Mammalian chimaeras have proved useful for investigating early steps in embryonic development. However, a complete clonal analysis of cell lineages has been limited by the lack of a marker which is ubiquitous and can distinguish parental cell types in situ. We have developed a cell marker system which fulfils these criteria. Chimaeric mice were successfully produced from two mouse species which possess sufficient genetic differences to allow unequivocal identification of parental cell types. DNA-DNA in situ hybridization with cloned, species-specific sequences was performed to distinguish the parental cell types. We have identified a cloned, Mus musculus satellite DNA sequence which shows hybridization differences between Mus musculus and Mus caroli DNA. This clone was used as a probe in in situ hybridizations to bone marrow chromosomes from Mus musculus, Mus caroli, and an interspecific F1 hybrid. The clone could qualitatively distinguish Mus musculus from Mus caroli chromosomes after in situ hybridization, even when they were derived from the same F1 hybrid cell. Quantitation of this hybridization to interphase nuclei from bone marrow spreads indicates that the probe can successfully distinguish Mus musculus from Mus caroli cells and can determine the percentage contribution of Mus musculus in mixtures of bone marrow cells of these species and in chimaeric bone marrow cell preparations.

Clonal analyses of embryonic lineages in mammals have been facilitated by the use of chimaeras. These animals are mosaics of two, or more, parental cell types, constructed by either blastocyst injection (reviewed by Gardner, 1978) or embryo aggregation (Tarkowski, 1961; Mintz, 1964). A cell marker specific for one parental cell type is required to trace the origin and fate of tissues and cell lineages in the development of these chimaeras. This marker must exist in variant forms which are (1) cell autonomous and not transferred between cells, (2) cell localized and never secreted into the extracellular environment, (3) selectively neutral so that their presence does not produce any adverse effects on developmental processes, (4) present and detectable in all cells throughout development, and (5) easily detected in histological sections (McLaren, 1976; Oster-Granite & Gearhart, 1981).

Many genetic markers have been developed for the analysis of mouse chimaeras, but few possess the property of being detectable in histological sections. The most widely used method for quantitating the mosaicism of chimaeric animals has been electrophoretic analysis of isozymes of glucose phosphate isomerase (GPI) (Chapman, Ansell & McLaren, 1972; Peterson, Frair & Wong, 1978). However, detection of these GPI differences by electrophoresis requires tissue homogenization and is limited to tissues which are readily isolated by dissection techniques. This property makes this system unsuitable for clonal analysis studies.

In situ markers which have been developed have all been limited to certain cell types in mouse chimaeras. Pigmentation phenotypes (reviewed by Mintz, 1974) are limited to skin, eye and inner ear. In addition, they can only provide an approximation of the patterns of mosaicism in certain regions where the pigment is extracellular. Colonization of hair follicles by several melanocytes can lead to ambiguous results, and the melanocyte phenotype is not always independent of its environment. The ichthyosis mutant shows altered nuclear morphology (Goldowitz & Mullen, 1982) which allows it to be distinguished from normal cells in neuronal tissues of chimaeras. Extension of this observation to other tissues and early embryos has not been demonstrated.

Activity variants of certain enzymes have been used as in situ cell markers. Differences in β-glucuronidase genetic activity have been used to analyse histological sections of chimaeric liver (Condamine, Custer & Mintz, 1971; West, 1976). This staining technique has been extended to several tissues, but is of limited usefulness because β-glucuronidase is transferred between cells and activity level differences are not widely detectable in all tissues (Feder, 1976). Similar studies of the central nervous system have been performed using β-galactosidase (BGS) activity variants (Dewey, Gervais & Mintz, 1976), but the use of BGS is limited to certain tissues where a two-fold difference in BGS activity is present.

Fluorescent staining of chimaeric Purkinje cell sections has been reported using an allozyme-specific antibody to GPI-1B (Oster-Granite & Gearhart, 1981). This technique shows great promise for clonal analyses of other cell types in the central nervous system. However, it appears that GPI expression may not reach equivalent levels in all tissues (Oster-Granite & Gearhart, 1980).

All of the existing cell markers are limited to particular stages of development or to specific classes of cells. Thus, no system has yet generated a cell marker which is both ubiquitous and can identify cells of either parental genotype in situ during all stages of development. Interspecific chimaeras provide an alternative approach to this problem, because of the greater diversity present between their genomes. Satellite DNA sequences show excellent promise as in situ cell markers because (1) they have diverged between various Mus species (Sutton & McCallum, 1972; Rice & Straus, 1973), (2) they are present in all nucleated cell populations, and (3) they are the most abundant sequences in the mouse genome (Britten & Kohne, 1968). Radiolabelled clones of these repetitive elements can be used as probes in in situ hybridizations to identify cells of a particular genotype. The abundance of these sequences in the Mus genome should facilitate their detection in situ.

