Download A Novel Mouse Chromosome 17 Hybrid Sterility Locus

Copyright 0 1991 by the Genetics Society of America
A Novel Mouse Chromosome 17 Hybrid Sterility Locus: Implicationsfor the
Origin of t Haplotypes
Stephen H. Pilder, Michael F. Hammer' and Lee M. Silver
Department of Molecular Biology, Princeton University, Princeton,New Jersey 08544-1014
Manuscript received January 2 1, 1991
Accepted for publication May 11, 1991
ABSTRACT
The effects of heterospecific combinations of mouse chromosome 17 on male fertility and transmission ratio were investigated through a series of breeding studies. Animals were bred to carry
complete chromosome 17 homologs, or portions thereof, from three differentsources-Mus domesticus,
Mus spretus and t haplotypes. These chromosome 17 combinations were analyzed for fertility within
the context of a M. domesticus or M . spretus genetic background. Two new forms of hybrid sterility
were identified. First, the heterospecific combination of M . spretus and t haplotype homologs leads to
complete male sterility on both M. spretus and M. domesticus genetic backgrounds. This is an example
of symmetrical hybrid sterility. Second, the presence of a single M. domesticus chromosome 17 homolog
within a M. spretus background causes sterility, however, the same combination of chromosome 17
homologs does not cause sterility within the M.domesticus background. This is a case of asymmetrical
hybrid sterility. Through an analysis of recombinant chromosomes, it was possible to map the M.
domesticus, M. spretus and t haplotype alleles responsible for these two hybrid sterility phenotypes to
the same novel locus (Hybrid sterility-4). Previous structural studies had led to the hypothesis that the
ancestral t haplotype originated through an introgression event from M. spretus or a related species.
If this were true, one might expect that (1) M. spretus homologs would be transmitted at a nonMendelian ratio within the M. domesticus background, and (2) t haplotypes would be transmitted at a
ratio closer to Mendelian within the M . spretus background. The functional data generated in the
current study indicate that neitherof these predictions is fulfilled, and thus, the M.spretus introgression
hypothesis appears to be unlikely.
M
OST naturally occurring populations of the
house mouse (whichinclude the sibling species
Mus musculus, Mus domesticus, as well as others) are
polymorphic for a selfish chromosomal entity known
as a t haplotype. A t haplotype occupies a 20 cM
region at the proximal end of chromosome 17 and it
maintains its structural integrity through a series of
four inversions that block recombination with the
wild-type homolog (Committee for Mouse Chromosome 17 1991; Figure 1A). t Haplotypes are maintained at relatively high levels in natural populations
through the expression of a male-specific phenotype
of transmission ratio distortion (TRD) by which 95%
or more of the offspring from heterozygous +/t males
receive the t-bearing chromosome (SILVER1985). Although TRD provides a powerful selectiveadvantage,
t haplotypeshave not become fixed in the mouse
genome because males homozygous for the t-form of
the chromosome are completely sterile. From mapping studies, it appears likely that this recessive sterility phenotype is a consequence of homozygosity for
the same t genes that are involved in the dominant
TRD phenotype (LYON1986). In addition, most (but
' Current address: Museum of Comparative Zoology,Harvard University,
Cambridge, Massachusetts021 38.
Genetics 149: 237-246 (September, 1991)
not all) naturally occurring t haplotypes carry recessive
embryonic lethal mutations that also counteract TRD
(BENNETT
1975). These three selective forces-TRD,
sterility and lethality-balance each other out giving
rise to equilibrium allele frequencies of5-1 5% in
most populations that have been studied (FICUEROA
et
al. 1988a).
Since their discovery over 60 years ago, the origin
and evolution of these unusual genetic entities has
intrigued many investigators. Until the last decade,
the question of t haplotype derivation hadbeen a
complete enigma. However,with the ability to use
DNA probes to compare t haplotypes withother forms
of chromosome 17, it becamepossible to begin t o
unravel this puzzle in the context of a model shown
in Figure 1B. The crux of this model is as follows:
first, all current-day t haplotypes derive from a single
ancestor; second, t haplotypes diverged apart from
the line leading to the M . domesticus form of chromosome 17 between 1 and 6 million years ago, prior to
the divergence of the various strongly commensal
mouse species from each other, but not prior to the
divergence of the M . domesticus line from the M .
spretus line (DELARBRE
et al. 1988; HAMMER,
SCHIMENTI and SILVER
1989). Finally, once the primordial
238
S. H. Pilder, M. F. Hammer and L. M. Silver
A
48
M.domesticus
119 66 Tcpl Sai2 66 Tu50122 54 Him Piml C y
bf
0:::;
Tai-1 T a l 4
t Haplotype!
B
fixation of
1.
;
T~Y Tai-3
I
I 7
89
bj
I
Tai-5
1 7
p"rima1 inversion
FIGURE1 .-(A) Diagram of three forms of the t region from mouse chromosome 17. The top line represents the M . domesticus t region,
the middle line, the t haplotype form, and the bottom line, the M. spretus form. Boxes represent t region-associated inversions, in(17)I
through in(l7)4, while arrows within the boxes illustrate their relative orientations. Circles at the far left of each line denote thepositions of
the centromeres. The 13 loci displayed above the topmost chromosomal segment have been placed in the order in which they would be
found in the M. domesticus t region. The loci indicated with numbers alone (48, 1 19, 66, 122, 54 and 89) are abbreviations for Dl 7Leh48,
Dl7Lehl19, D17Leh66,D l 7Leh122, D17Leh54and D17Leh89, respectively. Markers for all loci except D17Tu50 have been used in this study.
The position of DI7Tu5O has been shown to denote the distal boundary of the Hst-I hybrid sterility locus. Loci involved in t haplotypeassociated spermatogenic phenotypes (Tcd-I through Tcd-5and Tcr) are placed above the middle chromosomal segment and orderedas they
would be found in t haplotypes. (B) An evolutionary tree depicting the divergence of different forms of chromosome 17 carried by several
modernday species of strongly and weakly commensal mice. The complete range of possible ancestral origins for t haplotypes is denoted by
the shaded box and the question mark.
t chromosome had appearedwith a slight TRD advantage, it is possible to explain its evolution intocurrentday t haplotypes by invoking selective forces that acted
to increase the absolute level of TRD through (1)
internal single genemutations and (2) a series of
inversions that locked together all of the genes responsible for this phenotype (CHARLESWORTH
and
HARTL1978).
