J. Cell Sci. 54, 193-206 (1982) 193Printed in Great Britain © Company of Biologists Limited 1982

LOCALIZATION OF MITOTIC FACTORS ON

METAPHASE CHROMOSOMES

RAMESH C. ADLAKHA1, CHINTAMAN G. SAHASRABUDDHE1,DAVID A. WRIGHT3, WILLIAM F. LINDSEY* AND POTU N. RAO1

Departments of 1DevelopmentaI Therapeutics, ^Molecular Biochemistry, and sGenetici,The University of Texas System Cancer Center, M. D. Anderson Hospital and TumorInstitute, Houston, Texas 77030, U.S.A.

SUMMARY

The objective of this study was to determine whether the mitotic factors of HeLa cells, whichinduce meiotic maturation, i.e. germinal vesicle breakdown (GVBD) and chromosome con-densation, when injected into fully grown Xenopus laevis oocytes, were localized in the cyto-plasm or associated with the metaphase chromosomes. Cytoplasmic extracts Were prepared bylysing mitotic HeLa cells in low-salt hypotonic buffer and separating the chromosomes bycentrifugation. The mitotic factors bound to chromosomes were extracted with high-salt(0-2 M-NaCl) buffer. Both the cytoplasmic and chromosomal protein fractions were evaluatedfor their maturation-promoting activity (MPA) in the Xenopus oocytes. The results of this studyindicate that both the cytoplasmic and chromosomal fractions are identical in many respects,including their ability to induce GVBD, but the specific activity of the chromosomal fractionwas at least threefold greater than that of the cytoplasmic fraction. These data suggest thata major portion of the mitotic factors is localized on the metaphase chromosomes. This associa-tion does not appear to be due to adventitious binding of mitotic proteins to chromosomesduring the extraction procedures. Furthermore, when extracts were prepared in a similar wayfrom early- and mid-G,-phase HeLa cells, only the nuclear extracts had MPA and no activitywas found in the cytoplasmic fraction. Both the cytoplasmic and nuclear extracts of late-G! cellsexhibited MPA. These data support the conclusion that the mitotic factors become preferen-tially bound to chromatin as soon as they are synthesized, and as the cell synthesizes moreof these factors in preparation for mitosis, increasing amounts of them are retained in thecytoplasm.

INTRODUCTION

In eukaryotic cells in general, the chromosomes are visible as discrete entities onlyfor a brief period during mitosis or meiosis. However, when a mitotic cell is fused withan interphase cell using either ultraviolet light-inactivated Sendai virus or poly-ethylene glycol, the factors present in the mitotic cell induce breakdown of the inter-phase nucleus and the condensation of chromatin into discrete chromosomes. Thephenomenon has been called premature chromosome condensation (Johnson & Rao,1970). The morphology of the prematurely condensed chromosomes (PCC) indicatesthe position of an interphase cell in the cell cycle at the time of fusion with a mitoticcell (Johnson & Rao, 1970; Rao & Johnson, 1970). The mitotic factors have beenfound to migrate to the interphase nucleus where they subsequently become associatedwith the PCC (Rao & Johnson, 1974).

For the isolation and characterization of the mitotic factors, an in vitro bioassay

194 R.C. Adlakha and others

system became a necessity. More recent studies from this laboratory (Sunkara, Wright& Rao, 1979 a, b) have demonstrated that the immature frog oocyte system is idealfor this purpose. When extracts from mature amphibian oocytes (Masui & Markert,1971), mitotic HeLa cells (Sunkara et al. 1979a, b), and mitotic Chinese hamsterovary cells (Nelkin, Nichols & Vogelstein, 1980) are injected into Xenopus laevisoocytes, they exhibit maturation-promoting activity (MPA) as evidenced by germinalvesicle breakdown (GVBD) and condensation of chromosomes. Since the meioticmaturation in the amphibian oocytes appears to be similar to the induction of PCC inmammalian cells, particularly with regard to the breakdown of the nuclear membraneand condensation of chromosomes, these results suggest that the mitotic and meioticfactors are similar. Mitotic factors accumulate slowly in the beginning of G2 but thenproceed at a progressively rapid rate during the late Ga and reach a threshold at theG2-mitotic transition, when the diffused chromatin condenses into discrete chromo-somes (Sunkara et al. 1979a).

However, the extraction procedures employed in these studies disrupted not onlythe whole cells but also the structural integrity of the metaphase chromosomes, thusraising the possibility that proteins from the cytoplasm as well as those bound to thechromosomes were extracted. Hence, we raised the question of whether the factorsfrom the mitotic HeLa cells that induce GVBD in Xenopus oocytes were localized inthe cytoplasm or on the metaphase chromosomes. To answer this question we modifiedour extraction procedures so that the cytoplasmic protein fractions are not contamin-ated by the chromosomal proteins. The results of this study indicate that mitoticfactors are preferentially bound to metaphase chromosomes and can be released byextraction with high-salt solution.

