Skip Navigation


Human Molecular Genetics Advance Access originally published online on January 13, 2005
Human Molecular Genetics 2005 14(5):615-625; doi:10.1093/hmg/ddi058
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/5/615    most recent
ddi058v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Leder, A.
Right arrow Articles by Leder, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Leder, A.
Right arrow Articles by Leder, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, Vol. 14, No. 5 © Oxford University Press 2005; all rights reserved

{zeta}–/– Thalassemic mice are affected by two modifying loci and display unanticipated somatic recombination leading to inherited variation

Aya Leder1, Jennifer McMenamin1, Karen Fontaine1, Alexander Bishop1 and Philip Leder1,2,*

1Department of Genetics, Harvard Medical School, Boston, MA 02115, USA and 2Howard Hughes Medical Institute, 77 Avenue Louis Pasteur, Boston, MA 02115, USA

* To whom correspondence should be addressed. Email: leder{at}gentics.med.harvard.edu

Received November 9, 2004; Revised December 27, 2004; Accepted January 5, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Thalassemia is a disease caused by a variety of mutations affecting both the adult and embryonic {alpha}- and ß-globin loci. A mouse strain carrying an embryonic {zeta}-globin gene disrupted by the insertion of a PGK-Neo cassette displays an {alpha}-thalassemia-like syndrome. Embryonic survival of this {zeta}-null mouse is variable and strongly influenced by genetic background, the 129/SvEv mouse strain displaying a more severe phenotype than C57BL/6. We have identified two modifying loci on C57BL/6 chromosomes 2 and 5, which affect the penetrance of embryonic lethality in the 129/SvEv mouse. Through this work, we were able to observe an interesting effect on somatic recombination events in thalassemic embryos. We show that these events can occur on multiple chromosomes in very early embryonic cells, prior to their allocation to the germline. Our results demonstrate that somatic recombination events can be transmitted to subsequent generations.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The primary cause of {alpha}-thalassemia is insufficient production of functional {alpha}-globin due to a variety of mutations at the {alpha}-locus. Many of these mutations have been characterized at the molecular level and have been found to involve gene transcription, pre-mRNA splicing, polyadenylation, defects in translation and gross deletions of one or more of the globin genes (1Go). The reduction in {alpha}-globin chain production results in an imbalance between {alpha}- and ß-globins. This imbalance plays a crucial role in the pathophysiology of the disease, because excess ß-chains in {alpha}-thalassemia and excess {alpha}-chains in ß-thalassemia are detrimental to the integrity of the red cells and shorten their lifespan (1Go,2Go).

To understand further details of this heritable disorder, we generated two mouse models of {alpha}-thalassemia: one with a disrupted embryonic {zeta}-globin gene, an {alpha}-globin-like gene, and the other with a disruption in the neighboring {alpha}1-globin gene (3Go,4Go). Homozygous disruption of either gene or the double heterozygote, {zeta}+/–/{alpha}1+/–, resulted in the characteristic features of {alpha}-thalassemia (3Go). Interestingly, all three were influenced by their genetic backgrounds. On a pure 129/SvEv (hereafter referred to as ‘129’) genetic background, the mutants display a more severe phenotype when compared with the 129/C57BL/6 genetic background. This suggested that unlinked modifying genes existed and accounted for the variable severity of the phenotype. Modifying genes clearly play an important role in human diseases, including the thalassemias (5Go), but are obviously more amenable to study in the mouse. Indeed, using a classical genetic approach, we linked a gene that modified the severity of {alpha}1-thalassemia in an {alpha}1-null mouse to the ß-globin locus. Two 129 ß-globin alleles lead to a severe phenotype, whereas the presence of one 129 ß-globin allele and one C57BL/6 allele result in a milder phenotype (6Go).

In the current study, we focused on the effect of genetic background on the embryonic thalassemic phenotype we observe in {zeta}-globin null mice. This particular thalassemic mouse displays a highly penetrant embryonic lethality on the 129 genetic background. This lethality is ameliorated in the 129/C57BL/6 background. Given our experience with {alpha}1-null mice, we thought, at first, that the ß-globin locus might also be linked to the embryonic lethality in {zeta}-globin null mice. It soon became clear that although ß-globin plays a role in the severity of the adult anemia as well as in the longevity of adult {alpha}-thalassemic mice, it had no effect on embryonic survival. We then performed an unbiased genome-wide screen, which revealed a possible linkage between embryonic survival of {zeta}-null thalassemic mice and two modifying loci on chromosomes 2 and 5. In the process of this screen, however, a further, unexpected phenotype emerged. We noticed a high frequency of somatic recombination with loss of heterozygosity (LOH) early in embryogenesis in the surviving {zeta}-null thalassemic mice. In one instance, it was possible to show that the recombination events occurred before germline formation as the alteration was transmitted to progeny. This shows how mitotic recombination, like meiotic recombination, could be contributing to genomic variation, gene assortment and variation in evolution of multicellular organisms such as mammals.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Phenotypic variability in {alpha}-thalassemic mice
The {alpha}-globin locus in the mouse consists of three expressed genes: 5'->{zeta}, {alpha}1, {alpha}2->3' (Fig. 1A) (7Go,8Go). Whereas {alpha}1 and {alpha}2 differ from one another in only one amino acid, the embryonic globin gene {zeta} shows only 53% identity to {alpha}1 and {alpha}2 (9Go,10Go). Although all three genes are expressed early in embryogenesis, the embryonic {zeta}-globin is switched off around day 11 of gestational development (9Go,11Go–14Go). Despite the fact that the {zeta}-globin genes are highly conserved throughout evolution, this gene is not required for successful gestation on mixed genetic background (3Go). Apparently, {alpha}1- and {alpha}2-globins alone can support the needs of the embryonic mouse from the onset of erythropoiesis (at 7.5 dpc) throughout adulthood (3Go,14Go). In the particular strain of {zeta}-null mice we created, expression of the adjacent {alpha}1-globin gene is downregulated due to the presence of the PGK-Neo cassette that was inserted into the nearby {zeta}-globin gene. As a result, our {zeta}-null adults suffer from {alpha}-thalassemia, albeit not as severe as {alpha}1–/– mice (3Go,15Go).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 1. (A) Diagrammatic representation of the murine {alpha}-globin locus showing three thalassemic genotypes and their relative phenotypic severity. I. The upper diagram represents the {zeta}–/–, {zeta}+/–/{alpha}1+/– (double heterozygote) and {alpha}1–/– genotypes indicating disrupted (null), attenuated and genes normally expressed during embryogenesis, respectively. II. The lower diagram represents the two {alpha}-globin genes that are expressed in adults. The two wedges indicate the relative severity of the embryonic and adult phenotypes corresponding to each genotype. (B) Gestational survival of {zeta}–/– thalassemic mice as a function of genetic background. (*This mouse died at 2 months of age. NA, not applicable.) (C) Genetic crosses designed to produce F1, N2 and N3 thalassemic mice and littermate controls. Underline refers to the particular genotype used in the cross.