One critical question that remains unanswered concerns the natureof the initial set of events that led to
the origin of the primordial t chromosome (represented within the "shaded box" in Figure 1B). In a
previous attemptto better understand the parameters
involved in this problem, Hammer and co-workers
region of chromosome
compared the structure of tthe
17 present in threedifferent species of mice ( M .
domesticus, Mus macedonicus (previously Mus abbotti)
and M . spretus with that of t haplotypes (HAMMER,
SCHIMENTI
and SILVER1989). T h e resultsdemonstrated that the oldest of the four inversion polymorphisms that distinguishes t haplotypes from M . domesticus forms of chromosome 17 [in(l7)2] actually originated within the M . domesticus line and not on the
line leading to t haplotypes (Figure 1 A). Furthermore,
the M . spretus form of the chromosome, like t haplotypes, does not carry this inversion. This finding led
to the proposal of two alternative models to explain
the origin oft haplotypes.
According to the first model, the in(l7)2 inversion
became fixed within the M . domesticus ancestral line,
and subsequently, a noninverted chromosome introgressed into thispopulationfrom
M . spretus (or a
related species in existence at that time). T h e basic
tenet of this model is that at the time of this event,
the M . spretus-like species carried a normal form of
chromosome 17 that could expressmeiotic drive upon
transfer to the M. domesticus geneticenvironment
(HAMMER,
SCHIMENTI
and SILVER1989;SILVER1982,
1985). Precedents have been described for this type
of cryptic male-specific meiotic drive system (CAMERON and MOAV 1957;LOEGERINC
and SEARS1963).
According to the second model, the primordial t
haplotype originated entirelywithin the M . domesticus
line through the stepwise accumulation of mutations
on a normal noninverted form of the chromosome
that acquired an
initial TRD advantage. Subsequently,
Chromosome I 7 Hybrid Sterility
the in(l7)Z inversion arose on awild-type chromosome
within the same population, and was selected for in
+ / t heterozygotes because it reduced recombination
between two or more alleles required for the TRD
phenotype (CHARLESWORTH
and HARTL1978).
In order todistinguish betweenthese models, it was
essential to extend our studies beyond that of chromosomal organization to an analysis of the functional
relationships that exist betweenthe different forms of
chromosome 17. The original rationale of the studies
reported herein was to test certain predictions made
by the M . spretus-origin model. In particular, one
might expect current-day t haplotypes to exhibit a
reduced meiotic drive phenotype when placed within
a putative "species of origin."
In addition, we reasoned
that it might be possible to unmask a cryptic meiotic
drive system within the M . s p e t u s form of chromosome 17 by placing it experimentally intothe M.
domesticus genome. Here we reportthe effects on
fertility and transmission ratio of various heterospecific chromosome I7 combinations within the genetic
backgrounds of M . spretus and M . domesticus.
T o perform this analysis, it was necessary to intercross the two species that normally carried the three
forms of chromosome 17 analyzed. However, crosses
between M . spretus and M . domesticus result in a classic
case of hybrid sterility at the first generation, which
is defined as a situation where two parental forms,
each ofwhichis
fertile interse, produce a hybrid
MARTIN and THALER
which is sterile (BONHOMME,
1972; HALDANE1922). The hybrid sterility phenotype is often confined to the "heterogametic" sex, as
is the case here. Two major loci, Hst-2 and Hst-3,
appear to be involved in thisexample of hybrid sterility, and after threegenerations of backcrossingto M .
domesticus (through females), most males express full
fertility (BONHOMME,
GUENETand CATALAN
1982;
et al. 1990). Thus, toview the possible effects
GUENET
on sperm function of heterospecific forms of chromosome I7 in isolation from other genic effects, it
was necessary to backcross the heterospecific chromosome into each genetic background for at least
four generations.
The results obtained indicate that present-day t haplotypes are not functionally related to present-day chromosome I7 representatives from the M . spretus species. Thus, theM . spretus origin model for t haplotypes
is rendered unlikely. However, these studies have led
to the identification and genetic characterization of a
novel hybridsterility locus witha complex phenotypic
expression.
MATERIALS AND METHODS
Nomenclature: The species of mice used inthis study are
et al. 1991;
named according to standard practice (AUFFRAY
SAGE 1981). The sibling species M . domesticus and M . musculus are considered subspecies (M. m. domesticus and M. m.
239
musculus respectively) by some investigators (AUFFRAY
et al.
1991). The genotypes of all hybrid mice used in this study
are described by a notation which takes the following form:
X. [Y/Z]. The first letter, outside the brackets, is indicative
of the genetic background of the animal. Letters inside the
brackets, separated by a "/," symbolize the chromosome I7
homologs. D represents M. domesticus, S represents M. spretus, and t represents a t-form of chromosome 17. For example, D[S/t] represents an animal with a M. domesticus
genetic background harboring a M . spretus form of chromosome 17 and a complete t haplotype. The notation X
Y represents a recombinant chromosome I7 with a proximal
X-derived region and a distal Yderived region. For example, t t, D denotes a proximal partial t haplotype that
originated within a M . domesticus mouse, D c-, t denotes a
distal partial t haplotype, and S c* D and D c-, S denote
structurally reciprocal forms of recombinant M. spretus-M.
domesticus chromosome 17. Each unique class of recombinant M . spretus-M. domesticus chromosomes used in thisstudy
is named according to thealleles present at each of six DNA
Crya-I and
loci (DI7Leh66,DI7Leh54,Hba-ps4,Pim-I,
DI7Leh89) that were typed as shownin Figure 2. For
example, SSDDDD denotes a recombinant chromosome I 7
with M. spretus alleles at DI7Leh66 and DI7Leh54, and M.
domesticus alleles at Hba-ps4, Pim-I, Crya-I and D17Leh89.
When independent recombinant chromosomes were recovered with the same set of typed alleles, they were distinguished by numbers as indicated in Table 1.
Mice and crosses: All breeding experiments were conductedat Princeton Universityin accordance with NIH
regulations and guidelines. All t haplotypes have been maintained in the colony of L. SILVERat Princeton University.
Outbred CD1 mice were obtained from Charles River (Wilmington, Massachusetts). M. spretus animals originally collected from Cadiz, Spain, were obtained from M. POTTER
(Bethesda, Maryland).
