300
100
200
300
400
80
50
100
150
200
TE D
5 Low 2
400
EP
CD8α
+ + + -
600 0
10
AC C
6 High 3
Spleen
50
100
150 0
2
4
6
Numbers of lymphocytes (x104)
Fig. 1a
ACCEPTED MANUSCRIPT
29 days post transfer (= 28 d p i) Gill
PBL
0
1
RI PT
4
0
IgM
5 Low 2 4 0 1
+ + + - N.A. 0
6 3
IgT
5
High Low
2 4 1
0
15 0
200
+ + + - N.A. 0
10
20
30 0
20
40
60
80
TE D
3
High
10
400
600
800 0
5
10
15 0
200
400
600
800
300 0
1
2
30
50
100
150
200
AC C
6
5
SC
5 Low 2
M AN U
CD8α
+ + + - N.A.
EP
6 High 3
Spleen
100
200
Numbers of lymphocytes (x104)
Fig. 1b
ACCEPTED MANUSCRIPT
+
6 3
-
4
+
RI PT
High
CD8α
20
0
+
6 High 3
-
4
+
IgM 0
80
200
400
600
800
100
150
200
+
6
AC C
High
IgT
60
EP
-
1
40
28 dpi 21 dpi
M AN U
0
TE D
-
1
SC
0
3
-
4
+
0
1
0
50
Numbers of lymphocytes (x104)
Fig. 1c
ACCEPTED MANUSCRIPT 4.0
Serum
3.5 3.0 2.5
RI PT
2.0 1.5 1.0 0.5
0.15
0 2.0 1.6 1.2
EP
0.05
TE D
0.10
M AN U
Gill mucus
Skin mucus
AC C
OD value
0.20
SC
0 0.25
0.8 0.4 0
Vaccinated donor
+
-
0 1
+
-
Low 4
2
+ High
5
3
Vaccination Infection
6 Group
Fig. 2
ACCEPTED MANUSCRIPT
Highlights - Protection against Y. ruckeri can be transferred by Peripheral blood lymphocytes (PBL).
RI PT
- Numbers of CD8α+, IgM+ or IgT+ splenocytes were increased in recipients, which survived after high dosage infection with Y. ruckeri
AC C
EP
TE D
M AN U
SC
- IgM titers were highest in Y. ruckeri infected recipients that have previously been injected with immune cells from vaccinated donors
S1050-4648(18)30035-4
DOI:
10.1016/j.fsi.2018.01.035
Reference:
YFSIM 5089
To appear in:
Fish and Shellfish Immunology
Received Date: 26 October 2017 Revised Date:
18 January 2018
Accepted Date: 21 January 2018
Please cite this article as: Yamaguchi T, Quillet E, Boudinot P, Fischer U, What could be the mechanisms of immunological memory in fish?, Fish and Shellfish Immunology (2018), doi: 10.1016/ j.fsi.2018.01.035. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
2 x bath vaccine 10 + 2 weeks
AquavacERM
RI PT
non-vaccinated
M AN U
SC
adoptive transfer of PBL
mock inf.
Y.r. low
-
-
EP
TE D
bath challenge
Y.r. high
Y.r. low
mock inf.
+
-
-
-
CD8+, IgM+ and IgT+ splenocytes
Low
High
Antigen-specific IgM secretion
Low
High
Mortality
AC C
Y.r. high
Graphical abstract
ACCEPTED MANUSCRIPT
1 2 3 4
What could be the mechanisms of immunological memory in fish?
RI PT
5 6 7
Takuya Yamaguchi1, Edwige Quillet3, Pierre Boudinot2 and Uwe Fischer1*
SC
8 9
M AN U
10 11 12 13
1
14
10, Greifswald-Insel Riems 17493, Germany; E-Mails: [email protected] (T.Y.);
15
[email protected] (U.F.)
16
2
17
France
18
3
19
78350 Jouy-en-Josas, France
Laboratory of Fish Immunology, Institute of Infectology, Friedrich-Loeffler-Institut, Südufer
TE D
Virologie et Immunologie Moléculaires, INRA, Université Paris-Saclay, 78350 Jouy-en-Josas,
Génétique animale et biologie intégrative, INRA, AgroParisTech, Université Paris-Saclay,
EP
20 21 22
24
* Corresponding author
AC C
23
25
Key words
26
Rainbow trout; Vaccine; Yersinia ruckeri; Adaptive immunity; Adoptive transfer; Targetfish
ACCEPTED MANUSCRIPT
27
Abstract (256 words) Vaccination is the most effective strategy to control infectious diseases in species with
29
adaptive immunity. In human and in mouse, vaccination typically induces specific memory cells,
30
which can mediate a fast anamnestic response upon infection by the targeted pathogen. In these
31
species, successful vaccination induces a long-lasting protection, long after the titres of specific
32
antibodies and the frequency of specific T cells have returned to steady state. Vaccination is also
33
an important challenge in aquaculture, since alternative treatments are either too costly, or, in the
34
case of antibiotics, are harmful for the environment or may result in dangerous resistances.
35
However, the mechanisms of the long-term protection elicited by vaccines in fish remain poorly
36
understood. Although fish possess typical B- and T-cells expressing diverse repertoires of
37
immunoglobulins and T-cell receptors, many features of antigen specific responses are different
38
from what is known in mouse and in human. Memory is one of the most elusive properties of
39
fish adaptive immunity, and its basis is widely unknown. In this opinion article, we discuss the
40
concept of immune memory in the context of the fish immunity. We illustrate the complexity of
41
this question by discussing the results of experiments showing that protection can be passed
42
through adoptive transfer of leukocytes from vaccinated donor fish to naive histocompatible
43
recipients. Combined with tools developed in Targetfish and in previous projects, such as
44
monoclonal antibodies against B- and T-cell markers, we propose that such models of protection
45
transfer provide excellent systems to dissect the mechanisms of B- and T-cell memory in the
46
future.
