Proc. Nat!. Acad. Sci. USA Vol. 77, No. 10, pp. 6166-6170, October 1980 Medical Sciences

Complementation of A in somatic hybrids of metachromatic and multiple deficiency disorder fibroblasts (unit-gravity sedimentation/nonselective isolation technique/nonallelic mutation/inherited neurodegenerative diseases/ lipidosis) PATRICIA L. CHANG AND RONALD G. DAVIDSON Department of Pediatrics, McMaster University, Faculty of Health Sciences, Hamilton, Ontario, Canada L8N 3Z5 Communicated by Harry Harris, July 10, 1980

ABSTRACT Metachromatic leukodystrophy and multiple technique was applied to study the genetic defects of two inborn sulfatase deficiency disorder are severe neurodegenerative errors of , metachromatic leukodystrophy (MLD) diseases inherited as separate autosomal recessive traits. Aryl- and its O-variant, multiple sulfatase deficiency disorder sulfatase A (aryl-sulfate sulfohydrolase, EC 3.1.6.1) activity is deficient in both diseases but in multiple sulfatase deficiency (MSDD). disorder, activities of B and C and other Hereditary metabolic diseases have been useful in providing are also reported to be reduced. Somatic hybrid cell clones answers to important basic genetic questions. MLD and MSDD produced by fusing cultured fibroblasts from patients with these are severe neurodegenerative diseases both characterized by diseases were isolated by a nonselective technique based on a deficiency of arylsulfatase A (ARSA; aryl-sulfate sulfohy- unit-gravity sedimentation. Arylsulfatase A activity was restored drolase, EC 3.1.6.1) activity although they are inherited as in these hybrids. The complemented resembled the normal arylsulfatase A in heat stability, pH optimum, Ki, separate autosomal recessive traits. They have similar clinical electrophoretic mobility, and immunologic reactivity. Because symptoms but can be differentiated by subtle differences, such a structurally normal enzyme can be restored in a hybrid only as the presence of Alder-Reilly granules in leukocytes (10) and through intergenic comlementation, these results indicate that the accumulation of glycosaminoglycans in addition to cere- the mutations responsible for the deficiency of arylsulfatase A broside sulfates in patients with MSDD (11). In MLD, only activity in metachromatic leukodystrophy and multi le sulfa- ARSA activity is affected, but in MSDD, deficiencies of other tase deficiency disorder are nonallelic and that at least two genetic loci control the expression of arylsulfatase A activity in arylsulfatases occur-i.e., (ARSB), arylsulfatase the . Furthermore, arylsulfatase C activity was C (ARSC), and steroid sulfatases (11-14). This raised a number also restored to normal in the hybrids, indicating that a common of questions. What is the nature of the genetic defects involved sulfatase inhibitor is not the cause of the multiple sulfatase such that two distinct diseases are deficient in the same enzyme deficiency. activity? Are they due to allelic or nonallelic mutations? Fur- thermore, how does a single genetic defect in MSDD diminish Many inherited metabolic diseases are caused by specific en- the biochemical activity of so many apparently unrelated en- zyme deficiencies. However, among diseases characterized by zymes? Answers to these questions were sought by looking for the same enzyme deficiency, many variants exist with diverse complementation of ARSA activity in hybrid clones formed clinical manifestations. Frequently, heterokaryons formed by between cultured skin fibroblasts from MLD and MSDD pa- fusing cultured cells from these variants showed complemen- tients followed by further definition of the nature of such tation of the previously deficient enzyme activity, indicating complementation biochemically and that the genetic defects in these variants were not identical al- immunologically. though the biochemical manifestations were apparently the same. Complementation groups have been found among the MATERIALS AND METHODS' variants