by bioautography - Europe PMC

5 downloads 0 Views 934KB Size Report
Argininosuccinic aciduria: Assignment of the argininosuccinate lyase gene to the ..... expression of ASL and argininosuccinate synthetase. We thank Ms. J. K.
Proc. Natl. Acad. Sci. USA

Vol. 75, No. 12, pp. 6159-6162, December 1978 Genetics

Argininosuccinic aciduria: Assignment of the argininosuccinate lyase gene to the pter--q22 region of human chromosome 7 by bioautography (somatic cell hybrids/aminoaciduria)

S. L. NAYLOR*t, R. J. KLEBE*, AND T. B. SHOWSf *Division of Human Genetics, University of Texas Medical Branch, Galveston, Texas 77550; and *Biochemical Genetics Section, Roswell Park Memorial Institute, New York State Department of Health, Buffalo, New York 14263

Communicated by Victor A. McKusick, September 21, 1978

ABSTRACT Argininosuccinic aciduria, an autosomal recessive disorder of the urea cycle in humans, is associated with a deficiency of argininosuccinate lyase (ASL; L-argininosuccinate arginine-lyase, EC 4.3.2.1). ASL activity was visualized on gels after electrophoresis by a new method, termed bioautography. Bioautography involves the use of mutant bacteria to visualize the location of mammalian enzymes after zone electrophoresis. By this technique, human ASL migrated to a position different from mouse ASL, while a survey of mouse strains, tissues, and tissue culture cell extracts demonstrated the same electrophoretic form and no genetic variants of mouse ASL. Identifying human ASL by bioautography in humanmouse somatic cell hybrids has made it possible to regionally locate the ASL gene on human chromosome 7. The human ASL phenotype segregated concordantly with the human enzyme ,-glucuronidase (GUS; jlD-glucuronide glucuronosohydrolase, EC 3.2.1.31) in cell hybrids, but showed discordant segregation with 32 other enzyme markers representing 23 linkage groups. The gene for GUS has been assigned to chromosome 7 in humans, and cosegregation (synteny) of ASL and GUS demonstrates the assignment of ASL to chromosome 7. Regional location of ASL and GUS to the pter--q22 region of chromosome 7 was achieved in hybrids segregating a 7/9 translocation. Bioautography, a method of isozyme visualization, uses bacterial growth to locate an enzyme after zone electrophoresis. The auxotrophic bacteria used require a nutrient product that is generated by the enzyme (1). Since there is a wide variety of genetically marked bacterial strains (2), it has been possible to use bioautography to visualize several isozymes after electrophoresis that have not been amenable to study by standard procedures, such as branched chain aminotransferase (1) and aminoacylase-1 (3). These new enzyme markers, generated by bioautography, have immediate application in the detection of genetic electrophoretic polymorphisms. Bioautography should be instrumental in generating many new markers for gene mapping studies by somatic cell genetic techniques. Bioautography has been used to detect the human and rodent forms of argininosuccinate lyase in human-rodent somatic cell hybrids (Fig. 1). Argininosuccinate Iyase (ASL; L-argininosuccinate arginine-lyase, EC 4.3.2.1), an enzyme of the urea cycle, cleaves argininosuccinate to L-arginine and fumarate (4). Deficiency of ASL has been associated with argininosuccinic aciduria, an aminoaciduria that is characterized by mental retardation, seizures, ataxia, hepatomegaly, hyperammonemia, and large amounts of argininosuccinic acid in blood, urine, and cerebrospinal fluid (5). Although all patients with the disease show autosomal recessive inheritance, argininosuccinic aciduria appears to have several definable forms (6). Screening tests for

argininosuccinate aciduria with auxotrophic bacteria have been devised by Murphey et al. (7). In this study we have investigated the expression of human ASL in human-mouse somatic cell hybrids. By the bioautographic procedure for studying ASL (Fig. 1), evidence establishing the chromosome assignment and location of the ASL gene has been gathered from human-mouse somatic cell hybrids segregating human chromosomes. The human ASL structural gene is syntenic with the gene coding for ,B-glucuronidase (GUS; J3-D-glucuronide glucuronosohydrolase, EC 3.2.1.31), which has previously been assigned to chromosome 7 (8-10). A 7/9 chromosome translocation (11) has been used in cell hybrids to map ASL as well as GUS to the pter- q22 region of chromosome 7 in humans.