SC
M AN U
TE D
EP AC C
47
RI PT
28
ACCEPTED MANUSCRIPT
48
1. How to define immune memory?
49
Immune memory is a fundamental characteristic of the adaptive immune system in
51
vertebrates. However, the definition of memory is still changing, under the influence of new
52
mechanisms of long-term protection. A recent article [1] put in perspective different points of
53
view about the nature of immunological memory. According to Donna Farber, memory is
54
defined by three conditions: (1) memory immune cells are long-lived and maintained
55
independently of stimulation by the antigen, (2) memory immune cells are epitope specific and
56
(3) memory immune cells are intrinsically changed by the previous encounter with the antigen.
57
Andreas Radbruch and Klaus Rajewsky consider that immunological memory is defined by "a
58
stimulus-specific change of immune reactivity that persists in the absence of the stimulus". Rolf
59
Zinkernagel reminds the classical definition of a memory response as a faster and stronger
60
response against a previously encountered antigen, compared with the primary response, and
61
emphasizes that the key question is whether or not protective immunity is maintained in an
62
antigen-dependent or -independent manner. Finally, Mihail Netea proposes a wider definition of
63
memory as the capacity of the immune system to keep and recall information on encountered
64
antigens or pathogens, which also includes the concept of innate immune memory or trained
65
immunity [2]. While adaptive immune memory involves permanent genetic changes such as
66
mutations and recombinations, innate immune memory is characterized by epigenetic
67
reprogramming involving sustained changes in gene expression and cell physiology. These
68
multiple definitions and concepts underscore the complexity of the question, while these authors
69
essentially refer to the immunity of well-known species of mammals. Overall, however, it is
70
accepted in mammals that vaccination against an infectious pathogen is based on the
71
establishment of populations of memory cells after an intentional immunization with a
72
homologous attenuated/inactivated strain of the corresponding pathogen or with parts of it.
73
Beyond individual protection, mammalian immunological memory is certainly also important for
74
the survival of offsprings in utero and during the neonatal life. This is mainly due to antibodies,
75
since protecting antibodies can be transferred from the mother to offspring during the foetal life
76
in many species, and also via colostrum (reviewed in [3,4]).
AC C
EP
TE D
M AN U
SC
RI PT
50
77
Importantly, the modalities of secondary responses, which allow assessment of
78
immunological memory, vary considerably across vertebrates. Differences between primary and
ACCEPTED MANUSCRIPT
secondary responses have been observed from fish to mammals. Typical secondary responses,
80
similar to those described in mouse and human, are fast and lead to higher titres of specific
81
antibodies when compared to primary responses; they are found in mammals and birds, where Ig
82
gene hypermutation occurs in germinal centres [5-7]. Early works in frogs proved the existence
83
of secondary responses in "lower" vertebrates, but showed they were not accompanied with a
84
substantial increase in antibody affinity, though Ig genes get hypermutated even in tadpoles [8].
85
In fish, the modalities of secondary responses and long-term protection are even less similar to
86
the typical situations described in mammals.
88
SC
87
RI PT
79
2. Secondary response and long-term protection in teleost fish The existence of a true secondary response with typical characteristics in fish was initially
90
debated [9]: the primary B cell response is generally much slower and develops over several
91
weeks or even months especially in cold water. For example, the antibody response against the
92
hapten dinitrophenyl keyhole limpet haemocyanin (DNP-KLH) in rainbow trout peaked 50 days
93
post immunization, and antibody titres remained high for a long time, at least until 150 days post
94
immunization [10]. Thus, the maintenance of protection might be partly due to a long-term
95
primary response and long-lasting increase of specific antibody titres. However, significant
96
differences between primary and secondary responses were demonstrated in several fish species.
97
A faster secondary response along with elevated antibody titres (about doubled titres) were
98
reported after immunization with bovine serum albumin [11] and with sheep red blood cells
99
(SRBC) [12] in carp, as well as against the pathogenic bacteria Y. ruckeri [13] and Aeromonas
100
salmonicida [14] in rainbow trout. The dose dependency of the secondary antibody response was
101
studied in detail in carp, further supporting the presence of a cell based immune memory: fish
102
were immunized with 105, 107 and 109 SRBC by intramuscular injection and challenged 1, 3, 6
103
or 10 months later with the same dose; with 109 SRBC, a secondary response was already
104
observed at 3 months and was maintained until later time points, and even with the lowest dose,
105
titres reached comparable or higher levels. However, the high-dose, 109 SRBC, did not lead to a
106
clear secondary response with enhanced titres, while a high primary response was induced [12].
107
These data showed that the secondary response was at least quantitatively higher than the
108
primary response, supporting the existence of a certain level of immunological memory.
