Identification of Human RNA Transcripts Among Heterogeneous

Proc. Nat. Acad. Sci. USA Vol. 72, No. 5, pp. 1868-1872, May 1975

Identification of Human RNA Transcripts among Heterogeneous Nuclear RNA from Man-Mouse Somatic Cell Hybrids (human X chromosome/repetitive DNA sequences/DNA- RNA hybridization) R. A. DI CIOCCIO*, R. VOSS*, M. KRIM*, K. H. GRZESCHIKt, AND M. SINISCALCO* * Sloan-Xettering Institute for Cancer Research, New York, N.Y. 10021 and t Institute for Human Genetics, University of Mflnster, Munster, Germany Communicated by James E. Darnell, Jr., February 14, 1975

ABSTRACT In a man-mouse hybrid line from our X-linked in man (7). Consequently, when phosphoribosyl- cell library, the only cytological detectable portion of the cells are fused with phosphoribo- human genome is the X chromosome, and the only genetic transferase-negative mouse markers regularly expressed are coded by genes known to syltransferase-positive human cells, the resulting hybrid clones be X-linked. A component of the heterogeneous nuclear grown in HAT medium may lose every part of the human RNA of these cells was found to be complementary to genome but the X chromosome, or at least the portion of it repetitive human DNA sequences by means of RNA-DNA bearing the locus for the human enzyme (8, 9). hybridization on nitrocellulose filters. The same procedure also permitted the identification of hybrid cell The identification of human RNA transcripts in man- DNA sequences that are complementary to human hetero- mouse somatic cell hybrids of the type described above is geneous nuclear RNA. This experimental approach, based on RNA- DNA hybridization. Molecular hybrids coupled with hybridization studies in situ, is expected to formed by annealing RNA from A9/HRBC2 cells with human yield critical data on the distribution and the specificity to allow detection for human RNA syn- of the repetitive DNA sequences present in the human DNA are expected genome and to provide a new tool for cytological mapping thesized in A9/HRBC2 cells. Likewise, molecular hybrids ob- of human chromosomes. tained by annealing RNA from human cells with DNA from A9/HRBC2 should detect the human DNA carried by this Highly unstable somatic cell hybrids offer a unique op- hybrid cell line. To be meaningful, the levels of molecular portunity for basic experimentation of molecular biology at hybridization obtained in such experiments should be con- the level of single human chromosomes. The Sendai virus- stantly higher than those between the RNA of the parental mediated fusion between somatic cells of human and rodent mouse cells with human DNA and vice versa, since these origin usually yields interspecific somatic cell hybrids which measure the amount of interspecific crosshybridization. undergo the progressive loss of the human genome (1-5). Soeiro and Darnell (10) have demonstrated that hybridization When the rodent parental cell line is deficient in a metabolic between L-cell hnRNA and HeLa cell DNA and vice versa is function that is essential for survival in a given selective negligible. For this reason, hnRNA was chosen as experi- medium, these types of somatic cell hybrids evolve rapidly mental material. While this work was in progress, Coon et al. towards a reduced karyotype where the only component of (11) found that complementary RNA of human mitochondrial the human genome retained is the chromosome (or chromo- or nuclear DNA does not significantly hybridize with rodent some portion) carrying the human gene that codes for the DNA and vice versa. essential metabolic function. The present report describes the rationale and the method- MATERIALS AND METHODS ologies devised for the identification of human RNA tran- The A9/HRBC2 hybrid line was derived from the man- scripts among the heterogeneous nuclear RNA (hnRNA) mouse hybrid line established by Miller et al. (8) through the isolated from a man-mouse somatic cell hybrid line (A9/ fusion of hypoxanthine phosphoribosyltransferase-positive HRBC2) that has lost almost entirely the parental human human male diploid cells (HRBC2) with mouse A9-cells genome with the exception of the human X chromosome. that lack this enzyme (12) as well as adenine phosphoribosyl- The selective system that has permitted the isolation of this transferase (EC 2.4.2.7; AMP:pyrophosphate phosphoribo- hybrid cell line was devised by Szybalski and Szybalska (6) syltransferase) (13). The original hybrid line had been prop- to select against mammalian cells deficient in hypoxanthine agated in the HAT-selective medium for about 1 year, phosphoribosyltransferase (EC 2.4.2.8; IMP: pyrophosphate before being stored in liquid nitrogen. During that period it phosphoribosyltransferase). This is achieved by carrying out had lost almost the entire human genome, with the exception cell growth in the so-called HAT selective medium, containing of the human X chromosome and all or some of its markers a potent inhibitor of purine and pyrimidine synthesis (A for (8, 9). The present clonal derivative originated from the few aminopterin) together with preformed nitrogen bases (H for cells that recovered in HAT medium after 4 years of storage. hypoxanthine and T for thymidine). Cell survival in this The HeLa-S3 cell line (14) was used as a source of human selective system depends upon the ability of the cell to use the hnRNA. This line is known to have from 43 to 69 chromo- preformed bases as a salvage pathway for nucleic acid syn- somes with two or more doses of each human chromosome (15) thesis. This ability is provided by normal activity of the phos- and exhibited normal activity for all X-linked and autosomal phoribosyltransferase, whose structural gene is known to be enzyme markers mentioned in Table 1. All cells wvere grown in monolayer cultures in 1585 cm2 Abbreviations: hnRNA, heterogeneous nuclear RNA; HAT roller bottles. HeLa and A9 cells were constantly propagated medium, hypoxanthine/aminopterin/thymidine medium. in minimum essential medium, and A9/HRBC2 cells in HAT 1868 Downloaded by guest on September 29, 2021 Proc. Nat. Acad. Sci. USA 72 (1975) Human RNA in Man-Mouse Cell Hybrids 1869