MATERIALS AND METHODS Human and Rodent Parental Cells. Human primary cell strains for fusions were DUV [46,X,t(X;15)(pll;qll)] (12) and JoSt [46,XY,t(7;9)(q22;p24)] (11);,and CaVa [46,X,t(X;22p (q22;ql3)] (13) and AnLy [46,X,t(X;9)(q12;p24)] (13) leukocytes. Human fibroblasts were maintained on Eagle's basal medium (diploid) (GIBCO), 10% fetal calf serum, and antibiotics. Rodent parental lines with selectable markers were RAG (HPRT-) (14), A9 (HPRT-) (15), and LM/TK- (16) maintained on Dulbecco's modified Eagle's medium. Human-Rodent Somatic Cell Hybrids. Human and mouse cells were fused in suspension or as monolayers with inactivated Sendai virus (13) and the resulting hybrids were cloned and maintained as monolayers on HAT (hypoxanthine/aminopterin/thymidine) selection medium (17) consisting of Dulbecco's modified Eagle's medium (GIBCO), 10% fetal calf serum, and antibiotics. Primary hybrid clones were established from four fusion experiments with four different human parental cells. Human-mouse hybrid clones were designated DUA (DUV X A9) (18), JSR (JoSt X RAG) (11), ALR (AnLy X RAG) (13), and REX (CaVa X RAG) (13). Mouse Tissues and Strains. Bioautography of ASL was performed on homogenates of the following mouse tissues in C57BL/6J: liver, kidney, brain, heart, lung, stomach, skeletal muscle, testis, spleen, whole blood, salivary glands, and skin. In addition to the Tris/EDTA/borate buffer system (pH 8.6) (below), buffer systems at pH 7.2 [phosphate (1)] and pH 6.15 [4-morpholineethanesulfonic acid (Mes): 50 mM for electrode buffer and 60 mM for hydration of cellulose acetate plates] were tested. In these same electrophoretic systems, liver and kidney extracts of the following strains of mice were assayed for ASL Abbreviations: ASL, argininosuccinate lyase; GUS, f3-glucuronidase. Present address: Biochemical Genetics Section, Roswell Park Memorial Institute, New York State Department of Health, Buffalo, NY

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

t

14263.

6159

6160

Genetics:

Naylor et al.

Proc.

0)

Natl. Acad. Sci. USA 75 (1978)

Table 1. Segregation of ASL and 33 other enzyme markers in hybrid cell clones

ASL/enzyme marker(s) Arg-E. coli -*

L-Arginine

Bacterial growth

Argininosuccinate

agar

FIG. 1. Bioautography of ASL. A band of ASL is formed after gel electrophoresis of tissue or cell extracts. The gel is placed in contact with a layer of indicator agar that contains arginine-requiring (arg-) E. coli and argininosuccinate. At the site of the enzyme, arginine is produced from argininosuccinate, and bacteria proliferate to form a visible band.

by bioautography: AKR/J, A/HeJ, AU/SsJ, BALB/cJ, BDP/J, BUB/BnJ, CBA/J, CE/J, C3HeB/FeJ, C57BR/cdJ, C57BL/6J, C58/J, DBA/2J, LP/J, MA/MyJ, NZB/BnJ, P/J, PL/J, RIII/2J, RF/J, SJL/J, SM/J, ST/6J, SWR/J, and 129/J. Electrophoresis and Bioautography of ASL. Hybrid extracts (5 Al) were electrophoresed on cellulose acetate plates (6 X 15 cm) (Helena Laboratories, Beaumont, TX) in a Tris/ EDTA/borate buffer (pH 8.6) (1) for 1.5 hr at 250 V and 4 mA at 40C. The buffer system consisted of a stock solution (0.9 M Tris/20 mM tetrasodium EDTA/0.5 M boric acid at pH 8.6) which was diluted (i) 1:10 for hydrating the cellulose acetate plates, (ii) 1:2.5 for the cathode; and (iii) 1:3.5 for the anode. The enzyme was located after zone electrophoresis by bioautography. Briefly, after electrophoresis of the extract, the cellulose acetate plate was placed in contact with a bacteria-seeded layer of minimal agar lacking the required nutrient (arginine) and containing the substrate (argininosuccinate) for the enzyme. After incubation, a band of bacteria is generated wherever the enzyme is localized on the cellulose acetate plate. Details of the bioautographic method for ASL have been published (1). ASL was visualized by using the arginine-requiring Escherichia coli mutant AT753 (E. coli Genetic Stock Center, Yale University Medical School) as the microbial reagent. The indicator agar (1) contained 20,ug of argininosuccinate per ml as a substrate for production of L-arginine and 108 bacteria per ml in minimal medium (19). Because argininosuccinate was obtained as the barium salt (Sigma), barium (which interfered with bacterial growth) was precipitated with an equimolar concentration of sodium sulfate in a lOOX stock solution. The BaSO4 precipitate was removed by centrifugation at 2000 X g for 10 min. The argininosuccinate stock is stable at -20°C for several weeks. Electrophoresis of GUS. Human and rodent forms of GUS were separated on starch in Veronal buffer (pH 8.0) (9). The enzymatic activity of GUS was visualized histochemically with 4-methylumbelliferyl-(-D-glucuronide as described (9). Mouse GUS migrated anodally and the human enzyme migrated