A cloned, satellite DNA sequence of M. musculus has been used as a probe on bone marrow spreads from M. musculus and M. caroli. In situ hybridization has demonstrated that the clone can distinguish M. musculus from M. caroli chromosomes, even when they are derived from a single Fi hybrid cell. Hybridization of the clone to interphase nuclei can be detected and used to distinguish M. musculus and M. caroli cells in bone marrow mixtures as well as in chimaeric bone marrow preparations. Extension of this marker system to embryo and adult histological sections would generate a cell marker system which is both ubiquitous and able to distinguish parental cell types in situ (see Rossant, Vijh, Siracusa & Chapman, 1982). This marker system can then be used to generate a complete fate map of the embryo as well as to study cell-cell interactions during development.

Animals

M. musculus mice of the Ha/ICR random-bred strain and M. caroli mice originally trapped in Thailand were obtained from colonies maintained at Roswell Park Memorial Institute. M. musculus ↔M. caroli chimaeras were constructed by blastocyst injection of M. caroli inner cell mass into M. musculus blastocysts (Rossant & Freis, 1980). An interspecific hybrid male was obtained by mating a chimaeric M. musculus ↔M. caroli female with an Ha/ICR male (Rossant & Chapman, 1982).

Mus musculus repetitive DNA library

A DNA library was constructed by inserting repetitive M. musculus DNA sequences (average size 160 base pairs) into plasmid pBR322 and transforming E. coli strain X1776 (Pietras, 1981, 1982). All clones were numbered and given the prefix pMR (plasmid mouse repetitive).

DNA extractions

Genomic DNA was extracted from adult mouse livers. Tissues were homogenized on ice in buffer containing 5 % sucrose, 50mM-Tris-HCl, 24 mM-KC1 with concentrations of spermine, spermidine, EDTA and EGTA as outlined by Burgoyne, Waqar & Atkinson (1970). Nuclei were pelleted by layering the homogenization mixture on a 20% sucrose buffer followed by low-speed centrifugation. Nuclei were resuspended in 50 mM-Tris-base, 0-15 M-NaCl, 0-1 M -EDTA, pH 10, lysed by addition of NaDodSO4 to 0 ·8%, and treated with Ribonuclease A (100 μg/ml, Boehringer Mannheim) and pronase (100 μg/ml, P-L Biochemicals). DNA was phenol : chloroform : isoamyl alcohol (25 :24:1) extracted, precipitated, spooled, and dissolved (Pellicer, Wigler, Axel & Silverstein, 1978).

Purified plasmid DNA was extracted from selected clones by CsCl densitygradient centrifugation following chloramphenicol amplification of plasmids (Clewell & Helinski, 1969).

Nick translation of DNA

Recombinant plasmid DNA as well as M. musculus and M. caroli genomic DNA were labelled in vitro by the nick translation protocol of Rigby, Dieckman, Rhodes & Berg (1977). [α—32P]TTP (400Ci/mmol, Amersham) was used to generate probes for Grunstein-Hogness colony filter hybridizations and Southern blots. [3H]TTP (56 ·8Ci/mmol, New England Nuclear) was used to generate probe for in situ hybridizations. The resulting labelled DNA probes had specific activities of 1-1 ·5 x 108dpm//igand 1 ·7—3 ·5 × 107dpm/μg respectively.

Colony filter hybridizations

Colonies from the M. musculus repetitive DNA library were grown on nitrocellulose filters, lysed, and the DNA denatured and fixed (Grunstein & Hogness, 1975). The filters were prehybridized in 4 ×SSC (1 × SSC is 0 ·15 M-NaCl/0 ·015M-Na citrate), 1 x Denhardt’s solution (Denhardt, 1966), 0-1% NaDodSO4, 0 ·1 % Na pyrophosphate and 150/ig/ml denatured salmon sperm DNA for 2-4 h at 60 °C. M. musculus and M. caroli total liver DNA was 32P-labelled by nick translation and used as probes; these were added to separate filters and hybridized overnight. Filters were washed in 1 M-NaCl, 0 ·1% NaDodSO4, 0T % Na pyrophosphate, and 1 × Denhardt’s solution at 60°C. Autoradiographs were developed after exposure to Kodak XAR-5 film at -70 °C with one intensifying screen.

Southern blot analysis

Restriction endonucleases Alul and EcoRII were obtained from Boehringer Mannheim. Conditions for genomic DNA restrictions were those recommended by the suppliers. Restricted DNA from M. musculus and M. caroli was subjected to electrophoresis in 1 % agarose gels using a 36mM-Tris, 1 mM-EDTA, 30 mM-NaH2PO4, pH 7 ·4, buffer. The DNA was denatured, transferred to (Schleicher and Schuell No 85) nitrocellulose (Southern, 1975b) and prehybridized in 2 × SSC, 0 ·1 % NaDodSO4, 0 ·1 % Na pyrophosphate and 150 μg/ml herring sperm DNA for 4 h at 60 °C. 32P-labelled pMR196 was added and hybridized overnight. Filters were washed in 2 × SSC, 0 ·1 % NaDodSO4 and 0 ·1 % Na pyrophosphate at 60 °C, and radiolabelled bands visualized by autoradiography as described for colony filter hybridizations.