MATERIALS AND METHODS

Whenever possible, chemicals of analytical reagent quality were used. They were purchasedfrom Sigma Chemical Co. (St Louis, Mo.), except for 2-mercaptoethanol and CoomassieBrilliant Blue G250 protein binding dye, which were obtained from BioRad Laboratories(Richmond, Calif.).

Cells and cell synchrony

HeLa cells were grown as monolayer cultures in 150-mm culture plates at 37 °C Eagle'sminimum essential medium (MEM) (Grand Island Biological Co., Grand Island, N.Y.)supplemented with 5 % foetal calf serum, sodium pyruvate, non-essential amino acids, glut-amine, and penicillin/streptomycin mixture in a humidified, 5 % CO, atmosphere (Rao &Engelberg, 1965). These cells have a cell cycle time of 22 h, consisting of a io-5-h Gt period,a 7'O-h S period, a 3-5-h G, period and 1 h of mitosis (Rao & Engelberg, 1966).

To obtain mitotic cells, a random population of HeLa cells plated in 150-mm Lux culturedishes was partially synchronized by a single thymidine block (2-5 DIM) for 24 h. The thymidineblock was removed by washing and incubating cells in regular medium. Three to four hoursafter the reversal of the thymidine block, the dishes were further incubated in a chamber filledwith N,O at a pressure of 80 lb/in1 (543 kPa) and incubated at 37 CC for 10 h (Rao, 1968). Therounded and the loosely attached mitotic cells were selectively detached by gentle pipetting,which routinely yielded a population with a mitotic index of 98 %. In order to disrupt themitotic spindle, the mitotic cells were collected and kept on ice or incubated in the presence ofColcemid (0-05 /*g/ml) for 1 h at 37 °C, or both. Since the mitotic block induced by N,0 is

Localization ofmitotic factors 195

reversible, incubation of N,O-blocked mitotic cells under normal culture conditions for 3 hyielded highly synchronous populations of Gt cells.

To obtain a pure population of HeLa cells in Gi} we first synchronized exponentially growingcells into S phase by the excess thymidine double-block method (Bootsma, Budke & Vos, 1964;Rao & Engelberg, 1966). Colcemid (0-05 /*g/ml) was added 4 h after the reversal of the secondthymidine block and incubation continued. At 7, 8 and 9 h after reversal of the second thy-midine block, Colcemid-arrested mitotic cells were removed by selective detachment and thecells that remained firmly attached to the dish were harvested by trypsinization to yield early-,mid-,and late-Gt populations, respectively. The mitotic indices in all three populations wereless than 3 %.

Preparation of cell extracts

Mitotic HeLa cells collected in the cold as described above were washed 3 times with MEMwithout serum and once with low-salt buffer (buffer A) at 4 °C and cells were resuspended at aconcentration of 20 x io6 to 40 x io'/ml and incubated for 10 min at 37 °C in buffer A, amodification of the buffer described by Blumenthal, Dieden, Kapp & Sedat (1979), which con-sisted of 15 mM-Tris-HCl containing 15 mM-NaCl, 0-5 mM-spermidine, 0-15 mM-spermine,15 mM-/?-mercaptoethanol, 2 mM-EDTA, 0-5 mM-ethylene glycol (y?-aminoethyl ether)-N,N,N',N',-tetnaceUc acid (EGTA), 1 mM-ATP, 60 mM-KCl, 1 mM-phenylmethylsul-phonylfluoride (PMSF, from a fresh stock of 100 mM-PMSF in isopropylalcohol), 5 mM-NaF,5 mM-sodium /S-glycerol phosphate, and 5 % (v/v) glycerol (pH 7-4)- Some other buffersystems were also tried as indicated in the footnote to Fig. 1. The homogenate was lysed by4 passages through a syringe (23 gauge) and was centrifuged at 30000 g for 15 min. The super-natant was designated as cytoplasmic extract and stored at o °C. After the second extractionwith buffer A the chromosomal pellet was resuspended in a high-salt buffer (10 mM-Na,HPO«/NaHjPO4, 200 mM-NaCl, 5 % glycerol, 2 mM-EGTA, 10 mM-MgSO«, 1 mM-ATP, 1 mM-PMSF, s mM-NaF and 5 mM-sodium /?-glycerol phosphate), pH 6-5, and stirred at 4 °C for30 min. The homogenate was centrifuged at 30000J* for 15 min. The pellet was discardedand the supernatant was designated as chromosomal extract. The 3 supernatants designatedas cytoplasmic extracts I and II, and chromosomal extract (Fig. 1) were further centrifugedat icooooy for 1 h in a TI 50 rotor in a Beckman L5-50 ultracentrifuge. The pellets werediscarded and aliquots from the resulting clear supernatants were injected into X. laevisoocytes to determine their MPA. The extracts were always stored at —70 °C until furtheruse.

