Fish & Shellsh Immunology (2002) 12, 287301

doi:10.1006/fsim.2001.0371
Available online at http://www.idealibrary.com on
Differences in MHC class I genes between strains of
rainbow trout (Oncorhynchus mykiss)
CHUN XIA
1
*, IKUNARI KIRYU
1
*, JOHANNES MARTINUS DIJKSTRA
1
, TERUO AZUMA
2
,
TERUYUKI NAKANISHI
1
AND MITSURU OTOTAKE
1

1
Inland Station/National Research Institute of Aquaculture, Tamaki,
Mie 519-0423, Japan and
2
Nikko Branch, National Research Institute of
Aquaculture, Chugushi, Nikko, Tochigi 321-1661, Japan
(Received 28 February 2001, accepted after revision 4 July 2001, published
electronically 28 January 2002)
In rainbow trout there is only one dominant classical MHC class I locus,
Onmy-UBA, for which four very di#erent allelic lineages have been described.
The purpose of the present study was to determine if Onmy-UBA poly-
morphism could be used for strain characterisation. This was performed by
lineage-specic PCR investigation of 30 sh, each of the Nikko and Donaldson
strains, and by sequence analysis of 25 of the amplied DNA fragments. Two
new MHC class I lineages were detected in addition to the four previously
described lineages, thus six distinct lineages were observed within the sh
examined (Sal-MHCIa*AF). The distribution of lineages appeared to be
strain-specic. For example, the lineage Sal-MHCIa*A was very common in
the Nikko strain but could not be detected in the Donaldson strain. Analysis
of MHC class I variation may help to elucidate relationships between strains
and the roles of MHC alleles in disease resistance. 2002 Elsevier Science Ltd.
Key words: MHC, polymorphism, lineages, strain, rainbow trout.
I. Introduction
Classical MHC class I molecules are involved in cell surface presentation
of peptides, derived from intracellular proteins, for recognition by T cell
receptors (TCR) of CD8 positive T lymphocytes [1, 2]. When presented non-self
peptides T cells can be stimulated, this results in killing the presenting cell.
This screening mechanism is the primary immune defence against many viral
diseases [3, 4].
One characteristic of classical MHC class I is extensive polymorphism. The
polymorphism results in the presentation of di#erent sets of antigenic peptides
by each allomorph (protein expressed by an allele), and may result in
Present address: Department of Microbiology and Infection, College of Veterinary Medicine,
China Agricultural University, Beijing 100094, P.R. China.
Present address: Department of Veterinary Medicine, College of Bioresource Sciences,
Nihon University, Kameino, Fujisawa, Kanagawa 252-8510, Japan.
*Authors contributed equally to this work.
Corresponding author. E-mail: ototake@a#rc.go.jp
287
10504648/02/040287+15 $35.00/0 2002 Elsevier Science Ltd.
di#erences in disease resistance. Well-characterised correlations between
MHC class I alleles and disease resistance are described for Mareks disease in
chickens [5, 6], and malaria [7] and HIV [8] in humans.
In aquaculture there is extensive interest in the characterisation of unique
strains within di#erent sh species [9, 10]. Characterisation of strains with
respect to their histories and relationships is expected to guide strain
improvement, and knowledge regarding polymorphic genes could aid this
characterisation.
Classical MHC class I is known to be one of the most polymorphic genes
[11], and its polymorphism has been successfully used to characterise popu-
lations of several species including humans [12], horses [13], and cattle [14].
Natural populations of chinook and coho salmon have been distinguished by
their MHC class I polymorphism [1518]. One advantage of MHC class I as a
genetic trait marker is that allelic di#erences may relate to di#erences in
disease resistance.
Previously, it has been shown that there is only one dominant expressed
classical MHC class I locus in rainbow trout, designated Onmy-UBA [19].
Among 14 sh, of which ve were distinct homozygous clones, 10 di#erent
Onmy-UBA alleles were detected [19]. The alleles of the locus were shown to
be classical by conservation of key amino acids, high variability in the 1 and
2 domains and ubiquitous expression. A remarkable variability in the 3
domain was observed, whereas for MHC class I loci in higher vertebrates and
sharks [20] this region is quite well conserved. This may be due to the ancient
character of the Onmy-UBA locus or to di#erent stringencies with respect to
interactions with T lymphocytes. The 10 alleles in rainbow trout were placed
into four lineages (Sal-MHCIa*AD) based on similarity of 1, 2 and 3
domains [19]. MHC class I sequences of other salmonid species could also be
placed into these lineages [19]. In the present study signicant di#erences are
shown in the Onmy-UBA alleles of the Nikko and Donaldson rainbow trout
strains present in Japan.
II. Materials and Methods
FISH
The sh used in this study were randomly selected 30100 g Nikko or
Donaldson strain rainbow trout from the National Research Institute of
Aquaculture (Nikko, Japan). Fish were held in 9 C owing water. Both
strains were maintained by moderate inbreeding using 510 males and 2030
females for annual reproduction, which is a common practice in Japan. The
Nikko strain was imported from the U.S.A. before 1951, but its exact origin is
unclear [21]. The Nikko strain sh have classic rainbow trout characteristics.
The Donaldson strain [22] was imported from the U.S.A. in 1966 [21]. Large
size, rapid growth and high fecundity characterise these sh.
PRIMERS USED
The specicities of the primer combinations used for detection of the
Sal-MHCIa*AF lineages in Nikko and Donaldson strains are detailed below,
288 C. XIA ET AL.
followed by their specicities for other lineages detected by us (in other
strains) or by other research groups.
For the generic amplication of sequences belonging to lineages
Sal-MHCIa*-A, -C, -D, -E, or -F, a primer from the leader peptide encoding
region, pG-LPf, 5-GTATTATCTTGCTGGTGCTGGGAA (forward), was used
in conjunction with a primer from the 3UTR, pG-3UTRr, 5-TTATGTTCTTG
AGAAGTTCCTCTTC (reverse). This primer set was also shown to amplify
rainbow trout sequences belonging to lineages Sal-MHCIa*-G and -H (data not
shown).