medium (16), both supplemented with the standard additions TABLE 1. Characterization of enzymatic markers of sera and antibiotics reported in full elsewhere4I All cell in A9/HRBC2 cells stocks were checked for mycoplasma with Levine's method A B C (17) and frozen. Each experiment was conducted with a freshly thawed vial of mycoplasma-free frozen cells. Con- G6PD EC 1.1.1.49 X M ( hy ( H fluent monolayers of cells were labeled for 1 hr with 2 mCi of HGPRT EC 2.4.2.8 X - H [3H]uridine (27 Ci/mmol) in 25 ml of medium. The cells were PGK EC2.7.2.3 X M H chilled to 40 and harvested, and their nuclei were isolated NAD-MDH-1 EC1.1.1.37 2 M >>> hy >>> H (18, 19). Nuclear RNA was extracted (10, 19) and hnRNA, IDH-1 EC1.1.1.42 2 M >>> hy >>> H with a sedimentation constant greater than 45 S, was isolated (A) List of human enzyme markers (with trivial abbreviations (10). DNA was extracted (20) from isolated nuclei of cultured and EC numbers) retained by A9/HRBC2 hybrid cell. (B) cells (18, 19) and of human placenta (21). Minor modifica- Chromosomes carrying relevant loci in man. (C) Relative in- tions of these classical nucleic acid isolation procedures were tensity of mouse (M), human (H), and heteropolymeric (hy) applied and are described elsewhere.T RNA* DNA hybridiza- enzymatic bands in the electrophoretic pattern of A9/HRBC2 tion on nitrocellulose filters (22) was carried out as described cell lysates. The following additional 23 enzyme markers, identi- by Soeiro and Darnell (10). Background binding to blank fying 13 of the 22 human autosomes, were found to be regularly filters under these experimental conditions ranges from 1 to absent: AK2(EC 2.7.4.3); PGM-1(EC 2.7.5.1); Pep C(EC 3% of the radioactive RNA hybridized to DNA-bearing 3.4.3.-); PPH(EC 4.2.1.11); Me-l(EC 1.1.1.40); IPO-B(EC filters. 1.6.4.3); sGOT-1(EC 2.6.1.1); HK(EC 2.7.1.1); LDH-A(EC Multistep hybridization experiments were conducted as 1.1.1.27); LDH-B(EC 1.1.1.27); Pep B(EC 3.4.3.-); NP(EC described by and Szybalski (23) with the following 2.4.2.1); PK-3(EC 2.7.1.40); MPI(EC 5.3.1.8); APRT(EC Bovre 2.4.2.7); TK(EC 2.7.1.21); Pep A(EC 3.4.3.-); GPI(EC 5.3.1.9); modifications. The first step or preparative hybridization was ADA(EC 3.5.4.4); IPO-A(EC 1.6.4.3); ,-Gluc(EC 3.2.1.31); conducted as described by Soeiro and Darnell (10). After and Pep-D(EC 3.4.3.-). For details concerning these enzyme radioactivity determination, the filter was removed from the markers, their nomenclature, and the chromosomal assignment scintillation fluid, air-dried, and treated with iodoacetate of relevant human genes, see ref. 32. (23). Then 0.75 ml of 0.30 M NaCl/0.030 M sodium citrate (pH 7) containing 0.1% sodium dodecyl sulfate was added to phoresis. This procedure ensures that conclusions drawn from the vial containing the filter, and the solution with the filter enzyme and chromosomal studies can be confidently con- was heated in a boiling-water bath for 10 min. The solution sidered pertinent to the same population of A9/HRBC2 cells was immediately chilled to 40, and the filter was carefully re- used in the experiments of molecular hybridization, despite moved and dried for radioactivity measurements to determine possible variations of the hybrid genome known to occur the efficiency of elution. The eluted RNA was then cross- during prolonged growth in HAT selective medium (8). hybridized to a new set of filters under the same conditions Cell lysates of A9/HRBC2 were always studied in parallel used in the preparative hybridization. with lysates of A9 (mouse parental) and of normal human Electrophoretic characterization of enzyme markers was cells. All lysates were screened for the presence of 28 human performed by the techniques reported by Shin et al. (24), genetic markers which identify the human X chromosome and Meera Khan (25), Migeon et al. (26), and Grzeschik (27). 