Bone marrow preparations

Nuclei and metaphase chromosomes, used for in situ hybridizations, were prepared from bone marrow. Each mouse was given an intraperitoneal injection of demecolcine (Sigma) in physiological saline at a dose of 2 μg/g body weight. The mice were killed one hour after treatment; femurs were dissected and the bone marrow cells flushed out with sterile phosphate-buffered saline, pH 7 ·4. The cells were pelleted gently and subsequently swollen in hypotonic solution (75 mM-KCl). They were fixed in methanol : acetic acid (3:1) and dropped onto cold, wet, chromic-acid-cleaned slides. The slides were allowed to air dry completely prior to use.

In situ hybridization

The in situ hybridization techniques of Jones (1970) and Pardue & Gall (1970) were modified as follows: slides were passed through an ethanol dehydration series (70 %, 70 %, 90 %, 100 % EtOH) prior to heat denaturation in 2 ×SSC at 70°C for 30min. This was followed by another dehydration series and then alkali denaturation in 0 ·07 N-NaOH for 3 min. Slides were again dehydrated and allowed to air dry completely before application of 20 μl of a hybridization mixture which contained 2 ×SSC, 10 % dextran sulphate (Wahl, Stern & Stark, 1979), 1 μg yeast tRNA as carrier, and 7 × 105dpm of 3H-labelled pMR196 probe. Coverslips (18 × 18 mm) were applied and the slides placed in covered Petri dishes with 10 ml of 2 × SSC in the bottom to keep the ‘chamber’ moist. Dishes were sealed with Parafilm and placed in a 60 °C incubator overnight. The slides were washed 5 times, 30 min per wash, in 2 x SSC at 60 °C followed by at least 2 washes, 60min each, in 2 ×SSC at room temperature to remove non-specifically bound, radioactive probe. Coverslips were gently removed during the first wash. This was followed by an ethanol dehydration series prior to the application of Kodak NTB-2 emulsion which had been diluted 1:1 with water. Lightproof boxes containing desiccant were used to hold slides at 4 °C until appropriate exposure times had been reached. Slides were developed at 17-20 °C in Kodak D-19 developer for 2 ·5 min, rinsed in distilled water, and fixed in Kodak rapid fixer for 5 min. Slides were stained with Giemsa.

Isolation of Mus caroli highly repeated DNA sequences

Total M. caroli DNA was nick translated with 3H-TTP, to a specific activity of 2 –3 ×107 dpm/ μg, ethanol precipitated, dissolved in water, and denatured in 0 ·1 N-NaOH. This solution was adjusted to 0 ·2M-phosphate buffer (equimolar amounts of NaH2PO4 and Na2HPO4) and immediately fractionated by hydroxylapatite (Bio-Rad, DNA grade) column chromatography at 65 °C (Campo & Bishop, 1974). Under these conditions single-stranded DNA was separated from foldback sequences, which had renatured, by elution with 0 ·12M-phosphate buffer. Single-stranded fractions were pooled and allowed to reanneal at 65 °C (in 0 ·18M-Na+) to a Cot value of 7 × 10−3 mole sec/1; at this Cot, most highly repeated sequences were renatured. The reannealed DNA was fractionated by hydroxylapatite chromatography. The double-stranded fractions, which contained highly repeated sequences, were eluted with 0 ·4M-phosphate buffer. These fractions were pooled and run on a Sephadex-Chelex column (Campo & Bishop, 1974) to remove phosphate. Radioactive fractions were pooled, ethanol precipitated and dissolved in water.

Slides hybridized with the M. caroli probe received 20 μl of hybridization mix containing 3 ·5 × 104dpm M. caroli repetitive DNA. The amount of M. caroli probe used was 20 times less than that used when pMR196 was the probe in in situ hybridizations. This resulted in comparable amounts of radioactive repetitive sequences used on each slide, since the M. musculus-derived insert of clone pMR196 is only 5 % of the total recombinant plasmid DNA.

Species comparison of repetitive DNAs

An M. musculus repetitive DNA library (Pietras, 1981,1982) was screened by Grunstein-Hogness colony-filter hybridization to identify clones that showed a high degree of hybridization to M. musculus DNA (Fig. IB). The intensity of the autoradiographic signals produced is roughly proportional to a cloned sequence’s repetition number in the M. musculus genome (Pietras, 1981). Clone pMR196 exhibited the strongest signal of any clone after hybridization to total M. musculus DNA (Fig. IB) indicating that it was the most repeated sequence in the M. musculus genome which could be found in this DNA library. Liquid hybridization of clone pMR196 with total M. musculus DNA (Cot analysis) has shown that the insert is repeated at least 7 × 105 times per haploid genome in M. musculus (Pietras, 1981).

A comparative colony-filter hybridization was performed to reveal clones which showed minimal hybridization to M. caroli DNA. Five clones (pMR 124, 150,196, 238 & 286) of M. musculus origin showed intense hybridization to M. musculus DNA and minimal hybridization to M. caroli DNA. These five clones contain DNA sequences which are either less repeated or have diverged in the M. caroli genome. Clone pMR196 was chosen for further study because it exhibited striking differences in hybridization between M. musculus and M. caroli DNA and because of its great abundance in the M. musculus genome.