of GM1 gangliosidosis (1), the mucopolysaccharidoses Cell Culture. Cultured skin fibroblasts from a patient with (2), xeroderma pigmentosum (3), methylmalonicacidemia (4), the juvenile form of MLD were supplied by J. A. Lowden galactosemia (5), and the GM2 gangliosidoses, Tay-Sachs and (Hospital for Sick Children, Toronto, Ontario); those from a Sandhoff diseases (6-8). The lack of proper selection medium, patient with MSDD were supplied by H. Kihara (University however, precluded the isolation of somatic cell hybrids from of California, Los Angeles, CA). Skin fibroblasts for controls these heterokaryons for more precise identification of the were obtained from patients with normal karyotypes and no complemented enzyme and further biochemical and genetic known metabolic disease. Fibroblasts were maintained in Ham's studies. F-10 culture medium supplemented with L-glutamine (2 mM) A nonselective method to circumvent the above limitation and 10-20% (vol/vol) fetal calf serum under 5% CO2 in air and in isolating euploid human somatic hybrid clones was recently 100% humidity at 370C. Penicillin (100 units/ml) and strep- developed (9). It depended on separating a fusogen-treated tomycin (100 ,ug/ml) were included in the culture medium mixture of two fibroblast strains according to size by unit- during experiments. gravity sedimentation. The fractions highly enriched with Nonselective Isolation and Identification of Hybrids. double-nucleated cells were cloned to isolate the hybrid cells. Hybrid somatic cell clones were produced by fusion of fibro- Hybrids were identified by karyotyping and by an enzyme blasts with polyethylene glycol 1000 (15) and isolation by polymorphism characteristic of the two parent cell strains. This unit-gravity sedimentation as described by Chang et al. (9). Karyotyping was performed in situ on cells grown on cov- The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "ad- Abbreviations: MLD, metachromatic leukodystrophy; MSDD, multiple vertisement" in accordance with 18 U. S. C. §1734 solely to indicate sulfatase deficiency disorder; PGM, phosphoglucomutase; ARSA, ar- this fact. ylsulfatase A; ARSB, arylsulfatase B; ARSC, arylsulfatase C. 6166 Downloaded by guest on September 25, 2021 Medical Sciences: Chang and Davidson Proc. Natl. Acad. Sci. USA 77 (1980) 6167 erslips (2 cm2) by treating the cells with colchicine (5 Mg/ml of Immunoelectrophoresis. Crude fibroblast extracts were complete medium) for 4 hr at 370C, incubating them in 75 mM electrophoresed in agar plates with a barbitone buffer according KCl for 0.5 hr at 370C, fixing them with methanol/acetic acid, to standard procedure. An IgG fraction isolated from rabbit 3:1 (vol/vol), and staining them with Giemsa. antiserum against a partially purified ARSA preparation from Clones were judged to be diploid if three or more mitotic human liver was used to react against the electrophoresed fi- figures on the coverslip showed a diploid modal number of broblast extracts. The agar plates were then washed extensively . They were judged to be tetraploid if 10 mitotic with isotonic saline. Although no precipitin bands were visible, figures (or all the mitotic figures on the coverslip if the total was ARSA activity was not inhibited by the presence of the antibody fewer than 10) showed a tetraploid modal number of chro- and could be detected in normal fibroblasts when developed mosomes. with p-nitrocatechol sulfate as the substrate by the method of Cells were harvested by rinsing once with isotonic saline, Manowitz et al. (22). followed by trypsinization. Cells were washed twice with iso- Protein was determined by the method of Lowry et al. (23) tonic saline and the pellet was kept at -70'C until use. Fibro- with bovine serum albumin as the standard. blast extract was prepared by freeze-thawing in 50 mM acetate All chemicals were of analytical or reagent grade. Tissue buffer (pH 5.0) or lysis buffer (16) (10,ul/cell pellet from a culture reagents were from GIBCO. Polyethyleneglycol 1000 60-mm culture dish) five times on dry ice and centrifuging at for fusing cells was from BDH Chemicals, Toronto, Canada. 