AC C
EP
TE D
M AN U
89
ACCEPTED MANUSCRIPT
The existence of efficient “memory” in fish is also advocated by the (relatively) long-term
110
protection afforded by vaccination. For example, in rainbow trout, bath vaccination with a
111
bacterin of Y. ruckeri, the causative agent of enteric red mouth disease (ERM), was shown to
112
provide a significant protection of vaccinated fish 4 and 8 weeks post-vaccination [15]. DNA
113
vaccination of rainbow trout against infectious hematopoietic necrosis (IHN) has been shown to
114
confer protection over a period of two years, which would cover the life cycle of a commercial
115
fish [16]. It remains unclear whether a continuous stimulation of the immune system through
116
long-term expression of the viral G protein expression from the plasmid is required for this
117
protection. Importantly, reporter protein expression from injected plasmids can still be detected
118
over long periods [17, 18]. Another example is DNA vaccination against viral haemorrhagic
119
septicaemia (VHS), which was also shown to induce long-lasting protection post vaccination
120
[19]. While protection established shortly after vaccination was cross-protective in terms of a
121
heterologous viral pathogen this cross-protection was absent two months after vaccination, while
122
the specific response to the homologous virus persisted. In most cases, this protection seems to
123
be based on anamnestic secondary responses, which provide a better defence against the
124
pathogens compared to a naive immune system. Analysis of gene expression analysis showed
125
that such a protection was somehow associated with up- or down-regulation events of the
126
expression of immune related molecules such as cytokines and transcriptomic factors [15, 20,
127
21]. However, the detailed cellular basis of protection including T- and B-cell dynamics after
128
fish vaccination is often poorly understood and, most importantly, the persistence of the antigen
129
between the initial immunization and the challenge is most often not directly evaluated.
EP
TE D
M AN U
SC
RI PT
109
Other key features associated with immune B cell memory formation in mammals are class
131
switch and affinity maturation. While an Ig class switch seems to be absent in teleosts, at least in
132
all species studied in sufficient detail, it is generally believed that most lower vertebrates can
133
hypermutate Ig genes but lack the microenvironment required for the selection of B cells with
134
hypermutated Ig receptors of high affinity for the antigen [22, 23]. The enzymatic machinery
135
needed for somatic hypermutation of Ig genes is present in teleosts [24-26], and indeed
136
hypermutation was observed in zebrafish mature Ig repertoires [27]. Interestingly, long-term
137
studies of the antibody responses in rainbow trout showed that high affinity antibodies appear
138
late and become dominant, more than 25 weeks after (primary) immunization [28, 29]. The
139
kinetics of antibody affinity in trout immunized with TNP-KLH revealed that a relatively low
AC C
130
ACCEPTED MANUSCRIPT
affinity antibody subpopulation appears early, does not achieve high titres and is transient; an
141
intermediate affinity subpopulation appears later (five weeks after immunization), achieves
142
relatively high titres and persists longer than the low affinity antibodies; and the highest affinity
143
subpopulation emerges only after 15 weeks at still high titres [30, 31]. These studies suggest, that
144
a slow affinity maturation may actually occur in fish through alternative mechanisms.
RI PT
140
The vast diversity of fish (and their immune systems) should also be considered in
146
investigations about putative immune memory in this group. Thus, species related to cod
147
(Gadidae) and pipefish lack MHC class II and CD4 [32, 33], which implies that B cell responses
148
cannot rely on a T cell help as in mammals. In these species, B cell memory, if it exists, must be
149
based on particular pathways and this assumption should be considered with regard to the
150
successful vaccination protocols in these fish species.
M AN U
SC
145
In this general context, the definition and the putative location of fish memory cells also
152
remain quite elusive. B cells from rainbow trout head kidney still produced specific antibodies
153
later than 20 weeks post immunization with TNP-KLH [34]. In contrast, spleen or blood
154
leukocytes contained very few or no TNP-specific antibody secreting cells 10 - 15 weeks after
155
immunization. Thus, rainbow trout anterior kidney antibody secreting cells may resemble
156
mammalian long-lived plasma cells, and they are likely responsible for sustained serum antibody
157
titres [34]. In the channel catfish, it has been proposed that hypermutation of Ig genes may occur
158
in melanomacrophage centres of the spleen, since activation-induced cytidine deaminase (an
159
enzyme causing somatic hypermutation, gene conversion, and class-switch recombination of
160
immunoglobulin genes in mammalian B cells) is expressed in these structures [35], but the
161
cellular context remains to be further dissected in fish (discussed in [26]). Overall, fish memory
162
cells are very poorly defined and their existence remains debatable.
EP
AC C
163
TE D
151
164
3. A system to further define the basis of immune memory: transfer of
165
protection through cell transplant in rainbow trout.
166
If fish possess memory cells, their phenotypes, frequency, lifespan and functions remain
167
essentially unknown. The lack of antibodies against key surface markers has certainly hampered
168
progress in this field, making difficult the definition and isolation of relevant cell subsets.
169
Furthermore, the lack of syngeneic or congenic animals has hindered cell transfer experiments,
170
which would help defining a “memory” cell compartment. Taking advantage of the availability
ACCEPTED MANUSCRIPT
171
of clonal, isogenic rainbow trout, we could recently transfer protection by transplantation of
172
peripheral blood leukocytes (PBL) from vaccinated fish to histocompatible individuals. The
173
potential interest of this experimental set up for the study of immunological memory will be
174
discussed below. Isogenic fish were vaccinated by immersion using a commercial vaccine against ERM, and
176
boosted ten weeks later, also by immersion (material and methods in supplements). Control fish
177
of the same isogenic progeny were mock-vaccinated and -boosted. Two weeks after the boost,
178
PBL were transferred to histocompatible, naive recipients from the same fish clone, i.e. with the
179
same genetic background. Recipients were challenged by bath with different dosages (mock, low
180
and high) of Y. ruckeri twenty-four hours later and mortalities were recorded (Table 1). The high
181
dose challenge led to high mortality (63%) of recipient fish that had received PBL from non-
182
vaccinated donors (group3). While mortalities were observed between day 5 and 12 after
183
challenge, most fish died until day 9. In contrast, fish which had received PBL from vaccinated
184
donors (group 6), were fully protected, indicating that protection against Y. ruckeri was conferred
185
to the recipients by adoptive transfer of PBL from vaccinated donors.