M . domesticus females heterozygous for the complete t
haplotypes P5,Pb7,
tTUwz4were crossed to M . spretus males
to produce twotypesof F1 hybrids for chromosome I7
(D/S and t/S). t/S F1 females were crossed to M. spretus
males while D/S FI females were crossed to eitherM. spretus
or M. domesticus males to produce N2 backcross generation
offspring. Further backcrossing ofthe same type was carried
out through theN4 generation. N5 and successive backcross
generation D[S/D] offspring were derived by crossing to
either sex. Further backcrossing to produce S[S/D] offspring
was accomplished only through the maternal line. N5 and
N6D[S/t]males were produced by crossing N4 and N5
D[S/D] males to homozygous D[t"'/t"'] females maintained
routinely in our colony. N5, N6 and N7 generation D[S/t
c-, D] and D[S/D t, t] maleswere produced in the first
instance by crossing D[S/D] malesand females of the appropriate generation to animals that carried the desired partial
t haplotype. These same genotypes were also produced by
intercrossing. Males harboring recombinant S c-, D or D c-,
S chromosomes in conjunction with a variety of other chromosome 17 homologs were produced as described above
for all genotypes.
Definition of sterility: Males to be tested for sterility
were placed with at least four females of breeding age for
at least one month per pair of females. T o be classified as
sterile, a male had to produce vaginal plugs in at least two
different females without giving rise to any progeny. This
definition is absolute-the birth of a single offspring indicates
that the male parent is nonsterile.
Assay for chromosome 27 genotype: All experimental
animals were tested with eleven informative markers that
identify restriction fragment length polymorphisms (RFLPs)
-
240
S. H. Pilder, M. F. Hammer
and
L. M. Silver
DSt DSt DSt DSt DSt DSt DSt DSt DSt DSt DSt
7-
48
119
Tcpl Sod2 66
122
54
Hba
Piml
Cryal
89
FIGURE
2.-Southern analysis of RFLPs between different forms of chromosome 17. Each panel of three lanes represents a Southern
hybridization with one of eleven molecular probes to 10 gm per lane of blotted high molecular weight DNA digested with the restriction
endonuclease, Ta91. The letters D, S and t above each lane represent domesticus, spretus and t haplotype samples. Hybridiiration panels are
presented from left to right according to the order of corresponding loci along the chromosome. The loci are indicated below each panel
and areabbreviated as indicated in the legend to Figure 1. Clones used to identify RFLPs at these loci are indicated in the same order: T u 4 8
(FOX et al. 1985), Tu1 19 (HERRMANN
et al. 1986). 29x (WILLISON,
DUDLEYand POTTER1986), Sod2 (FIGUEROA
et al. 1988b). Cg3-100
(SCHIMENTI
et al. 1987). T u 1 22 (Fox et al. 1985), 54M ( B ~ C AetN al. 1987), Hba4 (Fox, SILVER
and MARTIN 1984), Pim 1A (NADEAU
and
PHILLIPS
1987), Crya-10 (KING, SHINOHARA
and PIATICORSKY
1982), and Tu89 ( B ~ C AetNal. 1987).
TABLE 1
Summary of breeding results
Total
tested
17-1
fmpl'u
S
tyz
D
1-2
t6
Tf6
Tr'
tr~3z
P
r'
P
r5
TP
PZ
T P
T P
T P
D
rZ
D
D
S
S
S1
S1
S
1
generation
17-2 Backcross
No. Fertile
Background
S
S
N4 (8)
S
S
control N4 ( 2 ) sibling
2
S
D
N 5 (1 2). N6 (2)
D
control N5 ( 3 )sibling
S
4
D
N6 (4)
S
D
7
N5 (6), N6 (1)
S
D
N5 (2), N6 (4), N7 (7)
S
D
N5 (1). N6 (7)
S
N6
D (I),
N5
(4). N7 (3)
S
D
N7 (6)
SSSDDD #1
1 D
N5 (1)
SSSDDD #1
D
N5 (1)
SSSDDD # I
D
N5 (2)
SSSDDD #2
1 D
N 5 (1)
SSSDDD #3
1 D
N 5 (1)
SSSDDD #4
1 D
N6 (1)
SSSDDD #5
D
N 8 (1)
SSDDDD
1
N 5 (1)
S
D
N4 (7), N5 (3). N6 (3). N7 (3)
S
S
17 N4 (1 6), N5 (1)
S
S
control N4 (6) sibling
1SSSDDD #6
1 S
N4 (1)
SSSSDD
(1)
S
N4
DDSSSS
(1)
S
N4
DSSSSS
S
N4 (1)
1
8
2
14
3
13
8
8
6
1
2
1
16
6
0
0
3
0
0
0
5
5
5
0
0
0
0
0
0
1
12
0
6
1
1
Columns 17-1 and 17-2 indicate the two forms of chromosome 17 present in a particular genotype within the background indicated in the
third column. S represents M . spretus, and D represents M . domesticus. Partial and complete t haplotypes are discussed in MATERIALS AND
METHODS. Classes of M . spretus-M.domesticus recombinant chromosomes are named as described i n MATERIALS AND METHODS. The number
of mice tested at each backcross generation is indicated in parentheses.
Chromosome 17 Hybrid Sterility
among the threeforms of chromosome 17 under analysis in
this report".
spretus, M. domesticus and t haplotypes. Markers and corresponding RFLPs are shown in Figure 2. High
molecular weight DNA, prepared from tail clippings ( H e
GAN,COSTANTINIand LACY1986), was cut to completion
with TaqI restriction endonuclease (New England Biolabs,
Beverly, Massachusetts), electrophoresed and blotted onto
nylon membranes (Genescreen, New England Nuclear, Boston, Massachusetts)according to themanufacturer's instructions. DNA was bound to the membrane by UV light and
hybridized according totheprocedure
of CHURCH
4 and
GILBERT(1984). Radioactive probes were produced by polymerization from a mixture of random oligonucleotides on
templates of denatured DNA (FEINBERG
and VOGEISTEIN
1984). Membranes were routinely stripped and reprobed
multiple times according to the proceduredescribed by the
manufacturer.