Southern analysis of clone pMRl 96

A Southern analysis was performed to further characterize clone pMR196 (Fig. 2). The restriction pattern produced by M. musculus DNA after hybridization to pMR196 is identical to that generated by CsCl-isolated satellite DNA which had been restricted with EcoRII or AluI (Southern, 1975a; Horz & Zachau, 1977). The type A (EcoRII) and type B (AluI) patterns are evident in lanes 1 and 2 respectively. Lanes 3, 4 and 5 show a different restriction pattern for M. caroli DNA hybridized with pMR196, indicating that sequences similar to, but diverged from, this particular M. musculus sequence are present in M. caroli DNA.

The Southern blot analysis provides a sensitive quantitation of the differences in hybridization of pMR196 to M. musculus and M. caroli DNA. The intensity of the autoradiographic signals is proportional to the amount of hybridization of the probe to restricted genomic DNA present in each lane. A comparison of the autoradiographic signals shows that the total amount of hybridization of the probe to 0 ·1 μg M. musculus DNA (lanes 1 and 2) is intermediate between the amount of hybridization seen when 1 and 10 μg of M. caroli DNA are used (lanes 4 and 5 respectively). Thus M. caroli DNA exhibits between 1 % and 10 % the amount of hybridization to pMR196 seen in M. musculus under these hybridization conditions. These results suggested that similar differences should be detected in in situ hybridizations of pMR196 to M. musculus and M. caroli cells.

In situ hybridization of clone pMR196 to metaphase chromosomes

Metaphase chromosome spreads, present in bone marrow preparations from M. musculus, M. caroli and an interspecific F1 hybrid, were hybridized in situ with nick-translated pMR196 (Fig. 3). The centromeres of most M. musculus chromosomes showed intense labelling (Fig. 3A). This pattern of hybridization is characteristically observed when mouse satellite DNA, which has been isolated from main band DNA by CsCl density gradients, is hybridized in situ to chromosome spreads (Jones, 1970; Pardue & Gall, 1970). The probe hybridized to a very limited extent with M. caroli chromosomes (Fig. 3B), as expected from the results of the colony filter hybridization (Fig. 1) and the Southern blot analysis (Fig. 2). Approximately half the chromosomes of the F1 hybrid showed intense labelling, and this labelling was specifically limited to centromeric regions (Fig. 3C). These results indicate that the probe qualitatively distinguished chromosomes of these two mouse species, even when they were derived from a single F1 hybrid cell.

Fig. 1.

Colony filter hybridization of the 129 clones from the M. musculus repetitive DNA library probed with 32P-labelled (A) M. caroli and (B) M. musculus male DNA. The arrowed and circled clones hybridized differently to DNA from these two mouse species.

Fig. 1.

Colony filter hybridization of the 129 clones from the M. musculus repetitive DNA library probed with 32P-labelled (A) M. caroli and (B) M. musculus male DNA. The arrowed and circled clones hybridized differently to DNA from these two mouse species.

Fig. 2.

Southern blot analysis of M. musculus and M. caroli DNA probed with clone pMR196. Lane (1) 0 ·1 μg M. musculus DNA, Eco RII restricted; (2) 0 ·1 μg M. musculus DNA, Alul restricted. Lanes (3), (4) and (5) have 0 ·1 μg, 1 μg and 10 μg M. caroli DNA, Alul restricted. Arrows indicate an ascending series of multimers based on a monomer length of 240 bp. ◯indicates the origin of the gel.

Fig. 2.

Southern blot analysis of M. musculus and M. caroli DNA probed with clone pMR196. Lane (1) 0 ·1 μg M. musculus DNA, Eco RII restricted; (2) 0 ·1 μg M. musculus DNA, Alul restricted. Lanes (3), (4) and (5) have 0 ·1 μg, 1 μg and 10 μg M. caroli DNA, Alul restricted. Arrows indicate an ascending series of multimers based on a monomer length of 240 bp. ◯indicates the origin of the gel.

Fig. 3.

In situ hybridizations to bone marrow preparations from (A) M. musculus male probed with pMR196, (B) M. caroli male probed with pMR196, (C) interspecific F1 hybrid probed with pMR196, (D) M. musculus/M. caroli mixture probed with pMR196, (E) M. musculus male probed with M. caroli repetitive DNA, and (F) M. caroli male probed with M. caroli repetitive DNA. All slides were developed after an 8-day exposure. Scale bar: 20μm.

Fig. 3.

In situ hybridizations to bone marrow preparations from (A) M. musculus male probed with pMR196, (B) M. caroli male probed with pMR196, (C) interspecific F1 hybrid probed with pMR196, (D) M. musculus/M. caroli mixture probed with pMR196, (E) M. musculus male probed with M. caroli repetitive DNA, and (F) M. caroli male probed with M. caroli repetitive DNA. All slides were developed after an 8-day exposure. Scale bar: 20μm.