For the specic amplication of Sal-MHCIa*A sequences, the primer pG-LPf
was used in conjunction with a primer overlapping the end of the 3 region
and the start of the transmembrane region pI-3/TMr, 5-GGGGACAACGTT
GGCGGCATCCKGG (reverse). Primer pI-3/TMr is specic for sequences
with a type I 3 region. This primer set also amplied sequences from the
lineage Sal-MHCIa*H (data not shown).
For the specic amplication of Sal-MHCIa*B sequences, the primer
pIV-LPf, 5-ATGAAGTCTTTCATCATTKTGCTCC (forward), which binds to
the leader peptide encoding region of all published sequences with a type IV 1
region, was used in conjunction with primer pG-3UTRr. This primer set is
also expected to amplify sequences from the Sal-MHCI*K lineage.
For the specic amplication of Sal-MHCIa*C sequences, the primer pI-1f,
5-CTACACCGSATCTTCTGAAGTTCCCA (forward), binding to type I 1
regions, was used in conjunction with primer p5-2r, 5-AATGTTTATCCCGCT
CAGTCAT (reverse). Primer p5-2r binds specically to the 2 region of the
Sal-MHCIa*C sequences thus far described for rainbow trout.
For the specic amplication of Sal-MHCIa*D sequences, the primer pI-1f
was used in conjunction with a primer overlapping the end of the 3 region
and the start of the transmembrane region pIII-3/TMr, 5-CATTTGAACCCC
TGTTGGTCTTT (reverse). Primer pIII-3/TMr is specic for sequences with a
type III 3 region.
For the specic amplication of Sal-MHCIa*E sequences, the primer
pIII-1f, 5-GCCAAGACTGAGGGGTCTGACTAC (forward), was used in
conjunction with primer pIII-3/TMr.
For the specic amplication of Sal-MHCIa*F sequences, the primer
pVI-1f, 5-TCCAGAAAGCTGAGTGGA TCAGTGG (forward), was used in
conjunction with a primer overlapping the end of the 3 region and the start
of the transmembrane region pII-3/TMr, 5-GGGGCTGGGTCATTCCAGT
TGGTCTT (reverse). Primer pII-3/TMr is specic for sequences with a type II
3 region.
CONDITIONS OF RT-PCR
Total RNA was isolated from whole blood with TRIzol reagent (Gibco BRL,
Life Technologies, Grand Island, U.S.A.) following the manufacturers recom-
mendations. Onmy-UBA expression was analysed by RT-PCR amplication
using a RT-PCR high-PLUS kit (Toyobo, Osaka, Japan). The 25 l RT-PCR
reaction mixtures were set up as suggested by the manufacturer, with 25 mM
Mn(OAc)
2
, 1 M of each primer and 05 g total RNA. The conditions for
MHC CLASS I TYPING OF RAINBOW TROUT STRAINS 289
amplication with the di#erent primer sets were: For the primer combinations
pG-LPf with pG-3UTRr, pG-LPf with pI-3/TMr, and pIV-LPf with
pG-3UTRr: First 60 C for 98 min, then 94 C for 2 min, then 35 cycles of 94 C
for 1 min followed by 55 C for 5 min, and nally 55 C for 7 min. For the primer
combinations pI-1f with p5-2r and pI-1f with pIII-3/TMr: First 60 C for
30 min, then 94 C for 2 min, then 25 cycles of 94 C for 1 min followed by 65 C
for 15 min, and nally 65 C for 7 min. For the primer combinations pIII-1f
with pIII-3/TMr and pVI-1f with pII-3/TMr: First 60 C for 98 min, then
94 C for 2 min, then 35 cycles of 94 C for 1 min, 60 C for 5 min, and nally
60 C for 7 min.
SEQUENCE ANALYSIS
Amplied RT-PCR fragments were cloned into the vector pGEM-T Easy
(Promega Corporation, Wisconsin, U.S.A.). The nucleotide sequences were
determined by the dideoxychain termination method using a CEQ Dye
terminator Cycle Sequencing Kit (Beckman Coulter, Inc., California, U.S.A.)
and suitable primers. Sequence analysis was performed with an automated
sequencer (CEQ 2000 DNA analysis system, Beckman Coulter, Inc.). Compari-
son of deduced amino acid sequences was performed using the programs
Search homology, Multiple alignment, and UPGMA of GENETYX version
10.1 (Software Development Co., Ltd, Tokyo, Japan) computer software.
GENBANK ACCESSION NUMBERS
New sequences described in this report have been deposited in GenBank
under the accession numbers AF318188AF318190.
III. Results
RT-PCR ANALYSIS
In a previous study four di#erent classical MHC class I lineages
Sal-MHCIa*-A, -B, -C and -D were detected in rainbow trout [19]. In the
present study lineage-specic RT-PCR assays were established (see Materials
and Methods). These assays were based on primers that were expected to bind
to all of the rainbow trout MHC class I sequences reported for each respective
lineage. Lineage denitions are discussed below.
Total RNA from the blood of 60 rainbow trout was tested with Sal-MHCIa
lineage-specic primer combinations following reverse transcription. A
number of the RT-PCR fragments (25) were sequenced (see below) and two
additional lineages, Sal-MHCIa*-E and -F, were detected. For these new
lineages specic RT-PCRs were established as well. The RT-PCR results are
summarised in Table 1. The results indicated that in 41 of the 60 sh at least
two di#erent MHC class I sequences were present, and for one Donaldson sh
three di#erent fragments were amplied. The main di#erence between the two
rainbow trout strains was that sequences belonging to lineage A were very
common in the Nikko strain but were not detected in the Donaldson strain. In
290 C. XIA ET AL.
addition, the C lineage was absent in the Nikko strain sh and the E lineage
was absent in the Donaldson strain sh.
