Intracellular localization of mouse DNA polymerase-alpha

Proc. Nat. Acad. Sci. USA Vol. 73, No. 4, pp. 1136-1139, April 1976

Biochemistry

Intracellular localization of mouse DNA polymerase-a [DNA polymerase-ft (DNA nucleotidyltransferase-#)/thymidine kinase/cytochalasin B enucleation]

GLENN HERRICK, BRIAN B. SPEAR*, AND GEORGE VEOMETT Department of Molecular, Cellular and Developmental Biology, University of Colorado, Boulder, Colo. 80302

Communicated by David M. Prescott, January 30,1976

MATERIALS AND METHODS Mouse L929 cells were grown in Eagle's minimum essential medium with 10% fetal calf serum (Kansas City Biological) and were judged to be mycoplasma-free by the following tests: cytoplasmic thymidine incorporation (19), surface appearance in the scanning electron microscope (performed by J. Meek and M. Clark), and conversion of uridine to uracil by cell extracts (20). Enucleation was performed by centrifugation at 370 in medium containing 10 .g/ml of cytochalasin B (Aldrich Chemical), after a pre-spin to remove weakly attached cells, as described elsewhere (21). Cytoplasts or untreated whole cells were removed by trypsin treatment, collected by a brief centrifugation 5.min at 300 X g), and gently resuspended in phosphate-b ffered saline. The karyoplasts were gently resuspended in the enucleation flasks in phosphate-buffered saline. Concentrations of whole cells, karyoplasts, and cytoplasts in the suspensions were determined by drying drops of known weight on microscope slides; these preparations were subsequently fixed, Feulgen stained, methyl green counter-stained, and all particles from each drop were scored as whole cells, karyoplasts, or cytoplasts. After samples were removed from the suspensions for counting and determination of intactness (trypan blue exclusion), the whole cells and cell parts were collected by a final centrifugation. Examination of supernatants indicated complete recovery (>99%) in the three pellets. For DNA polymerase extraction each pellet was resuspended and disrupted by sonication with a Branson Sonifier (model S125, lowest setting) in buffer at a final concentration of 200 mM potassium phosphate, pH 7.5, and 17 mM 2-mercaptoethanol; microscopic examination showed cell and nuclear disruption was complete. The lysates (300-350 ,ul) were clarified by ultracentrifugation (45 min at 100,000 X g) and the supernatant volumes were determined by weight. Recovery of DNA polymerases-a and -,B was complete. Preparation of thymidine kinase extracts was similar except the buffer was 50 mM Tris-HCl, pH 8.0, 10 mM 2mercaptoethanol, and 40,MM thymidine. DNA polymerase assays were performed at 370 essentially as described by Chang et al. (7). For polymerase-a assays 25 Ml of solution to be assayed was combined with 100 ,l of an assay solution to give final concentrations of 40 mM potassium phosphate, pH 7.5, 10 mM 2-mercaptoethanol, 6.7 mM MgCl2, 190 MuM of each deoxynucleoside triphosphate, one triphosphate being 3H-labeled (New England Nuclear, final specific activity, 120 cpm/pmol), and 180 ,ug/ml of calf thymus DNA activated with DNase I (6-8% trichloroaceticacid-soluble). For polymerase-( assays 50 ,l of solution to be assayed was combined with 75 Ml of an assay solution to give. final concentrations of 80 mM potassium phosphate, pH 7.5, 160 mM Tris-HCI, pH 8.6, 10 mM N-ethylmaleimide (Sigma), and MgCl2, dNTP's, and activated DNA as in the * Present address: Department of Biological Sciences, Northwestern polymerase-a assays; the final pH was 8.4. Aliquots were University, Evanston, Ill. 60201 1136