14 of the 22 autosomes (Table 1). These genetic markers in- The screening for human adenine phosphoribosyltransferase clude only constitutive enzymes whose electrophoretic mobil- was performed at the single cell level by autoradiography ity differs in the parental species and whose regular expression after growth of the hybrid cells in HAT medium containing is well documented in man-mouse somatic cell hybrids before 10 ,Ci/ml of [3Hjadenine; since the parental mouse cells lack they undergo the loss of the specific human chromosomes the enzyme, incorporation of labeled adenine is considered bearing the corresponding structural loci (31). evidence of the retention of the human gene coding for this Table 1 shows that hybrid line A9/HRBC2 has regularly enzyme. Standard techniques for centromeric (C-) banding retained only the three human X-linked markers: hypo- (28), Giemnsa (G-) banding (29), and quinacrine (Q-) banding xanthine phosphoribosyltransferase (HGPRT), glucose-6- (30) were applied for chromosome studies, sometimes with phosphate dehydrogenase (G6PD), and phosphoglycerate slight modifications as described in the legends to the figures. kinase (PGK). All human autosomal markers are absent, with the exception of those of chromosome no. 2: NAD-de- RESULTS AND DISCUSSION pendent malate dehydrogenase (NAD-MDH) and cyto- Characterization of Hybrid Line A9/HRBC2: Enzyme and plasmic isocitrate dehydrogenase (IDH-1) (32). The latter Chromosome Studies. Cellogel electrophoresis (25) of cell ones must, however, be present only in a small proportion of lysates and chromosome banding techniques were used to cells, as is suggested by the finding that the human bands as establish the amount of human genetic information residual well as the heteropolymeric mouse-human components of in the A9/HRBC2 hybrid line. Such studies were performed these enzymes have always been fo-und to be much fainter whenever a large batch of A9/HRBC2 cells was propagated than the mouse bands in all lysate preparations of A9/HRBC2 for isolation of nucleic acids. A small sample of the cell popula- cells. tion was grown in parallel in two Falcon flasks. When these Chromosome studies were regularly carried out on each were nearly confluent, cells in metaphase were removed by batch of thawed cells propagated for isolation of nucleic acids. shaking for chromosomal analysis, while the remaining at- Under the experimental conditions used, the mouse chromo- tached cells were harvested and lysed for enzyme electro- somes exhibited a distinct centromeric band while human ones did not. Giemsa and quinacrine banding were used for the $ R. A. Di Cioccio and M. Siniscalco (1975) Somatic Cell identification of individual human chromosomes. The A9 Genetics, submitted for publication. parental mouse line has a heteroploid genome with an average Downloaded by guest on September 29, 2021 1870 Genetics: Di Cioccio et al. Proc. Nat. Acad. Sci. USA 72 (1975)

