Monoclonal Antibodies against Human DNA Polymerase-a Inhibit ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY Q 1985 by The American Society of Biological Chemists, Inc,

Vol. 260, No. 1, Iesue of January 10, p : 134-138,1985 nnted &nU.S.A.

Monoclonal Antibodies against Human DNA Polymerase-a Inhibit DNA Replication in Permeabilized Human Cells* (Received for publication, March 8,1984)

Michael R. Miller+, RogerG. UlrichO, Teresa Shu-FongWangll, and David Kornllll From the $Department of Biochemistry, West Virginia University School of Medicine, Morgantown, West Virginia 26506, the $Department of Pathology and Toxicology, The Upjohn Co., Kalamazoo, Michigan 49001, and the (Laboratory of Experimental Oncology, Department of Pathology, Stanford University School of Medicine, Stanford, California 94305

Monoclonal neutralizing antibodies against DNA polymerase-a substantially inhibit nuclear DNA replication in lysolecithin-permeabilizedcultured human fibroblasts. The degree of inhibitionDNA of synthesis is proportional to antibody concentration, and the effect is specific in that RNA synthesis measured under the same experimental conditionsis unperturbed. Autoradiographic data demonstrate that the magnitude of inhibition measured in the mass culture reflects the uniformresponseof all the constituent cells inthe target population. These observations confirmparthe ticipation of DNA polymerase-a in replicative DNA synthesis and identify a versatile, novel approach to the dissection of mammalian processes of DNA replication and repair.

A variety of experimental approaches have been used to prepare subcellular mammalian systems with which to investigate processes of DNA replication and repair. A major goal of these efforts is to obtain preparations that are permeable to metabolic substrates and inhibitors that cannot readily penetrate intact cells and at the same time retain sufficient organizational integrity to support physiologically valid conclusions. We (1-3) have previously describedthe preparation of selectively permeabilized cells with lysolecithin and have documented that such cells exhibit acceptable preservation of gross structural integrity and the capacity to carry out synthesis of DNA, RNA, and protein at near-in vivo rates for reasonable lengths of time. In particular, such selectively permeabilized Chinese hamster cells and human fibroblasts have been shown to exhibit brisk rates of semiconservative, discontinuous DNA synthesis that is cell-cycle dependent and has been demonstrated to be localized essentially exclusively to the nuclear compartment (1,3-5). Among other purposes, wewish to use this method to examine the degree to which DNA polymerase-a, a putative replicative polymerase (6),and DNA polymerase-@,a putative repair enzyme (6, 7), may participate, individually or conjointly, in the repair of DNA damage inducedby a variety of

* These studies were supported in part by Grant NP-300 from the American Cancer Society, Grant 2507 RR 05433-22 from the National Institutes of Health, a research grant from the West Virginia University Medical Corp., and a West Virginia University Senate research grant to M. R. M. and by Research Grant CA-14835 from the National Cancer Institute anda gift from the George D. Smith Fund to D. K. The costs of publication ofthis article were defrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 11 To whom reprint requests should be addressed.

chemical and physical agents. Our studies to date (2, 3), like innumerable others that have been concerned withthe physiological roles of these polymerases (8),have been limited to the use of diverse polymerase inhibitors that are believed to be reasonably selective, but generally not absolutely specific, for the different polymerase classes, even whentested in the entirely dissimilar context of in vitro assay of partially or the extensively purified enzyme fractions. The recent establishment of a panel of murine hybridomas producing monoclonalantibodies against DNA polymerase-a (9.10) offers a setof specific reagents that could significantly extend the scope of inhibitor studies and, in principle, prove most useful inthe task of dissecting the biochemical mechanisms of DNA replication and repair. In an initial exploration of this potential application, we have succeeded in demonstrating that several of these monoclonal antibodies are in fact capable of entering lysolecithin-permeabilized human cells and of inhibiting nuclear DNA replication. EXPERIMENTAL PROCEDURES