 
In an effort to assess the phenotype brought about by these mutant genes, three genotypes were generated (Fig. 1A, I and II). The first two, {zeta}–/– and {alpha}1–/–, were created by homologous recombination and insertional mutagenesis, and the third, {zeta}+/–{alpha}1+/– (a double heterozygote), was created by crossing a {zeta}+/– heterozygote to an {alpha}+/– heterozygote. The three genotypically distinct mice display embryonic and adult differences in the severity of their {alpha}-thalassemia, and this correlates with their {alpha}-globin chain production as shown schematically in Figure 1A. For example, during development, {zeta}–/– embryos do not survive on the 129 genetic background (Fig. 1B), whereas the {alpha}1–/– and the double heterozygotes {zeta}+/–/{alpha}+/– survive easily (6Go) (data not shown). Lacking {zeta}-globin, the {zeta}–/– embryos depend solely on {alpha}2 and residual {alpha}1 for survival. In comparison, the {alpha}1–/– embryos that express normal amounts of both {zeta}- and {alpha}2-globins do well, whereas the double heterozygotes, expressing normal amounts of {alpha}2-globin but reduced levels of {zeta}- and {alpha}1-globin, are intermediate in their ability to survive gestation (6Go) (data not shown). In adults, however, the reverse is true (Fig. 1A, II). {zeta}–/– Adults expressing {alpha}2-globin and reduced {alpha}1-globin fare best, whereas {alpha}1–/– adults expressing only {alpha}2-globin are more profoundly affected. The double heterozygote is again intermediate in terms of the severity of its adult thalassemia phenotype.

Phenotypic variability is influenced by genetic background
As we followed gestational survival in the three genotypes, it was evident that while all three produced viable offspring on mixed genetic backgrounds (3Go), this was not the case on the 129 background (Fig. 1B). On 129, {alpha}1–/– and double heterozygotes survived gestation (6Go), whereas {zeta}–/– did not (Fig. 1B). In an effort to understand this gestational survival phenotype, we explored several inbred mouse strains for their ability to promote survival of {zeta}–/– embryos. The inbred C57BL/6 strain proved particularly effective and appropriate crosses between C57BL/6 and 129 thalassemic mice were initiated to assess the number of loci that might be involved in the modification of this embryonic lethality phenotype. As shown in Figure 1C, crosses between normal C57BL/6 and 129 {zeta}+/– heterozygotes produced 129/C57BL/6 F1 offspring heterozygous for the {zeta}+/– mutation. The F1 {zeta}+/– heterozygotes were then backcrossed to 129 {zeta}+/– heterozygotes, generating N2 {zeta}–/– mice homozygous for the insertional mutation (Fig. 1C). At weaning, {zeta}–/– mice are considerably smaller than their littermates, but with time, they gain weight and stature, eventually reaching a normal size. Both males and females are fertile and some, but not all, have a normal lifespan. Similarly, N3 {zeta}–/– mice were generated by appropriate crosses (Fig. 1C) and displayed a phenotype similar to that of the N2 {zeta}–/– mice.

Embyronic survival of {zeta}–/– mice is not linked to the ß-globin locus
In our previous study (6Go), we determined that ß-globin is a genetic modifier of {alpha}-thalassemia in {alpha}1-null mice. Therefore, we postulated that ß-globin may act in a similar fashion to affect gestational survival of {zeta}–/– mice. If so, {zeta}-null embryos having two 129 ß-globin genes would not survive, whereas {zeta}-null embryos with one 129 ß-globin and one C57BL/6 ß-globin allele (the permissive allele) would survive. This hypothesis was tested using a cohort of 91 {zeta}-null N2 mice (Fig. 2A, line 7). These mice, generated by backcrossing F1 {zeta}+/– heterozygotes to 129 {zeta}+/– heterozygotes (Fig. 1B), obviously inherit at least one 129 allele at every locus. The second allele could be either a 129 or C57BL/6 allele, depending on the contribution from the F1 parent. We took advantage of a sequence length polymorphism adjacent to the ß-globin gene, D7Mit40, to determine the strain origin of the DNA (Fig. 2B). As can be seen in Figure 2A, line 7, half of the cohort of live born {zeta}–/– mice had 129 ß-globin alleles (n=45), whereas the other half inherited both 129 and C57BL/6 ß-globin alleles (n=46). In other words, {zeta}–/– mice with 129 ß-globin could survive just as well as {zeta}–/– mice with both 129 and C57BL/6 ß-globin alleles. This result clearly excludes the possibility that ß-globin plays a role in gestational survival of the {zeta}-null mouse.



View larger version (38K):
[in this window]
[in a new window]
  		Figure 2. ß-Globin alleles affect the severity of {alpha}-thalassemia and the longevity of {zeta}–/– null mice, but not their gestational survival. (A) Assessment of the effect of different ß-globin alleles on the severity, longevity and gestational survival of {zeta}–/– thalassemic mice. Rows 1–4: Severity of the anemia, measured by the percentage of reticulocytes and hemoglobin concentration in peripheral blood (rows 1–2: wild-type control; rows 3–4: {zeta}–/–). Rows 5 and 6: Longevity (in months) of {zeta}–/– mice as a function of ß-globin allele. Row 7: Survival of {zeta}–/– mice as a function of ß-globin alleles. N, number of mice. (B) SSLP analysis of D7Mit40 using tail DNA from control mice (lane 1, C57BL/6; lane 2, 129) and backcrossed mice (lanes 3–10). D7Mit40 is the SSLP marker, which is adjacent to the ß-globin locus and differentiates between the 129 and C57BL/6 ß-globin alleles. The mice represented in lanes 3–5 are homozygous for 129 ß-globin, whereas those in rows 6–10 carry both 129 and C57BL/6 ß-globin alleles.

