EXPER:iUENTAL

CELL

202?

RESEARCH

167-173

(19%)

rotein Associated with Lethal Heat Shock J.

L.

~IPKIN,*~~,~

*Division

W.

G.

HINs~N~*

of Genetic

Toxicology and $Division Food and Drug Administration, and TUniversity

L. E.

LYN-COOK,*

E. R. BURNS,?

II.

of Reproductive and Developmental Toxicology, National Center for ToxicologicaZ Research, of Arkansas for Medical Sciences, Little Rock,

~HEEHAN,+$

AND

Department of Health and Human Jefferson, Arkansas 72079; Arkansas 72205

Services,

response are uniquely characteristic to specific cell hnes

P, 101.

The responses to stress in living cells are well known. Thermal stress causes decreased protein synthesis as well as rapid induction of heat shock proteins (hsps), or alternately termed stress proteins (sps). The exposure of cultured promyelocytic leukemia cells (HL-60) to a 45’C lethal heat shock for 1 h elicited synthesis and phosphorylation of a polypeptide A& 48,000 and pI 7.5 (p 48) as visualized by two-dimensional polyacrylamide gel ultramicroelectrophoresis. p 48, which was not observed at sublethal temperatures (39 and 4l“C), was synthesized during all phases of the cell cycle but was pbhosphorylated only in G,, + G1 and S-phases. The appearance of p 48 was marked by a concomitant and reciprocal reduction in hsps or sps 70 and 90. Distinct protease V8 fragment maps of p 48, hsps 70 and 90 in conjunction with immunochemical determination indicated vast differences in their primary structures. These facts suggest that p 48 was not formed from coalesced breakdown products of hsps 70 or 90. Western Matting showed that p 48 possessed the same immunochemical determinants as two other proteins with the same molecular mass but different isoelectric points. In an association assay, p 48 was shown to bind with actins and lisp 96 from HL-60 nuclei. 0 ma Academic Press,

Moderate heat (43’C) exposure of L-60 cells evoked synthesis and phosphorylation of classical hsps 7Q and 90 and a novel group of 48,OOQrelati lecular mass (A4J proteins [II-14], while at a Ie temperature (45’C) a major cell cycle regulated protein (p 48) was accentuated. The intent of this work was to analyze biochemical properties of a major M= 48$JOO~ expressed at a lethal temperature in cells and to examine its relationship to two minor teins.

CeU culture. Human leukemia cells, type 60 (HL-6G) were subcultured (37OC) at a density of 2.5 X 10’ cells in 75cm’ flasks with 19 ml RPMI 1640 medium (GIEXQ, Long Island, NY), supplemented with 10% fetal bovine serum as described elsewhere [4]. Flasks were pooled until a sufficient number of cells were coliected for fluorescent-activated cell sorting and analysis. Heat shock and isotope administration. L-60 cells in methionine or phosphate-deprived Hanks’balanced salt (HBSS) medium were hs in a water bath for 1 h at 39,41,43, or 45’C. Simultaneously with bs, cells were labe!ed by exposure to 100 ,~Ci/ml of ~-[‘5S]methionine (1170 Ci/mM, New Xngland Nuclear) or 5 &X/ml of [3zP]phosphoric acid (8500 CUmM, Amersham, Arlington Heights II>) for a 1-h pulse labeling period [2, 151. Cells were in HBSS only during labeling and washing procedures. Control cells were pulse labeled at 37’C for 1 h in HBSS. Cells were washed in isotope-free HRSS and resuspended in a DNA staining solution described below and furt Cell cycle analysis. Each sample of cultured Wed to 1.5 X 10’ cells/ml of Vindelov propidium iodine (PI) staining solution [4, 161 which lyses cells, releasing stained nuclei for DNA analysis and sorting by flow cytometry. Nuclei were sorted [17> 181 on the basis of DNA quantity (PI fluorescence) using the FACStar Plus (Becton-Dickinson, Mountain View, CA). These sorted samples were used for protein analysis [4, l9]. DNA analysis was performed with the FACScan flow cytometer (Becton-Dickinson) and the SFIT cell cycie analysis program. This ailowed qaantitation of nuclei in different phases of the cell cycle (i.e., G0 + G1, S, and Gz + M). Sorting of nuclei. Nuclei were sorted in the presence of a plmsphate-buffered sheath stream containing 0.14 M NaCl [ZO] directly into 250.~1 Beckman Airfuge tubes held in mini-McLeod chambers [16]. Five to ten million nuclei were sorted into the McLeod chambers from each partition of the HL-60 cell cycle; these constituted the nuclear sample used for protein extraction. A typical histogram of nuclear fIGorescence used as a standard for partition sorting is &own elsewhere [15].