Protein determinations

Protein concentration was determined by the use of a Coomassie Brilliant Blue G-250 bindingassay according to Bradford (1976), bovine serum albumin being the standard.

Preparation of Xenopus oocytes

Oocytes were obtained by surgically removing a portion of the ovaries from Xenopusfemales. With a small incision that can be closed by a few stitches, multiple harvests of oocytescan be obtained from the same animal. All operations on oocytes were conducted using modifiedBarth's medium (Woodland, 1974) supplemented with MgCl8.H,0 (0-12 g/l). Oocytes weremanually dissected from the follicles with watchmakers' forceps.

Assay for maturation-promotion activity

Cell extracts were assayed by injecting between 40 and 70 nl (nanolitre) into each oocyte.Injected oocytes were inspected for GVBD after 2 to 4 h. GVBD is detected by a depigmenta-tion of an area of the animal hemisphere (Merriam, 1972). The presence or absence of thegerminal vesicle for the questionable oocytes was also determined by fixation in 7-5 % trichloro-acetic acid and subsequent dissection for scoring of the oocyte. Some oocytes were fixed inSmith's fluid, dehydrated in ethanol and amyl acetate (Drury, 1941), embedded in paraffin,sectioned at 7 /tm, and stained with Feulgen and fast green (Subtelny & Bradt, 1963).

196 R. C. Adlakha and others

Ammonium sulphate precipitation

One volume of saturated ammonium sulphate was added to 4 vol. of cytoplasmic or chromo-somal extracts to yield a 20 % (NHi),SO4 precipitate fraction. The resulting precipitate wascentrifuged, more ammonium sulphate was added to the supernatant, and the above processrepeated to yield 40%, 60 %, 80% and 100% (NH4)1SO4 precipitate fractions. The resultingprecipitates were dissolved in 1 vol. of 15 mM-Tris'HCl (pH 7-4) buffer containing 1 mM-PMSF, 1 ITLM-ATP, 5 mM-NaF and 5 nvM-sodium /?-glycerol phosphate and dialysed over-night against 1000 vol. of the same buffer with one change of buffer.

Trichloroacetic acid (TCA) precipitation

The crude cytoplasmic and chromosomal extracts were made 2 % with respect to TCA byaddition of 100 % TCA. The precipitates formed were removed by centrifugation according toGoodwin, Sanders & Johns (1973). Six volumes of acetone were then added to the clear super-natants to precipitate the remaining proteins. The 2 precipitates thus obtained were washed inacetone/o-i M-HC1 (6:1, v/v) and then 3 times in acetone. All 4 precipitates were dissolved in0-5 ml of 15 mM-Tris-HCl (pH 7'4), containing the protease and phosphatase inhibitors, anddialysed against the same buffer for 4 h with one change of buffer.

RESULTS

Maturation-promoting activity of the cytoplasmic and chromosomal protein fractions frommitotic HeLa cells

The protocol employed in this study to separate the chromosomal proteins fromthe cytoplasmic proteins is presented in Fig. 1. By this procedure, the metaphasechromosomes remained intact as clusters when observed under a phase-contrastmicroscope with little or no cytoplasmic contamination (Fig. 2 A, B).

The maturation-promoting activities of the cytoplasmic and chromosomal fractionsare shown in Table 1. Although the total activity of the two fractions appeared to bethe same, the protein content of the cytoplasmic fraction was at least three timesgreater than that of the chromosomal fraction. Use of other low-salt hypotonic buffersystems (see legend to Fig. 1) for extraction yielded similar results. A second extrac-tion (cytoplasmic fraction II) of the chromosomal pellet with buffer A even for 30 mindid not recover any maturation-promoting activity (MPA) (Table 1).

Using crude extracts, we noted that the factors extracted from chromosomes didnot give entirely normal maturation in the oocytes. While a white spot typical ofmeiotic maturation initially appeared after injection of either the cytoplasmic orchromosomal extracts, the chromosomal-extract-injected oocytes rapidly becamegrossly depigmented. Histologically, the oocytes had undergone germinal vesiclebreakdown (GVBD) but the meiotic figure was not always at the surface of theoocyte. This observation raised the possibility that factors from the chromosomalfraction were different from those of the cytoplasm. To test this possibility, preliminarycharacterization of these factors was undertaken.