RESULTS
t Haplotypes cause male sterility when placed
the M. spretus genome: One of the predictions of the
M. spretus-origin model of t haplotypes is that these
chromosomes might not distort transmission ratios as
drastically ina heterozygous combination with the M .
spretus form of the chromosome as they do normally
with the M. domesticus chromosome, since they would
retain residual functional homology with a putative
chromosome of origin. T o examine this prediction,
we generated males that were heterozygous for a
complete t haplotype within a M . spretus background
(S. [S/t]) at the N4 generation of backcrossing, and
assessed their fertility according to the protocols described in the MATERIALS AND METHODS. All eight
males tested proved to be sterile, whereas the two
backcross sibling males tested without a t haplotype
were found to be fertile (Table 1). Thus, it appears
that one or more genes within or closely linked to t
haplotypes is causing male sterility within M . spretus
animals. The prediction of an altered transmission
ratio obviously cannot be tested within the context of
this genotype.
The M. spretus form of chromosome 17 does not
M.
distort transmission ratios when placed into the
domesticus genome: The second prediction of the M .
spretus-origin modelfor t haplotypes is that thenormal
M . spretus form of chromosome 17 might act to distort
transmission ratios when it is placed experimentally
into theM . domesticus genome. T o test this prediction,
we backcrossed the M. spretus form of chromosome
17 into the M . domesticus genome. At each generation,
offspring with an intact M . spretus chromosome across
the complete length of the t region (from D17Leh48
to Dl7Leh89) were identified for further breeding
(Figure 1A). Of the 16 males of the D - [D/S] typethat
were analyzed from the N4 through N7 generations,
12 were found to be fertile. When eight of these
fertile maleswere bred todeterminetheratio
of
transmission to their offspring of each chromosome
17 homolog, no significant distortion from 50% was
24 1
TABLE 2
M. dontesticus/M. spretus chromosome 17 transmission ratios for
D [S/D] males
.
Backcross
generation
Male No.
domesticus
offspring
spretus
domesfinrsjspretus
ratio
offspring
N5
48
5
6
7
26
15
52
N6
14
1.11
1.02
1.22
1.22
0.79
0.5
1.07
N6
8
49
27
1.81
1-8
252
253
0.996
N4
N4
N4
N4
N5
Total
1
10
2
3
48
9
47
22
18
44
38
36
observed (Table 2). This result implies that the present-day M . spretus chromosome 17 is functionally unintorelated to the primordial t haplotype. A caveat to this
interpretation is that a small distortion of transmission
ratio could go undetected in laboratory studies, but
nevertheless, have a profound effect in natural populations.
The M. spretus/t haplotype chromosome 17 combination causes hybrid sterility irrespective of genetic background: Two general genetic models can
be proposed to account for the sterility phenotype
caused by t haplotypes in a M. spretus background.
The first model positsa direct incompatibility between
the M . spretus and t alleles at a particular chromosome
17 locus that results insterility irrespective of the
genetic background of the animal. According to this
model, the chromosome 17 gene(s)involvedwould
represent a classic symmetrical hybrid sterility locus.
The second model posits an incompatibility between
one or more t haplotype gene(s) andother genes
present in the M . spretus genetic background. According to this model, the guilty t haplotype gene(s) would
be acting as a true dominant mutation only withinthe
context of the M . spretus genetic background, giving
rise to an asymmetrical hybrid sterility phenotype.
In order to distinguish between these two models,
we generated animals that carried the same chromosome 17 genotype as that described above (one homolog from M . spretus and a t haplotype) but within the
M . domesticus genetic background (D.[S/t]). If the
dominant mutation explanation of the original sterility phenotype were correct, these new males wouldbe
fertile. If this were the case, it would be possible to
investigate whether t haplotypescould continue to
distort transmission ratios against non-M. domesticus
forms of chromosome 17. On the other hand, if the
symmetrical hybrid sterility explanation were correct,
these new males would be sterile, and once again, it
would not be possible to obtain a value for TRD.
Fourteen D-[S/t] males were tested for fertility. All
14 proved to be sterile, whereas allthree sibling males
242
Pilder,
S. H.
M. F. Hammer
and
tested with a D. [D/t] genotype
showed fertility (Table
1). Thus, the M . spretuslt haplotype hybrid sterility
phenotype is independent of genetic background.
One or more M. domesticus chromosome 17 loci
acts as a dominant male
sterile mutation within the
M.spretus genome: To further investigate the nature
of chromosome 17 effects on fertility in animals with
hybrid genotypes, we backcrossed the M . domesticus
form of chromosome 17 into the M . spretus genome
in order toobtain S .[D/S] males. At each generation,
offspring with anintact M . domesticus chromosome
across thecompletelength
of the tregion(from
D17Leh48 to D l 7 L e h 8 9 )were identified for further
breeding. Unexpectedly, all 17 males of the S. [D/S]
type proved to be sterile, whereasall six sibling males
not carrying a M . domesticus chromosome 17 (S [S/S]
) were found to be fertile (Table 1). When these data
are viewed in conjunction with the observation described above that D[D/S] malesare fertile, it is clear
that the hybrid chromosome I7 combination of M .
domesticus and M . spretus is not in itself causing sterility. Rather, it appears that oneor more loci on theM .
domesticus form of chromosome I7 can act as a dominant male sterile mutation within the context of the
M. spretus genome. This hybrid sterility phenotype is
termed asymmetrical in contrast to the symmetrical
hybrid sterility result obtained with the M . spretuslt
haplotype chromosome 17 combination.
Mapping of a t haplotype locus responsible for
hybridsterility: We reasoned that since D. [D/S]
males are fertile, it should be possible to map the t
allele responsible for the sterility of D.[t/S] males by
replacing defined portions of complete t haplotypes
with M . domesticus DNA to produce D. [D c,t/S] and
D.[t c* D/S] animals. In practice, this can be accomplished through breeding protocols that make use of
existing stocks of mice carrying different recombinant
[ t c, M . domesticus] chromosomes known as partial t
haplotypes that have resultedfrom rare crossover
events in De [D/t] animals. Partial t haplotypes are
classified as "proximal" if they retain a proximal portion of the complete t haplotype from which they are
derived with a M . domesticus distal region [t c, Dl.
"Distal" partial t haplotypes retain only a distal portion
of a complete t haplotype with a proximal region from
M . domesticus [D c,t].