Inc.

INTRODUCTION

Heat shock proteins (hsps) are the premier group of polypeptides recognized as speciahzed cellular constituents responding to thermal stress [l]. Presently they are considered a lesser category of the larger family of more general stress proteins (sps) [2]. Stress proteins, documented in prokaryotic and eukaryotic cells of wideranging phylogenetic origin, respond to elevated temperatures and a constellation of other perturbating factors [e.g., chemicals ]3], drugs [4], teratogens [5,6], and toxicants [4, ?, S]]. The effect of heat shock (hs) on cells varies with cell types For example, temperatures of 3942’C elicit classical sps/hsps (i.e., 70 and 90) and reduce total protein synthesis; nevertheless, peculiarities in hs ’ To whom

correspondence

should

be addressed. 167

!0014-4827/92 $5.00 Copyright @ 1992 by Academic Press, Inc. AU rights of reproduction in any form reserved.

168

PIPKIN

ET

AL.

E~tructio~ of proteins from sorted m.&~. ~entrifugation of the McLeod chambers at 8OOg for IO min pelleted the various nuclear populations [16] which were solubilized with imidazole extracting buffer [21] in a concentration of 10 ~1/500,000 nuclei [17]. After centrifugation of the nuclear extract in a Beckman Air Fuge (Beckman Instruments, Palo Alto, CA) at 100,OOOg for 1 h at 2’C ]l6], the supernatant was dialyzed against a sampIe solution [4] in a microdialysis unit [2OJ. Protein concentration was determined [22] as described elsewhere [4, 61. Electrophoresis and extraction of nuclear proteins from gels. Twodimensional polyacrylamide gel electrophoresis (PAGE) was performed in either an ultramicroelectrophoresis apparatus (UMEA) manufactured in-house l23] or a conventional Bio-Rad 220 electrophoresis unit (Richmond, CA). The UMEA was used for analysis of protein synthesis and for biochemical evaluation of proteins, while the Bio-Rad 220 unit was used in preparative procedures requiring the extraction of proteins from second-dimension gels. In UMEA, proteins were separated in the first-dimension in 25.~1 capillary tubes by isoelectric points (pa, and in the second-dimension by &fr in 10% acrylamide-SDS [24] slab gels (35 X 25 mm) as modified by Ratz et al. [25]. Equi~bri~m isoelectric focusing gels contained 0.4% (pH 5-8) and 1.6% (pH 3.-9) ampholytes (LKB, Bromma, Sweden). Radiolabeled HL-60 proteins (500 ng of protein in 3 ~1) were applied to the first-dimension tube gel of UMEA. This was followed by second-dimension PAGE and fluorograpby of the slab gel [26]. One-dimension SDS-PAGE analysis in UMEA was accomplished by using sample “combs” designed for slab gels. These formed lanes 25 mm L X 2 mm W (8 per microslab) and a l-p1 sample was loaded by ultramicro pipet into each well (2 X 2.1 mm) which contained a 1.2~1 volume [21]. Individual proteins for biochemical analysis were extracted electrophoretically from gel plugs taken from the second-dimension slab (220 unit). The plugs were placed in ISCO protein isolation-concentrator cups (ISCO, Lincoln, NE) [19] in an extraction buffer [4]. The same gel system was used in UMEA and in the 220 unit. Proteins must be free of SDS for further gel electrophoresis (i.e., binding experiments, etc.). The detergent was removed from the proteins [27] during the extraction process [4J. Nondenaturing PAGE (for analysis of hsp 90 and actin proteinbinding studies) was performed as described elsewhere [21, 281. I&orography of slab ge&. Slab gels from either unit were impregnated with 2,5-diphenyloxazole, dried, and placed on DuPont Cronex (ASA 1500~ DuPont, Wilmin~on, DL) sheet fiIm for 14 days at -70°C [26]. V8 protease digests. Limited proteolysis of proteins obtained from gel plugs from previous two-dimensional electrophoresis (220 unit) was performed with a final concentration of 26 &g/ml of Sta&yZococcus aurem V8 protease (Sigma Chemical Co., St. Louis, MO) for 30 min at 37OC [29]. The hydrolyzed proteins were subjected to one-dimensional electrophoretic peptide mapping in SDS using UMEA according to the procedure of Takeda and Cone [30] as modified by Anson et aZ. [6] and Pipkin et al. [2, 151 for sps. Immunochemical determinationa. Unlabeled p 48, elicited at 45OC (most abundant at this temperature) was isolated from HL-60 cells from pooled gels [4, 191. Polyclonal antibody was prepared in-house following~he procedure of Pipkin et al. [31] as applied to sps by immunizing a rabbit with isolated p 48 according to the procedure of Tahourdin [32], The hsps 70 and 90 polyclona1 antibodies were obtained from Affinity BioReagents (Neshanic Station, NY). Unlabeled proteins were separated by two-dimensional PAGE in UMEA. After electrophoresis [32] and equilibration [33] of the microgel, the slab was covered with a nitrocellulose membrane (Hoefer Scientific instruments, San Francisco, CA) and introduced into an uhramicro trans-blot electrophoretic cell (manufactured in-house). The dimensions of the transfer cell accommodated slabs from UMEA