In situ hybridization of clone pMR196 to bone marrow nuclei

To be a useful in situ cell marker, the probe should abundantly label M. musculus interphase nuclei and show negligible hybridization to M. caroli nuclei. The hybridization of pMR196 to interphase nuclei was examined in the bone marrow preparations used for chromosome spreads. A strong autoradiographic signal over M. musculus nuclei was observed and minimal hybridization of the probe was seen to M. caroli nuclei (Fig. 3A, B). In addition, M. musculus nuclei showed areas with heavy concentrations of grains. This pattern of hybridization is indicative of satellite DNA, which shows concentrated labelling to the chromocentres of interphase nuclei (Jones, 1970; Pardue & Gall, 1970).

Quantitation of the hybridization of pMR196 to bone marrow preparations from M. musculus and M. caroli was performed by counting the number of grains over interphase nuclei (Fig. 4). There is a linear increase in average grain numbers as a function of exposure time for M. musculus nuclei. This increase in average grain numbers is consistent with a high amount of specific hybridization. No significant labelling of M. caroli nuclei could be detected. M. caroli nuclei averaged low grain numbers, less than seven per nucleus, even after a 16-day exposure of the emulsion. These results demonstrate that the difference in grain numbers between M. musculus and M. caroli nuclei was significant (P< 0 ·001) and readily detectable in both experiments after only a 5-day exposure. Although the extent of hybridization to M. musculus preparations varied between experiments, the difference between M. musculus and M. caroli ranged from approximately 10-fold in expt. 1 to 50-fold in expt. 2 (Fig. 4).

Fig. 4.

Quantitative analysis of in situ hybridization to interphase nuclei. Bone marrow preparations from M. musculus and M. caroli males were hybridized with equal counts of 3H-labelled pMR196 and developed after 2 –16 days exposure. Silver grains were counted over a minimum of 100 nuclei present in at least five separate microscopic fields on each slide. This graph shows averaged grain numbers from two experiments, expt. 1 •Af caroli vs. ◯ M. musculus’, expt. 2 M. caroli vs △ M. musculus.

Fig. 4.

Quantitative analysis of in situ hybridization to interphase nuclei. Bone marrow preparations from M. musculus and M. caroli males were hybridized with equal counts of 3H-labelled pMR196 and developed after 2 –16 days exposure. Silver grains were counted over a minimum of 100 nuclei present in at least five separate microscopic fields on each slide. This graph shows averaged grain numbers from two experiments, expt. 1 •Af caroli vs. ◯ M. musculus’, expt. 2 M. caroli vs △ M. musculus.

Quantitative analysis of grain distributions of interphase nuclei

Grain distributions from slides which showed the greatest hybridization differences (Fig. 4, expt. 2, 8 days exposure) were plotted as histograms (Fig. 5A). There is no overlap in the number of grains overlying M. musculus and M. caroli nuclei. Greater than 99% of M. caroli nuclei had less than four grains each, indicating a very low level of hybridization or background labelling. Less than 4% of M. musculus nuclei had grain numbers between 8 and 23; this small percentage of cells might be detected as M. caroli nuclei if the overall extent of hybridization of M. musculus nuclei were reduced.

Fig. 5.

The frequency distribution of grains/nucleus from in situ hybridizations of pMR196 to bone marrow nuclei. (A) M. caroli □ and M. musculus ▀. Both slides were exposed 8 days with Kodak NTB-2 emulsion. These slides are shown in Figs. 3B and 3A respectively. (B) M. musculus/M. caroli mixture exposed 8 days with Kodak NTB-2 emulsion. This slide is shown in Fig 3D. (C) M. musculus⟷M. caroli chimaera exposed 5 days with Kodak NTB-3 emulsion.

Fig. 5.

The frequency distribution of grains/nucleus from in situ hybridizations of pMR196 to bone marrow nuclei. (A) M. caroli □ and M. musculus ▀. Both slides were exposed 8 days with Kodak NTB-2 emulsion. These slides are shown in Figs. 3B and 3A respectively. (B) M. musculus/M. caroli mixture exposed 8 days with Kodak NTB-2 emulsion. This slide is shown in Fig 3D. (C) M. musculus⟷M. caroli chimaera exposed 5 days with Kodak NTB-3 emulsion.

Clone pMR196 was hybridized in situ to M. musculus and M. caroli bone marrow mixtures to determine whether this probe could detect the percentage contribution of M. musculus cells in these mixtures (Fig. 3D, 5B). An aliquot of this mixture was analysed for phosphoglycerate kinase, since these two mouse species express different PGK isozymes (Chapman & Shows, 1976). Results of the PGK analysis indicated that approximately 75 % of the mixture was of M. musculus origin. The grain distribution clearly delineates two classes of nuclei present in the mixture, one showing 0–15 grains/nucleus and the other exhibiting 24-80+ grains/nucleus. If the latter class of cells represents the M. musculus contribution, then 71 % of this mixture is M. musculus in origin; this agrees well with the PGK analysis.

The grain distribution for bone marrow nuclei from an M. musculus ⟷ M. caroli chimaera was also divided into two classes by the probe (Fig. 5C). In situ hybridization indicated a 29% M. musculus contribution to this tissue. An isozyme analysis of this preparation was not available.