M AN U
SC
RI PT
175
Following PBL transfer, dynamics of CD8α+, IgM+ and IgT+ lymphocytes in the blood, gills
187
and spleens of recipients that had survived the infection were analysed. We are aware that bath
188
challenge did not ensure that all fish had received the same amount of bacteria, and that the
189
survival might have been affected by this factor; however, these observations allow a first
190
comparison between immune cell populations in protected fish and in the other groups on 22
191
(Fig. 1a) and 29 days (Fig. 1b; Fig 1c) after transfer, which are corresponding to 21 and 28 days
192
post challenge, respectively.
EP
TE D
186
Twenty two days post transfer, the total number of CD8α+ T cells in non-challenged
194
recipients which had received PBL from non-vaccinated fish (group 1), was about 2 × 104 in
195
PBL and approx. ten times more in spleen and gills (Fig. 1a). In comparison, non-challenged
196
recipients, which were injected with PBL from vaccinated donors (group 4), had much more
197
CD8α+ T cells in the gills (about 2.5 times more than those of group 1), while this trend was not
198
observed in spleen and PBL. This suggests that T cells might have accumulated (or proliferated)
199
in gills three weeks after the transfer. In challenged recipients, twenty-one days after bath
200
infection, the number of CD8α+ T cells was higher in PBL of recipients, which had received
201
cells from non-vaccinated donors (group 2 and 3), but not in those transferred with cells from
AC C
193
ACCEPTED MANUSCRIPT
vaccinated donors (group 5 and 6). This trend was also observed in the spleen of recipients
203
challenged at low dose (compare group 2 to group 5), but not in other cases. This suggests that
204
although recipient groups challenged with low dose did not show mortality, different responses
205
and recruitments of T cells occurred between fish transferred with cells from vaccinated and
206
control fish. These patterns were not observed 28 days post challenge, the most obvious pattern
207
being then an increase (approx. 3.5 times) of CD8α+ T cells in the spleen of high dose
208
challenged recipient fish, which had received PBL from vaccinated donors (and were protected,
209
group 6). In the same recipient group, there was an increase (approx. 2.5 times) of CD8α+
210
splenocytes between day 21 and 28 (Fig. 1c) post challenge, while such an increase was not
211
observed in recipient fish which had received PBL from control donors (group 3, in which high
212
mortalities were recorded). To what extent this protection relates to a CD8α+ T cell response
213
remains to be analysed. Of note, no differences in CD8α+ T cell numbers were observed 28 days
214
post challenge, between non-challenged fish, which had received PBL from vaccinated (group 4)
215
and control donors (group 1). Additional studies will be necessary to establish firmly that this
216
kinetics is reproducible and to further elucidate the role of CD8α+ T cells in the transferred
217
protection to ERM.
M AN U
SC
RI PT
202
Similar to CD8α+T cells, twenty two days post transfer, IgM+ and IgT+ B cell numbers were
219
about three times increased in gills of non-challenged recipients, which had received PBL from
220
vaccinated donors (group 4) compared to those which had received PBL from non-vaccinated
221
control donors (group 1). Again, this pattern was not found twenty nine days post transfer. The
222
only remarkable trend regarding B cells was observed at the second time point of sampling. 28
223
days after high dose infection an accumulation (or proliferation) of both IgM+ and IgT+ B cells in
224
the spleen of challenged fish was observed, especially in recipients which had received PBL
225
from vaccinated donors (group 6). Combining the most obvious increase in B cell numbers in
226
high dose challenged recipients that have received immune cells from vaccinated donors with the
227
fact they are fully protected (Table 1) points on a role of B cells in the immune response against
228
ERM.
AC C
EP
TE D
218
229
An interesting observation was made in the spleen, where the numbers of CD8α+, IgM+ and
230
IgT+ cells of high dose challenged recipient fish, which had received PBL from vaccinated
231
donors (group 6) at 28 d p.i. were higher than in those at 21 d p.i. Such an increase was not
232
observed in low dose challenged (group 1 and 4) or mock challenged (group 2 and 5) recipients
ACCEPTED MANUSCRIPT
233
(Fig. 1C). This suggests that B- and T-cell responses might continue developing at this stage in
234
the spleen of group 6 recipients at 28 d p.i. Importantly, such trends should be taken with great caution, since they come from a pool of 3
236
- 6 survivors from a single campaign of experiments. Although a single individual with a
237
remarkable phenotype may induce a bias, we believe the strongest trends are interesting.
238
Although the modest numbers of fish used in this first set of experiment do not fully demonstrate
239
trends, they provide a first frame of analysis to be confirmed by further studies. The adoptive
240
transfer model appeared to be excellent to pursue the characterization of mechanisms involved in
241
the transfer of “memory” and/or protection.
SC
RI PT
235
In order to check the titres of secreted IgM antibodies against Y. ruckeri, sera as well as muci
243
from gills and skin were collected from surviving recipients 28 days after challenge and were
244
analysed by ELISA. Among all examined groups, the titres of anti-Y. ruckeri IgM in muci and
245
sera were clearly elevated only in recipients of group 6 that had received PBL from vaccinated
246
donors and that thereafter had been challenged with high dose of Y. ruckeri (Fig. 2) further
247
suggesting that a secondary response is dose dependent. Nevertheless, these results indicate that
248
transfer of PBL from vaccinated donors successfully transferred the capacity to produce high
249
titres of anti-Y. ruckeri antibodies upon infection in the recipients, hence to mount a "memory"
250
response. This was correlated with high levels of protection. In contrast, recipient fish that had
251
received PBL from non-immunized donors (group 3) showed high mortality after high dose
252
challenge, and survivors did not develop anti-Y. ruckeri IgM antibodies. Interestingly, not all
253
recipients of group 6 were seroconverted although full protection and high levels of IgM+ and
254
IgT+ cells were recorded. Group 6 can be divided into non- and high IgM responders (which is
255
partially also true for vaccinated donors). This implies that several mechanisms may explain
256
survival and that, although genetically identical fish were used in these experiments, immune
257
responses to a given pathogen may show considerable interindividual variability in fish. These
258
observations underline the importance of following these parameters when tracking immune
259
memory.