--W-

4P4C

0’ 0

1

2

DAYS

IN CULTURE

3

4

5

FIG. 1. Growth rate of HL-60 cells in culture following heat shock at various temperatures. HL-60 cells were washed in methionine and phosphate-free HBSS treated for 1 h at the indicated temperatures and cuhured at 37OC. Cells were quantitated with a Coulter counter at the appropriate time (days).

[23]. The description and operation will be published elsewhere. The transfer was performed using the same buffer system [32], spacers, and filter sequence as described by Towbin et al. [33]. Following electrophoretic blotting, the nitrocellulose sheets containing the hsps (antigens) were soaked in a saline solution with 3% bovine serum albumin [33] and treated with either rabbit anti-48 hsp, mouse anti70 hsp, or mouse anti-90 hsp primary antibodies, and with either donkey anti-rabbit or sheep anti-mouse ?!+labeled secondary antibodies (Amersham, Arlington Heights, IL). After incubation for 5 h with each antibody, sheets were washed and fluorography was performed [31,32]. The actins and hsp 90 from HL-60 nuclei Binding experiments. were isolated from gels following electrophoresis [4, 191. They were labeled with a 15 to 30-fold excess of fluorescent dye N(iodacetyl)JJ’-1-sulfa-5-naphthyiethylene-diamine (Sigma Chemical Co., St. Louis, MO) in a minimal amount of dimethylformamide at Z°C in the dark for 12 h [21,34]. The solvent-buffer consisted of 50 &KCI, 25 n&fTris (pH 7.5), and 2 m&r MgCl. ApproximateIy 10% by volume of the dye solution was added to the actin and hsp 90 sample. The labeled proteins were depolymerized by dialysis against 1000 vol of buffer (10 rnM imidazole-HCl, 0.2 m&f ATP, 0.2 m&f CaCla, pH 8.0) for 24 h at 2“C. The fluorescent tagged-proteins were depolymerized against 1000 vol of buffer in the microdialysis apparatus for 24 h at 2*C as described for rabbit actin 121,341. RESULTS

Temperature and HL-6Ogrowth. Heat shock of 45OC for 1 h dramatically inhibited proliferation of HL-60 cells in culture (Fig. 11 and resulted in their death as determined by trypan blue staining. Temperatures of 43 and 44Y revealed growth rates depressed from 42OC heat shock but were not lethal to the majority of cells (data not shown), Two-dimensional PAGE. Fluorography of [35Sjmethionine-labeled nucIear proteins from HL-60 cells of the Cl-phase of the cell cycle resolved a compIex array