Localization of mitotic factors

MITOTIC HeLa CELLS

197

Wash 3-4 times with serum-freeMEM and once with buffer A

PELLET

Incubate at 37 °C for 10 min in buffer A,lyse, centrifuge at 15000 rev./min for 15 min

PELLET

2nd extraction with buffer Aand centrifuge at 15000 rev./minfor 15 min

SUPERNATANT

Centrifuge*

PELLET SUPERNATANT(Discard) (Cytoplasmic fraction I)

PELLET SUPERNATANT

Extract with buffer B at 4 °Cfor 30 min and centrifuge at15000 rev./min for 15 min .

PELLET(Discard)

PELLET(Discard)

SUPERNATANT

Centrifuge*

Centrifuge*

SUPERNATANT(Cytoplasmic fraction I

PELLET(Discard)

SUPERNATANT(Chromosomal fraction)

Fig. 1. A protocol for the preparation of cytoplasmic and chromosomal extracts.Buffer A: 60 mM-KCl, 15 mM-NaCl, 15 mM-Tris-HCl, 0 5 mM-spermidine,

0-15 mM-8permine, 15 mM-/?-mercaptoethanol, 2 mM-EDTA, 0-5 mM-EGTA, 1 nffl-ATP, 1 triM-PMSF, 5 mM-NaF, 5 mM-sodium /?-glycerol phosphate and 5%glycerol (pH 7-4) (a modification of the buffer described by Blumenthal et al. 1979).

Buffer B: 10 mM-Na,HPO4/NaH,PO4, 200 mM-NaCl, 250 mM-sucrose, ioraM-MgSO4, 2 mM-EGTA, 1 mM-PMSF, 1 mM-ATP, 5 mM-NaF and 5 mM-sodiumyS-glycerol phosphate (pH 6-5).

Some other buffer systems tried in place of buffer A are as follows: (i) 5-10 mM-Na2HPO4, 1 mM-PMSF, 1 mM-NaF, 1 mM-sodium /?-glycerol phosphate, 1 mM-ATP, 1 mM-MgSO4, 1 mM-EGTA (pH 6-5). (ii) 70 mM-KCl, 1 mM-PMSF, 1 mM-ATP, 1 rtiM-MgSO4, 1 mM-NaF, 1 mM-sodium ^-glycerol phosphate, 1 mM-EGTA(pH 6-5). (iii) 60 mM-KCl, 15 mM-NaCl, is mM-HEPES, 0-5 mM-spermidine, 0-15mM-spermine, 15 mM-/?-mercaptoethanol, 2 mM-EDTA, 0-5 M - E G T A , I mM-ATP,1 mM-PMSF, s mM-NaF, 5 mM-sodium y?-glycerol phosphate, and 250 mM-sucrose(pH J:[) (a modification of the buffer described by Adolph, 1980).

• Centrifuge at 100000 g for 1 h at 4 °C.

R. C. Adlakha and others

• ' T - r '

Fig. 2. Giemsa-stained cytopreparations of mitotic HeLa cells synchronized by theN2O block method (A) ; chromosome clusters obtained by incubating at 37 °C andwashing the mitotic cells with the (low-salt) buflfer A as described in Materials andMethods (B); synchronized G2 cells (c); and nuclei obtained by incubating at 37 °Cand washing the synchronized G2 cells with buffer A (D). Note that the chromosomeclusters and the Ga nuclei are relatively free from any cytoplasmic remnants, x 1400.

Localization of mitotic factors 199

Preliminary characterization of mitotic factors from the cytoplasmic and chromosomalfractions

Ammonium sulphate precipitation. In both cases, the MPA was observed in fractionsprecipitated by 40% saturated (NH4)2SO4. No activity was observed in fractions pre-cipitated by 20%, 60% and 80% saturated (NH4)2SO4. The precipitation with 40%saturated (NH4)2SO4 resulted in at least threefold enrichment of the MPA.

Table 1. Maturation-promoting activities of cytoplasmic and chromosomal fractionsof mitotic HeLa cells

Oocytesinjectedwith:

Cytoplasmicfraction I

Cytoplasmicfraction II

Chromosomalfraction

Extractdilution

0

1 / 2

1 / 3i / 4

1/8

O

O

1 / 2

1 /31 / 41/8

Proteininjected

(ng*)

S8S2 9 2

195146

7369

1799 0

60

452 2

No.oocytesinjected

302525

25

25

3°25

252 0

25

No.oocytesshowingGVBD

302525

70

O

3O2523

50

%

inductionGVBD

1 0 0

1 0 0

1 0 0

280

0

1 0 0

1 0 0

92

250

• A total volume of 65 nl of the extracts was injected into each oocyte, and oocytes werescored for GVBD 2-3 h after injection. A concentration of 40 x 10" cells/ml was used forpreparation of these extracts as described in Materials and Methods.

The abnormal depigmentation that occurred with oocytes injected with crudechromosomal extract did not occur with the active ammonium sulphate-precipitatedfraction. We believe that the abnormal reaction of the oocytes may have been causedby another chromosomal protein contaminating the crude extract.