12,000 X g for 2 min at room temperature. The supernatant was The antiserum against ARSA was kindly provided by H. L. saved for ARSA, ARSB, and phosphoglucomutase (PGM) Nadler (Children's Memorial Hospital, Chicago, IL). analyses and protein determinations. The pellet was used for ARSC and protein determinations. RESULTS Enzyme Assays. PGM polymorphism was identified by a Isolation of hybrid clones procedure modified from that of Harris and Hopkinson (17) and After cultured skin fibroblasts from MLD and MSDD patients Khan (18) by use of Cellogel instead of starch gel for elec- were fused and separated by unit-gravity sedimentation, 123 trophoresis. Samples (1-5 Ml) containing 20,g of protein were clones were successfully identified by karyotyping. Of these, placed on Cellogel (17 X 17 cm, Kalex) and electrophoresed in 86 were diploid, 12 were mixed diploid and tetraploid, and 25 10 mM phosphate/citrate at pH 7.0 with a constant current of were pure tetraploid clones. 7 mA for 4 hr at room temperature. After the run, the Cellogel The origin of the tetraploid clones was established by the was dipped in the substrate solution, porous side down, for 1 min PGM1 electrophoretic mobility on Cellogel. The MLD fibro- and incubated at 37°C for 3-5 min in the dark. The reaction blasts had PGM11 and the MSDD fibroblasts had PGM12. Five was stopped immediately at the appearance of the first intense of the tetraploid clones were found to have the PGM11 pattern blue band by dipping the gel in 37% (vol/vol) formalde- (homologous MLD fusion), 17 had the PGM12 pattern (ho- hyde. mologous MSDD fusion), and 3 had the hybrid PGM12-1 pat- ARSA and ARSB were identified by Cellogel electrophoresis tern expected from a fusion between MLD and MSDD cells. by the method of Rattazzi et al. (19). Quantitative assays were Al PGM1 patterns in the tetraploids were confirmed again after performed with 4-methylumbelliferylsulfate (Koch Light, 6-11 passages in culture. Colnbrook, England) as the substrate, based on the principle that Ag+ selectively inhibits ARSA (20). The total activity of Screening for ARSA activity in tetraploid clones ARSA and ARSB was assayed as follows: an appropriate amount The soluble arylsulfatase activities (ARSA and ARSB) of these of fibroblast extract was made up in a total of 80 ,l with 50mM tetraploid clones were monitored by Cellogel electrophoresis acetate buffer (pH 5.6). A freshly prepared 30 mM lead acetate and by enzymatic assays on fibroblast extracts with 4-methyl- solution (20 Ml) was added just before the incubation. The re- umbelliferylsulfate as substrate. ARSA migrated anodally action was started by addition of 100 ,ul of a substrate solution during Cellogel electrophoresis and appeared as a bright fluo- containing 10 mM 4-methylumbelliferylsulfate in 50 mM ac- rescent band when stained with 4-methylumbelliferylsulfate. etate buffer (pH 5.6) and the mixture was incubated at 370C This band was not seen in the MLD or MSDD diploid fibroblasts for 15 min. The reaction was terminated by adding 1 ml of 0.2 or in the tetraploid cells formed from homologous fusions. In M carbonate/glycine buffer, pH 10.4, containing 1 mM EDTA. the three hybrid tetraploid clones, however, a distinct band of The activity of ARSB was assayed as above except for the in- activity corresponding to that of ARSA from a normal fibroblast clusion of 0.3 mM AgNOs in the assay mixture. The difference strain appeared (Fig. 1). between the two determinations was the activity of ARSA. The quantitative assays of ARSA, ARSB, and ARSC activities ARSC activity was assayed in the pellet by the method of Eto of normal fibroblasts, the diploid MLD and MSDD parent cell et al. (13) after it was washed once with 50mM Tris-HCl (pH strains, and their tetraploid clones that were