In Table 1 the lineages Sal-MHCIa*-G, -H, and -K are indicated between
brackets. These lineages were discovered in other rainbow trout strains
during the course of this study. Sequences belonging to these lineages would
probably be amplied by the RT-PCR systems used, but they were not detected
in the Nikko and Donaldson strains.
SEQUENCE ANALYSIS
The 25 amplied and sequenced RT-PCR fragments derived from 13 sh are
shown in Table 1. For each fragment the complete sequences from at least
three clones were determined to exclude PCR mistakes, the exact number
depending on the quality of the sequence information obtained and whether
the fragment contained one or two Onmy-UBA sequences. The sequences
Table 1. RT-PCR specic for Sal-MHCIa lineages and sequence analysis. Total RNA
samples from 30 Nikko strain and 30 Donaldson strain rainbow trout were investigated
by seven RT-PCR systems: primer pG-LPf with pG-3UTRr [A, C, D, F (G, H)], pG-LPf
with pI-3/TMr [A (H)], pIV-LPf with pG-3UTRr [B (K)], pI-1f with p5-2r (C), pI-1f
with pIII-3/TMr (D), pIII-1f with pIII-3/TMr (E), and pVI-1f with pII-3/TMr (F)
No. of
sh
RT-PCR for Sal-MHCIa lineages:
A, C, D, E, F (G, H) A (H) B (K) C D E F
Nikko
10 1401(A)2
a
+ 4901(F)1
a
6 + + +
4 1401(A)2
b
1401(A)1
b
401(B)1
b
4 + +
3 1401(A)1
c
+ 701(D)1
c
701(D)1
c
1 1401(A) +
1 +
1 0501(E) + +
Donaldson
8 +
8 701(D)2
d
401(B)1
d
701(D)1
d
4 + +
3 + + +
3 4901(F)1 + +
2 501(C)2
e
401(B)1
e
+
1 + +
1 701(D) 401(B) 701(D) 4901(F)
4901(F)
The + symbols or the names of sequenced fragments indicate positive RT-PCR results, whereas
open spaces indicate negative RT-PCR results. The number of sh showing the same PCR pattern
is indicated at the left. When sequencing analysis was performed, this is indicated by the name of
the Onmy-UBA sequence, followed by the Sal-MHCIa lineage to which it belongs (indicated
between parentheses) and the number of sh analysed. The superscripts ae indicate that
sequenced PCR fragments were derived from the same sh; superscripts following a 2 refer to
only one of the two sh for which the amplied fragment was sequenced.
MHC CLASS I TYPING OF RAINBOW TROUT STRAINS 291
detected were Onmy-UBA*-401, -501, -0501, -701, -1401, and -4901. The
sequences Onmy-UBA*-401, -701 and -4901 were present in both the Nikko and
Donaldson strains. Onmy-UBA*-401, -501 and -701 had previously been
detected in other rainbow strains [19]. The sequences Onmy-UBA*-0501 and
-1401 were recently submitted to the GenBank by Shum et al. [23] (GenBank
AF296362 and AF296371). The sequence Onmy-UBA*4901 has not been
previously published. The new sequences were numbered from 4901 down-
wards, since Shum et al. [23] recently submitted Onmy-UBA sequences
numbered through Onmy-UBA*1601 to the GenBank database, and it was
necessary to avoid giving the same designations. It is unfortunate that, more
or less simultaneously, sequences were designated as Onmy-UBA*-101 to -701
[19] while others designated di#erent Onmy-UBA sequences as Onmy-UBA*-
0101 to -0701 [23] (Genbank AF091785, AF296359AF296364).
In Fig. 1 the deduced amino acid sequences of the extracellular domains of
the di#erent alleles detected in the present study are compared, with a
description of features typical of classical MHC class I molecules.
HOMOLOGY COMPARISON
The new sequences obtained were placed into lineages based on similarities
in 1, 2 and 3 domains as done by Aoyagi et al. [19]. Sequences were placed
into the same Sal-MHCIa* lineage when (i) the 1 and 2 domains each had
over 67% amino acid identity with the bulk of the other lineage members and
were grouped together by UPGMA tree analysis, and (ii) the 3 domains had
a similar length.
A simplied form of the UPGMA tree analysis for the 1 and 2 domains, in
which homologous sequences are indicated as groups instead of individual
sequences, is shown in Fig. 2. In addition to salmonid sequences several
cyprinid MHC class I sequences were compared, in order to show conservation
of lineages between evolutionarily distant sh. The gure legend describes
which sequences were compared. Sequences within the 1 and 2 homology
groups, indicated with Roman numerals, share greater than 67% amino acid
identities with the bulk of the other members in the group. With this denition
six di#erent 1 groups and two di#erent 2 groups were established for
rainbow trout. The 1 homology group V was divided into Va and Vb, as the
UPGMA analysis clustered them close together, but the cyprinid sequence
Brre-UAA*01 showed greater than 67% amino acid identity with only a few of
the members of group Va. The 3 region was divided into three di#erent
groups, each of them with a di#erent length of the C-terminus: I has no
extension, II has a short extension, and III a long extension (compare Fig. 1
and Table 2). The C-termini of the 3 domains were based on the expected
intron-exon borders, in comparison with the sequences Onmy-UBA*-401, -501,
and -701 for which these borders had been determined (unpublished data).
The sequences compared in Fig. 2 that encompass the 1, 2 and 3 domains
are schematically compared in Table 2. The Roman numbers in Table 2 refer
to those in Fig. 2. By lineage denition the sequence Onmy-UBA*0501 can
be placed into a new lineage Sal-MHCIa*E, and Onmy-UBA*4901 can be
placed into a new lineage Sal-MHCIa*F. Table 2a shows the Sal-MHCIa
292 C. XIA ET AL.
F
i
g
.
1
.