TCA precipitation. Mitotic factors from both the fractions were found to be solublein 2 % TCA and were precipitated by acetone. All the biological activity was recoveredin acetone-precipitated fractions. This resulted in about fivefold enrichment.

Effect of dialysis and some positive and negative ions on the mitotic factors

Mitotic factors from both the cytoplasmic and chromosomal fractions were foundto be non-dialysable proteins, highly sensitive to Ca2+ (1 ITIM), and unaffected byMg2+ (Table 2). The activity lost in the presence of 1 mM-Ca2+ could be recovered byaddition of 1 to 2 mM-EGTA. Addition of 1 to 2 mM-EDTA to the mitotic factorsin the presence of 1 mM-Mg2+ resulted in slight inhibition of MPA. EGTA or EDTAat concentrations greater than 2 mM had significant inhibitory effect. Addition of10 mM-EGTA or EDTA resulted in complete inactivation of the factors.

2oo R. C. Adlakha and others

Stability of the mitotic factors

Mitotic factors from both the fractions were quite stable when stored at — 70 °C.Incubation of the mitotic factors at 37 °C for 15 min had no significant effect on theirMPA. In the presence of inhibitors of proteases, (1 mM-phenylmethylsulphonyl-fluoride (PMSF) and phosphatases (1 mM-ATP, 5 mM-NaF and 5 mM-sodium/?-glycerol phosphate), these factors were stable for about 24 h even at room tempera-ture. Extracts are now routinely prepared in the presence of these inhibitors.

Table 2. Effect of positive and negative ions on the maturation-promoting activity ofmitotic factors from cytoplasmic and chromosomal extracts

Oocytes injected with:*

Dialysed fraction (DF)JD F + i mM-CaCl,DF + s mM-CaCl,DF+ 1 mM-CaClj<for 10 min) + 1 mM-EGTADF+ 1 mM-CaCl, (for 10 min+ 2 mM-EGTAD F + i mM-MgSO4

DF + 5 mM-MgSO4DF+iomM-MgSO4D F + i mM-MgSO4 (for 10 min)+ 1 mM-EDTADF + 1 mM-MgSO^ (for 10 min) + 2 mM-EDTAD F + i mM-EDTADF + 2 mM-EDTADF + s mM-EDTADF +10 mM-EDTAD F + i mM-EGTADF +2 mM-EGTADF + s mM-EGTADF +10 mM-EGTA

Induction of GVBD (%) withmitotic factors from:f

A

Cytoplasmicextract

ICO

0

0

9090

1 0 0

1 0 0

1 0 0

7070

1 0 0

1 0 0

5°0

1 0 0

1 0 0

700

Chromosomalextract

1 0 0

0

0

1 0 0

901 0 0

ICO

1 0 0

807090

1 0 0

400

9090801 0

• A volume of 65 nl of each of the samples was injected into each oocyte.t 20 oocytes Were injected in each case and GVBD was scored after 2-4 h.% Mitotic extracts (cytoplasmic and chromosomal) were dialysed overnight against a buffer

free from Ca1+, Mg1+, EDTA, and EGTA. For details, see Materials and Methods.

However, we have noticed that the partially purified (12 to 15-fold) mitotic factors(20-40 % ammonium sulphate fractions) lost their activity in 24 h at both o °C and— 70 °C even in the presence of these inhibitors. To test if this loss of activity of themitotic factors was due to a decrease in the total amount of proteins, we added1-5 mg/ml of bovine serum albumin (BSA) as carrier protein to supplement thelevel of proteins. This resulted in a complete stabilization of the mitotic factors forover 2 months at both o °C and — 70 °C.

Localization of mitotic factors 201

Table 3. Maturation-promoting activity of the cytoplasmic and nuclear extracts fromearly-, mid- and late G2 HeLa cells

Oocytesinjectedwith:#

Early-G!Cytoplasmic extractNuclear extract

Mid-G,Cytoplasmic extractNuclear extract

Late G,Cytoplasmic extractNuclear extract

Buffer alone

Proteininjected

(ng)

686201

721218

545176

No.oocytesinjected

1515

2525

15151 0

No. oocytesshowingGVBD

0

10

0

23

12

13

0

GVBDinduction

(%)

0

667

092-0

800867

0

• A volume of 65 nl of the extracts was injected into each oocyte and GVBD was scored2-3 h after injections. In some cases GVBD was scored after overnight incubation. 40 x i o 'to 60 x 10' cells/ml were used for preparation of these extracts as described in Materials andMethods.