Although DNA polymerase-a (DNA nucleotiABSTRACT dyltransferase; deoxynucleoside triphosphate:DNA deoxynucleotidyltransferase; EC 2.7.7.7) probably functions in the nucleus, it is usually found predominantly in the nonnuclear fraction of disrupted cells. We have reexamined the intracellular location of this enzyme using cytochalasin-B-induced enucleation, a technique which avoids exposure of nuclei to extra-cellular conditions during cell fractionation. In conditions where viability of separated cell parts is high and recovery is quantitative, we find greater than 85% of total DNA polymerase-a (and DNA polymerase-P) activity in the nucleated cell fragments (karyoplasts), from which we conclude that the location in vivo of DNA polymerase-a is either nuclear or perinuclear. On the other hand, thymidine kinase (ATP:thymidine 5'-phosphotransferase, EC 2.7.1.75) is found primarily in the enucleated cell fragments (cytoplasts) The enucleation procedure used in this work should be of general use for intracellular location studies. Mammalian cells contain two major nonmitochondrial DNA polymerases (DNA nucleotidyltransferase; deoxynucleoside triphosphate:DNA deoxynucleotidyltransferase; EC 2.7.7.7), a and (3 (1). Polymerase-a is predominant in proliferating cells and appears to be involved in nuclear DNA replication, because its amount is coupled with cell proliferation in a variety of situations (2-10). However, most or all of polymerase-a is usually found in the nonnuclear fraction of cell lysates (11-13), although it has been reported recently that the enzyme is found in the nuclear fraction if nonaqueous isolation procedures are used (14, 15). With all such procedures it is possible for polymerase to leak from or adsorb to the isolated nuclei. Therefore, we have approached the question of the intracellular location of DNA polymerase-a using a procedure which avoids lysis of the cell during the separation of the nucleus from the cytoplasm. The procedure employs the fungal drug cytochalasin B, which causes cultured cells to extrude their nuclei into an outpocketing of the plasma membrane, presumably as the result of microfilament disruption (16). The effect is rapidly reversible, and, even in the presence of the drug, thymidine, uridine, and leucine incorporations are unperturbed (ref. 17, unpublished results of D. M. Prescott). Such drug-treated cells can be enucleated in a centrifugal field, leaving the enucleated cells (cytoplasts) still attached to the growth surface. The nucleus-containing fragments (karyoplasts, each consisting of a nucleus and a small portion of perinuclear cytoplasm enclosed by an intact plasma membrane) sediment away from the growth surface and can be recovered from the culture medium. Both cytoplasts and karyoplasts remain metabolically active and intact, and can, in fact, be fused to form cells capable of resuming the cell cycle (18). Here we report the relative activities of DNA polymerases-a and -( in karyoplasts and cytoplasts.

Proc. Nat. Acad. Sci. USA 73 (1976)

Biochemistry: Herrick et al.