Materials-Aphidicolin was supplied by Imperial Chemical Industries, Cheshire, England. [3H]dTTP, [‘HIUTP (10-20 Ci/mmol), and Biofluor scintillation fluid were purchased from New England Nuclear. NTB3 emulsion was obtained from Kodak. Cell culture media were purchased from Gibco, and fetal calf serum was from Microbiological Associates. Unless otherwise indicated, all other reagents were from Sigma. Protein was measured by the method of Bradford (ll), using bovine serum albumin as a standard. CeU Culture-Human diploid fibroblasts, designated Hs24F, were established by the Naval Biosciences Laboratory, Oakland, CA and were obtained from the American Type Culture Collection. The cells were propagated in 45% Dulbecco’s modifled Eagle’s medium, 45% Ham’s F-10 medium, 10% fetal calf serum, supplemented with gentamycin (10 pglml). DNA synthesis studies were performed with exponential cells harvested between cell passage numbers 8 to 20. Monoclonal Antibodies-Monoclonal antibodies were prepared and purified as described by Tanaka et 01. (9). The purified IgGs were adjusted to 2.0 mg/ml, and heat-treated albumin (10 mg/ml in 50% glycerol, heated at 85 “C for 1 h) was added to a final concentration of 5 mg/ml. The IgG solutions were then dialyzed exhaustively against Solution A (150 mM sucrose, 80 mM KCl, 5 mM MgC12,35 mM Hepes,’ pH 7.4) containing 0.02% NaNs to remove ammonium sulfate. CeU Permeabilizatwn and Assays of RNA and DNA SynthesisProtocols for permeabilizing cells with lysolecithin (Sigma, type I) and measuring DNA replication have been described in detail (1-5). Briefly, following trypsinization, cells were collected by centrifugation (5 min at 1000 X g), washed twice at 4 “C in Solution A containing 0.2 mM phenylmethylsulfonyl fluoride, and finally suspended in Solution A at 0 “C at. 8 X lO’/ml. Lysolecithin was added to a final concentration of 0.5 mg/ml, and permeabilization was monitored by trypan blue exclusion. Generally, >95% of the cells were permeabilized within 1 min. For assay of DNA replication, permeable cells were diluted to 2 X The abbreviations used are: Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; HF, human diploid fibroblast.

134

Inhibition of DNA Replication with Anti-polymerase-a Antibodies 107/mlin Solution A containing 5 mM phosphoenolpyruvate, 1.25 mM ATP, 0.1 mM each CTP, GTP, and UTP, 0.25 mM each M T P , dCTP, and dGTP, 50 p~ dThd, and 50 p~ ['HId'ITP at 2.0 Ci/mmol and were preincubated at 0 'C for 60 min in the presence or absence of dialyzed IgGs, as indicated. The samples were then incubated at 37 "C for 30 min, and incorporation of ['HIdTTP into DNA wasquantitated (1,495). To measure RNA synthesis, permeable cells were diluted to 2 X 107/ml in Solution A containing 1.25 mM ATP, 5 mM phosphoenolpyruvate, 200 p~ CTP and GTP, and 10 p~ ['HIUTP at 21.5 Ci/ mmol. Subaequent steps were exactly as described above, and incorporation of ['H)UTP into RNA was quantitated (1). Autoradiography-Permeable cells were allowed to perform DNA replication as described above, except that [*H]d'M'P was at 4.0 Ci/ mmol.Following incubation at 37 "C, the cells werecollected by centrifugation, washed twice in Solution A containing 60 GM d'M'P, suspended in 1 ml of 0.1 M sucrose, 1 mM CaC12, 0.3 mM NaCl for 5 min at 0 "C, and again sedimented. The cells werefixed in 75% methanol, 25% acetic acid for 10 min at 0 'C and then dripped onto glass slides. The slides were dried, coated with NTB3 emulsion, stored in the dark, and developed as indicated. Before determining the per cent labeled nuclei or number of grains/nucleus, the cells were stained with Giemsa. Polymerase-a Assays and Antibody Neutralization Titrations-HF cells were harvested by trypsinization, washed thrice in phosphatebuffered saline containing 0.2 mM phenylmethylsulfonyl fluoride, suspended in -9 volumes of 5 m M KPO., pH 7.5,2 mM MgC12,l mM mercaptoethanol, 1mM EDTA, 1mM phenylmethylsulfonyl fluoride, 10 mM NaHSO', and broken by gentle homogenization (12). The homogenate was clarified by centrifugation (15 min at 20,OOO X g ) (Fraction I), and DNA polymerase-a activity was measured (12). KB cell DNA polymerase-a Fraction IIA was prepared and assayed as described (13).Equal units of activity of KB polymerase Fraction IIA or HFpolymerase Fraction I were preincubated with nonimmune or immune IgGs at 0 "C for 60 min and then incubated at 37 'C for 10 min in the standard polymerase-a assay (12, 13). Results are expressed as per cent neutralization of polymerase-a activity relative to samples that were preincubated with the nonimmune control P3 IgG (9). Zrnrnuocytockrnistry-Permeable HF cells were preincubated with nonimmune or immune IgGs for 1h at 0 'C and thenincubated in DNA synthesis solution with d"P in place of ['HIdTTP for 20 min a t 37 "C. The cells were then washed thrice in Solution A, fixed in fresh paraformaldehyde, and processed for immunocytochemical eximinatibn. by light and electron microscopy, exactly as previously described (10); Electron microscopic examination was performed in collaboration with Dr. K. G. Bensch, Stanford University, to whom we acknowledge our gratitude. RESULTS