Table 4. MPA in cytoplasmic and chromosomal fractions obtained by different methods

Method of chromosomeisolation

Amount ofproteinsinjected

(ng)

No. ofoocytesinjected

No. ofoocytesshowingGVBD

GVBD

(1) Wray-StubblefieldCytoplasmic fraction* 205Chromosomal fraction 68

(2) MEM + 005 % NP-40Cytoplasmic fraction 193Chromosomal fraction 73

(3) Isotonic buffer withmechanical shearingCytoplasmic fraction 181Chromosomal fraction 65

2525

2525

2525

o24

2525

2223

o96

100100

8892

• The presence of 1 M-hexylene glycol in the chromosome isolation buffer seems to inhibitthe maturation-promoting activity of the mitotic factors.

Mitotic factors from G2 cells

Since our earlier studies indicated MPA in the extracts of G2 cells we wanted tofind out whether these factors are localized in the cytoplasm or the nucleus. Cytoplas-mic and nuclear extracts from early, mid- and late G2 cells were prepared as describedfor mitotic cells. The G2 nuclei used for preparing nuclear extracts were practicallyfree from cytoplasmic contamination (Fig. 2 c, D). The maturation-promoting activitiesof the cytoplasmic and nuclear extracts from early-, mid- and late G2 cells are pre-

202 R. C. Adlakha and others

sented in Table 3. The cytoplasmic extracts from early- and mid-G2 cells did not haveany MPA, while the nuclear extracts exhibited activity. A significant amount of MPAwas observed in the cytoplasmic extracts from the late G% cells, although the specificactivity was much lower than those from the nuclear extracts.

Localization of mitotic factors on metaphase chromosomes

The above experiments indicate that the mitotic factors are preferentially associatedwith the chromosomes. Since the chromosomes are prepared by lysing the mitoticcells in low ionic strength (hypotonic) buffer, it is possible that the mitotic factors

Table 5. The amount of radioactivity found in different fractions of the unlabelled mitoticcells lysed and extracted in the presence of cytoplasmic extracts from mitotic cellsprelabelled with tritiated amino acids

Total c.p.m./io' Relative c.p.m.Fraction cells (%)

Cytoplasmic extract from labelled mitotic 38166 ioo-ocells + unlabelled mitotic cells*

Extracts from the unlabelled mitotic cellsCytoplasmic extract I 35 571 93-2Cytoplasmic extract II 1706 4*4Chromosomal extract 313 o-8

• A total of 11 x io7 mitotic cells were used for the preparation of labelled cytoplasmic ex-tract and this extract was mixed with an equal number of unlabelled mitotic cells before theywere lysed for extraction of the various fractions.

became adventitiously associated with the chromosomes. To rule out this possibilitychromosomes were prepared by different methods: (1) using the Wray-Stubblefieldmethod (1970); (2) lysing mitotic cells in regular culture medium without serum inthe presence of a non-ionic detergent, NP-40; and (3) lysing mitotic cells in isotonicbuffer (0-125 M 8^0 by mechanical shearing (3-4 passages through a 23 gaugeneedle).

The results of these experiments are presented in Table 4. Since isotonic bufferswere used for isolation of chromosomes (in methods (2) and (3)) it is unlikely thatnon-specific ionic binding of cytoplasmic proteins with chromosomes would occur.

To rule out further the possibility of adventitious association of mitotic proteinswith the chromosomes the following experiment was performed. HeLa cells weresynchronized in S phase by the excess thymidine double-block method. Six hoursafter the reversal of the second thymidine block, 1 /iCi/ml each of [^HJleucine (sp. act.6o-o Ci/mmol and pHJtryptophan (sp. act. 5-4 Ci/mmol) (New England Nuclear,Boston, Mass.) and Colcemid (0-05 /ig/ml) were added and incubation continued.Eight to 10 h after the addition of Colcemid, labelled mitotic cells were collectedby selective detachment. Using the same procedures an equal number of unlabelledmitotic cells was also obtained at the same time. First, cytoplasmic extracts were pre-pared from the labelled mitotic cells was described in Materials and Methods. Then

Localization of mitotic factors 203

the unlabelled mitotic cells were washed and pelleted by centrifugation. The cell pelletwas resuspended in the cytoplasmic extract of the labelled mitotic cells and thenthe cells were lysed and extracted for cytoplasmic and chromosomal fractions. Theamounts of radioactivity found in the different fractions are shown in Table 5.These results clearly demonstrate that adventitious association of mitotic proteinswith the chromosomes is negligible according to the procedures employed.

DISCUSSION

The present study confirmed our earlier observation (Sunkara et al. 1979 a, b\Rao, Sunkara & Wright, 1981) that mitotic factors from mammalian cells can induceGVBD and chromosome condensation in X. laevis oocytes. However, the most sig-nificant aspect of this study lies in the use of low-salt hypotonic buffer to extract cyto-plasmic proteins preferentially and separate the intact metaphase chromosomes. Aftera second extraction with the low-salt buffer the chromosomal pellet (Fig. 2B) wasextracted with the high-salt buffer that was used in earlier studies (Sunkara et al.1979 a, b). The differential extraction of cytoplasmic and chromosomal fractions haspermitted the study of their maturation-promoting activities individually (Table 1).