Six different partial t haplotypes were used in this
experiment. The extent of t-DNA present in each is
shown in Figure 3. Fertility was observed in D. [t c,
D/S] males that carried any one of three different
proximal partial t haplotypes (tTuu32,th2, t')(Table 1).
No significant difference was observed between the
fraction of these males that are fertile and thefraction
of maleswith a control D-[S/D] genotype that are
fertile. In contrast, all of the D.[D c,t/S] males that
carried any one of three distal partial t haplotypes (t6,
-
-
L. M. Silver
c"---f Region
spretus
-
48SodTcp1196612251Hb.PhnDY.89~DNAy.rLnr
7-
? S l h v w . l w
rhdhwMh
5/8
518
48SodTcp1196612251HbaplmcrY.89
516
f3
014
f@
4 6 6 1 1 9 T ~ ~ s o d ~ 1 ~ ~ ~ , , ~ ~ 8 , ~
Oi7
0113
48661l#T~Sod6612289crY.pimHb.Y
FIGURE3.-Mapping the Hst-4' allele. Male mice with a M. domesticus background having one chromosome 17 homolog from M.
spretus (black box at top) and the other with one of six partial 1
haplotypes (mosaic boxes below the line) were tested for fertility as
indicated in the text. The shaded areas represent the extent of t
chromatin; the remaining portion of these chromosomes is derived
from M. domesticus. The small boxes below the M. spretus chromosome represent inverted regions in(17)J through in(17)4, and the
marker loci employed in the genetic analysis are shown above each
chromosome. The fertility of each genotypic class is presented as a
fraction of the males tested. The Hst-4' map position is indicated by
the solid bar at the bottom of the figure.
f6,th") were completely sterile (Table 1). These data
allow the localization of the t haplotype gene($ responsible for sterility to the distal in(l7)4 region defined by the markersD l 7Leh54 and DI7Leh89.
Mapping of the M. spretus hybrid sterility locus:
To map theM . spretus allele responsible for the sterility of D [t/S] males, we took advantageof the obvious
fact that D [t/D] males are fertile. Thus, it is possible
to replace portionsof the M . spretus chromosome with
M . domesticus DNA in D [t/D c* SI and D [t/S c* Dl
males and test for the restoration of fertility. Recombinant formsof chromosome 17 containing M . spretusand M . domesticus-derived portionsoccurred in the
offspring from a number of crosses described in this
report. These chromosomes were characterized at 11
molecular loci that span the t region (Figure 4). Further crosses were performed to obtain animals that
carried each recombinant D c, S or S t, D chromosome in conjunction with a t haplotype on a M . domesticus background. Since distal partial t haplotypes appear indistinguishable from complete t haplotypes in
the expression of the D.[t/S] sterility phenotype, we
used them interchangeablywith complete t haplotypes
in this experiment (data obtained with each are reported separately in Table 1).
Fertility was observed in a male thatcarrieda
recombinant chromosome with a breakpoint between
Dl7Leh54 and Hba-ps4 (D-[t/SSDDDD]) (Figure 4).
Since the sterility phenotypeexpressed by D-[t/S]
animals is absolute, the recovery of single fertile De
[t/D c,SI or D.[t/S t,Dl male of a particular genotype
-
-
-
-
Region
-t
SSSDDD#lSSSDDD#2
-
48
SSSSDD
No. of Fertlle
Males With
96
hlt
Tc 1 1 9 6 6 1 2 2 Y H b a P l m C m 8 9
011 n,d,
48SodTc 1196612254HbsPlmQywBB
SSSDDD#5
SSDDDD
t i
1 012 0l2
48Sod Tc 1196612254HbaPImcN.89
SSSDDD#3 I
)
SSSDDDM
Chromosome 17 Hybrid Sterility
t
n.d.
n.d.
n.d. n.d. 0,1
48SodTc 1 1 9 6 6 1 2 2 5 4 H b s P l m 9 y ~ 8 9
48SodTc
j n.d. n.d. 011
1196612254HbaPlmCm89
-
48SodTc 1 1 9 6 6 1 2 2 5 4 H b s P l m u Y . 8 9
n.d. n.d.
1111 n.d.
n.d.
FIGURE4.-Mapping the Hst-4' allele. Male micewith a M .
domesticus background. carrying either a complete t haplotype or
one of two disral partial t haplotypes (indicated at the top of the
figure with mosaic boxes) and one of six M. sprefus-M. domesticus
recombinant t regions (indicated belowwithmosaic boxes) were
tested for fertility as indicated in the text. Black portions of mosaic
boxes represent M.spretus DNA, white portionsrepresent M .
domesticus DNA, and shaded portions represent t haplotype DNA.
Marker loci used in the analysis of each male are shown above each
chromosome. Fertility values are indicated for each combination of
M . spretus-M.domesficus and f haplotype homologs. n.d. indicates
not done. The large solid bar at the bottom of the figure represents
the Hst-4' map position from Figure 3, while the smallsolid bar
indicates the Hst-4' map position.
is a significant outcome and can be used to indicate
the presence of the M . domesticus allele rather than
the M . spretus allele at thehybrid sterility locus. Thus,
the M . spretus allele responsible for the sterility of De
[t/S] males must map distal to DI7Leh54.
Five independent SSSDDD recombinant chromosomes were bred into a D.[t/SSSDDD] genotype. In
total, eight males were analyzed that carried one of
these five chromosomes. Seven males representing
four of the breakpoints proved to be sterile, whereas
one male representing the fifth breakpoint was fertile
(Figure 4). This result further defines the proximal
boundary of the sterility locus distal to the recombinationbreakpointpresent
in the one fertileD. [t/
SSSDDD] male (distal to Hba-ps4).
Although the fertility of a particular genotype can
be demonstrated with the birth of a single offspring,
the sterility of a particular genotype cannot be demonstrated unless data from enough animals are accumulated to be significantly differentfromthe
expected fraction of fertile males observed with control
genotypes. Consequently, the data from this class of
recombinants do not allow the determination of a
definitive distal boundary for this hybrid sterility lo-
DDSSSS
243
48SodTcp1196612254HbsPImC~
4866119TcpSod6612254HbaPlm
-
BB
1I1
1I1
1I1
FIGURE 5.-Mapping
the Hst-4' ;~lIc'Ic.\l;lle micewith a M .
sprefusbackground h;uhoring one complete M. spretuschromosome
I7 homolog in conjunction with one of four M . spretus-M. domesficus
recombinant t regions were tested for fertility as indicated in the
text. The black boxes represent M . spretus DNA. Fertility values
are presented a s a fraction of the total number of males tested.