toward the cathode in the Veronal buffer system. Procedures for other enzymes tested are indicated in Table 1. Chromosome Studies. Giemsa-trypsin banding procedures were used to distinguish rodent and human chromosomes (13). Chromosomes were identified as established in the 1971 Paris Conference.

Concordant Discordant

Chromosome

Enzyme marker

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 X

AK2/PGM1/PEPC MDHs/IDHs fGALA PEPS HEXB

% dis-

+/+ -/- +/- -/+ cordancy

7 8 4 5 4 5 MEs/SODM GUS 12 1 GSR AK1 4 7 GOTS LDHA/ACP2/ESA4 4 .5 PEPB ESD 11 NP 10 MPI/HEXA/PKM2 8 APRT* 6 GALK 1 PEPA 4 GPI 5 4 ADA 7 SODS ACONM 1 PGK/HPRT/G6PD 9

8 7 6 2 8 8 10 6 10 9 9 7 4 5 7 8 6 6 6 7 8 10 8

6 5 8 8 9 8 1 11 9 6 9 8 2 3 5 7 11 9 8 9 6 12 4

2 3 2 8 2 2 0 5 0 1 1 3 6 5 3 2 4 4 4 3 2 0 2

35 35 50 70 48 43 4 70 39 30 43 48 35 35 35 39 68 57 52 52 35 52 26

Electrophoretic and visualization procedures for each enzyme marker are described in the following references. Adenylate kinase-2 (AK2), phosphoglucomutase-1 (PGM,), peptidase-C (PEPC), malate dehydrogenase-S (MDHs), isocitrate dehydrogenase-S (IDHS), malic enzyme-S (MES), superoxide dismutase-M (SODM), adenylate kinase-1 (AK1), glutamate oxaloacetate transaminase-S (GOTS), lactate dehydrogenase-A (LDHA), esterase-A4 (ESA4), peptidase-B (PEPB), peptidase-A (PEPA), glucose phosphate isomerase (GPI), superoxide dismutase-S (SODs), and glucose-6-phosphate dehydrogenase (G6PD) (20, 21); f3-galactosidase-A (3GALA) and aconitase-M (ACONM) (22); peptidase-S (PEPS) (23); hexosaminidase-B (HEXB), hexosaminidase-A (HEXA), and mannosephosphate isomerase (MPI) (24); f3-glucuronidase (GUS) (9); glutathione reductase (GSR) (25); acid phosphatase-2 (ACP2) (26); esterase-D (ESD) (27); nucleoside phosphorylase (NP) (28); pyruvate kinaseM2 (PKM2) (29); adenine phosphoribosyltransferase (APRT) (30); galactokinase (GALK) (31); adenosine deaminase (ADA) (32); phosphoglycerate kinase (PGK) (33); hypoxanthine phosphoribosyltransferase (HPRT) (34). Chromosome assignments of the genes for the marker enzymes have been published (ref. 35; ACONM and 1#GALA, ref. 22; PEPS, ref. 23). Concordant segregation columns indicate the number of clones in which both enzymes were either present (+/+) or absent (-/-). Discordant segregation columns show the number of clones in which only the enzyme marker or ASL was present. * APRT (adenine phosphoribosyltransferase; EC 2.4.2.7) was assayed by both a bioautographic method (1) and by the standard autoradiographic technique (30) in 18 hybrids. Segregation of APRT in hybrid clones scored identically by the autoradiographic or the bioautographic procedure.

RESULTS ASL in Human and Mouse Parental Cells and in Somatic Cell Hybrids. Human and mouse ASL from cultured cell extracts were separated electrophoretically in a Tris/EDTA/ borate buffer system (pH 8.6) (1) (Fig. 2, channels 1 and 2). Mouse cell lines produced a single slow band of activity that migrated to the same position as in all mouse tissues tested (Fig. 2, channel 1). No electrophoretic variants of ASL were detected under several electrophoretic conditions in any of the mouse tissue culture lines or in 25 mouse strains tested. Human tissue

culture lines (WI-38 and KB) produced a faster anodally migrating band in the buffer system (Fig. 2, channel 2).