NUCLEAR

PROTEIN

ASSOCIATED

WITH

LETHAL

HEAT

§HOCK

169

[35S]methionine incorporation by hsp 70 mo hsp 90 dramatically (Fig. 2C)~ A further increase of two degrees (43’C) reduced incorpo ti~n into hsp 70 and the actins, while hsp 90 remaine mhanged (Fig. 2D). Additionally three labeled areas (“spots9’) of 48JIOO MT but varying in p1 (i.e.> 7.5, 6.4P 5.8) were noted at 43’C (Fig. 2C). Following a lethal bs of 45’C, p 48 (~17.5) increased while p 48 (~1 6.4) and p 48 (p1 5.8) di 3A). A slight reduction of hsp 70 was ob clear protein sample from the S-1 part substantial increase in p 48 (~17~5) co pattern, and a reci cal reduction in hsp 96 and the loss of hsp 70 (Fig. ). The synthesis of p 48 was further increased at mid S-phase (partition S-3) (Fig. 3C). Hsp 70 was not observed m S-phase? and hsp 90 disap-

FIG. 2* Two-dimensional PAG fluorographic patterns of imidazale buffer-extracted nuc!ear proteins from HL-60 cells cultured in the presence of [%]methionine. Nuclei were sorted from the Gi partition of the cell cycle following hs at three different temperatures for 1 h. Each population of nuclei was collected and extracted and the solubilized proteins were analyzed in UMEA (500 ng each sample). (A) Gi, 39’C hs, (B) Gi, 37’C control, ((2) G1, 4l’C hs, (D) G1, 43’C hs. The actins are represented with an A and vimentin with a V.

of labeled proteins fohowing exposure to hs for 1 h at 39’C. These proteins ranged from &I= 17,000 to 150,000 and from pI 4.0 to 7.0 (Fig. 2A). Hsps 70 and 90 were synthesized in modest amounts following hs at this temperature. They were not seen under normal culturing conditions (control) at 37’C (Fig. 2B). The nuclear actins (A) and vimentin (V) are indicated as reference proteins. Increasing the temperature by two degrees (4l’C) reduced general protein synthesis but enhanced

FIG. s0 Two-dimensionai PAG fluorographic patterns of imidazale buffer-extracted nticlear proteins from I%-60 ceh cukured in the presence of [a%]methionine, Nuclei were sorted from five partitions of the cell cycle following hs at 45’C for 1 h. Each population of nuclei was collected and extracted and the sokbihzed proteins were analyzed in UMEA (500 ng each sample). (A) G1 partition, (B) S-1 partition, (C) S-3 partition, (D) S-5 partition, (E) G2 partition.

170

FIG. presence partition.

PIPKIN

4.

Two-dimensional of [32P]phosphate.

PAG Nuclei

fluorographic were sorted

ET

AL.

patterns of imidazole buffer-extracted nuclear proteins from and processed as in Fig. 2. (A) G1 partition, (B) S-l partition,

peared by partition S-3. Protein 48 was reduced in late S-phase (partition S-5) (Fig. 3D) and GZ-phase of the cell cycle (Fig. 3E). As observed in these fluorographs, the actins were among the last proteins synthesized at a lethal hs temperature. Phosphorylation of p 48 and hsp 90 was substantial in GI-phase, and hsp 70 was weakly labeled following hs at 45’C (Fig. 4A). Phosphorylation of p 48 continued to increase in S-l (Fig. 4B) and S-3 (Fig. 4C) but was reduced substantially in S-5 partition (Fig. 4D), Hsp 90 phosphorylation was reduced dramatically in early Sphase and vanished by S-5. Hsp 70 was not phosphorylated in S-phase. There was no “P-labeling of p 48 in the GZ-phase (data not shown). V8 Protease digests. Protease maps of p 48 and hsp 90 depicted differences in the migratory rates of their respective major fragments (Fig. 5, lane 1 and 2). Hsp 70 (lane 3), however, showed fragments in the lower iUr range which were congruent with those of p 48 but not with hsp 90. Immunochemical determination. Samples of GI nuclear protein from cells exposed to 43’C were transferred from microslab gel to nitrocellulose paper and