P
r
e
d
i
c
t
e
d
a
m
i
n
o
a
c
i
d
s
e
q
u
e
n
c
e
s
o
f
t
h
e
e
x
t
r
a
c
e
l
l
u
l
a
r
d
o
m
a
i
n
s
o
f
t
h
e
O
n
m
y
-
U
B
A
*
-
1
4
0
1
,
-
4
0
1
,
-
5
0
1
,
-
7
0
1
,
-
0
5
0
1
,
a
n
d
-
4
9
0
1
m
o
l
e
c
u
l
e
s
d
e
t
e
c
t
e
d
i
n
N
i
k
k
o
a
n
d
D
o
n
a
l
d
s
o
n
s
t
r
a
i
n
r
a
i
n
b
o
w
t
r
o
u
t
.
T
h
e
s
e
q
u
e
n
c
e
s
O
n
m
y
-
U
B
A
*
-
4
0
1
,
-
5
0
1
a
n
d
-
7
0
1
h
a
v
e
b
e
e
n
a
l
r
e
a
d
y
d
e
s
c
r
i
b
e
d
b
y
A
o
y
a
g
i
e
t
a
l
.
[
1
9
]
(
G
e
n
B
a
n
k
A
F
2
8
7
4
8
7
,
A
F
2
8
7
4
8
8
,
A
F
2
8
7
4
9
2
)
a
n
d
t
h
e
s
e
q
u
e
n
c
e
O
n
m
y
-
U
B
A
*
-
0
5
0
1
a
n
d
-
1
4
0
1
b
y
S
h
u
m
e
t
a
l
.
[
2
3
]
(
G
e
n
B
a
n
k
A
F
2
9
6
3
6
2
a
n
d
A
F
2
9
6
3
7
1
)
.
A
l
i
g
n
m
e
n
t
w
a
s
p
e
r
f
o
r
m
e
d
b
y
c
o
m
p
u
t
e
r
s
o
f
t
w
a
r
e
a
n
d
m
o
d
i

e
d
b
y
h
a
n
d
.
N
u
m
b
e
r
i
n
g
o
f
a
m
i
n
o
a
c
i
d
s
w
a
s
p
e
r
f
o
r
m
e
d
f
o
r
t
h
e
O
n
m
y
-
U
B
A
*
1
4
0
1
s
e
q
u
e
n
c
e
,
s
t
a
r
t
i
n
g
f
r
o
m
t
h
e

1
d
o
m
a
i
n
.
D
a
s
h
e
s
i
n
d
i
c
a
t
e
i
d
e
n
t
i
c
a
l
a
m
i
n
o
a
c
i
d
s
;
a
s
t
e
r
i
s
k
s
i
n
d
i
c
a
t
e
g
a
p
s
i
n
t
h
e
s
e
q
u
e
n
c
e
;

l
l
e
d
c
i
r
c
l
e
s
i
n
d
i
c
a
t
e
t
h
e
c
o
n
s
e
r
v
e
d
p
o
s
i
t
i
o
n
s
b
e
l
i
e
v
e
d
t
o
i
n
t
e
r
a
c
t
w
i
t
h
a
n
t
i
g
e
n
i
c
p
e
p
t
i
d
e
t
e
r
m
i
n
i
i
n
m
a
m
m
a
l
s
(
M
a
d
d
e
n
[
3
2
]
)
;
P
i
n
d
i
c
a
t
e
s
p
o
s
i
t
i
o
n
s
b
e
l
i
e
v
e
d
t
o
i
n
t
e
r
a
c
t
w
i
t
h
t
h
e
a
n
t
i
g
e
n
i
c
p
e
p
t
i
d
e
a
n
d
d
e
t
e
r
m
i
n
e
i
t
s
s
p
e
c
i

c
i
t
y
i
n
m
a
m
m
a
l
s
(
M
a
t
s
u
m
u
r
a
e
t
a
l
.
[
3
3
]
)
;
C
i
n
d
i
c
a
t
e
s
c
o
n
s
e
r
v
e
d
c
y
s
t
e
i
n
e
r
e
s
i
d
u
e
s
.
T
h
e
b
o
x
a
t
p
o
s
i
t
i
o
n
8
6
i
n
d
i
c
a
t
e
s
a
c
o
n
s
e
r
v
e
d
N
-
g
l
y
c
o
s
y
l
a
t
i
o
n
s
i
t
e
.
MHC CLASS I TYPING OF RAINBOW TROUT STRAINS 293
III
II
0.248
I
0.372
VI
0.325
Va
0.224
Vb
0.123
IV
0.391
VII
0.473
1
I
0.365
2
II
(3rt; 0.006)
(4rt; 0.012)
(36rt, 6bt 1as, 1ps; 0.145)
(2rt, 2bt, 1ca; 0.145)
(5rt, 1bt; 0.166)
(1zb)
(8rt, 1bt; 0.052)
(1zb)
(38rt, 9bt, 1as, 1ps; 0.217)
(17rt, 1bt, 1ca, 2zb; 0.181)
Fig. 2. UPGMA analysis. The amino acid sequences of the 1 and 2 regions of
published rainbow trout (rt) MHC class Ia molecules and those of several other sh
[brown trout (bt), Atlantic salmon (as), pink salmon (ps), common carp (ca) and
zebrash (zb)] are compared. Amino acid substitutions per site is indicated on the
horizontal bars and is also represented by bar length. Division into 1 and 2 regions
is in agreement with Fig. 1. The clustering of the individual sequences is shown in
a simplied manner by depicting only the homology groups, indicated with Roman
numerals. The number of sequences included in each group, as well as the amino
acid substitutions per site between group members, is indicated between parenth-
eses. Compared are: the rainbow trout sequences Onmy-UBA*101701 by Aoyagi
et al. [19] (GenBank AF287483AF287492), Onmy-UBA*48014901 (this report; Gen-
Bank AF318188AF318190) and Onmy-UBA*01011601 by Shum et al. [23] (GenBank
AF091785, AF296359AF296373, AJ007847). Also the partial rainbow trout
sequences Onmy-A*- and Onmy-UA*1 reported by Miller et al. [17] (GenBank
AF1045234529 and AF1045794582), Onmy-UA*-, Onmy-UAA*-, Onmy-UCA*-, and
Onmy-UBA*- reported by Hansen et al. [30] (GenBank AF0021712179 and
AF1155185528), the brown trout sequences Satr-UBA*01011001 by Shum et al. [23]
(GenBank AF296374AF296383), the Atlantic salmon sequence Sasa-p30 by
Grimholt et al. [34] (GenBank L07606), the pink salmon sequence Ongo-UA-(92H)
by Katagiri et al. [35] (GenBank D58386), the common carp sequence Cyca-UA1*01
by van Erp et al. [31] (GenBank X91015), and the zebrash sequences Brre-UAA*01
and Brre-UBA*01 by Takeuchi et al. [36] (GenBank Z46776 and Z46777).