Genetics: Naylor et al.

Proc. Natl. Acad. Sci. USA 75 (1978) (+)

Human -up

Mouse

Origin (-)

1

2

3

4

5

FIG. 2. Electrophoretic patterns of ASL from parental and human-mouse somatic cell hybrids. Channel 1, mouse cell line, LM/TK-; channel 2, human cell line, WI-38 VA13; channels 3 and 5, human-mouse hybrid cell clones showing negative expression of ASL; channel 4, human-mouse hybrid clones showing positive expression of ASL.

Two ASL phenotypes were observed in cell hybrids. One phenotype was identical to the mouse pattern (channels 3 and 5). However, in some cell hybrids ASL was expressed as a rodent band, three intermediate bands, and a faint or missing human enzyme (channel 4). The five molecular forms can be explained as the random assembly of two homologous subunits with different charges into two homotetramers and three heterotetramers (36). Since two copies of the mouse gene and only one copy of a human gene are usually present in hybrid clones, there is a progressive decrease in activity from the heteropolymers towards the human anodal enzyme due to relatively less human enzyme (9). A tetrameric structure for ASL was observed by Lusty and Ratner (37) by physical methods. Clones were scored positive for ASL if any of the heteropolymers were present and negative if only the mouse enzyme was expressed (Fig. 2). Cosegregation of ASL and GUS. Cosegregation of ASL and another enzyme marker in cell hybrids would indicate assignment of the genes coding for the two enzymes to the same chromosome (synteny). The syntenic relationships of human ASL were investigated by examining segregation patterns of ASL and other human enzyme markers coded by genes on all chromosomes except the Y chromosome. The Y chromosome was excluded for assignment because female cells express ASL. Twenty-three hybrid clones were established from three different fusions of human and rodent cells (DUV, ALR, and REX). The parents of these human-mouse hybrids were three

6161

different human cells and two different mouse lines. Hybrid clones were tested for ASL and 33 enzyme markers representing 23 of the 24 different human chromosomes (Table 1). The ASL marker showed concordant segregation with GUS but was discordant for all other enzymes tested. As indicated in Table 1, only one hybrid deviated from this pattern. This discordant clone could result from either chromosome breakage or a difference in sensitivity of the enzyme assays. Since the percent of discordancy is low, ASL and GUS could be closely linked on the same chromosome. The human gene for GUS has been confirmed to be assigned to chromosome 7 (8-10) and localized in the pter-*q22 region (11). It follows that the gene coding for ASL is also located on chromosome 7. Since only the GUS linkage group was necessary for expression of human ASL, more than one gene is not implicated for expression of human ASL. Chromosome Assignment of ASL. ASL was assigned to a chromosome in seven additional human-mouse cell hybrids whose chromosome analyses are given in Table 2 (11). To localize ASL, we studied ASL in. hybrids segregating a 7/9 translocation (7pter -7q22::9p24---i-9pter) that were produced from the fusion of RAG and a human 7/9 translocation [46,XY,t(7;9)(q22;p24)] fibroblasts (11). Concordant segregation of ASL and the pter- q22 region of human chromosome 7 was observed (Table 2), confirming the assignment of ASL to chromosome 7 previously made from enzyme segregation data. ASL and GUS segregated concordantly in these hybrids and with the pter--q22 region of chromosome 7 (Table 2), confirming the assignment and regional location of ASL. All other human chromosomes were excluded for assignment due to discordant segregation in chromosome and enzyme studies. DISCUSSION Bioautography can be used as a general method of isozyme visualization. Several enzymes that have not been visualized before are those involved in aminoacidurias. These enzymes of amino acid metabolism can be studied by bioautography because a large number of amino acid-requiring bacteria have been isolated (2). ASL, the enzyme deficient in patients with argininosuccinic aciduria, was visualized by bioautography in this study. The mapping of the ASL gene was investigated with human-mouse somatic cell hybrids. ASL segregated concordantly with GUS, but not with 32 other markers representing 23 human chromosomes. GUS has been assigned to chromosome 7 in humans (8-10), and ASL has been shown by chromosome analysis to be assigned to chromosome 7. Further studies with hybrids with a 7/9 translocation localized both ASL and GUS to the pter--q22 region of chromosome 7. The gene assigned to chromosome 7 is probably the structural gene for ASL, since human and heteropolymer forms of ASL