 
Severity of the adult thalassemia phenotype is linked to the ß-globin locus in the {zeta}-null mouse
Given that {alpha}1-globin is downregulated in {zeta}–/– mice (Fig. 1A) (3Go), we expected that ß-globin alleles might affect the severity of the anemia in {zeta}-null adults, as they do in the case of {alpha}1–/– adult mice. We, therefore, examined hemoglobin and reticulocyte levels, because reduced hemoglobin and high levels of reticulocytes are measures of thalassemic severity. Blood samples from 10 backcrossed N2 {zeta}-null mice, five with 129 ß-globin and five with both 129 and C57BL/6 ß-globin alleles were analyzed (Fig. 2A, lines 3 and 4). Although all 10 mice suffered from microcytic anemia with red cell parameters characteristic of human thalassemia (Fig. 2A, lines 1 and 2 compared with 3 and 4), as expected, mice homozygous for the 129 ß-globin locus were more severely affected than {zeta}–/– mice with both 129 and C57BL/6 ß-globin loci. {zeta}-Null mice with 129 ß-globin (line 3) show lower levels of hemoglobin and higher percentages of reticulocytes in peripheral blood when compared with {zeta}-null mice that inherited both 129 and C57BL/6 ß-globins (Fig. 2A, line 4).

We next examined the lifespan of {zeta}-null survivors (Fig. 2A, lines 5-6). Again, we noted a connection between ß-globin alleles and longevity in the adult mice. Mice with both 129 and C57BL/6 ß-globin alleles live to an average age of 23.7 months, comparable to the normal lifespan. {zeta}-Null mice that inherit two 129 ß-globin alleles exhibit a shortened lifespan, averaging 12.7 months. When {alpha}-globin expression is compromised due to an insertion of the PGK-Neo cassette in the {zeta} locus, the specific structure of the excess adult ß-globin chain, 129 versus C57BL/6, affects the severity of the disease, which in turn affects longevity.

Modification of the gestational survival phenotype reveals a possible linkage to loci on chromosomes 2 and 5
As a first step in understanding the gestational survival of {zeta}–/– mice in a C57BL/6 background, we tried to determine the number of loci that, when inherited from C57BL/6 mice, promotes survival in a dominant fashion. As can be seen in Figure 1B, crosses between two {zeta}+/– heterozygotes on the 129 genetic background, generally, fail to generate live {zeta}–/– offspring. Crosses between the 129 {zeta}+/– heterozygous parent and an F1 129/C57BL/6 {zeta}+/– heterozygous parent generated a cohort of N2 {zeta}–/– mice. Assuming that survival was not an issue, we would have expected that one mouse out of four would have had the {zeta}–/– genotype as a result of these {zeta}+/– heterozygote crosses and that the number of {zeta}+/+ mice (323) would be approximately equal to the number of live born {zeta}–/– mice. This was not the case, and the number of surviving {zeta}–/– mice was only one-fourth of this number (80), consistent with a requirement for the inheritance of two C57BL/6 survival genes (Fig. 1B). The ratio of {zeta}–/– mice among the N3 offspring also suggested the involvement of two survival genes (Fig. 1B). To generate the N3 generation, N2 {zeta}–/– mice were backcrossed to {zeta}+/– 129 heterozygotes (Fig. 1C). Assuming survival was not an issue, one out of two offspring should be {zeta}–/–, and the number of surviving {zeta}–/– mice should have been approximately equal to the number of surviving {zeta}+/– mice (238). As shown in Figure 1B, we found only 59 {zeta}–/– mice, a number consistent with the participation of two C57BL/6 survival genes. Of course, more complex scenarios with respect to genetic interactions can be imagined.

To map and identify these postulated gestational survival genes, a genome-wide screen was performed on cohorts of 97 N2 {zeta}–/– and 105 N2 {zeta}+/+ mice and {zeta}+/– littermate controls. A smaller cohort of N3 {zeta}–/– mice was also examined. Genomic DNAs were prepared, and DNA amplification reactions were run using appropriate oligonucleotide primers to detect simple sequence length polymorphorisms (SSLPs) between the two inbred strains, 129 and C57BL/6 (16Go–18Go). Genotyping 97 DNAs of the {zeta}–/– survivors and 105 DNAs from littermate controls with over 100 SSLP markers dispersed along each of the 20 chromosomes provided information concerning the possible chromosomal location of the survival genes. When a SSLP marker shows a non-random distribution of the F1 alleles such that a high proportion of the N2 and N3 {zeta}–/– survivors carry a C57BL/6 allele, a linkage between survival and the marker is likely because gestational survival depends on genes derived from the C57BL/6 genetic background. At the same time, this marker is expected to show a random distribution among N2 {zeta}+/– and {zeta}+/+ control mice.

Indeed, our screen revealed that a high proportion of the N2 {zeta}–/– gestational survivors carry C57BL/6 alleles on chromosomes 2 and 5 (Fig. 3). This was not the case in the N2 {zeta}+/+ and {zeta}+/– control group. As can be seen in Figure 3, there is no significant difference between the percentage of mice with 129 marker alleles (shaded columns) and the percentage of mice with both 129 and C57BL/6 marker alleles (black columns) in the control group. However, a clear difference is evident in the percentage of mice with 129 marker alleles and the percentage of mice with 129 and C57BL/6 marker alleles in the {zeta}–/– cohort. Numerous markers on chromosome 2 and chromosome 5, which are polymorphic between 129 and C57BL/6, were tested in our screen. Figure 3 depicts a single marker on chromosome 2, D2Mit48, at position 87 cM, and a single marker on chromosome 5, D5Mit161, at position 70 cM. Because large portions of chromosomes 2 and 5 showed the same trend, namely a high proportion of the C57BL/6 marker alleles among the {zeta}–/– survivors but not in the control group, markers other than the ones chosen could have been shown to illustrate this point. Comparison of the N2 {zeta}–/– gestational survivors with the control group revealed significant differences between the segregation pattern of chromosome 5 and, especially, chromosome 2 (Z=2.14, P≤0.032 and Z=3.20, P≤0.0014, respectively). These results suggest a possible association between gestational survival and C57BL/6 alleles on chromosomes 2 and 5, although the significance with respect to chromosome 5 does not meet that required in order to declare linkage.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 3. Loci that modify gestational survival of {zeta}-null mice show a high proportion of C57BL/6 alleles on chromosomes 2 and 5. Genotyping the N2 {zeta}–/– cohort with D2Mit48, a marker on chromosome 2, and with D5Mit161, a marker on chromosome 5, shows a non-random distribution of the C57BL/6 alleles, whereas the control cohort N2 {zeta}+/+ and {zeta}+/– shows a random distribution of these markers. Shaded columns indicate the percentage of mice with 129 alleles, and black columns indicate the percentage of mice with both 129 and C57BL/6 alleles. These consisted of 92 mice in the {zeta}–/– cohort, 101 mice in the {zeta}+/+ and {zeta}+/– control group of chromosome 2 and 97 mice in the {zeta}+/+ and {zeta}+/– control group of chromosome 5.