400

S.0

0

2

~~~~~~~~~~~~~C B DNA

polymerose-fi

.0 k+c V

E

5 X150

o

1137

E

E 4 21

0

z

z

a

al

-

-

100

Fraction 12

24

36

Minutes at 370

FIG. 1. DNA polymerases-a and -ft from mouse L929 cells, cytoplasts and karyoplasts. Logarithmically growing cells (3.2 X 104 per cm2; total 4.8 X 106) were enucleated in medium containing cytochalasin B, and extracts from the resulting particles were made as described in Materials and Methods. Cytoplast recovery was 110%, and karyoplast recovery was 104%, when compared to the yield of whole cells from sister cultures. Whole cells and cytoplasts were >99% intact, and karyoplasts were 89% intact, as judged by trypan blue exclusion. Extracts were assayed (Materials and Methods) in duplicate, and 29 ul aliquots were removed at 0, 12, 24, and 36 min; results are expressed as total nucleotides incorporated per 106 particles: whole cell, wc; karyoplast, k; cytoplast, c. Cytoplast activities shown have been corrected for whole-cell (2%) and karyoplast (5.4%) contamination; thus, the a-activity shown represents 62%, and the ,-activity, 71% of the activity actually found in the cytoplast extract. (A) DNA polymerase-a assay conditions. Karyoplast activity is per 106 intact karyoplasts. (B) DNA polymerase-,6 assay conditions.

spotted on glass fiber filters and analyzed for acid-precipitable radioactivity. It should be noted that phosphate in the polymerase-f assays causes considerable inhibition, as noted by others (22); thus, the polymerase-fl activities reported represent about 25% of activities assayed in the absence of phosphate. Thymidine kinase (ATP:thymidine 5'-phosphotransferase, EC 2.7.1.75) assays were performed as described by Stubblefield and Mueller (23), but dTMP formation was detected using DEAE paper as described by Migeon et al. (24). RESULTS In an experiment where 5 X 106 rapidly growing mouse L929 cells were centrifuged in the presence of cytochalasin B, 98% were enucleated. The recoveries of cytoplasts and

karyoplasts were 110% and 104%, respectively. Crude extracts were prepared, and the levels of DNA polymerases-a and -jB were assayed (Fig. 1). In both cases most of the polymerase activity (88% of whole-cell control) was found in the karyoplast extract, with a small portion recovered in the cytoplast extract (8.5% of a, 15% of fl). We have observed re-

FIG. 2. Sedimentation analysis of DNA polymerase activities in karyoplast and whole-cell extracts. Equal volumes of extracts (140 1l) were layered on 3.6 ml, 5-20% (wt/vol) sucrose gradients containing 200 mM potassium phosphate, pH 7.5, and 10 mM 2mercaptoethanol. Centrifugation was for 13.5 hr at 40 and 53,500 rpm in a Beckman SW60 rotor. Fractions were collected from the bottom and assayed as described in Materials and Methods; the entire assay volume was precipitated after a 1 hr reaction at 37°. Incubated no-enzyme controls showed no incorporation above background. Bovine serum albumin (BSA) was sedimented in a third gradient as a marker (4.4 S, ref. 25), and its position was determined by absorbance at 280 nm. Sedimentation was from right to left. (A) Whole-cell extract. (B) Karyoplast extract.

peatedly that recovery of polymerase-a is dependent of intactness of the karyoplasts, where intactness is judged by trypan blue exclusion. In the experiment shown in Fig. 1, 89% of the karyoplasts were intact, and the karyoplast polymerase-a level is expressed per intact karyoplast. The activity per cytoplast plus the activity per intact karyoplast equals 96% of the activity per whole cell. Even without a correction for the karyoplasts that are not intact, polymerase-a recovery is 86%. The use of this correction assumes that trypan blue-permeable karyoplasts lose their polymerase-a, which is consistent with the idea that this polymerase is only loosely associated with the nucleus. This correction is not necessary to show full recovery of the polymerase-fl (103%), as might be expected, since most of this enzyme appears to be tightly bound to the chromatin (11). In the enzyme assays used, the differentiation between the two polymerases depends on the known sensitivity of polymerase-a, and the known insensitivity of polymerase-ft, to sulfhydryl-blocking reagents such as N-ethylmaleimide, and on the alkaline pH optimum of polymerase-fl (7). To assure that the assays actually perform as intended, and to test for various effects of our procedures on the polymerases, portions of whole-cell and karyoplast extracts were subjected to sucrose gradient sedimentation (Fig. 2). In.both extracts the polymerases behave in a typical fashion, polymerase-a sedimenting rapidly (about 6.4 S), and polymerase-fl more slowly (about 3.4 S). Having demonstrated that the assay conditions used do discriminate between the two polymeras-