Table 2. Segregation of ASL and human chromosomes in human-mouse cell hybrids ASL GUS 1 2 3 4 5 6 7 7q-* 8 9 10 11 12 13 14 15 16 17 18

19 20 21 22 X Y Hybrid JSR-6D + - - - - + + + _ _ _ - + + + - - + + + + + + - + - + - - + _ _ + + JSR-17F + + + + _ + + - + +_ + - + - + + _ + + + + + + - + - + - - + - _ + + + JSR-17G + + + + + + + - + - - + + JSR-24C + + + + - + + + + + +- ++ + + + + + + + + - + - + + - + + - + + + +_ + JSR-24D + - - + - +-+---REX-11 BsAgB + + t + ++ - -+ - + + --+- -- + t DUA-6 Enzyme analyses and karyotyping of clones were performed on the same cell passage. JSR clones were hybrid clones from the fusion of a human cell containing a 7/9 chromosome translocation (10). * 7q- is the pter- q22 region of human chromosome 7. t These hybrids were derived from female parental cells.

6162

Genetics:

Naylor ei al.

were expressed in cell hybrids and no other chromosome was necessary for expression. ASL is absent from all tissues examined in most patients with argininosuccinic aciduria (38). This form of the disease may result from a change in the ASL structural gene assigned to chromosome 7. However, we have not excluded the possibility of a closely linked ASL control gene from the somatic cell genetic data. There is evidence suggesting that other genes may be involved in the expression of ASL since rare forms of argininosuccinic aciduria have been described in which some tissues retain ASL activity while others do not

(6).

Argininosuccinate synthetase (EC 6.3.4.5) and ASL, sequential enzymes in the conversion of citrulline to arginine, are repressed coordinately in the presence of arginine in aneuploid fibroblasts (39). Upon removal of arginine from the medium, activities of both enzymes increase S- to 20-fold (39). Therefore, it is important to determine if the two structural genes are closely linked due to their coordinate control. In a study by Carritt et al. (40), an argininosuccinate synthetase gene was syntenic with adenylate kinase-1 (AK-1) and was assigned to chromosome 9. In a later study, the gene assigned to chromosome 9 was proved to be the structural gene for argininosuccinate synthetase (41). Therefore, the argininosuccinate synthetase and ASL structural genes are not linked. These two genes, argG (argininosuccinate synthetase) and argH (ASL), are not closely linked in E. coli although both genes are repressed coordinately in the presence of arginine (42). ArgR, the regulatory gene for all the arginine biosynthetic genes in E. coli, controls the expression of genes at several locations on the E. coli chromosome (42, 43). Perhaps a similar control mechanism for genes at two different locations can be elucidated for juman expression of ASL and argininosuccinate synthetase. We thank Ms. J. K. Townsend, Ms. M. Byers, Mr. R. Eddy, Ms. L. Haley, Ms. A. Goggin, and Ms. L. Scrafford-Wolff for their excellent technical assistance. We are also grateful to Ms. C. Young for preparing the manuscript. This work was supported by National Institutes of Health Grants GM 21433 and HD 05196. 1. Naylor, S. L. & Klebe, R. J. (1977) Biochem. Genet. 15, 11931211. 2. Bachmann, B. J., Low, K. B. & Taylor, A. L. (1976) Bacteriol. Rev.

40, 116-167. 3. Naylor, S. L., Shows, T. B. & Klebe, R. J. (1978) Somat. Cell Genet., in press. 4. Ratner, S. (1973) Adv. Enzymol. 39, 1-90. 5. Westall, R. G. (1960) Biochem. J. 77, 135-144. 6. Glick, N. R., Snodgrass, P. J. & Schafer, I. A. (1976) Am. J. Hum. Genet. 28, 22-30. 7. Murphey, W. H., Patchen, L. & Guthrie, R. (1972) Biochem. Genet. 6, 51-59. 8. Lalley, P. A., Brown, J. A., Eddy, R. L., Haley, L. L. & Shows, T. B. (1976) Cytogenet. Cell Genet. 16, 184-187. 9. Lalley, P. A., Brown, J. A., Eddy, R. L., Haley, L. L., Byers, M. G., Goggin, A. P. & Shows, T. B. (1977) Biochem. Genet. 15, 367-382. 10. Grzeschik, K. H. (1976) Somat. Cell Genet. 2, 401-410. 11. Shows, T. B., Brown, J. A., Haley, L. L., Byers, M. G., Eddy, R.