vigorous enough 8.4). to propagate in culture are shown in Table 1. MLD cells had Heat Stability and Enzyme Kinetic Studies. Fibroblast 15% of normal ARSA activity, whereas ARSB and ARSC were extracts were diluted with dialyzed bovine serum albumin (1 within normal range. In the single tetraploid clone from ho- mg/ml of 50 mM acetate buffer, pH 5.0), then dialyzed first mologous fusion of MLD cells, the same enzyme activity pattern against 1000 vol of 25 mM acetate buffer (pH 5.0) overnight, was maintained as in the parent diploid cells. In the MSDD followed by 1000 vol of deionized distilled water for 1-2 hr diploid cells, only ARSB was within the normal range; ARSA before use. Enzyme preparations were inactivated at various and ARSC were about 25% of normal. In the five tetraploid temperatures for 10 min each. The residual activities were clones from homologous fusion of MSDD cells, all three aryl- transformed into logarithmic functions by the method of sulfatases showed increased activities. ARSA was about 50% of Feinstein et al. (21). A straight line was fitted to these trans- normal. ARSC reached normal levels, and ARSB attained al- formed functions by linear regression from which the tem- most 300% of the normal value. In the hybrid clone from the perature at which 50% of the activity remained (TM0) was cal- heterologous fusion of MLD and MSDD cells, the activity of culated. For estimations of Km, incubations were for 5 instead ARSA reached normal levels; the activities of ARSB and ARSC of 15 min. were elevated similar to the MSDD self-fusion tetraploid. These Downloaded by guest on September 25, 2021 6168 Medical Sciences: Chang and Davidson Proc. Natl. Acad. Sci. USA 77 (1980)

FIG. 1. Cellogel electrophoresis of ARSA and ARSB activities of diploid fibroblasts and tetraploid clones from MLD and MSDD fusion. Lane 1, normal; lane 2, MSDD; lanes 3-5, MSDD X MSDD; lane 6, MLD x MSDD. 0, origin. Arrows, ARSA activity. quantitative results were confirmed in a separate experiment and the hybrid clone. From a Lineweaver-Burk plot of these (data not shown). data, the Km was calculated by linear regression to be 2.90 + 0.77 mM for the three normal fibroblast strains and 1.53 mM Characterization of restored ARSA in hybrid clones for the hybrid clone. The apparent Km for the normal cell Heat Stability. The percentage of ARSA activity remaining strains measured in different experiments varied from 1.3 to after heating at 50, 55, 60, 65, and 700C for 10 min each in three 8.9 mM. Those of the hybrid clone were from 0.815 to 2.09 mM. different normal fibroblast cell strains and one hybrid clone is Within the same experiment, the Km of the hybrid ARSA ac- shown in Fig. 2. The temperature at which 50% of the activity tivity was usually about half of the normal value. remained (T50) was 59.0 + 1.40C for the three normal cell Immunological Identification. Extracts from a normal fi- strains and 60.00C for the hybrid clone. broblast strain and the hybrid clone were electrophoresed and Kinetics. pHoptimum. The relative activity of ARSA in the then treated with a partially purified preparation of antibody hybrid clone at different pH values was compared with that of a normal fibroblast strain (Fig. 3). The optimal pH for both t hybrid and normal strains was identical. With acetate as the 100- buffering ion, the highest activity observed was at pH 5.6 (range tested, 4.5-5.6). With maleate as the buffering ion, the highest 90- activity was at pH 6.0 (range tested, 5.3-6.5). Michaelis constant. The activity of ARSA at different con- centrations (1-10 mM) of the substrate 4-methylumbellifer- 80- ylsulfate was plotted for three different normal fibroblast strains :. NX p70- .\ \ Table 1. Arylsulfatase activities of diploid fibroblasts and A tetraploid clones from MLD and MSDD fusion 60- Strain* ARSAt ARSBt ARSCt d.; Diploid Normal (2) 176 203 29.8 ~3 1*R (121-242) (171-231) (15.0-44.6) MLD (1) 28.4 187 23.3 MSDD (1) 47.4 189 4.3 1'.