1138

Biochemistry: Herrick et al.

Proc. Nat. Acad. Sci. USA 73 (1976)

Table 1. Intracellular distribution of L929-cell thymidine kinase

Thymidine kinase activity, %*

Karyo-

Whole

Exp.

Cytoplast

plast

Recovery

Cell

1 2 3 Average

56 56 60 57

28 18 20 22

84 74 80

100 100 100

79

* VVhole-cell extract activity was 0.40 nmol of dTMP formed per hr per 106 cells in Exp. 2.

es and that no activity has been lost during enucleation (Fig. 1), we conclude that both polymerases are predominantly associated with the nucleust. The regulation of nonmitochondrial thymidine kinase activity, like that of DNA polymerase-a, appears to be coupled to nuclear DNA replication (see ref. 31 for review). In this case, however, there is no reason a priori to assume that it must act at the site of DNA synthesis, and it is, in fact, found in the non-nuclear fraction of cell homogenates (32, 33). Table 1 shows the results of an intracellular location study of thymidine kinase using the cytochalasin-enucleation procedure. Much of the activity (57% of whole-cell) is recovered in the cytoplasts. Note that complete recovery of activity is not achieved. We suggest that this is due to the known rapid turnover of this enzyme (34, 35) and impaired protein synthesis in cytoplasts (17). Karyoplasts contained 22% of the kinase activity; while some of this activity presumably reflects the cytoplasmic content of karyoplasts (see ref. 36), it may in addition represent enzyme in the nucleus. In any case, it can be concluded that this enzyme is predominantly in the cell cytoplasm, in agreement with the results of cell fractionations in vitro (32, 33). Whether karyoplast thymidine kinase is only perinuclear, or is also intranuclear, could be determined by comparison of karyoplast activity with the cytoplasmic content of the karyoplast preparation by using assays for known cytoplasmic enzymes (36) or RNAs (14).

DISCUSSION Of total DNA polymerase-a activity, 88% was found in the karyoplasts (Fig. 1). Thus, we conclude that this enzyme is associated with the nucleus; because the karyoplast has some cytoplasm, we cannot distinguish between a perinuclear and an intranuclear location. Some polymerase activity, both a tThere is considerable N-ethylmaleimide-sensitive activity sedimenting near the 4.4S marker (Fig. 2). Activity has been observed repeatedly at this position, but its amount varies from experiment to experiment. Of several possible origins of this activity, we feel it unlikely to be a mycoplasma polymerase (see Materials and Methods), or a fragment of polymerase-a generated either by trypsin exposure (compare karyoplast activity to whole-cell activity in Fig. 2, where karyoplasts are not exposed to trypsin) or by sonic disruption during extraction (osmotic lysis gives an identical yield-not shown). This activity could, however, represent a lowmolecular-weight form of polymerase-a (26, 27). Alternatively, it could be either a cellular reverse transcriptase (RNA-dependent DNA polymerase) (28) or possibly a tumor virus reverse transcriptase (29, 30). Finally, this shoulder could represent in part DNA polymerase-# active in the polymerase-a assay conditions. Whatever the nature of this activity, its extent of association with the nucleus is indistinguishable from that of the faster-sedimenting DNA polymerase-a.