AC C
EP
TE D
M AN U
242
260 261
4. Perspectives and discussion
262
The available literature illustrates and the case study described above exemplifies that much
263
remains to be understood and defined regarding the protection against pathogens - including the
ACCEPTED MANUSCRIPT
protection induced by vaccination - and a putative immunological memory in fish. An important
265
issue would be to identify the cellular compartment responsible for the long-term protection.
266
Whether the mechanisms triggered by such cells may be named "memory" would be the next
267
question. Several mechanisms can be proposed on how protection has been transferred from
268
vaccinated donor fish to recipient fish. One possibility is that effector cells were transferred in an
269
activated stage induced by vaccination plus boost, and that thus have directly contributed to
270
rescue recipients from infection. Such activated effector cells could have acted specifically for
271
example via specific antibody production, neutralising the pathogen. It is also reasonable to
272
consider that transferred donor effector cells mounted a more efficient immune response in
273
concert with the hitherto naïve immune system of the recipient. Thus, innate immunity of the
274
recipients complemented by the action of transferred cells could be efficient enough to promote
275
survival. Last but not least, bona fide memory cells could have been transferred to recipients
276
allowing a fast and effective response to the infection and leading to high survival. These
277
different hypotheses could be tested by performing transfer of particular sorted cell subsets, and
278
detailed characterization of the response to different challenges (homologous versus
279
heterologous pathogens, multiples routes). Using a similar experimental set-up in isogenic
280
ginbuna crucian carp [36] where Yamazaki et al. have shown that adoptive transfer of CD4+ and
281
CD8α+ cells from vaccinated donors protects recipients against a homologous Edwardsiella
282
tarda challenge. The carriers of protection were CD8α+ cells and, to a lesser extent, CD4+ cells.
283
However in this study, ginbuna lymphocytes were isolated from donor fish only 8 days after
284
immunization, which is probably enough to mount a specific immune responses in this species,
285
but clearly too short to study long-term, persistent "memory" cells.
EP
TE D
M AN U
SC
RI PT
264
To get more insights in the mechanisms of long-term protection and memory, transfer
287
experiments of sorted cell subsets should be performed in models similar to the one described
288
above in trout. When the size of the fish and the time line allows, it would be very powerful tool
289
to follow immune parameters such as antigen specific antibody titres in individual fish, using
290
non-lethal sampling, before and after challenge.
AC C
286
291
In conclusion, models such as the cell transfer between histocompatible fish presented in this
292
article will be instrumental to understand the mechanisms and impact of anamnestic memory
293
responses in fish. These future analyses will benefit from antibodies specific for lymphocytes
294
markers. A better characterization of markers such as CD27, CD38, CD45RO, CD45RA and
ACCEPTED MANUSCRIPT
CD62L (and production of the corresponding reagents) in fish would be also extremely useful.
296
Besides, the development of new high-throughput transcriptomic approaches will likely be
297
instrumental for the definition of memory cell subsets in fish. Modifications of lymphocyte
298
populations in lymphoid organs after PBL transfer and or challenge suggests that transferred
299
cells might be prone to recruitment into defined territories, or to re-localisation of immune cells.
RI PT
295
AC C
EP
TE D
M AN U
SC
300
ACCEPTED MANUSCRIPT
301
Acknowledgements We are grateful to Prof. Oriol Sunyer, University of Pennsylvania, USA, for providing us the
303
monoclonal antibody against IgT, to Prof. Inger Dalsgaard, Danish Technical University for
304
providing us with the Yersinia ruckeri O1 strain and to Prof. Kurt Buchmann, University of
305
Copenhagen, Denmark for technical advice regarding ELISA. We also thank Mrs. Susann
306
Schares for excellent technical assistance. This work was supported by the EU FP7 grant 311993
307
(TargetFish) for Pierre Boudinot, Takuya Yamaguchi and Uwe Fischer.
RI PT
302
AC C
EP
TE D
M AN U
SC
308
ACCEPTED MANUSCRIPT
References
310
[1] D.L. Farber, M.G. Netea, A. Radbruch, K. Rajewsky, R.M. Zinkernagel, Immunological
311
memory: lessons from the past and a look to the future, Nat. Rev. Immunol. 16 (2016) 124-8,
312
https://doi.org/10.1038/nri.2016.13.
RI PT
309
313
[2] M.G. Netea, L.A. Joosten, E. Latz, K.H. Mills, G. Natoli, H.G. Stunnenberg, L.A. O’Neill,
314
R.J. Xavier, Trained immunity: a program of innate immune memory in health and disease,
315
Science 352 (2016) aaf1098, 1-9, https://doi.org/10.1126/science.aaf1098
[3] A. Marchant, M. Sadarangani, M. Garand, N. Dauby, V. Verhasselt, L. Pereira, G. Bjornson,
317
C.E. Jones, S.A. Halperin, K.M. Edwards et al., Maternal immunisation: collaborating with
318
mother nature. Lancet Infect. Dis. 17 (2017) e197-e208. https://doi.org/10.1016/S1473-
319
3099(17)30229-3.