The exposure of permeabilized human fibroblasts to each of three neutralizing monoclonal anti-polymerase-a antibodies results in theconcentration-dependent inhibition of DNA synthesis (Fig. 1 and Table I). In contrast, thesame levels of identically prepared monoclonal control (P3)IgG are without effect on the reaction. The neutralizing antibodies appear to 19.2 differ in their respective inhibitory potencies as measured by this assay, and these relative potencies are different from those measured in conventional neutralization titrations with purified KB DNA polymerase-a (9).Thus, by the latter assays, the rank order of antibody potencies is SJK 132 > SJK 287 = SJK 211;whereas in the permeabilized cellassays, the order is SJK 287 > SJK 132 > SJK 211, whether expressed as t h e concentration of antibody required to inhibit DNA synthesis by 50% or as themaximum extent of inhibition that can be attained. The data in Table I compare the extent of inhibition produced by the antibodies with that resulting from treatment with aphidicolin, a known potent chemical inhibitor of DNA polymerases-a. By this approximation, it is apparent that the magnitude of replication inhibition produced by antibody SJK 287 is comparable to that achieved with aphidicolin. It is also evident (Table I) that onlymodest incremental inhibition is accomplished either by combining

70

-

60

-

50

-

135 2

""0

P-

40-

,

"-

"_"

"" """" ""

,.a'

"""""""""""D

0

100

200

300

500

400

ug MAB/ml

FIG. 1. Effect of monoclonal antibodies against DNA polymerase-a on DNA replication in permeable cells. Exponential HF cells were permeabilized with lysolecithin, preincubated for 1 h at 0 "C in DNA replication solution with the indicated concentrations of monoclonal antibodies (MAB) or with heat-treated albumin (see "Experimental Procedures"), and then incubated at 37 "C for 30 min for assay of ['HId'ITP incorporation into DNA. In typical control reactions, 14 pmol of dTTP were incorporated per 10' cells/30 min. The values shown are averages of duplicate determinations. SJK 287-38; 0,SJK 132-20;0,SJK 211-14; 0,P3 (nonimmune) IgG.

.,

TABLE I Effect of Aphidicolin and of anti-DNA polymerase-a antibodies on DNA replication in permeabilized cells Exponential HF cells were permeabilized as described under "Experimental Procedures," preincubated for 1 h at 0 "C in DNA replication solution containing the indicated additions, and then further incubated for 30 min at 37 "C for assay of incorporation of ['HIdTTP into DNA. All immunoglobulins were present at 250 pg/ml; aphidiColin was at 2 pglml. SJK 287-38 and 211-14 are neutralizing antiDNA polymerase-a monoclonal antibodies; P3 is a nonimmune, control monoclonal IgG (9). Per cent inhibition of DNA replication is expressed relative to the P3IgG control value.

DNA Replication

Addition

None P3 IgG SJK 287-38 Aphidicolin SJK 287-38 + aphidicolin 92 SJK 211-14 40 SJK 211-14 287-38

+

dlTP/lod cells/tO min

Inhibition

Pml

x

20.4 6.9 4.2 1.6 12.2 4.7

0 66 80

77

these two different inhibitor reagents or from combination of two independent antibodies. The only other monoclonal antibody that we tested in these experiments was a non-neutralizer of veryhigh binding affinity, SJK 237-71 (9,13), which produced only a minimal (-10%) degree of inhibition ofDNA synthesis at these IgG concentrations (data not shown). Because of the apparentdifferences inantibody potency as measured in the permeabilized cell systemuersus traditional