 
Although the majority of our cohort of {zeta}–/– survivors carry C57BL/6 marker alleles on regions of chromosome 2 and chromosome 5, some survivors did not. Seven mice (of 97) had only 129 alleles across chromosome 2, and five mice (of 97) had only 129 alleles across chromosome 5. One mouse (out of 97) had only 129 alleles across both chromosome 2 and chromosome 5. It is possible that although the major survival genes reside on chromosome 2 and chromosome 5, minor survival genes reside elsewhere. We are now generating a cohort of N4 and N5 {zeta}–/– survivors, all descendents of one N3 {zeta}–/– mouse. Mice in this cohort are expected to carry the same survival genes, making it possible to narrow the relevant region on chromosome 2 (data not shown).

Evidence for unexpected somatic LOH in surviving {zeta}–/– mice
In screening N2 and N3 backcrossed mice for SSLPs, we expected two possible genotypes at each locus, either homozygous for the 129 allele or heterozygous for both 129 and C57BL/6 alleles. For either genotype, one 129 allele is inherited from the 129 homozygous parent and the other allele, either 129 or C57BL/6, is inherited from the F1 or N2 parent. Progeny with two C57BL/6 alleles are not expected. However, we occasionally observed only a single polymerase chain reaction (PCR) product corresponding to the C57BL/6 allele, which was initially interpreted as a failure of the PCR to detect the 129 template. To minimize the possibility of this type of artifact, the PCR reaction was repeated several times and, when available, adjacent polymorphic markers were examined. If two adjacent markers appeared to have only C57BL/6 alleles, then we interpreted this as a true representation of the genotype at these loci.

An independent confirmation for the unexpected appearance of C57BL/6 alleles presented itself when a non-agouti black {zeta}–/– mouse was discovered among agouti littermates. Because the agouti coat color is a dominant trait over the black coat of C57BL/6 and each locus was expected to have at least one 129 allele (an agouti allele), all offspring from these crosses were expected to be agouti (19Go). Thus, a non-agouti black mouse was completely unexpected. Fortunately, the parents of this mouse, its siblings and 10 offspring were available for further study. PCR amplification of the polymorphic markers on chromosome 2 of the non-agouti black mouse and his four agouti siblings is shown in Figure 4A. D2Mit48, a marker on chromosome 2 at position 87 cM in close proximity to the agouti locus, in addition to D2Mit113, a more distal marker located at position 103 cM, shows that this black mouse (Fig. 4, lane 5) has only C57BL/6 alleles across this region of chromosome 2 (87–103 cM), which includes the agouti locus. As expected, the agouti siblings have 129 alleles or both 129 and C57BL/6 alleles at each locus. The agreement between the PCR genotyping at the agouti locus and the black coat color of this mouse rule out a potential PCR artifact.



View larger version (62K):
[in this window]
[in a new window]
  		Figure 4. Genotype of the non-agouti black mouse, his siblings and his parents. (A) Marker analysis of the non-agouti black mouse (lane 5) and his four siblings (lane 1–4) with three chromosome 2 SSLP markers: D2Mit494 at 73 cM, D2Mit84 at 87 cM and D2Mit113 at 103 cM. (B) Marker analysis of the non-agouti black mouse and his agouti parents using seven SSLP markers distributed along chromosome 2. Their positions are given in centimorgans (cM), and the position of the agouti locus is indicated.

 
The genotype of the parents of the non-agouti mouse with markers on chromosome 2 is shown in Figure 4B. The 129 {zeta}+/– mother, as anticipated, is homozygous 129 for all loci examined across chromosome 2. The N2 {zeta}–/– father had both 129 and C57BL/6 alleles at all loci. The N3 {zeta}–/– non-agouti black mouse had both 129 and C57BL/6 alleles along most of chromosome 2 except at markers in distal regions where genotyping showed only C57BL/6 alleles as previously mentioned (Fig. 4A). To further reassure ourselves that the mother was 129 {zeta}+/–and not F1 or N2, we genotyped the non-agouti black mouse with 26 SSLP markers at which his father was 129 homozygous. At all these loci, the son, the non-agouti black mouse, was 129 homozygous as well (data not shown). This is an extremely unlikely outcome (G=9.9, P≤0.0016) were the mother not 129 homozygous. Thus, this analysis supports the conclusion that the mother is of homozygous 129 genetic background. Further, it tends to rule out any possible issue with mouse husbandry, and further suggests that mitotic recombination may be responsible for the homozygosity or LOH seen on a region of chromosome 2 of the non-agouti black mouse. In addition, as shown in Figure 5, LOH was also evident with respect to markers on chromosome 4, as well as on chromosome 19 of the non-agouti black mouse. It seems that LOH occurred mitotically on three different chromosomes of this mouse.



View larger version (23K):
[in this window]
[in a new window]
  		Figure 5. Germline transmission of regions of LOH on chromosomes 2, 4 and 19 from the non-agouti black mouse to its progeny. A pedigree is represented indicating the inheritance of chromosomes bearing regions of LOH. Conventional symbols represent each mouse analyzed, black being C57BL/6, white being 129, and striped portions being unknown (UNK). Chromosomes are not represented to scale.

 
The non-agouti black mouse displays LOH in the pigment cells and in the cells of the tail from which DNA for genotyping was obtained. Examination of the offspring of the non-agouti mouse revealed that the recombination events resulting in the observed LOH must have occurred early in development, before germline formation, as the regions of LOH were inherited by all of his progeny (Fig. 5). This was established by crossing the male non-agouti black mouse to a 129 agouti female and genotyping the offspring with markers for which the black father carried two C57BL/6 alleles (Fig. 5). All progeny inherited one C57BL/6 allele from the father and one 129 allele from the mother (Fig. 5). None of the 10 offspring was 129 homozygous for any locus on chromosome 2, 4 or 19, where the father was homozygous for C57BL/6, indicating that all the father's germ cells carried C57BL/6 alleles at these loci. Because primordial germ cells have been identified as early as 7–7.5 dpc, the somatic recombination events in the non-agouti black mouse must have occurred earlier (20Go). As noted, mosaicism was not evident in the non-agouti black mouse with respect to his germ cells, nor was it evident with respect to coat color, because the mouse was uniformly black. Furthermore, mosaicism was not apparent as PCR products corresponding to 129 alleles from the tail DNAs were absent, despite the sensitivity of the assay.