Tetraploid clones 20- MLD X MLD (1) 38.2 236 14.0 MSDD X MSDD 92.6 ± 38.4 (4) 604 ± 159 22.3 + 12.5 (5) (54.8-146) (362-808) (4.9-39.8) 10- MLD X MSDD (1) 172 532 48.8

All activities are expressed as nmol of 4-methylumbelliferylsulfate U 50 60 7b hydrolyzed per mg of protein per hr at 370C. Temperature * Number of strains or clones used are in parentheses. (C) t Values in parentheses are the range obtained from different cell FIG. 2. Heat stability of ARSA in normal fibroblast and hybrid strains. Each value was determined in duplicate or triplicate; the MLD x MSDD clone. O *, .. .A, and *- -*,- Normal fi- means 4: SD from different strains are shown. broblasts; 0-0, hybrid clone. Downloaded by guest on September 25, 2021 Medical Sciences: Chang and Davidson Proc. Natl. Acad. Sci. USA 77 (1980) 6169

10c The restored ARSA activity exhibited properties indistin- guishable from the normal enzyme when tested for heat sta- bility (Fig. 2), optimal pH (Fig. 3), and Km. Its electrophoretic mobility, as shown by Cellogel electrophoresis (Fig. 1), and immunologic reactivity, as tested by immunoelectrophoresis 80O on agar gel (Fig. 4), were also similar to those of the normal enzyme. In assessing the heat stability of this enzyme, it was 70- important to dialyze the samples thoroughly and to inactivate the enzyme over a range of temperatures in order to compute T5o, the temperature at which 50% of the activity remained. If the heat stability were tested at only a single temperature, there was too much variation even among different normal cell > strains to allow for unequivocal identification. In contrast, the T50 was a more consistent parameter (see Fig. 2). 40- .1 Because heat stability is one of the most sensitive techniques A,' .% * b for detection of structural changes among , the iden- 11 tical T50 in the normal and complemented ARSA enzymes . ba strongly suggested that the restored enzyme was structurally 20- a / normal. This was corroborated by the similarities in kinetic, a/ electrophoretic, and immunologic properties of the normal and the complemented ASA activities. These data showed clearly that the criteria for intercistronic (24, 25) complementation I- were met in that a structurally normal ARSA was restored in 5:5 6.0 6 4.5 5.0 hybrids formed between MSDD and MLD cells. pH The detection of complementation between MLD and FIG. 3. Activity profile ofARSA in normal fibroblast and hybrid MSDD cells depended on the successful isolation of the hybrid MLD X MSDD clone at different pH values. A -*A, Normal fi- clones. In unsorted mixtures of fused and unfused cells from broblast in acetate buffer; * - - *, hybrid clone in acetate buffer; previous fusion experiments involving MLD cells and two other .... ., normal fibroblast in maleate buffer; ---a, hybrid clone MSDD cell strains, we failed to detect significant increases in in maleate buffer. ARSA activity. Absence of complementation between MLD and MSDD cells was also reported by workers in two other against ARSA. They were stained for arylsulfatase activity; both laboratories (14, 26) using similar techniques of cell fusion showed immunoprecipitated activity bands of identical elec- without isolation of heterokaryons or hybrids. This points out trophoretic mobility (Fig. 4a). Under similar conditions, the the advantage of isolating pure hybrid clones, especially when MLD cells did not show any enzymatically active precipitin a result is obtained in an unsorted fusion mixture. band, whereas the MSDD cells showed a weak trace of activity