HL-60 cells cultured in the (C) S-3 partition, (D) S-5

were incubated in the presence of primary anti-p 48 and labeled secondary antibodies and analyzed by fluorography. While labeling of all three 48 polypeptides was clearly seen, no other labeling of nuclear proteins was detected (Fig. 6A). Fluorography of the GI sample from cells exposed to 45’C following transfer from microslab gel to nitrocellulose paper and incubated in the presence of anti-hsp 70 (Fig. 6B) and anti-hsp 90 (Fig. 6C) showed the specificity of each antibody for its respective hsp. No other protein was recognized by these antibodies, including p 48. Protein binding studies. Binding experiments with isolated p 48 (Mr 48,000), fluorescent tagged-actin (iW= 43,000), and hsp tagged-90 (&I= 90,000) were conducted in one-dimensional nondenaturing PAG. Fluorescent tagged-actin and hsp tagged-90 were run alone (Fig. 7, lanes 1 and 2). Mixing of p 48 with tagged-actin (lane 3) and p 48 with tagged-90 (lane 4), in separate samples, showed weak fluorescence in the area of the gel which indicated binding of these proteins. Nonfluorescent tagged-bovine serum albumin fraction V (IO ng) and phosphorylase b (10 ng) were mixed

NUCLEAR

PROTEIN

ASSOCIATED

WITH

LETHAL

HEAT

171

SHOCK

FIG. 5. One-dimezisional chromatograph (silver stained) of V8 protease fragments of hsps. Lane 1, HL-60 hsp 90 (200 ng); Lane 2, HL-60 p 48 (200 ng); Lane 3, HL-60 hsp ‘70 (200 ng). Reference markers of 17 and 43 X 10m3 Mr are indicated.

separately with fluorescent tagged-p 48 (25 ng) in solvent-buffer and run in different PAG lanes as controls. There was no additional fluorescence (only p 48 fluoresa lack of nonspecific bindcence) which demonstrated ing (data not &own).

Lethal hs for 1 h at 45’C elicited the synthesis and phosphorylation of a Mr 48,000 and pI 7.5 protein which was cell cycle regulated and most abundant during Sphase of the HL-60 cell cycle. Synthesis and phosphorylation of p 48 was concomitant through most of the cell cycle except during the Gz-phase where “P-labeling was absent. Differences in sensitivity levels of these isotopes could, in part, account for 35S-labeling of ~48 in Gzphase with no phosphorylation of p 48. Hsp 70 was only labeled in G1, while hsp 90 was labeled in G1, and S-1 at sublethal and lethal temperatures. Stress protein synthesis is usually confined to the G1-phase [35], with exceptions [15, X9]* Cells recovered within 24 h from a hs of 43OC and within 36 h at 44’C, but culture death ensued at 45’C. Inhibition of general cellular protein synthesis accompanied by a reciprocal expression of hsps by moderately elevated temperatures has been reported as a common event [9], while p 48 synthesis induced by excessive heat was reported in several tissues [g-13]. In conjunction with p 48 (p17.5), two additional radioactive spots of identical iMr, but different pI (i.e., 5.8, 6.4) were reported and assumed as isoforms of the former polypeptide [9]. We observed two polypeptides in HL60 nuclei with similar pFs at a hs of 43OC, but not at lethal hs, which were confirmed as isoforms of p 48 (~1 7.5) by immunochemical determination in microslab gels (Fig. 6). Lethal hs completely inhibited general pro-

A

FIG. 6. ImmLmochemical blotting of p 48 (~17.4) and isoforms p1 6.4 and pZ 5.8. HL-60 G1 nuclear proteins (500 ng) extracted from cells hs at 43’C for 1 b were separated in UIVIEA by two-dimensional PAGE (A). Polyclonal primary antibodies were made against hsp 48 in rabbits. Western blot determination was done with unlabeled primary anti-hsp 48 and “S-labeled commercially obtained secondary antibodies. Immunochemical blotting of bsps 70 and 90. HL-60 G.l nuclear proteins (500 ng) extracted from ceilshs at 4:3’C for 1 h were separated in UREA by two-dimensional PAGE. Anti-&p polyclonaI primary and antibodies were obtained commercially for hsps 70 and 90. Nitrocellulose blot of hsp 7G (B) and hsp 90 (C). The arrows represent &f- . and w1 of the actins. ~.~~aonroximate ~-~..~~~~ hvoothetical -.~ . coordinates

172

PIPKIN

ET

AL.

espoused as a mechanism of repair and thus safeguarding against traumatic events [38-401. The appearance of the family of 48 hsps (synthesis and phosphorylation) is possibly used as a mechanism to protect vulnerable cellular structures from extreme temperatures, where the other classical hsps have failed. The significance of p 48 is currently unknown. However, if p 48 is a ubiquitous protein in mammalian systems and is evoked by terminal stressors other than hs, then it could serve as a biomarker for lethality. REFERENCES FIG. 7.