la.

~* AS > AWWK;Pt E *~~~~~~~~~~~~~~v., j~

H: ) ,f

FIG. 1. Metaphase chromosomes stained with Giemsa diluted with phosphate buffer pH 6.8 after various pretreatments. (A and B) C-banded A9 (A) and A9/HRBC2 (B) chromosomes treated for 3 min in saturated Ba(OH)2, 2 min in 0.07 M NaOH, and 1 hr in 0.90 M NaCl/0.090 M Na citrate at 600. Arrows indicate non-C-banded chromosomes, the submetacentric in panel B is presumably the human X chromosome. (C and D) A9/HRBC2 chromosomes (C, complete metaphase; 1), 5 human X chromosomes cut out of five different A9/HRBC2 metaphases) after a treatment with a mixture solution of 0.05% trypsin and 0.02% EI)TA for 8-10 min, which give a combination of C- and G- banding on these chromosomes. Arrow indicates human X chromosome. number of 51 chromosomes, 28 of which are telocentric and 23 exposed individually. The results of these experiments can be biarmed believed to be the result of centromeric fusion (33). summarized as follows (Fig. 2). All these chromosomes showed a distinct centromeric band (i) Under the given experimental conditions the percent of except one or two telocentric chromosomes per metaphase hybridization between hnRNA from cell hybrid, mouse, or analyzed. The A9/HRBC2 hybrid line used in this study has a human cells with their homologous DNA ranges from 2.60 to karyotype which would be indistinguishable from that of the 3.00 (Fig. 2A, B, and C). § A9 parental cell line, except for the regular presence of one (ii) The hnRNA from A9/HRBC2 hybrid cells or from A9 biarmed chromosome without centromeric banding. From parental mouse cells binds, as expected, much more efficiently G- and C- banding, this chromosome was identified as the to their homologous DNA or to quasihomologous DNA (as is human X in 26 out of 27 karyotyped metaphases (Fig. 1). the A9 T)NA for the A9/HRBC2 RNA and vice versa) than However, this conclusion could not be confirmed in 19 addi- they do to human placental DNA (Fig. 2A, B, D, and E). tional metaphases which were studied for C- and Q- banding. Correspondingly, human HeLa hnRNA binds to human DNA In these metaphases the non-C-banded human X-like chromo- much more efficiently than to A9 or to A9/HRBC2 D)NA some failed to show the characteristic Q-banding pattern of (Fig. 2C and F). the human-X, while the mouse chromosomes were banded as (iii) The A9/HRBC2 hnRNA anneals better to human expected. We have no explanation for this inconsistency. DNA than does the A9 hnRNA. Namely, 0.18% (0.33 - 0.15 We are presently investigating whether this is a phenomenon = 0.18%) more of hybrid hnRNA binds to filters bearing of general occurrence in man-mouse hybrids that have re- human DNA than does mouse hnRNA (Fig. 21) and E). Cor- tained very few human chromosomes. The C- and G- banding respondingly, human HeLa hnRNA binds to A9/HRBC2 studies revealed also the presence of an extra biarmed non-C- DNA better than to A9 DNA. Namely, 0.22% (0.28 X 2 - banded chromosome with an atypical G-banding pattern in 0.17 X 2 = 0.22%) more of human hnRNA anneals to hybrid six out of the 27 above-mentioned metaphases karyotyped. DNA than it does to mouse DNA (Fig. 2F). In view of the enzyme data reported above, it is conceivable (iv) Both A9 and A9/HRBC2 hnRNA were found to bind that this chromosome may be human autosome no. 2, prob- slightly better to A9 DNA than to A9/HRBC2 DNA when ably modified in its banding pattern as result of de novo each of the RNA preparations was exposed simultaneously to chromosomal rearrangements occurring in culture. both types of DNA (Fig. 2A and B). Since the chromosomal In conclusion, both the enzymatic and the cytological content of these two cell lines is almost identical, their hn- studies suggest that the only human chromosome regularly RNA is expected to anneal to A9/HRBC2 DNA and A9 present in the A9/HRBC2 hybrid cells is the X chromosome. DNA identically. However, because the differences in hybrid- However, the occasional presence of additional human genetic ization are small, it is difficult to ascribe significance to them, information in a small proportion of these hybrid cells cannot especially in view of the finding that differences of the same be excluded. magnitude have been found also when two sets of filters Identification of Human hnRNA and of Human DNA in bearing the same 1)NA (A9/IIRBC2) were exposed to the A9/HRBC2 Hybrid Cells. [3HlUridine-labeled hnRNA from homologous RNA preparation. At any rate, whatever its ex- hybrid cells, parental A9 mouse cells, and human HeLa cells were separately exposed to nitrocellulose filters bearing A9/ HRBC2, A9, or human placental DNA. Throughout these § When two filters bearing the same amount of identical or hybridization experiments, the filters bearing A9 or A9/ quasihomologous 1)NA are exposed simultaneously to the same HRBC2 DNA were exposed simultaneously to each RNA RNA preparation, the true percent of input RNA hybridized preparation, while filters bearing human placental D)NA were is to be considered twice as much as the one observed per filter. Downloaded by guest on September 29, 2021 Proc. Nat. Acad. Sci. USA 72 (1975) Human RNA in Man-Mouse Cell Hybrids 1871