It seems that recombination events take place in an early progenitor cell capable of differentiating not only into primordial germ cells, but also into other tissue types. Genotyping the 10 offspring with several markers on chromosomes 2, 4 and 19 permitted closer examination of the individual homologs of the non-agouti black mouse. For example (Fig. 5), one group of four mice inherited one of the father's chromosome 2 homologs, whereas another group of four inherited the other homolog. The remaining two mice are recombinants, a product of meiotic recombination in the father's germ cells. A similar situation was seen in chromosome 4, where four offspring inherited one of the father's homologs and three inherited the father's other homolog. The remaining three were meiotic recombinants. This further analysis allows us to rule out deletion as a possible mechanism for the observed LOH and confirms the presence of two C57BL/6 alleles in the non-agouti black mouse.

A high number of {zeta}-null mice occur among the non-agouti black descendents
Of the 10 offspring of the non-agouti black mouse shown in Figure 5, four were {zeta}–/– homozygotes and six were {zeta}+/– heterozygotes. The number of homozygous survivors exceeds that expected in crosses between N2 {zeta}–/– homozygotes and {zeta}+/– heterozygotes (Fig. 1B, last line). This raises the possibility that regions of LOH in the non-agouti black mouse harbor survival genes, which are transmitted to all progeny. As our genome-wide screen linked survival to chromosome-2 (Fig. 3), we questioned whether the region of LOH at the distal portion of this chromosome is, in fact, the site of a survival gene.

To explore this possibility, a brother–sister mating between the N4 offspring of the non-agouti black mouse was initiated. A non-agouti black female {zeta}+/– heterozygote was recovered. Like the non-agouti black grandfather, she carried two C57BL/6 alleles at the distal region of chromosome-2, encompassing the agouti locus, where a putative survival gene may reside. She was then crossed to an N4 {zeta}–/– homozygote agouti male, a son of the non-agouti black father. Four litters yielded a total of 23 mice. Eleven were {zeta}+/– heterozygotes and 12 were {zeta}–/– homozygotes, clearly a high number of {zeta}-null survivors. This reinforced the notion that a gene essential for survival resides at the distal region of chromosome 2. Although these numbers are too small to be significant, it is noteworthy that nine out of 12 {zeta}–/– mice are black (i.e. carry two copies of the C57BL/6 allele), consistent with a gene dosage advantage.

Occurrence of somatic LOH in additional {zeta}–/– mice
In addition to the non-agouti black mouse, which displays LOH on chromosomes 2, 4 and 19 (Fig. 6, mouse 5), LOH was detected in two other mice (Fig. 6, mouse 8 and 90). The latter two mice were members of a cohort of N2 {zeta}–/– gestational survivors, and, like the non-agouti black mouse, displayed LOH on several chromosomes. Fifteen SSLP markers were tested on chromosome 2 and indicated that LOH in the non-agouti black mouse (no. 5) spanned four markers over a chromosomal segment of 18 cM. Mouse 8 showed LOH at two markers on chromosome 2, spanning a chromosomal segment of 2 cM. Interestingly, this region happens to overlap in the two mice. In addition, mouse 8 displays a 6 cM LOH over four markers on chromosome 5, in addition to two markers encompassing 7 cM on chromosome 6. Mouse 90, like the non-agouti black mouse, shows LOH on chromosome 19 with one marker (adjacent markers were not available) and also with two markers on chromosome 10 encompassing 0.5 cM. This mouse also displays LOH with all five markers tested on chromosome 13, suggesting a possible chromosomal non-disjunction or mitotic recombination near the centromere, where recombination occurs at an unusually high frequency (21Go).



View larger version (22K):
[in this window]
[in a new window]
  		Figure 6. Regions of LOH in backcrossed {zeta}–/– null mice. Note that mouse #5, the non-agouti black mouse, displays regions of LOH on chromosomes 2, 4 and 19 (indicated as overlapping black segments on each chromosomal pair). Mouse #8 displays LOH on chromosomes 2, 5 and 6. Mouse #90 displays LOH on chromosomes 10, 13 and 19. Arrows represent markers used to establish regions of LOH. The number of markers that show LOH is indicated in each case. White bars indicate 129 alleles. Black bars indicate C57BL/6 alleles. Striped bars indicate that the identity of alleles is unknown. As indicated, white portions of the represented chromosomes are 129 in origin, black portions are C57BL/6 and striped portions are unknown.

 
{zeta}-Null mice display increased occurrence of LOH
We questioned whether the frequency of somatic recombination seen in {zeta}–/– gestational survivors is somehow driven or linked to the thalassemia phenotype or whether this apparent high frequency of mitotic recombination is, in fact, a normal occurrence in mice. To address this question, we turned our attention to the N2 backcrossed mice that were littermates of N2 {zeta}–/– gestational survivors (Table 1). This control group, consisting of DNA from 54 {zeta}+/+ mice and 51 {zeta}+/– mice, was assembled and subjected to a genome-wide screen similar to that described earlier for the N2 {zeta}–/– survivors. Although the earlier screen of {zeta}–/– mice revealed three out of 97 mice with LOH on multiple chromosomes, none was found among 105 mice in the control group (Table 1). PCR analysis of 115 SSLP markers distributed across all 20 chromosomes, totaling 10 654 reactions, was performed on DNA from 97 of the N2 {zeta}–/– cohort. A similar analysis of 118 SSLP markers totaling 10 712 reactions was carried out on DNA from 105 of the {zeta}+/+ and {zeta}+/– control group.