M AN U
SC
316
320
[4] M.F. Jennewein, B. Abu-Raya, Y. Jiang, G. Alter, A. Marchant, Transfer of maternal
321
immunity and programming of the newborn immune system, Semin. Immunopathol. 39
322
(2017) 605-13, https://doi.org/10.1007/s00281-017-0653-x
324 325 326
[5] L. Du Pasquier, Meeting the demand for innate and adaptive immunities during evolution, Scand. J. Immunol. 62 (2005) 39-48, https://doi.org/10.1111/j.1365-3083.2005.01608.x.
TE D
323
[6] M. Flajnik, L. du Pasquier, Evolution of the immune system in Fundamental immunology 7th Edn. (Paul, W. E., ed.) Lippincott-Williams & Wilkins, Philadelphia, (2012) 67-128. [7] R. Mehr, H. Edelman, D. Sehgal, R. Mage, Analysis of mutational lineage trees from sites of
328
primary and secondary Ig gene diversification in rabbits and chickens, J. Immunol. 172
329
(2004) 4790-6, https://doi.org/10.4049/jimmunol.172.8.4790.
AC C
EP
327
330
[8] M. Wilson, A. Marcuz, L. du Pasquier, Somatic mutations during an immune response in
331
Xenopus tadpoles, Dev. Immunol. 4 (1995) 227-34, http://dx.doi.org/10.1155/1995/38639.
332
[9] F. Rothe, H. Ambrosius, Contributions on the immunobiology of poikilothermic vertebrates.
333
V. The proliferation of antibody-producing cells in tortoises, Acta. Biol. Med. Ger. 21 (1968)
334
525-36.
335
[10] M. Cossarini-Dunier, F.X. Desvaux, M. Dorson, Variability in humoral responses to DNP-
336
KLH of rainbow trout (Salmo gairdneri). Comparison of antibody kinetics and
ACCEPTED MANUSCRIPT
337
immunoglobulins spectrotypes between normal trouts and trouts obtained by gynogenesis or
338
self-fertilization, Dev. Comp. Immunol. 10 (1986) 207-17, https://doi.org/10.1016/0145-
339
305X(86)90004-2. [11] R.R. Avtalion, Temperature effect on antibody production and immunological memory, in
341
carp (Cyprinus carpio) immunized against bovine serum albumin (BSA), Immunology 17
342
(1969) 927-31.
RI PT
340
[12] G.T. Rijkers, E. M. H. Frederix-Wolters, W. B. Van Muiswinkel, The immune system of
344
cyprinid fish. The effect of antigen dose and route of administration on the development of
345
immunological memory in carp (Cyprinus carpio), M.J. Manning (ed.), Phylogeny of
346
Immunological Memory, Elsevier/North-Holland, Amsterdam, (1980) 93-102.
M AN U
347
SC
343
[13] M. Cossarini-Dunier, Secondary response of rainbow trout (Salmo gairdneri Richardson) to
348
DNP-haemocyanin
and
Yersinia
ruckeri,
349
https://doi.org/10.1016/0044-8486(86)90027-X.
Aquaculture
52
(1986)
81-6,
[14] M.F. Tatner, A. Adams, W. Leschen, An analysis of the primary and secondary antibody
351
response in intact and thymectomized rainbow trout, Salmo gairdneri richardson, to human
352
gamma globulin and Aeromonas salmonicida. J. Fish. Biol. 31 (1987) 177-95,
353
https://doi.org/10.1111/j.1095-8649.1987.tb05224.x.
TE D
350
[15] M.K. Raida, K. Buchmann, Bath vaccination of rainbow trout (Oncorhynchus mykiss
355
Walbaum) against Yersinia ruckeri: effects of temperature on protection and gene expression,
356
Vaccine 26 (2008) 1050-62, https://doi.org/10.1016/j.vaccine.2007.12.029.
EP
354
[16] G. Kurath, K.A. Garver, S. Corbeil, D.G. Elliott, E.D. Anderson, S.E. LaPatra, Protective
358
immunity and lack of histopathological damage two years after DNA vaccination against
359
infectious
360
https://doi.org/10.1016/j.vaccine.2005.07.068.
AC C
357
hematopoietic
necrosis
virus
in
trout,
Vaccine
24
(2006)
345-54,
361
[17] J.M. Dijkstra, H. Okamoto, M. Ototake, T. Nakanishi, Luciferase expression 2 years after
362
DNA injection in glass catfish (Kryptopterus bicirrhus), Fish Shellfish Immun. 11 (2001)
363
199-202, https://doi.org/10.1006/fsim.2000.0303.
ACCEPTED MANUSCRIPT
364
[18] T.C. Tonheim, J. Leirvik, M. Løvoll, A.I. Myhr, J. Bøgwald, R.A. Dalmo, Detection of
365
supercoiled plasmid DNA and luciferase expression in Atlantic salmon (Salmo salar L.) 535
366
days
367
https://doi.org/10.1016/j.fsi.2007.03.015
injection,
Fish
Shellfish
Immun.
23
(2007)
867-76,
RI PT
after
368
[19] N. Lorenzen, E. Lorenzen, K. Einer-Jensen, S.E. LaPatra, Immunity induced shortly after
369
DNA vaccination of rainbow trout against rhabdoviruses protects against heterologous virus
370
but not against bacterial pathogens, Dev. Comp. Immunol.
371
https://doi.org/10.1016/S0145-305X(01)00059-3.
SC
26 (2002) 173-9,
[20] M.K. Purcell, G. Kurath, K.A. Garver, R.P. Herwig, J.R. Winton, Quantitative expression
373
profiling of immune response genes in rainbow trout following infectious haematopoietic
374
necrosis virus (IHNV) infection or DNA vaccination, Fish Shellfish Immun. 17 (2004) 447-
375
62, https://doi.org/10.1016/j.fsi.2004.04.017.