% of A9 HRBC2 % of As hn RNA % of Human (HeLa) hn RNA hybridized hybridized hn RNA hybridized 3: 1~~~~~~~ 2- I 2 _ (A) (B) .1 (A) 8} -!(C) 20 40 (0 80 'I 20' (4I60 I0 (N#g 0 40 Iisiig ioog

.3 .3

~~ ~ ~

(D) F 01 20 to4 60) 811 iioi1.tg 20 40 (A) P1A) I(N)lg

Micrograms of DNA immobilized on filters bearing Ao/HRBC2-DNA (o--o), or As-DNA (a--a), or human placenta-DNA (x--x) FIG. 2. RNA-DNA filter hybridization experiments carried out with ['H]uridine-labeled hnRNA of various kinds. The percents of RNA hybridized at each DNA concentration have been computed from the ratios between cpm bound per filter (average of two deter- minations) and total cpm of RNA preparation. (A) Mouse (A9) or hybrid cell (A9/HRBC2) DNA exposed to 4.7 X 104 cpm of A9/ HRBC2 hnRNA (1.3 X 105 cpm/gg). (B) Same DNA filters as in A, exposed to 4 X 104 cpm of A9 hnRNA (1.6 X 105 cpm/,Og). (C) Human placental DNA exposed to 2.2 X 104 cpm of HeLa hnRNA (9.84 X 104 cpm/.Ug). (D) Same DNA filters as in C, exposed to 3.8 X 106 cpm of A9/HRBC2 hnRNA (1.3 X 105 cpm/,gg). (E) Same DNA filters as in C and D, exposed to 4 X 105 cpm of A9 hnRNA (1.6 X 105 cpm/.ug). (F) Same DNA filters as in A and B, exposed to 8.4 X 105 cpm HeLa hnRNA (9.9 X 104 cpm/Atg). All experiments were performed in duplicate. The percent of A9/HRBC2 RNA that is specifically hybridizable to human DNA can be obtained by subtracting the single values reported in E from the corresponding ones reported in D. For 100 jug of DNA per filter, this amounts to 0.18% (i.e., 0.33 - 0.15). Correspondingly, the percent of HeLa RNA that is specifically hybridizable to cell hybrid DNA can be obtained from the difference of the two sets of values reported in F (interrupted line). For 100 ,g of DNA per filter, this amounts to 0.22% (i.e., 0.28 X 2 - 0.17 X 2).