The homology groups to which the individual sequences belong can be found in
Table 2, except for the single domain sequences of Miller et al. [17] and the
two-domain sequences of Hansen et al. [30]. For the sequences not depicted in Table
2 the following groupings were made: the 1 domain sequences Onmy-A*-1, -2, -3, -4,
-5, -6 and -7 to 1 group I, the 2 domain sequences Onmy-A*-1, -2, and -3 to 2 group
I and Onmy-UA*1 to 2 group II. The 1 domains of Onmy-UA*-b13, -K19, -b4.10 and
Onmy-UCA*KD2.11 belong to 1 group I, the 1 domains of Onmy-UA*A4.10,
Onmy-UAA*KD4.5 and -SP1.3 belong to 1 group II, the 1 domains of Onmy-
UA*A4.3 and Onmy-UCA*KD2.9 belong to 1 group III, the 1 domains of Onmy-
UA*-B3, -K18, -A5.1 and Onmy-UBA*Spu3.1 belong to 1 group IV and the 1
domains of Onmy-UA*A4.11 and Onmy-UAA*KD1.5 belong to 1 group V; whereas
the a2 domains of Onmy-UA*-b13, -K19, -A4.10, -B3, -K18, -A4.11, Onmy-UAA*KD4.5,
-SP1.3, -KD1.5 and Onmy-UBA*Spu3.1 belong to 2 group I, and the 2 domains of
Onmy-UA*b4.10, -A4.3, -A5.1, Onmy-UCA*KD2.9 and -KD2.11 belong to 2 group II.
lineages detected in the Nikko and Donaldson strains in Japan, 2b shows
lineages detected only in other rainbow trout strains, 2c shows brown
trout lineages for which no rainbow trout sequences were found and 2d shows
several cyprinid sequences. The lineages Sal-MHCIa*AK can be distin-
guished when comparing all rainbow trout sequences, and the lineages
Sal-MHCIa*AO when comparing all salmonid sequences.
The 3 domains of the cyprinid sequences are not indicated in Table 2
(despite being known for Cyca-UA*01 and Brre-UBA*01) since they are very
Table 2. Organisation of the full length classical MHC class I sequences detected in
salmonids into lineages
1 2 3
Lineage
Sal-MHCIa*-
(a)
Rainbow trout: Onmy-UAA*KD6, -HC-01,
-OSU-01, Onmy-UBA*101, -102, -201, -301, -0101,
Onmy-UBA*1001, -1101, -1201, -1301, -1401.
Brown trout: Satr-UBA*-0101, -0201, -0301.
Pink salmon: Ongo-UA-(92H)
I I I A
Rainbow trout: Onmy-UBA*-SP3, -401 IV I I B
Rainbow trout: Onmy-UBA*-501, -502, -503, 1501,
-1601. Atlantic salmon: Sasa-p30. Brown trout:
Satr-UBA*0401
I I II C
Rainbow trout: Onmy-UBA*-601, -701, -0301, -0401,
-0601, Onmy-UCA*C32. Brown trout: Satr-UBA*0901
I II III D
Rainbow trout: Onmy-UBA*0501 III II III E
Rainbow trout: Onmy-UBA*4901 VI I II F
(b)
Rainbow trout: Onmy-UBA*-4801, -4802 V II III G
Rainbow trout: Onmy-UBA*0901 II I I H
Rainbow trout: Onmy-UBA*0801 V I I I
Rainbow trout: Onmy-UBA*0701. Brown trout:
Satr-UBA*0701
VI I I J
Rainbow trout: Onmy-UBA*-0201, -0202 IV II III K
(c)
Brown trout: Satr-UBA*0601 I I III L
Brown trout: Satr-UBA*0801 V I II M
Brown trout: Satr-UBA*0501 VI I III N
Brown trout: Satr-UBA*1001 IV I II O
(d)
Zebrash: Brre-UAA*01 V II
Carp: Cyca-UA*01 VI II
Zebrash: Brre-UBA*01 VII II
The Roman numerals for the 1 and 2 domains refer to the homology groups indicated in
Fig. 2. The sequences of the 3 groups I, II and III have no extension at the C-terminus, a short
extension and a long extension respectively. (a) Sal-MHCIa lineages detected in the present study,
(b) Sal-MHCIa lineages detected in other rainbow trout, (c) Sal-MHCIa lineages detected only in
brown trout and (d) cyprinid sequences. The salmonid sequences are organised into lineages
Sal-MHCIa*AO based on having their extracellular domains placed into the same groups. The
sequences detected in this study are underlined.
MHC CLASS I TYPING OF RAINBOW TROUT STRAINS 295
di#erent from the salmonid sequences [19]. Table 2 clearly shows a mosaic
structure of domains, some which are shared between cyprinid and salmonid
MHC class I sequences.
IV. Discussion
The present study investigated MHC class I variability between two
rainbow trout strains. The purpose was to investigate the distribution of
Onmy-UBA alleles between rainbow trout strains maintained in captivity, and
to determine whether Onmy-UBA polymorphism can be used as a genetic
marker.