negative In the in which a normal rabbit serum was The enzyme deficiency in MLD was shown to be a structural (Fig. 4b). control, mutant in an was used instead of the anti-ARSA preparation, no bands of activity which abnormal ARSA molecule immu- were observed in of the cell extracts. nologically crossreactive with normal ARSA but enzymatically any deficient (27). In MSDD, however, various hypotheses have DISCUSSION been put forward to explain the multiple sulfatase deficiencies Cultured skin fibroblasts from patients with MLD or MSDD (11, 13). Whatever the genetic defect is, this report has shown expressed the biochemical defects of these diseases in part as that it must involve a locus distinct from that responsible for a deficiency of ARSA activity. By fusing these two types of fi- MLD. broblasts and isolating the hybrid clones, we have shown that One hypothesis suggested that in MSDD a structural ARSA activity was restored by complementation. coding for a common subunit of all the deficient sulfatases was defective. If that were the case, all the residual sulfatase ac- a tivities in MSDD would be due to structurally abnormal en- zymes. This was not supported by recent studies which showed HYB that the residual ARSA in MSDD was due to a reduced amount of an otherwise normal enzyme (28) and that its level can be modulated by the pH of the culture medium (29). Another hypothesis suggested that MSDD was due to the NOR presence of a common inhibitor of this group of sulfatases. This was supported neither by the usual type of mixing experiments b (13) nor by our results. We showed that although ARSC was MSDD deficient in the MSDD diploid parent cells, it was not at all inhibited in the MLD X MSDD hybrids. A third hypothesis suggested that MSDD was due to a defect of a regulatory gene that controlled the level of all the sulfatases. MLD In this case, the positive complementation of ARSA activity between MLD and MSDD cells may be interpreted as follows. FIG. 4. Immunoelectrophoresis of ARSA in normal, MLD, and In MLD, the structural gene coding for ARSA is defective but MSDD diploid fibroblasts and a hybrid MLD X MSDD tetraploid the regulator gene controlling the synthesis of all the other clone. Fibroblast extracts in the wells were electrophoresed and then reacted against anti-ARSA IgG in the trough. After extensive washing, sulfatases is normal. Therefore, structurally normal ARSB and ARSA activity was revealed as Hatchett's brown bands. HYB, hybrid; ARSC continue to be synthesized in normal amounts. In MSDD, NOR, normal. the regulatory gene is defective but the structural gene coding Downloaded by guest on September 25, 2021 6170 Medical Sciences: Chang and Davidson Prqc. Nati. Acad. Sci. USA 77 (1980)

for ARSA is intact. Therefore, structurally normal sulfatases 11. Murphy, J. V., Wolfe, H. J., Balazs, E. A. & Moser, H. W. (1971) were synthesized but in much reduced amounts. In the hybrids, in Lipid Storage Diseases, eds. Bernsohn, J. & Grossman, H. J. the normal regulator gene from the MLD genome and the (Academic, New York), pp. 67-110. normal structural gene from the MSDD genome complement 12. Eto, Y., Rampini, S., Wiesmann, U. & Herschkowitz, N. N. (1974) each other to produce ARSA molecules that are normal in J. Neurochem. 23, 1161-1170. structure and amount. We also noted that even in the tetraploids 13. Eto, Y., Wiesmann, U. N., Carson, J. H. & Herschkowitz, N. N. produced by homologous fusion of MSDD cells, ARSA, ARSB, (1974) Arch. Neurol. 