View this table:
[in this window]
[in a new window]
  		Table 1. Increased occurrence of LOH in {zeta}-null mice
 
The difference in frequency of LOH between the thalassemic and control groups is summarized in Table 1. The first two columns show the number of mice in each cohort and the number that demonstrate LOH. Although each of the mice that display LOH do so on multiple chromosomes, one can argue that they do not represent independent events. In this strictest comparison, the two groups of mice are significantly different from one another [P(G=4.5)≤0.0348]. The next two columns present the same data, but here we consider the possibility that each chromosome with a region of LOH is brought about by an independent recombination event. We, therefore, compare the frequency of LOH for the total number of chromosomes examined in either group (number of micex20 chromosomes) and determine them to be significantly different [P(G=11.8)≤0.0006]. In Figure 6, nine chromosomes are shown to display LOH. In the data shown in Table 1, chromosome 19 of mouse 90 was excluded because there were neither offspring nor nearby markers available to rule out PCR artifact. The second to last column of Table 1 shows the total number of PCR reactions that were performed on tail DNAs of each cohort. The total number of apparently independent LOH events that were identified in this analysis is shown in the last column. To compare our two cohorts, it was important that the two compared groups were similar. Indeed, as can be seen in Table 1, the {zeta}–/– group and the littermate control group had comparable numbers of mice, and the two groups were subjected to virtually the same number of PCR reactions. Upon comparison of the two groups, a significant difference was evident with respect to the three criteria mentioned earlier, with P-values of ≤0.0348, ≤0.0006 and ≤0.0008, respectively. This result confirmed our suspicion that the frequency of somatic recombinations observed in our cohort of {zeta}–/– gestational survivors is significantly higher than the frequency of recombination that might normally occur during early development.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Whereas the severity of {alpha}-thalassemia in the three mutants we describe depends on the nature of the adult ß-globin genes, the gestational survival of {zeta}–/– embryos does not. We suggest that this can be attributed to the fact that the embryonic ß-globin genes and the counterparts of {zeta}-globin gene (the embryonic {alpha}-globin gene) are expressed only during early embryogenesis before the expression of adult ß-globin genes takes place (22Go). These embryonic ß-globin genes display no functional difference between the two inbred lines and, therefore, do not differentially affect the embryonic phenotype. None the less, the differential survival effect of the 129 and C57BL/6 genetic backgrounds was useful in allowing us to identify modifying loci, in addition to the ß-globin locus.

Possible mechanisms responsible for the increased frequency of LOH
Our genome-wide screen revealed a possible linkage between survival and chromosomes 2 and 5. In addition, the screen revealed an unexpectedly high frequency of somatic recombination among {zeta}–/– survivors. The apparent linkage between the null mutation in {zeta}-globin and the high frequency of somatic recombination in the early development of {zeta}-null mice is not easily explained. Initially, we speculated that the mutation in the {zeta}-globon gene might result in increased oxidative stress due to erythroid destruction and iron release. This would subsequently produce DNA damage and, hence, an increased frequency of somatic recombination (23Go–26Go). However, in at least one mouse we recovered, namely, the non-agouti black mouse, LOH was observed in at least three different tissues: melanocytes of the fur, connective tissue of the tail and in germ cells. The derivation of both the melanocytes and germ cells are known to take place before day 7 of embryogenesis (20Go), that is, prior to functional activation of the hematopoietic system (11Go–14Go) and before impairment of this system would have produced hypoxic stress. Indeed, as mentioned earlier, these recombination events likely occurred in progenitor cells before commitment to melanocytes, skin and connective tissue and before allocation of germ cells. Such early events would occur before the emergence of embryonic hemoglobins, thereby ruling out DNA damage resulting from erythroid destruction as the mechanism responsible for increased frequency of recombination.

An alternative, more plausible explanation focuses on positive selection. Although the C57BL/6 gestational survival loci can act in a dominant fashion, inheriting a double dose of a C57BL/6 survival gene might be advantageous. Interestingly, two of the three affected mice display LOH on an overlapping region of chromosome 2, a region that is likely to harbor a survival gene (see Results). One of the two mice also displays LOH on chromosome 5. These are precisely the two chromosomes that our screen showed as possibly linked to gestational survival. The third mouse (Fig. 6, 90) displays LOH on other chromosomes and, as such, may nevertheless harbor recessive survival genes. Recombination that can lead to LOH occurs spontaneously and represents genetic transfer between homologous chromosomes (27Go–29Go). The absence of LOH in our control (non-thalassemic) group is not surprising considering the relatively small number of mice (105), the number of markers tested (115) and the frequency of spontaneous recombinations, which is estimated to be roughly 10–5 per cell (30Go–34Go). Although these parameters are similar for the {zeta}-null cohort as well, the two groups differ in two major respects. First, although there is no attrition during development for the non-thalassemic group, the majority of {zeta}-null embryos do not survive embryogenesis (Fig. 1B). Thus, one can argue that the original {zeta}-null embryonic cohort is in effect larger than the actual number of survivors. Furthermore, although recombination events are presumably neutral in the control group, neither advantageous nor disadvantageous with respect to embryonic survival, this is not the case for the {zeta}-null cohort. For this group, LOH on regions of chromosomes that harbor genes contributing to gestational survival might provide a selective advantage. We suggest that these two factors might account for the apparent high frequency of recombination seen in the surviving {zeta}-null mice. Indeed, mitotic recombination during early embryogenesis is likely to occur in the control groups as well, but at a reduced frequency.

The fact that LOH occurs on multiple chromosomes in these mice is intriguing. It suggests that for the rare cell capable of undergoing mitotic recombination, crossing-over events occur simultaneously on multiple chromosomes. It remains to be seen whether this is unique to early embryonic cells or whether it is a general feature of mitotic recombination. As most studies in this field follow chromosomal recombination events at specific reporter loci, potential recombination events occurring elsewhere in the genome would be overlooked (27Go,35Go–39Go). In contrast, this study tested numerous genomic sites in an unbiased manner. Such a global screen would be expected to identify recombination events on multiple sites, the result we see in our three {zeta}-null survivors.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Animals
The {zeta}+/– heterozygotes, the {alpha}1+/– heterozygotes and the double heterozygotes {zeta}+/–/{alpha}1+/– on 129/SvEv used for genetic crosses were generated in our laboratory (3Go). The inbred strain C57BL/6 mice were obtained from Taconic Farms (Germantown, NY 12526, USA) and housed in our animal facility, until they were old enough to breed.

Genotyping knockout mice
DNA was prepared from tail biopsies as previously described (3Go) and then subjected to PCR reactions. To distinguish between {zeta}+/+, {zeta}+/– and {zeta}–/–, three primers were used: F—5'-GAGAGAGCTATCATCATGTCCATGTGG-3', R—5'-CCTCTCTGCCTCTGTCCTTACTGTCC-3' and Neo—5'-GCAGCCTCTGTTCCACATACACCTT-3'. To distinguish between {alpha}+/+ and {alpha}+/–, two primers were used: Neo—5'-GCAGCCTCTGTTCCACATACACTT-3' and F—5'-GACAGACTCAGGAAGAAACCATGGTGC-3'.