Characterisation of sh strains in order to understand their historical
relationships and di#erences in properties such as disease resistance has long
been of interest in aquaculture, resulting in the discovery of a variety of
markers including allozymes, DNA-ngerprints and microsatellites [2427].
For rainbow trout specically, DNA nger printing has been established [28]
and a number of microsatellites and allozyme markers have been described
[29].
MHC class I polymorphism has been used previously for the characteris-
ation of salmonid populations by Miller et al. [1518]. They found that for
British Columbia coho and chinook populations, the MHC class I variability
was similar to the variability found in microsatellites [17]. However, their
analysis was based on genomic PCR analysis of single 1 and 2 domains and
no distinction between expressed and non-expressed genes could be made.
Their PCR system led them to the possibly false assumption that genes with
2 group I and II sequences can be expressed from di#erent loci. A problem
when only 1 or 2 domains are analysed is that non-classical MHC class I loci
in rainbow trout have domains very similar to those in some Onmy-UBA
alleles. Besides the Onmy-UBA locus at least two closely related non-classical
MHC class I loci (other than Onmy-UAA [23] which shows only low homology
with Onmy-UBA) can be detected, and the 1 domains of these loci may be
indistinguishable from some Onmy-UBA sequences (unpublished observa-
tions). Therefore, for MHC class I analysis of relatively small populations as
maintained in hatcheries, full length gene analysis is preferable, in order to
avoid unnecessary complexity due to the existence of closely related non-
classical MHC class I sequences. For the screening of large populations, the
method of Miller et al., who rapidly analysed short PCR fragments by
denaturing gradient gel electrophoresis (DGGE), is certainly the better
method of choice; they succeeded in analysing samples from 20 000 coho and
chinook salmon [18].
In the present study the Nikko and Donaldson rainbow trout strains
maintained in Japan were shown to have distinguishing di#erences in MHC
class I sequence lineages. Specically, MHC class I sequences belonging to
lineages Sal-MHCIa*-A and -E were common in the Nikko strain but not in the
Donaldson strain. Although more than 30 classical MHC class I sequences
have been described for rainbow trout, probably representing alleles (see
below), in the Nikko and Donaldson strains only ve and four di#erent
sequences were detected respectively. It must be recognised that these results
296 C. XIA ET AL.
pertain only to these strains as maintained in one hatchery, and that due to
inbreeding practices the Onmy-UBA alleles in sh from other hatcheries may
be di#erent. The variability in MHC class I genes is probably representative
for rainbow trout strains maintained in Japan, since the breeding practice
applied to maintain the strains used in this study is in general use throughout
the country.
The denitions used in this study (also in Aoyagi et al. [19]) for classical
MHC class I lineages in salmonids were based on similarities of 1, 2 and 3
domains. These distinctions seem to represent natural divisions for the 1 and
2 domains, since all reported classical MHC class I sequences of salmonids t
in easily. The 3 region is more problematic, since we could not nd a division
for all salmonid 3 regions in which length, homology and UPGMA analysis
agree, although these parameters match rather well when 3 domains of only
the rainbow trout sequences are compared (not shown). The transmembrane
and cytoplasmic regions are not used in this denition, since natural-
appearing lineages could not be detected and the functional importance of the
variability in this region is questionable. The lineage denition was based on
individual domains, rather than on overall homology, since recombination
events creating di#erent combinations of 1 and 2 groups seem to have been
a major cause of polymorphism. For example, the sequences Onmy-UBA*-101
and -701 have a completely identical 1 region, but a very di#erent 2 region,
with di#erences beginning at the domain border [19]. And for example
Onmy-UBA*601 [19] and Onmy-UA*A51 [30] have an identical 2 domain, but
a very di#erent 1 domain. A probable reason for this pattern is that the intron
between the 1 and 2 domains is very large, which would stimulate recombi-
nation events. This large intron size has not been proven, but is expected
based on the genome analysis of the carp MHC class I sequence Cyca-12 [31]
and the fact that several groups, including ours, have been unsuccessful to
amplify the complete intron at this location in Onmy-UBA.
The lineage denition was kept similar to that in Aoyagi et al. [19] in order
to be consistent, but in the future lineage denitions may need improve-
ment with respect to the 3 domain. It is fascinating, however, that only three
discrete lengths of 3 domains are found for the growing number of salmonid
classical MHC class I sequences. The sequences of Atlantic salmon,
brown trout, pink salmon and rainbow trout compared in Table 2 have 3
domains with C-termini lengths similar to Onmy-UBA*401 (no extension),
Onmy-UBA*501 (6 aa extensions) or Onmy-UBA*701 (11 aa extensions) (Fig. 1)
instead of continuous variation between absent and 11 aa extensions. Future
research is necessary to determine if there is a functional reason for these
discrete length di#erences, since there are presently no studies on other
species that show the same phenomenon.
The primer sets used in this study were developed to distinguish among
sequences of the lineages Sal-MHCIa*AF, but rainbow trout sequences
belonging to the Sal-MHCIa*-G, -H and -K lineages would also have been
amplied (as indicated in the Materials and Methods section); therefore those
lineages are likely to be absent in the Nikko and Donaldson strain rainbow
trout investigated. A few of the primers used in the present study bind
upstream or downstream from the reported rainbow trout sequences of
MHC CLASS I TYPING OF RAINBOW TROUT STRAINS 297
lineages Sal-MHCIa*-I and -J (Table 2), and therefore it cannot be predicted if
they would have been amplied.
Since the lineages denitions do not include precise sequences, and the
classical MHC class I sequences in salmonid species are highly variable,
lineage determination can never be performed by PCR alone. Nevertheless,
the lineage-specic PCR system indicated in Table 1 would amplify most of the
rainbow trout sequences published and give substantive clues regarding their
identity. Exceptions do occur, however, such as the Nikko strain sh for
which the partial Onmy-UBA*1401 sequence could be amplied by the primer
set pG-LPf/pI-3/TMr but not by the primer set pG-LPf/pG-3UTRr (Table 1).