The important observation resulting from this study is that both the cytoplasmic andthe chromosomal extracts from mitotic HeLa cells exhibit MPA in Xenopus oocytes,but the specific activity of the chromosomal fraction is at least three times greaterthan that of the cytoplasmic fraction (Table 1). However, the second extraction ofchromosomes with low-salt buffer does not recover any MPA. These results suggestthat a major portion of the mitotic factors in the cell is associated with the metaphasechromosomes. This association does not appear to be due to adventitious binding ofmitotic proteins to the chromosomes during the extraction procedures (Tables 4and 5). These observations are in agreement with those of Wray, Elgin & Wray (1980).They have reported that incubation of unlabelled isolated chromosomes with cyto-plasmic extracts of CHO cells prelabelled with [14C]leucine did not result in anysignificant adventitious association of the labelled cytoplasmic proteins with thechromosomes. Preliminary characterization of the mitotic factors from the cyto-plasmic and chromosomal fractions reveal that they are identical in many ways:both are heat labile, non-dialysable, and Ca2+-sensitive proteins that are precipitablewith 40% (NH4)aSOi and soluble in 2% TCA (Table 2).

The chromosomal affinity of the mitotic factors becomes even more apparent whenthe MPA from cytoplasmic and nuclear extracts of early-, mid- and late G2 HeLa cellsare compared (Table 3). No measurable MPA was observed in the cytoplasmicextracts of early- and mid-GE cells, whereas considerable activity was detected in thenuclear extracts. Further concentration of the cytoplasmic extracts of early- andmid-G2 cells by 40% (NH4)2SO4 precipitation did not increase the ability of theseextracts to induce GVBD. Even though both the cytoplasmic and nuclear extractsof the late G2 cells exhibited MPA, the nuclear extract had 3 to 4 times greaterspecific activity, as was the case for the mitotic cells. These results suggest that asthe mitotic factors are synthesized during G2 they readily and preferentially associate

204 R- C- Adlakha and others

with the chromatin. As a cell synthesizes more of these factors in preparation formitosis, increasing amounts of these factors are retained in the cytoplasm. This isconsistant with our earlier observation that PCC of late G2 cells are more condensedthan those of early G2 (Sperling & Rao, 1974).

In most cells the synthesis of nuclear proteins occurs primarily, if not exclusively,in the cytoplasm, and it has been suggested that some special mechanism may beinvolved in transporting these proteins from the cytoplasmic site of synthesis to theirultimate location in the nucleus (Kuehl, 1974; Kuehl, Barton & Dixon, 1980). Formitotic factors, however, no such special mechanism need be invoked. It has alreadybeen demonstrated by the cell fusion experiments of Rao & Johnson (1974), that thesefactors have free access to the nucleus, presumably through the nuclear pores. Thepresent data confirm the earlier observation (Roa & Johnson, 1974) that mitoticfactors that have moved into the nucleus by diffusion will bind to the appropriate siteson the chromatin and thus become associated with the PCC.

However, in the case of Xenopus oocytes, there is no evidence for direct associa-tion of mitotic factors with the Xenopus chromosomes. When cytoplasmic extracts ofprogesterone-stimulated mature oocytes are injected into immature oocytes theyinduced maturation (Drury & Schorderet-Slatkine, 1975; Wasserman & Masui, 1976)even in the absence of protein synthesis. These extracts not only induce maturationin immature oocytes but also stimulate the synthesis of more maturation-promotingfactor. If MPF of this oocyte is injected into a third oocyte this factor is furtheramplified, suggesting an autocatalytic nature for this process (Schorderet-Slatkine,1972; Drury & Schorderet-Slatkine, 1975). Cytoplasmic extracts of oocytes stimulatedby the injection of mitotic extracts of HeLa cells also exhibit a similar autocatalyticreaction (unpublished data). These results suggest that the mitotic factors may beactivating the MPF that already exists stored in an inactive form in the cytoplasmof the oocyte. It is possible that the mitotic factors that stimulate meiotic maturationin the oocyte may be different from those involved in premature chromosome con-densation. This dilemma can be resolved only when premature chromosome condens-ation can be induced by introducing mitotic factors into mammalian cells. Our attemptsso far have been negative. Until we develop an alternative mammalian cell systemfor bioassay, the oocyte system, in spite of its indirect nature, is the only one availablefor further investigation of the nature of the mitotic factors of mammalian cells.