Fluorescence analysis of PAGE of isolated nuclear proteins. Actin, hsp 90, and p 48 from HL-60 were isolated and purified from preparative gels (220 unit). Actin and hsp 90 were tagged with fluorescent dye (see Materials and Methods). Lane 1, actin (25 ng); Lane 2, hsp 90 (25 ng); Lane 3, actin (25 ng) + p 48 (10 ng); Lane 4, hsp 90 (25 ng) + p 48 (10 ng).

tein synthesis and reduced synthesis and phosphorylation of hsps 70 and 90 in the HL-60 cells. A similar scenario was observed elsewhere for hsp 70 [9]. The concomitant increase in p 48 and decrease of hsp 70 and hsp 90 appeared initially as if p 48 were formed from a reciprocal coalescing of breakdown products of hsps 70 or 90 resulting from excessive hs. This assumption seemed incorrect by the diverse polypeptide maps of p 48 and hsp 90; however, hsp 70 showed some homology with p 48. Therefore the micro-Western blotting technique was employed to resolve this question. There was no cross-reactivity between p 48 and anti-hsp 70 and antihsp 90 (Figs. 6B and 6C). Furthermore, following immunological determination by Western blotting of PAGEseparated V8 protease polypeptides of p48 reacted with hsp 70 antibodies, no bands were seen (data not shown). This indicates that p 48 is not derived from hsps 70 or 90. Polypeptide maps of other hsps from varied tissues in the approximate range of 48,000 M= were reported [9131; the fragment distributions of these proteins, including those from HL-60 nuclei, appear dissimilar. In theory, a family of closely related hsps (approximately 48,000 MJ may exist but must await other biochemical analyses to determine this relationship. The binding characteristics of sp 90 with other proteins is well documented [36,37], and we observe in this work a similar event with p 48. Not only does p 48 bind with hsp 90, but it forms a complex with the actins. Similar binding properties of a chick fibroblast 47 hsp to collagen was reported [12]. The theory of sps/hsps acting as “molecular chaperones” by binding and removing aberrant or damaged proteins from the cell has been

1.

2.

Schesinger, M. J. M., Ashburner, M., and Tissieres, A. (Eds.) (1982) in Heat Shock from Bacteria to Man, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Pipkin, J. L., Anson, J. F., Hinson, W. G., Burns, E. R., and Casciano, D. A. (1989) Appl. Theor. Electrophoresis 1, 77-84.

3.

Coureon, A. M., Maisonhaute, C., (1984) Exp. Cell Res. 153, 515-521.

4.

Pipkin, J. L., Anson, J. F., Hinson, W. G., Burns, E, R., Casciano, D. A., and Sheehan, D. M. (1987) Biochem. Biophys. Actu 927,334-344.

5.

Buzin, C. H., and Bournias-Vardiabasis, W. (1984) PFOC. N&l. Acad. Sci. USA 81,4075-4079. Anson, J. F., Hinson, W. G., Pipkin, J. L., Kwarta, R. F., Hansen, D. K., Young, J. F., Burns, E. R., and Casciano, D. A. (1987) Dev.Biol. 121,542-547. Brown, I. R., and Rush, S. J. (1984) Biochem. Biophys. Res. Commun. 120,150-155.

6.

7. 8.

Pipkin, J. L., Anson, Casciano, D. A. (1986)

9.

Margulis, Cytologia

J. F., Hinson, Electrophoresis

and

Best-Belpomme,

W. G., Burns, 7, 463-471.

B. A., Barlovskaja, V. V., and Semenova, (St. Petersburg) 29, 1196-1200.

E. R., and E. G. (1987)

10.

Smagina, L. V., and Margulis, burg) 30, 478-482.

11.

Giallongo, A., Fee, S., Showe, L. C., and Grace, C, M. Biochem. Biophys. Res. Commun. 134, 1238-1244. Nagata, K., Saga, S., andyamada, K. M. (1986) J. CellBiol.

12.

B. A. (1988)

M.

Cytologiu

(St. Peters(1986)

103,

223-229. 13.

Nakai, A., Hirayoshi, 265,992-999.

14.