As the forward primer binding to the leader peptide region is identical in both
primer sets, the lack of amplication by the latter set is probably due to a
mutation in the 3UTR (where the reverse primer should bind). Di#erences
in the 3UTR have been previously observed: Onmy-UBA*401 [19] and
Onmy-UBA*SP3 [30] are identical, except that Onmy-UBA*SP3 has a 274 bp
insertion in its 3UTR.
Expression from the rainbow trout locus Onmy-UBA had previously been
shown only for sequences of the lineages Sal-MHCIa*-A, -B, -C, and -D [19].
Since Onmy-UBA is most likely the only classical MHC class I locus in
rainbow trout, we gave the new sequence belonging to the Sal-MHCIa*F
lineage a name starting with Onmy-UBA, and followed the Onmy-UBA
designation of Shum et al. [23] (Genbank) for the detected sequence belong-
ing to the lineage Sal-MHCIa*E. The fact that in this study only one or two
sequences were amplied per individual from 59 sh, despite an intensive
search for presence of di#erent sequences, strongly supports the idea that
all classical MHC class I sequences are derived from the same locus.
However, for one Donaldson strain sh, three sequences belonging to the
lineages Sal-MHCIa*-B, -D, and -F were obtained. This might be due to
triploidy in this individual since it is not consistent with the other results,
but further studies on locus identication of the Sal-MHCIa*F lineage are
necessary. Although the allomorph Onmy-UBA*401 has a somewhat di#erent
1 domain, lacking tyrosine at position 59 (Fig. 1; [30]), locus identication
and expression analysis suggested that it functions as a classical MHC class
I molecule [19].
From the nomenclature for their sequences released to GenBank (Table 2),
it appears that Shum et al. [23] share the view that there is most likely only
one classical MHC class I locus. However, Miller et al. [17] and Hansen
et al. [30] suggested that the di#erent classical MHC class I lineages
might be derived from di#erent loci, and designated sequences accordingly
[17] or state that the designation is based on lineage rather than on locus
[30]. This discussion on loci and lineages is disturbing clear nomenclature
shared by di#erent groups (as seen in Table 2), and should change in the
future.
Despite the confusing nomenclature, all investigators seem to agree on
several important aspects of MHC class I molecules in rainbow trout: that
they are highly variable and that both their structure and the detection of
other genes involved in the MHC class I pathway [30] indicates that the MHC
class I molecules play an important role similar to their counterparts in higher
298 C. XIA ET AL.
vertebrates. This makes research on trout MHC class I genes an important
and promising eld.
In conclusion this study employed a convenient system to analyse
Onmy-UBA polymorphism in rainbow trout. The Nikko and Donaldson
rainbow trout strains appear to have di#erent Onmy-UBA alleles. Future
studies will address relationships between Onmy-UBA alleles and di#erences
in disease resistance.
This study was supported by the promotion of basic research activities for
innovative biosciences funded by Bio-oriented Technology Research Advancement
Institution (BRAIN), Japan, and by postdoctoral fellowships for C. Xia and J. M.
Dijkstra from the Science and Technology Agency of Japan. We thank Dr James D.
Moore, Bodega Marine Laboratory, U.S.A. for his help to edit this manuscript.
References
1 Townsend, A. & Bodmer, H. (1989). Antigen recognition by class I-restricted T
lymphocytes. Annual Review of Immunology 7, 601624.
2 Germain, R. N. (1994). MHC-dependent antigen processing and peptide
presentation: providing ligands for T lymphocyte activation. Cell 76, 287299.
3 Oldstone, M. B. (1997). How viruses escape from cytotoxic T lymphocytes: molecular
parameters and players. Virology 234, 179185.
4 Rodriguez, F., Zhang, J. & Whitton, J. L. (1997). DNA immunization: ubiquitination
of a viral protein enhances cytotoxic T-lymphocyte induction and antiviral
protection but abrogates antibody induction. Journal of Virology 71, 84978503.
5 Pazderka, F., Longenecker, B. M., Law, G. R. J., Stone, H. A. & Ruth, R. F. (1975).
Histocompatibility of chicken populations selected for resistance to Mareks
disease. Immunogenetics 2, 93100.
6 Kaufman, J. & Wallny, H. J. (1996). Chicken MHC molecules, disease resistance and
the evolutionary origin of birds. Current Topics in Microbiology and Immunology
212, 129141.
7 Hill, A. V., Allsopp, C. E., Kwiatkowski, D., Anstey, N. M., Twumasi, P., Rowe,
P. A., Bennett, S., Brewster, D., McMichael, A. J. & Greenwood, B. M. (1991).
Common west African HLA antigens are associated with protection from severe
malaria. Nature 352, 595600.
8 Hendel, H., Caillat-Zucman, S., Lebuanec, H., Carrington, M., OBrien, S., Andrieu,
J. M., Schachter, F., Zagury, D., Rappaport, J., Winkler, C., Nelson, G. W. & Zagury,
J. F. (1999). New class I and II HLA alleles strongly associated with opposite
patterns of progression to AIDS. Journal of Immunology 162, 69426946.
9 Chevassus, B. & Dorson, M. (1990). Genetics of resistance to disease in shes.
Aquaculture 85, 83107.
10 Wiegertjes, G. F. (1995). Immunogenetics of disease resistance in sh. Ph.D. Thesis,
Wageningen Agricultural University, Netherlands.
11 Klein, J. (1987). Origin of major histocompatibility complex polymorphism: the
trans-species hypothesis. Human Immunology 19, 155162.
12 Parham, P. & Ohta, T. (1996). Population biology of antigen presentation by MHC
class I molecules. Science 272, 6774.