planation, it is clear that this finding strengthens the signifi- mouse or -cell hybrid DNA (Table 2, Exps. 7-9) and cell cance of the above-mentioned higher affinity of the human hybrid to mouse DNA (Table 2, Exps. 6 and 7). RNA for filters bearing A9/HRBC2 DNA when the former is The data presented provide preliminary evidence that it is exposed simultaneously to filters bearing the hybrid cell or the possible to identify human RNA and DNA in human-mouse mouse DNA (Fig. 2F). somatic cell hybrids with a greatly reduced human chromo- Multistep Hybridization. The nature of the cell hybrid somal content. Additional studies, to be reported elsewhere, hnRNA that hybridizes to human DNA and of HeLa hnRNA provide further evidence that molecular hybrids between that hybridizes to cell hybrid DNA, was further analyzed by A9/HRBC2 and human nucleic acids, detected by nitro- multistep hybridization experiments, which confirm the ob- cellulose filter assay, are more specific than molecular hybrids servations reported in the preceding paragraph. Labeled RNA between A9 and human nucleic acids. These studies also pro- was eluted from filters each bearing a given combination of vide data identifying the presence of human RNA in A9/ RNA- DNA molecular hybrid (Table 2A) and reannealed to HRBC2 cells that is complementary to "few gene copy" a second set of filters bearing 100 gg of human, mouse, or human DNA, as well as data identifying the presence of hybrid cell DNA (Table 2B). The percent of eluted RNA human RNA in HeLa cells that is complementary to "few ranged from 85 to 91. In Table 2, the radioactivity bound to gene copy" A9/HRBC2 DNA.4 each filter in the series of cross-hybridizations has been cor- Since the enzyme and chromosomal data suggest that the rected for background by subtracting the radioactivity bound only human chromosome regularly present in A9/HRBC2 to filters bearing Escherichia coli DNA. The data indicate that cells is the human X chromosome, it can be concluded that the A9/HRBC2 hnRNA eluted from an A9/HRBC2 hnRNA- hnRNA identified by the experiments described in Fig. 2D human DNA molecular hybrid reanneals 1.9 times more to and in Table 2, Exp. 2, must evidently be transcribed from human than to A9 DNA (Table 2, Exp. 2). In contrast, RNA repetitive DNA sequences of the human X chromosome. How- isolated from an A9 hnRNA-human DNA molecular hybrid ever, despite the evidence favoring this conclusion, we cannot exhibits a slight preference for A9 DNA (Table 2, Exp. 4). as yet exclude the possibility that additional human genetic Correspondingly, HeLa hnRNA eluted from a HeLa hnRNA- information, besides the one identified through enzymatic and A9/HRBC2 RNA molecular hybrid anneals 4.9 times more to chromosomal studies, may be present in A9/HRBC2 hybrid A9/HRBC2 DNA than to A9 DNA, while HeLa RNA iso- cells in the form of undetectable chromosomal fragments lated from a HeLa hnRNA-A9 DNA molecular hybrid translocated to mouse chromosome. We intend to test this binds only 2.3 times more (Table 2, Exp. 5). In the control possibility with DNA preparations from human strains with experiments, RNA isolated from molecular hybrids of various multiples of the X-chromosome. If the human-X is essentially combinations of A9/HRBC2 and A9 nucleic acids anneals the only human chromosome in A9/HRBC2 cells, the human preponderantly to A9 DNA (Table 2, Exps. 1 and 3). Cor- component of A9/HRBC2 hnRNA should hybridize pro- respondingly, HeLa RNA eluted from HeLa hnRNA-human portionately to human DNA derived from XY, XXY, or DNA molecular hybrids and re-exposed simultaneously to XXXXY human cells. Should this prove to be the case, then various combinations of human, mouse (A9), and cell hybrid the in situ hybridization of this human component of A9/ (A9/HRBC2) DNA, prefer-as expected-human to either HRBC2 cells to human metaphase chromosomes would pro- Downloaded by guest on September 29, 2021 1872 Genetics: Di Cioccio et al. Proc. Nat. Acad. Sci. USA 72 (1975)