Marker loci employed i n the genetic analysis of eachmale are shown
above each chromosome. The Hst-4' map position is indicated by
the largest of the three solid bars at the bottom of the figure, the
Hst-4' map position by the smallest of the three bars. and the Ifst4d map position by the middle sized bar.
cus. However, the results just described can be used
to circumscribe the most probable region harboring
the locus. First, only one of the eight males from the
entire D.[t/SSSDDD] class was fertile. This fraction
(1/8) is significantly different from the lowest fertility
value obtained for any other fertile genome analyzed
(5/8) when the chi-square test,correctedfor small
sample size, is employed ( P < 0.0 15). Second, when
the fertile recombinant chromosomeis removed from
the remaining chromosome members
of this class, the
difference from other fertile genotypes is significant
with a P value of less than 0.005. Finally, four males
that carry one particular recombinant chromosome
from this class (SSSDDD#l) were all determined to
be sterile, yielding a P value of less than 0.05. Taken
together, these results strongly suggest that the M .
spretus sterility locus lies between Hba-ps4 and Pim-1.
Mapping of the M. domesticus hybrid sterility locus: T h e M . domesticus allele responsible for the sterility of S.[D/S] males was mapped according to the
samegeneralprotocol usedin themapping studies
described above. T w o recombinant [D * SI chromosomes (DSSSSS and DDSSSS) were recovered and
bred into the M . spretus genetic background to produce S. [D * S/S] males that were tested and found
to be fertile (Figure 5).
This result indicates a proximal
boundary for the M. domesticus sterility locus distal to
Dl7Leh54. Recombinant chromosomes of the [S t,
Dl type with breakpoints between Hba-ps4 and Pim-1
in one case, and between Pim-I and Crya-I in the
second case, were also found to produce fertile S-[S
* D/S] males. Thus the M. domesticus sterility locus
must map between D l 7Leh54 and Pim-I.
DISCUSSION
Present-day samples of M. spretus chromosome I7
do not express functional properties characteristic
244
Pilder,
S. H.
M. F. Hammer
and
of presentday t haplotypes: Comparative studies of
chromosome structure have led to the formulation of
two models for the origin of t haplotypes (HAMMER,
SCHIMENTI
and SILVER1989).According
toone
model, t haplotypes originated through theintrogression of a normal M. spretus (or M. spretus-like) chromosome I7 into an ancestral M. domesticus population.
T h e major assumption underlying this model is that
although the M. spretus chromosome I7 would have
functioned normally within its host species, it might
have carried uniquealleles that allowed a preferential
transmission relative to thenormal M. domesticus chromosome in heterozygous animals. A precedent of this
type has been described in tobacco (CAMERONand
MOAV 1957), and in further support of this model,
the current M. spretus range overlaps that of M. domesticus, and occasional formation ofhybrids between
these two species has been observed in the wild (F.
personal communication).
BONHOMME,
To determine whetherpresent-day t haplotypes and
M . spretus forms of chromosome I7 might continue
to have the original functional relationship predicted
by this model, a series of genetic studies were carried
out. In the first experiment, aM . spretus chromosome
I 7 was bredintoa M. domesticus background and
males heterozygous for this chromosome were tested
for transmission ratios. The results demonstrate the
equal transmission of the two chromosome I7 homologs, ruling out the possibility that the present-day
wild-type M. spretus chromosomecarriesacryptic
meiotic drive system.
The second experiment was designed to determine
whether t haplotypes would lose their meiotic drive
activity when they were placed "back" into aM. spretus
genome, as might be expected if these two forms of
the chromosome were closely related. With the demonstration that t haplotype/M. spretus combinations of
chromosome 17 caused hybrid sterility, it was not
possible to test this hypothesis directly. However, one
conclusion can bedrawnfrom
this result-it is extremely unlikely that complete t haplotypes would be
found as a polymorphism within present day populations of M. spretus because chromosomescarrying
sterility factors tend to be lost from populations.
Finally, the transmission results obtained with a
male that carries a third heterospecific combination
of chromosome I 7 homologs sheds further light on
the relative functional properties of M . spretus alleles
at four of the genes involved in the M . domesticuslt
haplotype TRD phenotype. T h e informative male
carries a completet haplotype (t"")
and a recombinant
[S c* Dl homolog (SSDDDD in Table 1) with M.
spretus alleles at the Tcr, Tcd-I, Tcd-3 and Tcd-4 loci
involved in the expression of the TRD phenotype by
t haplotypes. This male is fertile and expresses a
transmission ratio of 100% in favor of the t-carrying
L. M. Silver
chromosome. If the M . spretus alleles at the threeTcd
loci functioned like present-day t haplotype alleles,
one would expect male sterility (LYON 1986). If the
M. spretus allele at Tcr functioned like a t allele, then
this genotype would be effectively homozygous at the
Tcr locus and meiotic drive activity would be suppressed (LYONand MASON1977).
This last result provides further evidence for the
lack of a functional relationship between present-day
M. spretus and t haplotype alleles at loci involved in
the TRD phenotype. Rather, the M. spretus alleles at
Tcr, Tcd-I, Tcd-3and Tcd-4 appear to be functioning
in a manner expected of M. domesticus alleles at these
loci. In conclusion, the accumulated results described
in this report argue against the likelihood that t haplotypes were derived from aM. spretus form of chromosome 17. Nevertheless, this possibility cannot be
ruled out completely because all traces of a relationship between the two chromosomes could have been
eliminated through a long periodof divergence. Furthermore, as discussed earlier, a small distortion of
transmission ratio may have gone undetected in our
studies.
The hybrid sterility-1 (Hst-I) locus is not responsibleforthe M. spretuslM. domesticuslt haplotype
hybrid sterility phenotype: T o date, three other hybrid sterility loci (Hst-I, Hst-2, Hst-3) have been deGUENET
scribed in the mouse genome (BONHOMME,
and CATALAN1982; FOREJTand IVANYI1975; GUENET et al. 1990). One of these loci-Hst-l-maps to
chromosome 17 and is responsible for a classic hybrid
sterility phenotype in males that carry certain heterospecific combinations of chromosome 17 from the
sibling species M. domesticus and M. musculus (FOREJT
and IVANYI 1975). Recently obtained genetic data
have allowed the high resolution mapping of Hst-1
between Sod-2 and Dl 7Tu50 proximal to the in(l7)4
inversion (FOREJTet al. 1991) (Figure 1A).