Margulis, B. A., Zhivotovski, B. D., Pospelova, T. V., and Smagikna, L. V. (1991) Exp. Cell Res. 193, 219-222. Pipkin, J. L., Anson, J. F., Hinson, W. G., Burns, E. R., and Casciano, D. A. (1989) Appl. Theor. Electrophoresis 1, 85-94.

15. 16. 17.

K., and Nagata,

K. (1990)

J. Biol.

Chem.

Hinson, W. G., Pipkin, J. L., Hudson, J. L., Anson, J. F., and Tyer, H. (1982) Cytometv 2,390-394. Herzenberg, L. A., Sweet, R. G., and Herzenberg, L. A. (1976) Sci. Am. 23, 108-117.

18.

Pipkin, J. L., Anson, J. F., Hinson, W. G., Schol, H., E. R., and Casciano, D. A. (1986) Cytometry 7, 147-156.

19.

Pipkin, sciano,

J. L., Anson, J. F., Hinson, D. A. (1984) J. Biochem. 95,

W. G., Schol, 323-333.

20.

Pipkin, (1982)

J. L., Hinson, W. G., Anson, Cell Biol. Int. Rep. 6, 205, 214.

J. F., and

21.

Pipkin, Cook, Theor.

J. L., Hinson, W. G., Anson, J. F., Schol, H. M., LynL. E., Burns, E. R., and Casciano, D. A. (1991) Appl. Electrophoresis 2, 31-41.

Burns,

H., and Hudson,

CaJ. L.

NUCLEAR 22.

Bramhall, S., Noacb, N., Wu, And. Biochem. 31,146-14%

23.

ASSOCIATED

HEAT

I.73

SHOCK

Hinson, W. G., Pipkin, J. L., Anson, J. F., Burns, E. R., and Casciano, D. A. (1987) Rev. Sci. Instrum. 58,1736-1742.

33.

Towbin, II., Staehelin, T.> and Acad. Sci. USA T6,4350-4354.

24.

O’Farrell,

34.

25.

Ratz,

DOS Remedios, C. G., and Cooke, Acta 788, 193-205. Tanguay, R. M., and Vincent, M.

J. Viol.

Chem.

A., and Celis,

250,

J. R. (1969)

LETHAL

Tahourdin, C. S. M., Neihart, (1981) Biochem. J, 2O,SlO-915.

I’~ H. (1975)

and Leowenburg,

WITH 32.

G. P., Celis,

M.,

PROTEIN

4007-4021.

J. E. (1988)

Leukemia

2, 602-

615. 26. 27.

35.

Banner, 88. Hager:

W., and Laskey, D. A., and

FL. A. (1974)

Burgess,

Eur.

R~ R. (1980)

J. Biochem. Anal.

J. (1979)

R. (1984) (1982)

Bivchem.

Proc.

M. Notl.

Biophys.

Gun. Jo &o&em.

36.

Gate& M. G., Binart, N., Jung-Testas, E. E., Feramisco, J. R., and Welch,

37.

Koyasu, S., Nishida, E., Kadowaki, T., Matsuzaki, Harada, F., Kasuga, M., Sakai, II., and Sahara, Natl. Acad. Sci. USA M,8054-8058.

109,

GO?

I., Renoir. J. M., Bavheu, W. J. (1985) E&%0. ,X 4,

3131-3135.

76-86. 28.

Safer,

29.

Cleveland, D. W., Fisher, S. G., Kirecher, M. W., and Laemmlik, V. K. (1977) J. Viol. Chem. 252, 1102-1106.

30.

Takeda, A., and Cone, mu.. 122,932-937.

FL E. (1984)

Pipkin, J. L., Anson, Casciano, D. A. (1988)

J. F., Hinson, W. G., Burns, E. R., and Camp. Biochem. Physiol B 89, 43-50.

Received Revised

Gordon,

I., and Bustin,

306-315.

46,83-

Biochem.

N. K., Isenburg,

D. (1989)

Anul.

Biochem.

February 24, 1992 version received May

1’78,

21, 1992

32-37.

Biochem.

Biophys.

Res. Com-

38.

Hightower,

39.

Young,

L. E. (1991)

40.

Sadis, S., Raghavendra, chemistry .29? 8199-82Q6.

R. A., and Elliott,

F., Iida, K., I. (1986) F’roc.

Cell 66? 191-197. T. J. (1989) K., and

Cell 5

Higbtower,

L. E. (1990)

B~Q-