In situ hybridization of Mus caroli repetitive DNA to metaphase chromosomes

A probe specific for M. caroli DNA would be useful for clonal analyses in M. musculus ⟷M. caroli chimaeras where M. caroli cells are in the minority. Preliminary results of in situ hybridization of radioactive M. caroli highly repeated sequences to M. musculus and M. caroli chromosome spreads (Fig. 3E,F) indicated that the M. caroli probe showed abundant hybridization to M. caroli chromosomes (Fig. 3F) and minimal labelling of M. musculus chromosomes (Fig. 3E). The hybridization of this DNA was not limited to centromeric regions of M. caroli chromosomes.

A cell marker system has been developed which has the potential to be used for clonal analyses of cell lineages in Mus musculus ⟷Mus caroli chimaeras (see Rossant et al. 1982). This marker system depends on the use of cloned, highly repetitive Mus musculus DNA sequence as a probe in in situ hybridizations to detect Mus musculus chromosomes and interphase nuclei in bone marrow preparations. The use of a cloned DNA sequence as a probe ensures that only a single family of sequences will be consistently detected in in situ hybridizations. The fidelity of in situ hybridizations to metaphase chromosomes, using clone pMR196 as a probe, has been demonstrated. Nick-translated pMR196 showed centromerically localized, intense hybridization to most Mus musculus chromosomes (Fig. 3A) and no hybridization to Mus caroli chromosomes (Fig. 3B). The fact that the probe only labelled half the chromosomes of an F1 hybrid cell, and that this hybridization was limited to the centromeric regions of those chromosomes (Fig. 3C), provides additional evidence that pMR196 detects only Mus musculus chromosomes, even when they are present in hybrid cells.

In addition, the specificity observed in the in situ chromosomal hybridizations was maintained in hybridizations of the probe to interphase nuclei (Fig. 3A,B,D). This result is of prime importance because it establishes the usefulness of the probe in distinguishing interphase nuclei in situ. The ability to detect hybridization of the probe to interphase nuclei suggests that it can be used to identify Mus musculus cells in a variety of tissues and that it can provide a basis for clonal analyses in Mus musculus⟷Mus caroli chimaeras (see Rossant et al. 1982).

Quantification of hybridization to bone marrow nuclei revealed significant (P < 0·001) differences in the number of grains between Mus musculus and Mus caroli interphase nuclei after only a 5-day exposure (Fig. 4). These differences were used to determine the percentage contribution of Mus musculus cells to bone marrow mixtures and chimaeric bone marrow spreads. The percentage contribution determined by in situ hybridization was similar to that established by electrophoretic analysis of PGK or GPI isozymes. Thus in situ hybridization using pMR196 as a probe has the potential of identifying Mus musculus cells in haematopoietic lineages, even if they contribute only a small percentage to that tissue.

The use of species-specific, repetitive DNA sequences as probes in in situ hybridizations fulfils most of the criteria described by McLaren (1976) and Oster-Granite & Gearhart (1981) for an ideal cell marker in chimaeras. The marker does exist in variant forms which are (1) cell autonomous - this DNA sequence is not transferred between mouse cells, (2) cell localized - these sequences are not transcribed into RNA (Flamm, Walker & McCallum, 1969; Pietras, 1981) and they are not secreted from cells. This feature ensures that hybridization is not only cell localized, but that it is limited to the nucleus of the cell, (3) selectively neutral, Mus musculus ⟷Mus caroli chimaeras develop normally, show balanced mosaicism, and are fully fertile (Rossant & Freis, 1978; Rossant & Chapman, 1982), (4) it is present and should be detectable in all nucleated progeny of Mus musculus cells in interspecific chimaeras because it is part of the chromosomal complement of each cell, and is mitotically transmitted, and (5) it is easily detectable in histological sections (see Rossant et al. 1982).

Several lines of evidence lead to the conclusion that the Mus mus cuius-derived insert of clone pMR196 is a satellite sequence. First, hybridization of the probe to Mus musculus chromosomes is centromerically localized and detectable on most chromosomes (Fig. 3A). This is consistent with published reports of CsCl-isolated satellite DNA which had been hybridized in situ to mouse chromosomes (Jones, 1970; Pardue & Gall, 1970). Secondly, a Cot analysis indicated that the insert was present 7 × 105 times per haploid genome in Mus musculus (Pietras, 1981); previously, a copy number of 106 had been established (Waring & Britten, 1966), using satellite DNA isolated on CsCl gradients. Thirdly, a Southern blot analysis of restriction patterns revealed by hybridization of the probe to restricted Mus musculus DNA showed a ladder of 240bp multimers (Fig. 2). This pattern is identical to the type A (EcoRII, lane 1) (Southern, 1975a) and type B (Alul, lane 2) (Horz & Zachau, 1977) restriction patterns of Mus musculus satellite DNA. These data strongly suggest that the repetitive DNA insert in clone pMR196 is a Mus musculus major satellite DNA sequence.