M AN U
372
[21] N.O. Harun, T. Wang, C.J. Secombes, Gene expression profiling in naïve and vaccinated
377
rainbow trout after Yersinia ruckeri infection: insights into the mechanisms of protection seen
378
in vaccinated fish, Vaccine 29 (2011) 4388-99, https://doi.org/10.1016/j.vaccine.2011.04.003.
379
[22] M. Wilson, E. Hsu, A. Marcuz, M. Courtet, L. Du Pasquier, C. Steinberg, What limits
380
affinity maturation of antibodies in Xenopus--the rate of somatic mutation or the ability to
381
select mutants?, EMBO J. 11 (1992) 4337-47
TE D
376
[23] E. Bengtén, M. Wilson, Antibody repertoires in fish. In Pathogen-Host Interactions:
383
Antigenic Variation v. Somatic Adaptations, Springer International Publishing (2015) 193-
384
234, https://doi.org/10.1007/978-3-319-20819-0.
AC C
EP
382
385
[24] V.M. Barreto, Q. Pan-Hammarstrom, Y. Zhao, L. Hammarstrom, Z. Misulovin, M.C.
386
Nussenzweig, AID from bony fish catalyzes class switch recombination, J. Exp. Med. 202
387
(2005) 733-8, https://doi.org/10.1084/jem.20051378.
388
[25] L. Kato, A. Stanlie, N. A. Begum, M. Kobayashi, M. Aida, T. Honjo, An evolutionary view
389
of the mechanism for immune and genome diversity, J. Immunol. 188 (2012) 3559-66,
390
https://doi.org/10.4049/jimmunol.1102397.
ACCEPTED MANUSCRIPT
391 392
[26] B.G. Magor, Antibody Affinity Maturation in Fishes—Our Current Understanding, Biology, 4 (2015) 512-24, https://doi.org/10.3390/biology4030512. [27] N. Jiang, J.A. Weinstein, L. Penland, R.A. White, D.S. Fisher, S.R. Quake, Determinism
394
and stochasticity during maturation of the zebrafish antibody repertoire. P. Natl. Acad. Sci.
395
USA, 108 (2011) 5348-53. https://doi.org/10.1073/pnas.1014277108.
RI PT
393
[28] S.L. Kaattari, H.L. Zhang, W. Khor, I.M. Kaattari, D.A. Shapiro, Affinity maturation in
397
trout: clonal dominance of high affinity antibodies late in the immune response, Dev. Comp.
398
Immunol. 26 (2002) 191-200, https://doi.org/10.1016/S0145-305X(01)00064-7.
SC
396
[29] K.D. Cain, D.R. Jones, R.L. Raison, Antibody–antigen kinetics following immunization of
400
rainbow trout (Oncorhynchus mykiss) with a T-cell dependent antigen. Dev. Comp. Immunol.
401
26 (2002) 181-90, https://doi.org/10.1016/S0145-305X(01)00063-5.
M AN U
399
402
[30] J. Ye, E.S. Bromage, S.L. Kaattari, The strength of B cell interaction with antigen
403
determines the degree of IgM polymerization, J. Immunol. 184 (2010) 844-50,
404
https://doi.org/10.4049/jimmunol.0902364.
[31] J. Ye, I.M. Kaattari, S.L. Kaattari, The differential dynamics of antibody subpopulation
406
expression during affinity maturation in a teleost, Fish Shellfish Immun. 30 (2011) 372-7,
407
https://doi.org/10.1016/j.fsi.2010.11.013.
TE D
405
[32] B. Star, A.J. Nederbragt, S. Jentoft, U. Grimholt, M. Malmstrom, T.F. Gregers, T.B.
409
Rounge, J. Paulsen, M.H. Solbakken, A. Sharma et al. The genome sequence of Atlantic cod
410
reveals
411
https://doi.org/10.1038/nature10342.
unique
immune
system,
Nature
477
(2011)
207–10,
AC C
a
EP
408
412
[33] D. Haase, O. Roth, M. Kalbe, G. Schmiedeskamp, J.P. Scharsack, P. Rosenstiel, T.B.
413
Reusch, Absence of major histocompatibility complex class II mediated immunity in pipefish,
414
Syngnathus typhle: evidence from deep transcriptome sequencing, Biol. Lett. 9 (2013),
415
20130044, https://doi.org/10.1098/rsbl.2013.0044.
416
[34] E.S. Bromage, I.M. Kaattari, P. Zwollo, S.L. Kaattari, Plasmablast and plasma cell
417
production and distribution in trout immune tissues, J. Immunol. 173 (2004) 7317-23,
418
https://doi.org/10.4049/jimmunol.173.12.7317.
ACCEPTED MANUSCRIPT
419
[35] H.L. Saunders, A.L. Oko, A.N. Scott, C.W. Fan, B.G. Magor, The cellular context of AID
420
expressing cells in fish lymphoid tissues, Dev. Comp. Immunol. 34 (2010) 669-76,
421
https://doi.org/10.1016/j.dci.2010.01.013. [36] M. Yamasaki, K. Araki, T. Nakanishi, C. Nakayasu, A. Yamamoto, Role of CD4+ and
423
CD8α+ T cells in protective immunity against Edwardsiella tarda infection of ginbuna crucian
424
carp, Carassius auratus langsdorfii. Fish shellfish Immun. 36 (2014) 299-304,
425
https://doi.org/10.1016/j.fsi.2013.11.016.