13 Horin, P., Cothran, E. G., Trtkova, K., Marti, E., Glasnak, V., Henney, P., Vyskocil,
M. & Lazary, S. (1998). Polymorphism of Old Kladruber horses, a surviving but
endangered baroque breed. European Journal of Immunogenetics 25, 357363.
14 Ellis, S. A., Holmes, E. C., Staines, K. A., Smith, K. B., Stear, M. J., McKeever,
D. J., MacHugh, N. D. & Morrison, W. I. (1999). Variation in the number of
expressed MHC genes in di#erent cattle class I haplotypes. Immunogenetics 50,
319328.
15 Miller, K. M. & Withler, R. E. (1997). Mhc diversity in Pacic salmon: population
structure and trans-species allelism. Hereditas 127, 8395.
MHC CLASS I TYPING OF RAINBOW TROUT STRAINS 299
16 Miller, K. M., Withler, R. E. & Beacham, T. D. (1997). Molecular evolution at Mhc
genes in two populations of chinook salmon Oncorhynchus tshawytscha. Molecular
Ecology 6, 937954.
17 Miller, K. M. & Withler, R. E. (1998). The salmonid class I MHC: limited diversity
in a primitive teleost. Immunological Reviews 166, 279293.
18 Miller, K. M., Ming, T. J., Schulze, A. D. & Withler, R. E. (1999). Denaturing
gradient gel electrophoresis (DGGE): a rapid and sensitive technique to screen
nucleotide sequence variation in populations. Biotechniques 27, 10161030.
19 Aoyagi, K., Dijkstra, J. M., Xia, C., Denda, I., Ototake, M., Hashimoto, K. &
Nakanishi, T. (2002). Classical MHC class I genes composed of highly divergent
sequence lineages share a single locus in rainbow trout (Oncorhynchus mykiss).
Journal of Immunology 168, 260273.
20 Okamura, K., Ototake, M., Nakanishi, T., Kurosawa, Y. & Hashimoto, K. (1997).
The most primitive vertebrates with jaws possess highly polymorphic MHC class I
genes comparable to those of humans. Immunity 7, 777790.
21 Masaoka, T., Muto, K., Oda, S. & Iwata, M. (1997). Origin and characteristics in the
salmonid pedigree extending species at Nikko Branch, National Research Institute
of Aquaculture. NRIA Technical Report 14, 126.
22 Hines, N. O. (1976). Fish of Rare BreedingSalmon and Trout of the Donaldson
Strains. Washington, D.C.: Smithsonian Institute Press.
23 Shum, B. P., Rajalingam, R., Magor, K. E., Azumi, K., Carr, W. H., Dixon, B., Stet,
R. J., Adkison, M. A., Hedrick, R. P. & Parham, P. (1999). A divergent non-classical
class I gene conserved in salmonids. Immunogenetics 49, 479490.
24 Hallerman, E. M. & Beckman, J. S. (1988). DNA-level polymorphism as a tool in
sheries science. Canadian Journal of Fisheries and Aquatic Sciences 45, 10751087.
25 Utter, F. & Ryman, N. (1993). Genetic markers and mixed stock sheries. Fisheries
18, 1121.
26 Taniguchi, N., Takagi, M. & Matsumoto, S. (1997). Genetic evaluation of quantita-
tive and qualitative traits of hatchery stocks for aquaculture in red sea bream.
Bulletin of National Research Institute of Aquaculture, Supplement 3, 3541.
27 Hendry, A. P., Wenburg, J. K., Bentzen, P., Volk, E. C. & Quinn, T. P. (2000). Rapid
evolution of reproductive isolation in the wild: evidence from introduced salmon.
Science 290, 516519.
28 Taggart, J. B. & Ferguson, A. (1990). Minisatellite DNA ngerprints of salmonid
shes. Animal Genetics 21, 377388.
29 Sakamoto, T., Danzmann, R. G., Gharbi, K., Howard, P., Ozaki, A., Khoo, S. K.,
Woram, R. A., Okamoto, N., Ferguson, M. M., Holm, L. E., Guyomard, R. &
Hoyheim, B. (2000). A microsatellite linkage map of rainbow trout (Oncorhynchus
mykiss) characterized by large sex-specic di#erences in recombination rates.
Genetics 155, 13311345.
30 Hansen, J. D., Strassburger, P., Thorgaard, G. H., Young, W. P. & Du Pasquier, L.
(1999). Expression, linkage, and polymorphism of MHC-related genes in rainbow
trout, Oncorhynchus mykiss. Journal of Immunology 163, 774786.
31 van Erp, S. H., Dixon, B., Figueroa, F., Egberts, E. & Stet, R. J. (1996). Identication
and characterization of a new major histocompatibility complex class I gene in carp
(Cyprinus carpio L.). Immunogenetics 44, 4961.
32 Madden, D. R. (1995). The three-dimensional structure of peptide-MHC complexes.
Annual Review of Immunology 13, 587622.
33 Matsumura, M., Fremont, D. H., Peterson, P. A. & Wilson, I. A. (1992). Emerging
principles for the recognition of peptide antigens by MHC class I molecules. Science
257, 927934.
34 Grimholt, U., Hordvik, I., Fosse, V. M., Olsaker, I., Endresen, C. & Lie, O. (1993).
Molecular cloning of major histocompatibility complex class I cDNAs from Atlantic
salmon (Salmo salar). Immunogenetics 37, 469473.
35 Katagiri, T., Hirono, I., Aoki, T. & Sakai, M. (1996). Isolation of major histo-
compatibility complex class I cDNA from pink salmon (Oncorhynchus gorbuscha).
Developmental & Comparative Immunology 20, 217228.
300 C. XIA ET AL.
36 Takeuchi, H., Figueroa, F., OhUigin, C. & Klein, J. (1995). Cloning and characteri-
zation of class I Mhc genes of the zebrash, Brachydanio rerio. Immunogenetics 42,
7784.
MHC CLASS I TYPING OF RAINBOW TROUT STRAINS 301