Inhibition of DNA Replication with Anti-polymerase-aAntibodies

136

neutralizationtitrations (9) and because very little is yet known about the intraspecific and interspecific reactivity of these anti-KBpolymerase-a antibodies, we performed standard, comparative neutralization experiments with similar crude polymerase-a fractions prepared from KB and HF cells. The results are presented inFig. 2. As we previously reported (9), the neutralizing titer of antibody SJK 132 is considerably greater than those of SJK 287 and 211 when measured against the KB enzyme (Fig. 2A). Surprisingly, however, this difference is not detected in assays with the HF polymerase (Fig. 2B), in which the three neutralizing monoclonal antibodies generate almost superimposable titration curves. Indeed, from comparison of the data inFig. 2, A and B, it appears that the “anomalous” behavior is exhibited by SJK 132 in that its neutralizing activity against KB polymerase-a appears to be uniquely enhanced by an order of magnitude. From titration experiments like those in Fig. 2, we have obtained preliminary evidence suggesting an additional difference between the properties of the HF and KB enzymes. Specifically, at very high concentrations of neutralizing antibody that inhibit the KB polymerase by >95%, almost 20% of the HF polymerase-a activity remains detectable. The putatively resistant HF enzyme fraction appears to be typically a-like in its sensitivity to aphidicolin and its relative insensitivity to dideoxy-TTP (14);we can offer no explanation at this time. Since the maximum degree of inhibition of replication obtained with SJK 287 was between 70 and 80%, we per-

formed an autoradiographic population study to discriminate whether this “resistant” DNA synthesis was due to a resistant subpopulation of cells (e.g. insufficiently permeabilized) or to a uniform but incomplete suppression of DNA replication involving all of the target cells. The results are presented in Fig. 3 and Table I1 and indicate that the latterexplanation is the correct one. To confirm that the effects of these antibodies on DNA synthesis were appropriately specific, i.e. not due to general metabolic perturbations, we performed the experiments presented in Table I11 and observed that exposure of the permeabilized cells to neutralizing anti-polymerase-a antibodies had no detectable effect on the cells’ capacity to synthesize RNA. The data presented in Figs. 1 and 3 and Tables I and 11, together with previous characterization of the lysolecithinpermeabilization system (1, 4, 5) which demonstrated both the essentially complete retention of DNA within the nuclei of the permeabilized cells as well as the exclusively intranuclear localization of the DNA synthesis measured in these experiments, strongly argue that atleast some fraction of the monoclonal antibodies must be entering the nuclei of these

0.

0

000 00

\

oo

A

n n o

0

00000

00

”-&

,A*

/

I’

80-

o,

10

20

30

40

60

50

Grains per Nucleus

B “-a

A”- / o

ao6040-

-

20

0.5

5

50

500

5000

ng MAB/rnl FIG. 2. Neutralization of €IF and KB cell DNA polymerasea activities by monoclonal antibodies (MAB). Neutralization titrations were performed as described (9) using 0.3 unit of DNA polymerase-cy prepared (9) from KB cells (Fraction IIA) ( A ) or from HF cells (Fraction I) ( B ) ,respectively. A, SJK 132-20; A, SJK 28738; 0, SJK 211-14. Control reactions were run in all experiments with nonimmune P3 I&, which had no effect on either polymerasecy preparation.

FIG. 3. Autoradiographic analysis ofthe effect ofanti-DNA polymerase-a antibody SJK 287-38on DNA synthesis in permeabilized HF cells. Exponential HF cells were permeabilized, preincubated at 0 “C for 1 h in DNA replication solution with or without SJK 287-38 (250 pg/ml), further incubated at 37 “C for 30 min to permit incorporation of [‘HIdTTP, and processed for autoradiography, as described under “Experimental Procedures” (see also Refs. 4 and 5). To determine average number of grains/nucleus, slides were developed after 6 h and 20 nuclei were examined. The figure is a histogram that shows the distribution of numbers of grains measured in cell nuclei with (0)or without (0)SJK 287-38. The mean number of grains/nucleus and standard deviation are 62.3 4.8 and 19.5 f 4.6 in the absence and presence of antibody, respectively. These data are statistically significant at P < 0.01 by two-tailed T analysis. Under the experimental conditions described here and in Table I1 and with these short durations of autoradiographic exposure, no background grains are observed in control nuclei that have not been exposed to the tritiated dTTP.All of the grains detected in the labeled cells are confined to thenuclei, in agreement with the results of previous studies (1,4, 5) with cells prepared by this permeabilization protocol.