AC C
EP
TE D
M AN U
SC
426
RI PT
422
ACCEPTED MANUSCRIPT
427
Legends
428
Table 1
429 430
Summary of the experimental design, including groups, recorded mortalities and numbers of sampled fish.
RI PT
431 432 433
Figure 1
Numbers of CD8α+, IgM+ and IgT+ leukocytes in six recipient groups. 21 (A) and 28 (B) days
435
after bath (or mock) challenge, percentages of three leukocyte subpopulations in PBL, gill
436
leukocytes and splenocytes were analysed by flow cytometry, and the total numbers of leukocyte
437
subpopulations per fish were calculated. Pooled samples from each group (three to six
438
fish/group; shown in Table 1) were analysed. Data from PBL of group 1 are not available for 28
439
d p.i. (N.A.) due to a technical failure (blood clotting). In these figures, the results are shown as
440
follows; left vertical panels depict PBL, centre panels gill and right panels spleen; upper
441
horizontal panels depict CD8α+, middle panels IgM+ and lower panels IgT+ cells. (C) Here
442
percentages of the three leukocyte subpopulations in splenocytes from group 1, 3, 4 and 6 were
443
compared on 21 and 28 d p.i. (note the increase of CD8α+ cells in group 6 at day 28). Black bars
444
show the data from day 28 p.i., white bars show the data from 21 d p.i., respectively. The X-axis
445
corresponds to the total numbers of the respective leukocyte subpopulations per fish (× 104).
446 447 448
Figure 2.
EP
TE D
M AN U
SC
434
Titres of anti-Y. ruckeri IgM antibodies in the vaccinated donor fish (left dot column) and in
450
the six recipient groups 1 to 6. 28 days after challenge, muci from gill and skin as well as sera
451
were collected from survived fish (three to six fish per group as shown in Table 1) and were
452
individually analysed by ELISA except skin- and gill- muci from vaccinated donor fish. For
453
skin- and gill- muci from vaccinated donor fish, samples from 2 fish were pooled. The Y-axis
454
shows OD values. Dots (●) represent individual OD values. Hyphens (-) represent mean OD
455
values of each group. To distinguish between positive and negative samples, cut-off levels were
456
calculated (threefold standard deviations of negative control ODs plus their corresponding mean
457
OD values) and are shown as red lines.
AC C
449
Total No. of fish
Infection status
22 0 (mock)
1 nonvaccinated
3
0
0%
6
6
6
14
63.6 %
4
4
4
0
0%
3
3
3
Low
0
0%
6
6
6
High
0
0%
6
6
6
4
22 0 (mock)
EP
AC C
22
TE D
High
6
28 dpi
3
22 22
28 dpi
3
3 AquavacERM vaccinated
21 dpi
0%
Low
5
No. of FCM-analyzed fish
Mortality
No. of ELISAanalyzed fish
0
22
2
No. of dead fish
SC
Donor
M AN U
Group
RI PT
ACCEPTED MANUSCRIPT
Table 1
ACCEPTED MANUSCRIPT
22 days post transfer (= 21 d p i) Gill
PBL
0
6 3
IgM
High
5 Low 2 4 0 1
3
IgT
5
High Low
2 4 1
0
RI PT 4
6
+ + + 0
6
2
200
+ + + 0
80
20
SC
0
1
M AN U
4
40
600
10
20
30
40
20
300
100
200
300
400
80
50
100
150
200
TE D
5 Low 2
400
EP
CD8α
+ + + -
600 0
10
AC C
6 High 3
Spleen
50
100
150 0
2
4
6
Numbers of lymphocytes (x104)
Fig. 1a
ACCEPTED MANUSCRIPT
29 days post transfer (= 28 d p i) Gill
PBL
0
1
RI PT
4
0
IgM
5 Low 2 4 0 1
+ + + - N.A. 0
6 3
IgT
5
High Low
2 4 1
0
15 0
200
+ + + - N.A. 0
10
20
30 0
20
40
60
80
TE D
3
High
10
400
600
800 0
5
10
15 0
200
400
600
800
300 0
1
2
30
50
100
150
200
AC C
6
5
SC
5 Low 2
M AN U
CD8α
+ + + - N.A.
EP
6 High 3
Spleen
100
200
Numbers of lymphocytes (x104)
Fig. 1b
ACCEPTED MANUSCRIPT
+
6 3
-
4
+
RI PT
High
CD8α
20
0
+
6 High 3
-
4
+
IgM 0
80
200
400
600
800
100
150
200
+
6
AC C
High
IgT
60
EP
-
1
40
28 dpi 21 dpi
M AN U
0
TE D
-
1
SC
0
3
-
4
+
0
1
0
50
Numbers of lymphocytes (x104)
Fig. 1c
ACCEPTED MANUSCRIPT 4.0
Serum
3.5 3.0 2.5
RI PT
2.0 1.5 1.0 0.5
0.15
0 2.0 1.6 1.2
EP
0.05
TE D
0.10
M AN U
Gill mucus
Skin mucus
AC C
OD value
0.20
SC
0 0.25
0.8 0.4 0
Vaccinated donor
+
-
0 1
+
-
Low 4
2
+ High
5
3
Vaccination Infection
6 Group
Fig. 2
ACCEPTED MANUSCRIPT
Highlights - Protection against Y. ruckeri can be transferred by Peripheral blood lymphocytes (PBL).
RI PT
- Numbers of CD8α+, IgM+ or IgT+ splenocytes were increased in recipients, which survived after high dosage infection with Y. ruckeri
AC C
EP
TE D
M AN U
SC
- IgM titers were highest in Y. ruckeri infected recipients that have previously been injected with immune cells from vaccinated donors