Each amplification reaction contained 1 µl of diluted or undiluted genomic DNA, 0.2 mM dNTPs, 10x PCR buffer (Roche Molecular Systems, Branchburg, NJ 08876, USA), 0.5 µl each of forward (F) and reverse (R) primers and 0.1 µl of Taq polymerase (0.5 mg/ml) in a final volume of 25 µl. The thermal cycle conditions were: denaturation at 94°C for 2 min, followed by 40 cycles of 94°C for 30 s, 55°C for 30 s and 72°C for 1 min, with a final extension of 72°C for 10 min. The PCR reactions were analyzed by gel electrophoresis using 1.5% agarose gel.

Genotyping ß-globin
D7Mit40 primers were used in a PCR reaction to distinguish between ß-globin of 129/SvEv and ß-globin of C57BL/6. The PCR conditions were the same as mentioned earlier. The PCR reactions were analyzed on 4% NuSieve GTG agarose in TBE buffer.

Genome-wide screen
A cohort of 97 {zeta}–/– and a cohort of 105 littermate controls were analyzed with 115 and 118 microsatellite markers, respectively. Markers were chosen that are polymorphic between 129 and C57BL/6. The PCR and gel electrophoresis conditions were the same as those used to genotype ß-globin.

Peripheral blood analysis
Blood was collected from tails in potassium EDTA treated microtubes. Hematologic indices, including hemoglobin and percentage of reticulocytes, were determined using standard hospital methods in the clinical laboratory at The Children's Hospital, Boston.

Statistical analysis
Comparison between numbers of events was done by a standard G-test. The G-test is equivalent to a contingency {chi}2-test, but allows for classes with zero events (40Go). The frequency of chromosomal segregation in {zeta}–/– mutant mice was compared with that in control mice using a standard normal Z-test.


   ACKNOWLEDGEMENTS
 
We are grateful to David Beier and William Dietrich for their advice, encouragement and helpful suggestions. We are also grateful to Terri Broderick for her expert editorial assistance. This work was supported in part by the Howard Hughes Medical Institute.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Nathan, D. and Orkin, S. (1998) The thalassemias. In Nathan, D. and Oski, F. (eds), Hematology of Infancy and Childhood, 5th edn. Saunders, Philadelphia, PA, pp. 811–886.

  2. Nathan, D.G. and Gunn, R.B. (1966) Thalassemia: the consequences of unbalanced hemoglobin synthesis. Am. J. Med., 41, 815–830.[CrossRef][Web of Science][Medline]

  3. Leder, A., Daugherty, C., Whitney, B. and Leder, P. (1997) Mouse zeta- and alpha-globin genes: embryonic survival, alpha-thalassemia, and genetic background effects. Blood, 90, 1275–1282.[Abstract/Free Full Text]

  4. Chang, J., Lu, R.H., Xu, S.M., Meneses, J., Chan, K., Pedersen, R. and Kan, Y.W. (1996) Inactivation of mouse alpha-globin gene by homologous recombination: mouse model of hemoglobin H disease. Blood, 88, 1846–1851.[Abstract/Free Full Text]

  5. Weatherall, D. (1995) The molecular basis for phenotypic variability of the common thalassaemias. Mol. Med. Today, 1, 15–20.[CrossRef][Web of Science][Medline]

  6. Leder, A., Wiener, E., Lee, M.J., Wickramasinghe, S.N. and Leder, P. (1999) A normal beta-globin allele as a modifier gene ameliorating the severity of alpha-thalassemia in mice. Proc. Natl Acad. Sci. USA, 96, 6291–6295.[Abstract/Free Full Text]

  7. Leder, A., Swan, D., Ruddle, F., D'Eustachio, P. and Leder, P. (1981) Dispersion of alpha-like globin genes of the mouse to three different chromosomes. Nature, 293, 196–200.[CrossRef][Medline]

  8. Zhao, Q.Z., Liang, X.L., Mitra, S., Gourdon, G. and Alter, B.P. (1996) Cloning and characterization of the mouse alpha globin cluster and a new hypervariable marker. Mamm. Genome, 7, 749–753.[CrossRef][Web of Science][Medline]

  9. Leder, A., Weir, L. and Leder, P. (1985) Characterization, expression, and evolution of the mouse embryonic zeta-globin gene. Mol. Cell. Biol., 5, 1025–1033.[Abstract/Free Full Text]

  10. Nishioka, Y. and Leder, P. (1979) The complete sequence of a chromosomal mouse alpha—globin gene reveals elements conserved throughout vertebrate evolution. Cell, 18, 875–882.[CrossRef][Web of Science][Medline]

  11. Craig, M.L. and Russell, E.S. (1964) A developmental change in hemoglobins correlated with an embryonic red cell population in the mouse. Dev. Biol., 10, 191–201.[CrossRef][Web of Science][Medline]

  12. Barker, J.E. (1968) Development of the mouse hematopoietic system. I. Types of hemoglobin produced in embryonic yolk sac and liver. Dev. Biol., 18, 14–29.[CrossRef][Web of Science][Medline]

  13. Wilkinson, D.G., Bailes, J.A., Champion, J.E. and McMahon, A.P. (1987) A molecular analysis of mouse development from 8 to 10 days post coitum detects changes only in embryonic globin expression. Development, 99, 493–500.[Abstract/Free Full Text]

  14. Leder, A., Kuo, A., Shen, M.M. and Leder, P. (1992) In situ hybridization reveals co-expression of embryonic and adult alpha globin genes in the earliest murine erythrocyte progenitors. Development, 116, 1041–1049.[Abstract]

  15. Pham, C.T., MacIvor, D.M., Hug, B.A., Heusel, J.W. and Ley, T.J. (1996) Long-range disruption of gene expression by a selectable marker cassette. Proc. Natl Acad. Sci. USA, 93, 13090–13095.[Abstract/Free Full Text]

  16. Bailey, D.W. (1971) Recombinant-inbred strains. An aid to finding identity, linkage, and function of histocompatibility and other genes. Transplantation, 11, 325–327.[Web of Science][Medline]

  17. Dietrich, W., Katz, H., Lincoln, S.E., Shin, H.S., Friedman, J., Dracopoli, N.C. and Lander, E.S. (1992) A genetic map of the mouse suitable for typing intraspecific crosses. Genetics, 131, 423–447.[Abstract]

  18. Dietrich, W.F., Miller, J.C., Steen, R.G., Merchant, M., Damron, D., Nahf, R., Gross, A., Joyce, D.C., Wessel, M., Dredge, R.D. et al. (1994) A genetic map of the mouse with 4006 simple sequence length polymorphisms. Nat. Genet., 7, 220–245.[CrossRef][Web of Science][Medline]