TABLE I1 Autoradiographic analysis of pattern of response of permeabilized cells to anti-DNA polymerase-a antibody Autoradiographic analysis of the effects of monoclonal antibody on DNA synthesis inpermeabilized, exponential HF cella wmcarried out asdescribed under “Experimental Procedures” and in the legend for Fig. 3. The data for average number of grains/nucleus are obtained from Fig. 3. The per cent of labeled nuclei was assessed by scoring 280 cells after 3 days of exposure to the emulsion. Addition

None SJK 287-38

% of labeled nuclei

24.7 22.5

A

~

~

62.3 19.5

~

;

~

~

~

s

o

Inhibition of DNA Replication Anti-polymerase-a with

Antibodies

137

protein antibodies, but notof preimmune IgGs, after red blood cell-mediated cytoplasmic microinjection. The results we have obtained in the present experiments do not directly address this controversy since it is possible, on the one hand, that some of the normal permeability properties of the nuclear membrane are altered by the lysolecithin protocol and, on the other hand, that the active reagents in our studies were not intact IgG species but were active antibody fragments generated, for example, by endogenous proteases in the lysolecithinpmo1/30 minlI0' cells treated cells. Whatever the explanation may be, the result is None 46.6 important since it opens, for the first time, the possibility of P3 I g G 45.8 analyzing complex macromolecular processes with reagents 52.4 SJK 287-38 of unambiguous specificity in subcellular systems that retain 44.7 SJK 211-14 acceptable levels of morphological and physiological integrity. 50.3 SJK 132-20 Actinomycin D 2.9 Second, it is also not predictable, irrespective of considerations of nuclear membrane traverse, that catalytically active permeable cells to produce the observed inhibition results. On DNA polymerase-a molecules, participating in chromosomal theother hand, itis wellrecognized that relatively vast replication in situ,would necessarily prove to be accessibleto quantities of DNA polymerase-a are leached into the cyto- the active antibody molecules (or fragments). From the perplasmic fraction of aqueously extracted cells (see Ref. lo), spective of a putative organized replication complex containand similar substantial leaching of polymerase-a activity has ing a number of interactive protein components, the efficacy been shown to occur (4) in cells subjected to the lysolecithin of even the single most effective antibody, SJK 287, may be protocol. At the very least, therefore, these considerations considered remarkable, even if fortuitous. We presently have make it impossible to estimate even crudely what fraction of insufficient information about the structure either of mamthe added IgG might be available to enter the nuclei by malian replication complexes in general or of the polymeraseescaping immune complexation with free polymerase-a either LY epitopes in particular to offer more than speculation as to in theextracellular phase orin the membranous cytoplasmic reasons for the differences in inhibitory potency (Fig. 1) among three antibodies that are in fact equipotent in standard meshwork (1)of the permeabilized cells. Notwithstanding these difficulties, we attempted by ultra- neutralization assays with HF polymerase-a (Fig. 2B). Finally, the data in Fig. 2, A and B, that illustrate the structural cytochemistry (10) to obtain evidence of intranuclear accumulation of immune IgG, and we were able to striking difference in neutralization titer of a single monoclodemonstrate unequivocally the presence of significant nal antibody, SJK 132, for two human DNA polymerase-a amounts of peroxidase reaction product in nuclei treated with preparations, one obtained from near-normal fibroblasts (HF) immune IgG as compared with those exposed to identical and the other from a highly malignant epithelial line (KB), concentrations of the nonimmune IgG, which were essentially raise numerous intriguing questions about the nature of the negative (data not shown). Since the permeabilized cells are target epitope that we presume must be situated in or near fragile to begin with and because they do not toleratewell the the enzyme's active center. In additional experiments with repeated washings that are necessary to remove the large these two polymerase preparations (data notshown), we have quantities of IgG that are present in the incubation solution, observed another difference with the non-neutralizing antithe morphological preservation of the specimens is very poor, body (9) SJK 237, uiz. this antibody, which tightly binds KB andthe quality of the electron micrographs that can be polymerase-a and minimally (510%) enhances catalytic acobtained is correspondingly extremely limited. Nonetheless, tivity, produces a pronounced -2-fold stimulation of the it is of interest to note that both nucleolar remnants and activity of the HFpolymerase-a fraction when assayed under blocks of residual peripheral heterochromatin in the nuclei identical conditions. Whetherthese immunological differprepared from cells exposed to immune IgG are largely free ences reflect biologically significant differences in polymerase of the reaction product, in agreement with the diffuse mid- structure that relate to the neoplastic state, to suspension zonal intranuclear distributionof polymerase-a that was dem- versus surface-attached growth, to differences in tissue of onstrated in our earlier immunocytochemical studies with origin, etc. remains to be explored. However, we note that the intact cells (10). results of solid-phase competitive radioimmunoassays' indicate that the KB polymerase-a epitopes that are recognized DISCUSSION by each of the threeneutralizing monoclonal antibodies (SJK The results described in thisreport demonstrate that DNA 132,SJK 211,SJK 287) appear to be entirely independent of replication in lysolecithin-permeabilized human cells can be one another with respect to antibody binding. This observasubstantially inhibitedby neutralizing monoclonal antibodies tion is entirely consistent with the data in Fig. 2, A and B, specific for DNA polymerase-a. Although these observations with respect to the uniqueness of the difference in neutraliare not intrinsically surprising inlight of current concepts of the role of polymerase-a in chromosome replication, neither zation potency against the KB andHF enzymes that is are theypredictable a priori for at least two different reasons. exhibited by SJK 132,but not by SJK 211 or SJK 287. First, thequestion is currently controversial whether antibodREFERENCES ies that have been introduced into the cellular cytoplasm by 1. Miller, M. R.,Castellot, J. J., Jr., and Pardee, A. B. (1978) one or another microinjection technique can or cannot traBiochemistry 17,1073-1080 verse the nuclear membrane. For example, Einck and Bustin 2. Miller, M. R., and Chinault, D. N. (1982)J. Biol. Chem. 267, observed (15) that anti-histone F(ab)zfragments, but not the 46-49 intact IgG molecules, could accumulate within nuclei after cytoplasmic injection; while Bennett et al. (16)reported suc* J. Wehner, T. S.-F. Wang, and D. Korn, unpublished observacessful transnuclear passage of intact antinucleolar phospho- tions. TABLEI11 Anti-DNA polymerase-a antibodies do not affect RNA synthesis in permeabilized cells Exponential HF cells were permeabilized, preincubated for 1 h at 0 "C in RNA synthesis solution with additions as indicated, and then incubated for 30 min at 37 "C for assay of UTP incorporation into RNA. All immunoglobulins were present at 250 pg/ml; actinomycin D was at 1 pg/ml. Additions UTP incorporation