TABLE 2. Multistep hybridization We acknowledge the expert technical assistance of Aia Hersh- berg and Paula DeBona and express our gratitude to Drs. J. E. B. Cross- Darnell and R. Soeiro for their advice and criticism during the planning and performance of the work. This work was sup- A. Preparative hybridization hybridization ported in part by NCI Grant CA-08748. R.A.D. was recipient of eluted RNA of a postdoctoral fellowship from the American Cancer Society cpm % of (PF-877). hnRNA bound eluted H H hy (cpm) DNA and % RNA M hy M 1. Weiss, M. C. & Green, H. (1967) Proc. Nat. Acad. Sci. USA 58, 1104-1111. 1. A9/HRBC2 Mouse (A9) 750 88 0.03 2. Kao, F. T. & Puck, T. T. (1970) Nature 228, 329-332. (5 X 104) (.150) 3. Meera Khan, P., Westerveld, A., Grzeschik, K. H., Deys, Cell hybrid 651 85 B. F., Garson, 0. M. & Siniscalco, M. (1971) Amer. J. (A9/HRBC2) (1.31) 0.05 Hum. Genet. 23, 614-623. 2. A9/HRBC2 Human 1515 86 1.90 4. Grzeschik, K. H., Allderdice, P. W., Grzeschik, A. M., Opitz, J. M., Miller, 0. J. & Siniscalco, M. (1972) Proc. (4 X 105) (placenta) (0.38) Nat. Acad. Sci. USA 69, 69-73. 3. A9 Mouse (A9) 720 88 0.04 5. Croce, C. M., Litwack, G. & Koprowski, H. (1973) Proc. (4 X 104) (1.80) Nat. Acad. Sci. USA 70, 1268-1272. Cell hybrid 624 86 0.10 6. Szybalski, W. & Szybalska, E. H. (1962) Univ. Mich. Med. (A9/HRBC2) (1.56) Bull. 28, 277-293. 7. Seegmiller, J. G., Rosenbloom, F. M. & Kelley, W. N. 4. A9 Human 675 91 0.70 (1967) Science 155, 1682-1684. X (4 106) (placenta) (0.17) 8. Miller, 0. J., Cook, P. R., Meera Khan, P., Shin, S. & 5. HeLa Mouse (A9) 1428 91 2.3 Siniscalco, M. (1971) Proc. Nat. Acad. Sci. USA 68, 116- (8.4 X 105) (0.17) 120. Cell hybrid 2436 87 4.9 9. Siniscalco, M. (1974) in Somatic Cell Hybridization, eds. (A9/HRBC2) (0.29) Davidson, R. L. & de la Cruz, F. (Raven Press, New York), pp. 35-48. 6. HeLa Human 1848 88 3.6 10. Soeiro, R. & Darnell, J. E. (1969) J. Mol. Biol. 44, 551- (8.4 X 104) (placenta) (2.20) 562. 7. HeLa Human 1763 85 27.0 3.0 9.0 11. Coon, H. J., Horak, I. & David, I. B. (1973) J. Mol. Biol. (8.4 X 104) (placenta) (2.10) 81, 285-298. 12. Littlefield, J. W. (1964) Nature 203, 1142-1144. 8. HeLa Human 1626 89 2.7 13. Cox, R. P., Kruss, M. R., Balis, M. E. & Dancis, J. (1972) (8.4 X 104) (placenta) (2.00) Exp. Cell Res. 74, 251-268. 9. HeLa Human 1931 90 26.0 14. Gey, G. O., Coffmann, W. D. & Kubicek, M. T. (1952) (8.4 X 104) (placenta) (2.30) Cancer Res. 12, 264-265. 15. Miller, 0. J., Miller, D. A., Allderdice, R. W., Dev, V. G. & Grewal, M. S. (1971) Cytogenetics 10, 338-346. (A) Preparative hybridization performed with various prepara- 16. Littlefield, J. W. (1966) Exp. Cell Res. 41, 190-196. tions of ['H]uridine-labeled RNA (1-hr pulse) and filters bearing 17. Levine, E. M. (1972) Exp. Cell Res. 74, 99-109. 