Since the hybrid sterility phenotypes described in
this report are also caused by a gene(s) mapping to
chromosome 17, it was importanttodetermine
whether the Hst-I locus was involved as well. TWO
independent lines of evidence indicate that this is not
the case. First, the loci involved in the M. spretus/M.
domesticuslt haplotype phenotypes have been mapped
definitively to a region distal to the D17Tu50 IOCUS.
Second, preliminary studies indicate that the physiological basisfor thesterility phenotypes describedhere
is different from that caused by Hst-I (S. H. PILDER
and P. OLDS-CLARKE,
unpublished data). The Hst-1
phenotype is associated with a significant reduction in
testes weight and the absence of epididymal sperm
(FOREJTand IVANYI 1975). Incontrast,thesterile
males reported in our study have normal testes
weights and display no abnormalities in sperm count.
Thus, we have identified anovel locus, Hybrid sterility-
Chromosome 17 Hybrid Sterility
4 (Hst-4),which is responsible for the sterility of males
that carry heterospecific combinationsof chromosome
17 with one M . spretus homolog andeithera
M.
domesticus or t haplotype homolog.
The M. spretus, M. domesticus and t haplotype
genes responsible for hybrid sterility are likely to
represent alternatealleles at the same locus:
In three
separate sets ofexperiments, the hybrid sterility genes
of M. spretus, M . domesticus, and t haplotypes were
mapped to overlapping intervals in the distal portion
of the t complex on chromosome 17. The highest
resolution mapping was accomplished with M . spretus
to the 3-cM region between Hba-ps4 and Pim-1. The
next highest resolution was achieved with M. domesticus to the 5-cM region between D l 7Leh.54 and Pim-1.
Finally, the least resolution was obtained with t haplotypes to the 9-cM region between D17Leh122 and
DI7Leh89. The simplest interpretation of these data
is that a single hybrid sterility locus (Hst-4) is entirely
responsible for all forms of hybrid sterility described
in this report. Thislocus would map betweenHba-ps4
and Pim-1, and would havethree alternatealleles, Hst4d in M . domesticus, Hst-4" in M . spretus, and Hst-4' in
t haplotypes. More complex scenarios, however, cannot be ruled out at thepresent time and it is possible
thatthe sterility phenotypes result from dominant
interactions between closely linkedbut multiple genes
present within or among the three different chromosometypes (DUTCHER
and LUX 1989; SANO1990).
Notwithstanding, until proven to thecontrary, we will
assume that a single Hst-4 locus is responsible for all
of the sterility phenotypes described herein.
Speculation on the functional basis for the Hst-4
hybrid sterility phenotypes: Two types of Hst-4-associated hybrid sterility phenotypes havebeen described in this report. In the first, the Hst-4*/Hst-4"
genotype causes sterility in a background-independent
fashion. This result can be explained most simply by
an incompatibility between the M . spretus and t haplotype products of the Hst-4 gene. In the second form
of hybrid sterility, the H~t-4~/Hst-4"
genotype causes
sterility onlyin the M . spretus genetic background,
and not in the M. domesticus background. This result
cannot be explained by allelic incompatibility.Rather,
it would appear that the product of the H ~ t - 4allele
~
mustbeincompatiblewith
one or more products
expressed by unlinked M . spretus-specific genes.
A biochemical explanation of the first phenotype
would hold that the product of the Hst-4 gene normally forms homodimers or higher order multimeric
proteins, and Hst-4'/Hst-4" hybrid animals form nonfunctional dimers (or multimers) that interfere with
some aspectof normal sperm cell differentiation. The
second phenotype canbe explained biochemically
through the formation of heterodimeric proteins that
245
include the Hst-4 product as well as non-chromosome
17-encoded polypeptides.
An advantage of these two biochemical
explanations
is that they are unified through the proposition that
Hst-4 products function within the context of larger
multimeric units. Nevertheless, it is certainly possible
to formulate more complex models,and it is only with
the cloning of the Hst-4 gene or the biochemical and
physiological characterization of its product that any
model will be confirmed.
Finally, it is intriguing that the Hst-4 locus is not
separable genetically from the t complex distorter-2
(Tcd-2) locus involved in t haplotype effects on male
transmission ratio distortion and sterility. Furthermore, it has been demonstrated that Tcd-2 must also
function within the context of interacting protein
products (LYON 1984, 1986). Nevertheless, it is not
possible to further resolve the mapping of Tcd-2
within a 7-cM region because of an associated inversion (in(l7)4). Therefore, it is likely that Hst-4 will be
cloned and characterized first, at which point, it will
be possible to test its functional identity with Tcd-2.
This research was supported by grants from the National Institutes of Health to L.M.S., and by postdoctoral fellowships from the
National Institutes of Health to S.H.P. and M.F.H. We thank
CHRISTINE
BUCKfor technical assistance, andJuov CEBRA-THOMAS
and JEN-YUE TSAIfor animal assistance. S.H.P. thanks JOANNA
WIUON fortechnical advice and comments on the manuscript.
LITERATURE CITED
AUFFRAY,
J.-C., J. T. MARSHALL,
L. THALER
and F. BONHOMME,
1991 Focus on the nomenclature of european species of mus.
Mouse Genome 88: 7 4 .
BENNETT,
D., 1975 The T-locus of the mouse. Cell 6: 441-454.
BONHOMME,
F., J.-L.GUENET andJ. CATALAN,
1982 Prksence
d'un facteur de st6rilitk mile, Hst-2, segrkgant dans les croisements interspkcifiques M. musculus L. X M. spretus Lastaste et
lik iMod-I et Mpi-I sur le chromosome 9. C. R. Acad. Sci. 294
(Ser. 111): 691-693.
F., S. MARTIN and L. THALER,1972 Hybridization
BONHOMME,
between Mus musculus L. and Mus spretus Lataste under laboratory conditions. Experientia 34: 1140-1 141.
M., B. G. HERRMANN,
A.-M. FRISCHAUF,
V. L. BAUTCH,
V.
B~JCAN,
BODE,
L.
M.