138

Inhibition of DNA Replication with Anti-polymerase-aAntibodies

3. Miller, M.R., and Chinault, D.N. (1982)J. Bwl. C k m . 257, 10204-10209 4. Myers, C. A., Patel, P. I., and Miller, M.R. (1983)Exp. CeU Res. 143,227-236 5. Patel, P. I., Myers, C . A., and Miller, M.R. (1983)Exp. Cell Res. 149,347-358 6. Weissbach, A. (1977)Annu. Rev. Bwckm. 46.25-47 7. Hanawalt, c.9 Cooper? p*K.9 GaneWn, A* K.,and Smith, cA. (1979)Annu. Rev. Biochem. 48,783-836 8. Fry, M. (1982)in Enryms of Nucleic Acid Synthesis and Modification (Jacob, S. T., ed), Vol. I, pp. 39-92,CRC Series, CRC Press, Boca Raton, FL

9. Tanaka, S.,Hu,S.-Z.,Wang, T.S.-F., and Korn, D. (1982)J. Bwl. Chem. 257,8386-8390 10. Benxh, K. G.9 Tanaka9 s.9 HU, Wan& T.S.-F-, and KOm, D.(1982)J. Biol. Chem. 257,8391-8396 11* Bradford* M*(1976) Biockm. 729 248-254 12. Fisher, P. A., and Korn, D.(1977)J. Biol. Chem. 252,6528-6535 13. Wang, T. S.-F., Hu, S.-Z.,and Korn, D. (1984)J. BioL Chem. 259,1854-1865 14. Fisher, P. A*,Wang, T. S.-F., and Kom, D.(1979)J. BwL C h m . 2w.6128-6137 15. Einck, L., and Bustin, M. (1984)J. CeU BwL 98,205-213 16. Bennett, F.C., Busch, H.,Lischwe, M. A., and Yeoman, L.C. (1983)J. CeU BioL 97, 138a