and (3, was also found in the cytoplasts. This activity may represent nascent enzymes, or possibly enzyme turn-over products which have passed out of the nucleus but which retain the capacity to function in the assay conditions in vitro. However, the cytoplast and karyoplast activities have indistinguishable sedimentation profiles (not shown). Alternately, at a particular stage in the cell cycle (e.g., G2) the polymerases may migrate to the cytoplasm, as suggested by others (37), and the measured polymerase activity in cytoplasts would reflect the fraction of the total asynchronous cell population which is at this stage. Because karyoplast preparations contain some cytoplasm, cytochalasin enucleation does not allow as precise a determination of intracellular location as does subcellular fractionation by conventional techniques. Enucleation does, however, avoid the dangers inherent in any procedure employing separation of cell constituents in nonphysiological solutions, and enucleation should prove useful for a variety of intracellular localization studies. Using this procedure we have found greater than 85% of both DNA polymerases-a and -(3 associated with the nucleus. In contrast, thymidine kinase was found predominantly in the cytoplasm. We thank Drs. Lucy M. S. Chang and Arthur Weissbach for helpful advice. We would also like to thank Dr. David M. Prescott for his encouragement and support. This work was funded by the Jane Coffin Childs Memorial Fund for Medical Research (G.H. and B.B.S.), by the National Institutes of Health (G.H. and G.V.), and by the National Institutes of Health Program Project in Basic Oncology Grant no. CA13419, and was carried out in the laboratory of Dr. D. M. Prescott.

1. Weissbach, A., Baltimore, D., Bollum, F., Gallo, R. & Korn, D. (1975) Science 190, 401402. 2. Iwamura, Y., Ono, T. & Morris, H. P. (1968) Cancer Res. 28, 2466-2476. 3. O'Neill, M. & Strohman, R. L. (1970) Biochemistry 9, 28322839. 4. Stockdale, F. E. (1970) Dev. Biol. 21, 462-474. 5. Loeb, L. A. & Agarwal, S. S. (1971) Exp. Cell Res. 66, 299304. 6. Chang, L. M. S. & Bollum, F. J. (1972) J. Biol. Chem. 247, 7948-7950. 7. Chang, L. M. S., Brown, M. & Bollum, F. J. (1973) J. Mol. Biol. 74, 1-8. 8. Lynch, W. E. & Lieberman, I. (1973) Btochem. Biophys. Res. Commun. 52,843-849. 9. Coleman, M. S. & Hutton, J. J. (1973) Fed. Proc. 32, 451a. 10. Baril, E. F., Jenkins, M. D., Brown, 0. E., Laszlo, J. & Morris, H. P. (1973) Cancer Res. 33, 1187-1193. 11. Chang, L. M. S. & Bollum, F. J. (1971) J. Btol. Chem. 246, 5835-5837. 12. Weissbach, A., Schlabach, A., Fridlender, B. & Bolden, A. (1971) Nature New Btol. 231, 167-170. 13. Sedwick, W. D., Wang, T. S. & Korn, D. (1972) J. BMol. Chem.

247,5026-5033. 14. Foster, D. N. & Gurney, T., Jr. (1973) J. Cell Biol. 59,103a. 15. Lynch, W. E., Surrey, S. & Lieberman, I. (1975) J. Btol. Chem. 250, 8179-8183. 16. Shay, J. W., Gershenbaum, M. R. & Porter, K. R. (1975) Exp. Cell Res. 94, 47-55. 17. Follett, E. A. C. (1974) Exp. Cell Res. 84,72-78. 18. Veomett, G., Prescott, D. M., Shay, J & Porter, K. R. (1974) Proc. Nat. Acad. Sci. USA 71,1999-2002. 19. Nardonne, R. M., Todd, J., Gonzalez, P. & Gaffney, E. V. (1965) Science 149, 1100-1101. 20. Levine, E. M. (1972) Exp. Cell Res. 74,99-109. 21. Veomett, G., Shay, J., Hough, P. V. C. & Prescott, D. M. (1976) in Methods in Cell Biology, ed. Prescott, D. M. (Academic Press, New York), Vol. 13, in press.

Biochemistry: Herrick et al. 22. Chang, L. M. S. & Bollum, F. J. (1973) J. Biol. Chem. 248, 3398-3404. 23. Stubblefield, E. & Mueller, G. C. (1965) Biochem. Biophys. Res. Commun. 20,535-58. 24. Migeon, B. R., Smith, S. W. & Leddy, C. L. (1969) Biochem. Cenet. 3,583-590. 25. Loeb, G. I. & Scheraga, H. A. (1956) J. Phys. Chem. 60, 1633-1644. 26. Holmes, A. M., Hesslewood, I. P. & Johnston, I. R. (1974) Eur. J. Biochem. 43,487-499. 27. Bollum, F. J. (1975) Prog. Nucleic Acid Res. Mol. Biol. 15, 109-144. 28. Fry, M. & Weissbach, A. (1973) J. Biol. Chem. 248, 26782683. 29. Schaifer, W., Fischinger, P. J., Lange, J. & Pister, L. (1972) Virology 47, 187-209.

Proc. Nat. Acad. Sci. USA 73 (1976)

1139

30. Grandgenett, D. P., Gerard, G. F. & Green, M. (1972) J. Virol. 10, 1136-1142. 31. Hauschka, P. V. (1973) in Methods in Cell Biology, ed. Prescott, D. M. (Academic Press, New York), Vol. 7, pp. 362-462. 32. Berk, A. J. & Clayton, D. A. (1973) J. Biol. Chem. 248, 27222729. 33. Brown, R. L. & Stubblefield, E. (1974) Exp. Cell Res. 87, 400-402. 34. Bresnick, E. & Burleson, S. S. (1970) Cancer Res. 30, 10601063. 35. Adelstein, J. J., Baldwin, C. & Kohn, H. I. (1971) Dev. Biol. 26,537-546. 36. Ege, T., Hamberg, H., Krondahl, U., Ericsson, J. & Ringertz, N. R. (1974) Exp. Cell Res. 87, 365-377. 37. Bollum, F. J. & Potter, V. R. (1958) J. Biol. Chem. 233, 478482.