The evolutionary divergence of satellite sequences in the Mus genome (Sutton & McCallum, 1972; Rice & Straus, 1973) makes these sequences especially suited for use as in situ probes in M. musculus ⟷M. caroli chimaeras. Their abundance in the genome makes them very’easy to detect with in situ hybridization techniques. Although the absolute amount of hybridization to each slide showed some variation (Fig. 4), the differences in amount of hybridization to Mus musculus and Mus caroli cells is clearly maintained at exposures greater than 5 days. Sequences which are less abundant in the genome or exhibit less divergence between mouse species would be very difficult to use as in situ markers. No in situ marker of this type could be developed for use with different mouse strains because of the similarities of their genomes.

Another useful advantage of the divergence of satellite sequences in the Mus genome is that a reciprocal marker system, specific for Mus caroli cells, can be similarly developed. Preliminary data have already demonstrated the hybridization specificity of Mus caroli highly repeated sequences to Mus caroli chromosomes (Fig. 3E,F). Screening of a Mus caroli repetitive DNA library by Grunstein-Hogness colony filter hybridization should expedite identification of any highly repeated, Mus caroli DNA sequence which shows little homology to Mus musculus DNA. This probe will be useful for chimaeric analyses where Mus caroli cells are in the minority and also to study the nature of repetitive DNA in this species.

We wish to thank Mary Bodis, Mary Byers, Lesley Forrester and Patricia Ryan for excellent technical assistance, Dr Thomas Shows for the generous use of his Zeiss photomicroscope, and Gail Walsh for typing the manuscript.

This work was supported by grants from the General Medical Sciences of N.I.H. (V.M.C. and N.D.H.) and the Canadian Natural Sciences and Engineering Research Council (J.R.).