Proc. Natl. Acad. Sci. USA 75 (1978)

12.

13. 14.

15. 16. 17. 18.

19. 20. 21. 22. 23. 24. 25.

26. 27. 28. 29.

30. 31. 32.

33. 34.

35.

36. 37. 38. 39. 40. 41. 42.

L., Cooper, E. S. & Goggin, A. P. (1978) Cytogenet. Cell Genet. 21,99-104. Solomon, E., Bobrow, M., Goodfellow, P. N., Bodmer, W. F., Swallow, D. M., Povey, S. & Noel, B. (1976) Somat. Cell Genet. 2, 125-140. Shows, T. B. & Brown, J. A. (1975) Proc. Natl. Acad. Sci. USA 72,2125-2129. Klebe, R. J., Chen, T. R. & Ruddle, F. H. (1970) J. Cell. Biol. 45, 74-82. Littlefield, J. W. (1964) Nature (London) 203, 1142-1144. Kit, S., Dubbs, D. R., Piekarski, L. J. & Hsu, T. C. (1963) Exp. Cell Res. 31,297-312. Littlefield, J. W. (1964) Science 145,709-710. Champion, M. J., Brown, J. A. & Shows, T. B. (1978) Cytogenet. Cell. Genet., in press. Gray, C. H. & Tatum, E. L. (1944) Proc. Natl. Acad. Sci. USA 30,404-410. Shows, T. B. (1972) Proc. Natl. Acad. Sci. USA 69,348-352. Shows, T. B. (1972) Biochem. Genet. 7, 193-204. Shows, T. B., Scrafford-Wolff, L., Brown, J. A. & Meisler, M. (1979) Somatic Cell Genet., in press. Shows, T. B., Brown, J. A., Eddy, R. L., Byers, M. G., Haley, L. L., Cooper, E. S. & Goggin, A. P. (1978), Hum. Genet. 43, 119-125. Lalley, P. A., Rattazzi, M. C. & Shows, T. B. (1974) Proc. Natl. Acad. Sci. USA 71, 1569-1573. Kaplan, J. C. & Beutler, E. (1968) Nature (London) 217, 256258. Shows, T. B. & Lalley, P. A. (1974) Biochem. Genet. 11, 123141. Shows, T. B. (1975) in Isozymes III, ed. Markert, C. L. (Academic, New York), pp. 619-636. Riccuiti, F. & Ruddle, F. H. (1973) Nature (London) New Biol. 241, 180-182. Shows, T. B. (1974) in Somatic Cell Hyboidization, eds. Davidson, R. L. & de la Cruz, F. (Raven, New York), pp. 15-23. Mowbray, S., Watson, B. & Harris, H. (1972) Ann. Hum. Genet. 36, 153-162. Elsevier, S. M., Kucherlapati, R. S., Nichols, E. A., Creagan, R. P., Giles, R. E., Ruddle, F. H., Willecke, K. & McDougall, J. K. (1974) Nature (London) 251, 633-635. Creagan, R. P., Tischfield, J. A., Nichols, E. A. & Ruddle, F. H. (1973) Lancet ii, 1449. Chen, S. H., Malcolm, L. A., Yoshida, A. & Giblett, E. R. (1971) Am. J. Hum. Genet. 23,87-91. Shin, S., Kahn, M. & Cook, P. R. (1971) Biochem. Genet. 5, 91-99. Shows, T. B. & McAlpine, P. J. (1978) in Winnipeg Conference (1977): Fourth International Workshop on Human Gene Mapping. Birth Defects: Original Article Series (The National Foundation, New York), in press. Markert, C. L. & Moller, F. (1959) Proc. Natl. Acad. Sci. USA 45,753-763. Lusty, C. J. & Ratner, S. (1972) J. Biol. Chem. 247, 70107022. Hsia, Y. E. (1974) Prog. Gastroent. 67,347-374. Schimke, R. (1964) J. Biol. Chem. 239, 136-145. Carritt, B., Goldfarb, P. S. G., Hooper, M. L. & Slack, C. (1977) Exp. Cell Res. 106, 71-78. Carritt, B. (1977) Cytogenet. Cell Genet. 19, 44-48. Maas, W. K., Maas, R., Wiame, J. M. & Glansdorff, N. (1964) J.

Mol. Biol. 8,359-364.

43. Maas, W. K. & Clark, A. J.

(1964) J. Mol. Biol. 8,365-370.