100 ,gg of a given type of DNA. Types and amount of input RNA, 18. Kumar, A. & Lindberg, J. (1972) Proc. Nat. Acad. Sci. type of DNA on filters, cpm bound, and percent of eluted RNA USA 69, 681-689. are specified. Exps. 1, 3, and 5 were conducted with two DNA 19. Penman, S. (1966) J. Mol. Biol. 17, 117-130. filters simultaneously exposed to the same RNA preparation. 20. Marmur, J. (1961) J. Mol. Biol. 3, 208-218. Specific activities of the hnRNA preparations are the same as 21. Wang, T. Y. (1968) in Methods of Enzymology, eds. Gross- those reported in the legend to Fig. 2. (B) Cross-hybridization of man, L. & Moldave, K. (Academic Press, New York), the eluted RNA to filters 100 of human mouse Vol. XII, Part B, pp. 115-120. bearing /Ag (H), 22. Gillespie, D. & Spiegelman, S. (1965) J. Mol. Biol. 12, (M), or hybrid cell (hy) DNA. The ratios H/M, H/hy, and hy/M 829-842. express the relative cpm of eluted RNA bound to various filters 23. B0vre, K. & Szybalski, W. (1971) in Methods in Enzy- in the second step hybridization experiments. mology, eds. Grossman, L. & Moldave, K. (Academic Press, New York, London), Vol. XXI, Part D, pp. 350-383. 24. Shin, S., Meera Khan, P. & Cook, P. R. (1971) Biochem. vide critical information about the distribution the Genet. 5, 95-99. and/or 25. Meera Khan, P. (1971) Arch. Biochem. Biophys. 145, 470- chromosomal specificity of repetitive DNA sequences of the 483. human genome. 26. Migeon, B. R., Smith, S. & Leddy, C. (1969) Biochem. If a regional chromosomal specificity of the human repeti- Genet. 3, 583-590. tive DNA is demonstrated, the experimental approach de- 27. Grzeschik, K. H. (1973) Habilitationsschrift (Universitat Munster). scribed may be used to gather indirect information about the 28. Arrighi, F. E. & Hsu, T. C. (1971) Cytogenetics 10, 81-86. precise chromosomal location of given human structural loci 29. Seabright, M. (1972) Chromosoma 36, 204-210. with particular reference to those for human hypoxanthine 30. Caspersson, T., Zech, L., Johansson, C. & Modest, E. J. phosphoribosyltransferase, glucose-6-phosphate dehydro- (1970) Chromosoma 30, 215-227. whose 31. Ruddle, F. H. (1974) in Somatic Cell Hybridization, eds. genase, and phosphoglycerate kinase, cytological Davidson, R. L. & de la Cruz, F. (Raven Press, New mapping is hitherto still a matter of dispute (9). This could be York), pp. 1-13. achieved by annealing in situ to normal and abnormal human 32. Shows, T. B., Bias, W. B. (1974) in Human Gene Mapping, metaphases (notably those with X-autosomal translocations) ed. Bergsma, D. (Birth Defects; original series, The Na- different preparations of hnRNA derived from man-mouse tional Foundation March of Dimes), Vol. 10, Part 3, pp. hybrid cell lines that are known to have retained different 35-48. 33. Allderdice, P. W., Miller, 0. J., Miller, D. A., Warburton. segments of the X-chromosome and its genetic markers as A., Pearson, P. O., Klein, G. & Harris, H. (1973) J. Cell the only residual of the human genome (8, 9). Sci. 12, 263-274. Downloaded by guest on September 29, 2021