30, 153-156. and ARSC were generally elevated compared to the parent 14. Basner, R., von Figura, K., Glossl, J., Klein, U., Kresse, H. & diploid cells. It could be argued that in the tetraploid genome Mlekusch, W. (1979) Pediatr. Res. 13, 1316-1318. of MSDD the amount of defective, or partially active, regulator 15. Davidson, R. L., O'Malley, K. A. & Wheeler, T. B. (1976) Somatic gene product is now doubled, hence allowing for the increased Cell Genet. 2, 271-280. levels of these enzymes. 16. van Someren, H., van Henegouwen, H. B., Los, W., Wurzer- Two criteria were proposed for identifying a regulatory gene Figurelli, E., Doppert, B., Vervloet, M. & Khan, P. M. (1974) mutation (30). First, the mutation must alter the rate of enzyme Humangenetik 25,189-201. must 17. Harris, H. & Hopkinson, D. A. (1976) Handbook of Enzyme synthesis or breakdown. Second, the regulatory gene locus Electrophoresis in Human Genetics (North-Holland, Amster- be separate from the structural gene locus. The intergenic dam), 2.7.5.1. complementation between MSDD and MLD has fulfilled at 18. Khan, P. M. (1971) Arch. Biochem. Biophys. 145,470-483. least the second requirement. It remains to be shown whether 19. Rattazzi, M. C., Marks, J. S. & Davidson, R. G. (1973) Am. J. the first requirement will be met in the case of MSDD. Hum. Genet. 25,310-316. Assistance in tissue culture by Dr. N. E. Rosa is gratefully ac- 20. Christomanou, H. & Sandhoff, K. (1977) Clin. Chim. Acta 79, knowledged. This work was supported by a grant from the Medical 527-531. Research Council of Canada (MA 5789). 21. Feinstein, R. N., Sacher, G. A., Howard, J. B. & Braun, J. T. (1967) Arch. Biochem. Biophys. 122,338-343. 1. Hoeksema, H. L., Van Diggelen, 0. P. & Galjaard, H. (1979) 22. Manowitz, P., Goldstein, L. & Bellomo, F. (1978) Anal. Biochem. Blochim. Biophys. Acta 566,72-79. 80, 423-429. 2. Kato, T., Okada, S., Yutaka, T., Inui, K., Yabuuchi, H., Chiyo, 23. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. H., Furuyama, J. & Okada, Y. (1979) Biochem. Biophys. Res. (1951) J. Biol. Chem. 193,265-275. Commun. 91, 114-117. 24. Cabin, I. & Villarejo, M. R. (1975) Annu. Rev. Biochem. 44, 3. Cleaver, J. & Bootsma, D..(1975) Annu. Rev. Genet. 9, 19-38. 295-313. 4. Gravel, R. A., Mahoney, M. J., Ruddle, F. H. & Rosenberg, L. E. & C. Annu. Rev. Microbiol. (1975) Proc. Nati. Acad. SW. USA 72,3181-3185. 25. Schlesinger, M. J. Levinthal, (1965) 5. Nadler, H. L., Chacko, C. M. & Rachmeler, M. (1970) Proc. Nati. 19,267-284. Acad. Sci. USA 67,976-982. 26. Horwitz, A. L. (1979) Proc. Natl. Acad. Sci. USA 76, 6496- 6. Rattazzi, M. C., Brown, J. A., Davidson, R. G. & Shows, T. B. 6499. (1976) Am. J. Hum. Genet.- 28,143-154. 27. Shapira, E., DeGregorio, R. R. & Nadler, H. L. (1978) Pediatr. 7. GaIjaard, H., Hoogeveen, A., de Wit-Verbeek, H. A., Reuser, A. Res. 2, 199-203. J. J., Keijzer, W., Westerveld, A. & Bootsma, D. (1974) Exp. Cell 28. Fiddler, M. B., Vine, D., Shapira, E. & Nadler, H. L. (1979) Na- Res. 87,444-448. ture (London) 282,98-100. 8. Thomas, G. H., Taylor, H. A., Miller, C. S., Axelman, J. & Migeon, 29. Fluharty, A. L., Stevens, R. L., de la Flor, S. D., Shapiro, L. J. & B. R. (1974) Nature (London) 250,580-582. Kihara, H. (1979) Am. J. Hum. Genet. 31; 574-580. 9. Chang, P. L., Joubert, G. I. & Davidson, R. G. (1979) Nature 30. Paigen, K. (1971) in Enzyme Synthesis and Degradation in (London) 278, 168-170. Mammalian Systems, ed. Rechcigl, M., Jr. (Univ. Park Press, 10. Austin, J. H. (1973) Arch. Neurol. 28, 258-264. Baltimore), pp. 1-46. Downloaded by guest on September 25, 2021