Britten
,
R. J.
&
Kohne
,
D.
(
1968
).
Repeated sequences in DNA
.
Science
161
,
529
540
.
Burgoyne
,
L. A.
,
Waqar
,
M. A.
&
Atkinson
,
M. R.
(
1970
).
Calcium-dependent priming of DNA synthesis in isolated rat liver nuclei
.
Biochem. Biophys. Res. Comm
.
39
,
254
259
.
Campo
,
M. S.
&
Bishop
,
J. O.
(
1974
).
Two classes of messenger RNA in cultured rat cells: repetitive sequence transcripts and unique sequence transcripts
.
J. molec. Biol
.
90
,
649663
.
Chapman
,
V. M.
,
Ansell
,
J. D.
&
McLaren
,
A.
(
1972
).
Trophoblast giant cell differentiation in the mouse: expression of glucose phosphate isomerase (GPI-1) electrophoretic variants in transferred and chimeric embryos
.
Devl Biol
.
29
,
48
54
.
Chapman
,
V. M.
&
Shows
,
T. B.
(
1976
).
Somatic cell genetic evidence for X-chromosome linkage of three enzymes in the mouse
.
Nature
259
,
665
667
.
Clewell
,
D.B.
&
Helinski
,
D.R.
(
1969
).
Supercoiled circular DNA-protein complex in Escherichia coli: purification and induced conversion to an open circular DNA form
.
Proc, natn. Acad. Sci., U.S.A
.
62
,
1159
1166
.
Condamine
,
H.
,
Custer
,
R. P.
&
Mintz
,
B.
(
1971
).
Pure-strain and genetically mosaic liver tumors histochemically identified with the β-glucuronidase marker in allophenic mice
.
Proc. natn. Acad. Sci., U.S.A
.
68
,
2032
2036
.
Denhardt
,
D.T.
(
1966
).
A membrane-filter technique for the detection of complementary DNA
.
Biochem. Biophys. Res. Comm
.
23
,
641
646
.
Dewey
,
M. J.
,
Gervais
,
A. G.
&
Mintz
,
B.
(
1976
).
Brain and ganglion development from two genotypic classes of cells in allophenic mice
.
Devl Biol
.
50
,
68
81
.
Feder
,
N.
(
1976
).
Solitary cells and enzyme exchange in tetraparental mice
.
Nature
263
,
67
69
.
Flamm
,
W. G.
,
Walker
,
P. M. B.
&
McCallum
,
M.
(
1969
).
Some properties of the single strands isolated from the DNA of the nuclease satellite of the mouse (Mus musculus)
.
J. molec. Biol
.
40
,
423
443
.
Gardner
,
R. L.
(
1978
).
Production of chimeras by injecting cells or tissue into the blastocyst
.
In Methods in Mammalian Reproduction
(ed.
J. C.
Daniel
), pp.
137
166
.
New York
:
Academic Press
.
Goldowitz
,
D.
&
Mullen
,
R. J.
(
1982
).
Nuclear morphology of ichthyosis mutant mice as a cell marker in chimeric brain
.
Devl Biol
.
89
,
261
267
.
Grunstein
,
M. G.
&
Hogness
,
D. S.
(
1975
).
Colony hybridization: a method for the isolation of cloned DNAs that contain a specific gene
.
Proc. natn. Acad. Sci., U.S.A
.
72
,
3961
3965
.
Horz
,
W.
&
Zachau
,
H. G.
(
1977
).
Characterization of distinct segments in mouse satellite DNA by restriction nucleases
.
Eur. J. Biochem
.
73
,
383
392
.
Jones
,
K. W.
(
1970
).
Chromosomal and nuclear location of mouse satellite in individual cells
.
Nature
225
,
912
915
.
McLaren
,
A.
(
1976
).
Mammalian Chimaeras
.
Cambridge
:
Cambridge University Press
.
Mintz
,
B.
(
1964
).
Formation of genetically mosaic mouse embryos, and early development of’lethal (t12/t12)-normal’ mosaics
.
J. exp. Zool
.
157
,
273
292
.
Mintz
,
B.
(
1974
).
Gene control of mammalian differentiation
.
Ann. Rev. Genet
.
8
,
411
470
.
Oster-Granite
,
M.L.
&
Gearhart
,
J.
(
1980
).
Immunofluorescence and histochemical localization of glucosephosphate isomerase in neural tissue
.
J. Histochem. Cytochem
.
28
,
250
254
.
Oster-Granite
,
M. L.
&
Gearhart
,
J.
(
1981
).
Cell lineage analysis of cerebellar Purkinje cells in mouse chimeras
.
Devl Biol
.
85
,
199
208
.
Pardue
,
M. L.
&
Gall
,
J. G.
(
1970
).
Chromosomal localization of mouse satellite DNA
.
Science
168
,
1356
1358
.
Pellicer
,
A.
,
Wigler
,
M.
,
Axel
,
R.
&
Silverstein
,
S.
(
1978
).
The transfer and stable integration of the HSV thymidine kinase gene into mouse cells
.
Cell
14
,
133
141
.
Peterson
,
A. C.
,
Frair
,
P. M.
&
Wong
,
G. G.
(
1978
).
A technique for detection and relative quantitative analysis of glucose phosphate isomerase isoenzymes from nanogram tissue samples
.
Biochem. Genet
.
16
,
681
690
.
Pietras
,
D.F.
(
1981
).
Cloning and characterization of mouse repetitive DNA sequences
.
Ph.D. Thesis, State University of N.Y. at Buffalo
.
Pietras
,
D. F.
,
Siracusa
,
L. D.
,
Chapman
,
V. M.
,
Gross
,
K. W.
,
Bennett
,
K. L.
,
Kane-Haas
,
C.
&
Hastie
,
N. D.
(
1982
).
Construction of a Mus musculus repetitive DNA library: Identification of a new satellite sequence in Mus musculus (submitted)
.
Rice
,
N. R.
&
Straus
,
N. A.
(
1973
).
Relatedness of mouse satellite deoxyribonucleic acid to deoxyribonucleic acid of various Mus species
.
Proc. natn. Acad. Sci., U.S.A
.
70
,
3546
3550
.
Rigby
,
P.
,
Dieckman
,
M.
,
Rhodes
,
C.
&
Berg
,
P.
(
1977
).
Labelling deoxyribonucleic acid to high specific activity in vitro by nick-translation with DNA polymerase I
.
J. molec. Biol
.
113
,
237
251
.
Rossant
,
J.
&
Chapman
,
V. M.
(
1982
).
Somatic and germ-line mosaicism in interspecific chimaeras between Mus musculus and Mus caroli
.
J. Embryol. exp. Morph, (in press)
.
Rossant
,
J.
&
Frels
,
W. I.
(
1980
).
Interspecific chimeras in mammals: successful production of live chimeras between Mus musculus and Mus caroli
.
Science
208
,
419
421
.
Rossant
,
J.
,
Vijh
,
M.
,
Siracusa
,
L. D.
&
Chapman
,
V. M.
(
1982
).
Identification of embryonic cell lineages in histological sections of M. musculus ++M. caroli chimaeras
.
J. Embryol. exp. Morph, (in press)
.
Southern
,
E. M.
(
1975a
).
Long range periodicities in mouse satellite DNA
.
J. molec. Biol
.
94
,
51
69
.
Southern
,
E. M.
(
1975b
).
Detection of specific sequences among DNA fragments separated by gel electrophoresis
.
J. molec. Biol
.
98
,
503
517
.
Sutton
,
W. D.
&
McCallum
,
M.
(
1972
).
Related satellite DNAs in the genus Mus
.
J. molec. Biol
.
71
,
633
656
.
Tarkowski
,
A. K.
(
1961
).
Mouse chimaeras developed from fused eggs
.
Nature (London)
190
,
857
860
.
Wahl
,
G. M.
,
Stern
,
M.
&
Stark
,
G. R.
(
1979
).
Efficient transfer of large DNA fragments from agarose gels to diazobenzyloxymethyl-paper and rapid hybridization by using dextran sulfate
.
Proc. natn. Acad. Sci., U.S.A
.
76
,
3683
3687
.
Waring
,
M.
&
Britten
,
R. J.
(
1966
).
Nucleotide sequence repetition: a rapidly reassociating fraction of mouse DNA
.
Science
154
,
791
794
.
West
,
J. D.
(
1976
).
Patches in the livers of chimaeric mice
.
J. Embryol. exp. Morph
.
36
,
151
161
.