  19. Wolff, G.L. (2003) Regulation of yellow pigment formation in mice: a historical perspective. Pigment Cell Res., 16, 2–15.[CrossRef][Web of Science][Medline]

  20. McLaren, A. (2003) Primordial germ cells in the mouse. Dev. Biol., 262, 1–15.[CrossRef][Web of Science][Medline]

  21. Liang, L., Deng, L., Shao, C., Stambrook, P.J. and Tischfield, J.A. (2000). In vivo loss of heterozygosity in T-cells of B6C3F1 Aprt(+/–) mice. Environ. Mol. Mutagen., 35, 150–157.[CrossRef][Web of Science][Medline]

  22. Edgell, M.H., Weaver, S., Jahn, C.L., Padgett, R.W., Phillips, S.J., Voliva, C.F., Comer, M.B., Hardies, S.C., Haigwood, N.L., Langley, C.H. et al. (1981) The mouse beta hemoglobin locus. In Stamatoyannopoulos, G. and Nienhuis, A.W. (eds) Organization and Expression of Globin Genes. Alan R. Liss, Inc., New York, New York, pp. 69–88.

  23. Comporti, M., Signorini, C., Buonocore, G. and Ciccoli, L. (2002) Iron release, oxidative stress and erythrocyte ageing. Free Radic. Biol. Med., 32, 568–576.[CrossRef][Web of Science][Medline]

  24. Chiu, D. and Lubin, B. (1989) Oxidative hemoglobin denaturation and RBC destruction: the effect of heme on red cell membranes. Semin. Hematol., 26, 128–135.[Web of Science][Medline]

  25. Chiu, D.T. and Liu, T.Z. (1997) Free radical and oxidative damage in human blood cells. J. Biomed. Sci., 4, 256–259.[CrossRef][Web of Science][Medline]

  26. Lang, K.S., Roll, B., Myssina, S., Schittenhelm, M., Scheel-Walter, H.G., Kanz, L., Fritz, J., Lang, F., Huber, S.M. and Wieder, T. (2002) Enhanced erythrocyte apoptosis in sickle cell anemia, thalassemia and glucose-6-phosphate dehydrogenase deficiency. Cell Physiol. Biochem., 12, 365–372.[CrossRef][Web of Science][Medline]

  27. Hendricks, C.A. and Engelward, B.P. (2004) Recombomice: the past, present, and future of recombination-detection in mice. DNA Repair, 3, 1255–1261.[Medline]

  28. Bishop, A.J. and Schiestl, R.H. (2000) Homologous recombination as a mechanism for genome rearrangements: environmental and genetic effects. Hum. Mol. Genet., 9, 2427–2334.[Abstract/Free Full Text]

  29. Gupta, P.K., Sahota, A., Boyadjiev, S.A., Bye, S., Shao, C., O'Neill, J.P., Hunter, T.C., Albertini, R.J., Stambrook, P.J. and Tischfield, J.A. (1997) High frequency in vivo loss of heterozygosity is primarily a consequence of mitotic recombination. Cancer Res., 57, 1188–1193.[Abstract/Free Full Text]

  30. Saintigny, Y., Rouillard, D., Chaput, B., Soussi, T. and Lopez, B.S. (1999) Mutant p53 proteins stimulate spontaneous and radiation-induced intrachromosomal homologous recombination independently of the alteration of the transactivation activity and of the G1 checkpoint. Oncogene, 18, 3553–3563.[CrossRef][Web of Science][Medline]

  31. Liang, L., Shao, C., Deng, L., Mendonca, M.S., Stambrook, P.J. and Tischfield, J.A. (2002) Radiation-induced genetic instability in vivo depends on p53 status. Mutat. Res., 502, 69–80.[Web of Science][Medline]

  32. Shao, C., Deng, L., Henegariu, O., Liang, L., Raikwar, N., Sahota, A., Stambrook, P.J. and Tischfield, J.A. (1999) Mitotic recombination produces the majority of recessive fibroblast variants in heterozygous mice. Proc. Natl Acad. Sci. USA, 96, 9230–9235.[Abstract/Free Full Text]

  33. Shao, C., Stambrook, P.J. and Tischfield, J.A. (2001) Mitotic recombination is suppressed by chromosomal divergence in hybrids of distantly related mouse strains. Nat. Genet., 28, 169–172.[CrossRef][Web of Science][Medline]

  34. Cervantes, R.B., Stringer, J.R., Shao, C., Tischfield, J.A. and Stambrook, P.J. (2002) Embryonic stem cells and somatic cells differ in mutation frequency and type. Proc. Natl Acad Sci. USA, 99, 3586–3590.[Abstract/Free Full Text]

  35. Schiestl, R.H., Khogali, F. and Carls, N. (1994) Reversion of the mouse pink-eyed unstable mutation induced by low doses of X-rays. Science, 266, 1573–1576.[Abstract/Free Full Text]

  36. Lebel, M. (2002) Increased frequency of DNA deletions in pink-eyed unstable mice carrying a mutation in the Werner syndrome gene homologue. Carcinogenesis, 23, 213–216.[Abstract/Free Full Text]

  37. Dare, E., Zhang, L.H. and Jenssen, D. (1996) Characterization of mutants involving partial exon duplications in the hprt gene of Chinese hamster V79 cells. Somat. Cell Mol. Genet., 22, 201–210.[CrossRef][Web of Science][Medline]

  38. Van Sloun, P.P., Wijnhoven, S.W., Kool, H.J., Slater, R., Weeda, G., van Zeeland, A.A., Lohman, P.H. and Vrieling, H. (1998) Determination of spontaneous loss of heterozygosity mutations in Aprt heterozygous mice. Nucleic Acids Res., 26, 4888–4894.[Abstract/Free Full Text]

  39. Shao, C., Yin, M., Deng, L., Stambrook, P.J., Doetschman, T. and Tischfield, J.A. (2002) Loss of heterozygosity and point mutation at Aprt locus in T cells and fibroblasts of Pms2–/– mice. Oncogene, 21, 2840–2845.[CrossRef][Web of Science][Medline]

  40. Sokal, R.R. and Rohlf, F.J. (1969) Biometrics. W.H. Freeman, San Francisco.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/5/615    most recent
ddi058v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Leder, A.
Right arrow Articles by Leder, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Leder, A.
Right arrow Articles by Leder, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?