SILVER,G. R. MARTIN and H. LEHRACH,
1987 Deletion and duplication ofDNA sequences is associated with the embryonic lethal phenotype of the t9 complementation group of the mouse t complex. Genes Dev. 1: 376385.
CAMERON,
D. R., and R. M. MOAV,1957 Inheritance in Nicotiana
tabacum. XXVII. Pollen killer, an alien genetic locus inducing
abortion of microspores not carrying it. Genetics 42: 326-335.
CHARLESWORTH,B.,and D. HARTL,1978 Population dynamics
of the segregation distorter polymorphism of Drosophila melanogaster. Genetics 89: 171-192.
CHURCH,G. M., and W. GILBERT,1984 Genomic sequencing.
Proc. Natl. Acad. Sci. USA 81: 1991-1995.
Committee for Mouse Chromosome 17, 1991 Maps of mouse
chromosome 17: first report. Mammal. Genome 1: 5-29.
P. KOURILDELARBRE.
C., Y . KASHI, P. BOURSOT, J. S. BECKMANN,
SKY, F. BONHOMME
and G. GACHELIN,1988 Phylogenetic
distribution in the genus Mus of t-complex-specific DNA and
246
S. H. Pilder, M. F. H a m m e r and L. M. Silver
protein markers: inferences on the origin of t-haplotypes. Mol.
Biol. Evol. 5: 120-33.
DUTCHER,
S., and F. LUX,1989 Genetic interactions of mutations
affecting flagella and basal bodies in Chlamydomonas. Cell
Motil. Cytoskeleton 1 4 104-1 17.
FEINBERG,
A. P., and B. VOGELSTEIN,
1984 A technique for
radiolabeling DNA restriction endonuclease fragments to high
specific activity. Anal. Biochem. 137: 266-267.
FIGUEROA,
F., E. NEUFELD,U. RITTE and J. KLEIN, 1988a tSpecific DNApolymorphisms among wild mice from Israel and
Spain. Genetics 119: 157-60.
FIGUEROA,
F., V. VINCEK,M. KASAHARA,G. I. BELLand J. KLEIN,
198% Mapping of the Sod-2 locus into the t complex on
mouse chromosome 17. Immunogenetics 28: 260-264.
FOREJT,
J., and P. I V ~ N Y1975
I,
Genetic studies on male sterility
of hybrids between laboratory and wild mice (Mus musculus
L.). Genet. Res. 24: 189-206.
FOREJT,J., V. VINCEK,
J. KLEIN,H. LEHRACH
and M. LOUDOVA199 1 Genetic mapping of the t complex region on
MICKOVA,
mouse chromosome 17 including the hybrid sterility-1 gene.
Mammal. Genome 1: 84-9 1 .
Fox, H., L. M. SILVER and
G. R. MARTIN,1984 An alpha globin
pseudogene is located within the mouse t complex. Immunogenetics 19: 125-130.
Fox, H. S., G. R. MARTIN, M. F. LYON,B. HERRMANN,
A.-M.
FRISCHAUF,
H. LEHRACH
and L. M. SILVER, 1985 Molecular
probes define different regions of the mouse t complex. Cell
4 0 63-69.
GUENET,
J.-L., C. NAGAMINE,D.SIMON-CHAZOTTES,X . MONTAGUTELLI and F. BONHOMME,
1990 Hst3: an X-linked hybrid
sterility gene. Genet. Res. 56: 163.
HALDANE,
J. B. S., 1922 Sex ratio and unisexual sterility in hybrid
animals. J. Genet. 12: 101-109.
HAMMER,
M. F., J. SCHIMENTI
and L. M. SILVER,
1989 Evolution
of mouse chromosome 17 and the origin of inversions associated with t haplotypes. Proc. Natl. Acad. Sci. USA 86: 32613265.
HERRMANN,
B., M. B ~ ~ c AP.
N ,E. MAINS,A.-M. FRISCHAUF,
L.M.
SILVER and
H. LEHRACH,1986 Genetic analysis of the prox-
imal portion of the mouse t complex: evidence for a second
inversion within t haplotypes. Cell 44:469-476.
HOGAN,B., F. COSTANTINI
and E. LACY,1986 Manipulating the
Mouse Embryo: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
KING, C.R., T. SHINOHARA
and J. PIATIGORSKY,
1982 Alphaacrystallin messenger RNA of the mouse lens: more noncoding
than coding sequences. Science 2 1 5 985-987.
LOEGERING,
W. Q., and E. R. SEARS,1963 Distorted inheritance
of stem-rust resistance of timstein wheat caused by a pollen
killing gene. Can. J. Genet. Cytol. 5 65-72.
LYON,M. F., 1984 Transmission ratio distortion in mouse thaplotypes is due to multiple distorter genes acting on a responder locus. Cell 37: 621-628.
LYON,M. F., 1986 Male sterility of the mouse t-complex is due to
homozygosity of the distorter genes. Cell 4 4 357-363.
LYON,M. F., and I. MASON,1977 Information on the nature of
t-haplotypes from the interaction of mutant haplotypes in male
fertility and segregation ratio. Genet. Res. 2 9 255-266.
NADEAU,
J. H., and S. J. PHILLIPS,1987 The putative oncogene
Pim-I in the mouse: its linkage and variation among t haplotypes. Genetics 117: 533-541.
SAGE,R. D., 1981 WILDMICE, PP. 1-39 IN The Mouse in Biomedical
J. D. SMALL and
J. G.
Research, Vol. I , edited by H. L. FOSTER,
Fox. Academic Press, New York.
SANO,Y., 1990 The genic nature of gamete elimination in rice.
Genetics 1 2 5 183-191.
SCHIMENTI,
J., L. VOLD,D. SOCOLOW
and L. M. SILVER,1987 An
unstable family of large DNA elements in the center of the
mouse t complex. J. Mol. Biol. 1 9 4 583-594.
SILVER,L. M., 1982 Genomic analysis of the H - 2 complex region
associated with mouse t haplotypes. Cell 2 9 961-968.
SILVER,L. M., 1985 Mouse t haplotypes. Annu. Rev. Genet. 1 9
179-208.
WILLISON,K. R., K. DUDLEYand J. POTTER,1986 Molecular
cloning and sequence analysis of a haploid expressed gene
encoding t complex polypeptide-I. Cell 4 4 727-738.
Communicating editor: R. E. GANSCHOW