Since the phosphatase inhibitors (NaF, ATP and sodium /?-glycerol phosphate)seem to prevent the loss of activity of the mitotic factors, a role for phosphorylation-dephosphorylation reactions may be postulated. Interestingly, Wu & Gerhart (1980)have shown the possible involvement of protein phosphorylation in the activationand inactivation of maturation-promoting factor from X. laevis eggs. Whether themitotic factors from mammalian cells are phosphoproteins still remains to beelucidated

We thank Dr K. L. Satya-Prakash for his assistance in making the photomicrographs. Thisinvestigation was supported in part by research grants CA-11520, CA-27544, CA-23878 andCA-28153 from the National Cancer Institute; GM-23965 from the Institute for GeneralMedical Sciences, DHHS; and grant CH-152 from the American Cancer Society.

Localization of mitotic factors 205

REFERENCES

ADOLPH, K. W. (1980). Isolation and structural organization of human mitotic chromosomes.Chromoioma 76, 23-33.

BLUMENIHAL, A. B.( DIEDEN, J. D., KAPP, L. N. & SEDAT, T. W. (1979). Rapid isolation ofmetaphase chromosomes containing high molecular weight DNA. J. CellBiol 81, 255-265.

BOOTSMA, D., BUDKE, L. & Vos, O. (1964). Studies on synchronous division of tissue culturecells initiated by excess thymidine. Expl Cell Res. 33, 301-309.

BRADFORD, M. (1976). A rapid and sensitive method for the quantitation of microgram quan-tities of protein utilizing the principle of protein-dye binding. Analyt. Biochem. 72, 248-254.

DRUKY, H. F. (1941). Amyl acetate as a clearing agent for embryonic material. Stain Technol.16, 21-22.

DRURY, K. C. & SCHORDERET-SLATKINE, S. (1975). Effects of cycloheximide on the 'auto-catalytic ' nature of the maturation promoting factor (MPF) in oocytes of Xenopus laevis.Cell 4, 269-274.

GOODWIN, G. H., SANDERS, C. & JOHNS, E. W. (1973). A new group of chromatin-associatedproteins with a high content of acidic and basic amino acids. Eur. J. Biochem. 38, 14-19.

JOHNSON, R. T. & RAO, P. N. (1970). Mammalian cell fusion: induction of premature chromo-some condensation in interphase nuclei. Nature, Lond. 236, 717-722.

KUEHL, L. (1974). Nuclear protein synthesis. In The Cell Nucleus, vol. 3 (ed. H. Busch.),PP- 345~375- New York: Academic Press.

KUEHL, L., BARTON, D. J. & DIXON, G. H. (1980). Binding of the high mobility group protein,H6, to trout testis chromatin. J. biol. Chem. 255, 10671-10675.

MASUI, Y. & MARKERT, C. L. (1971). Cytoplasmic control of nuclear behaviour during meioticmaturation of frog oocytes. J. exp. Zool. 177, 129-146.

MERRIAM, R. W. (1972). On the mechanism of action in gonadotropic stimulation of oocytematuration in Xenopus laevis. J. exp. Zool. 180, 421-426.

NELKIN, B., NICHOLS, C. & VOOELSTEIN, B. (1980). Protein factor(s) from mitotic CHO cellsinduce meiotic maturation in Xenopus laevis oocytes. FEBS Lett. 109, 233-238.

RAO, P. N. (1968). Mitotic synchrony in mammalian cells treated with nitrous oxide at highpressure. Science, N. Y. 160, 774-776.

RAO, P. N. & ENGELBERG, J. (1965). HeLa cells: effects of temperature on the life cycle.Science, N.Y. 148, 1092-1094.

RAO, P. N. & ENGELBERG, J. (1966). Effect of temperature on the mitotic cycle of normal andsynchronized mammalian cells. In Cell Synchrony - Studies in Biosynthetic Regulation (ed.I. L. Cameron & G. M. Padilla), pp. 332-352. New York: Academic Press.

RAO, P. N. & JOHNSON, R. T. (1970). Mammalian cell fusion: studies on the regulation ofDNA synthesis and mitosis. Nature, Lond. 225, 159-164.

RAO, P. N. & JOHNSON, R. T. (1974). Regulation of cell cycle in hybrid cells. In Control ofProliferation in Animal Cells, Cold Spring Harbor Conference on Cell Proliferation (ed. B.Clarkson & R. Baserga), vol. 1, pp. 785-800. New York: Cold Spring Harbor Laboratory.

RAO, P. N., SUNKARA, P. S. & WRIGHT, D. A. (1981). Chromosome condensation factors ofmammalian cells. In Genes, Chromosomes and Neoplasia, 33rd M. D. Anderson Symposia onFundamental Cancer Research (ed. F. E. Axrighi, P. N. Rao & E. Stubblefield) pp. 48-60.New York: Raven Press.

SCHORDERET-SLATKINE, S. (1972). Action of progesterone and related steroids on oocytematuration in Xenopus laevis. An in vitro study. Cell Different. 1, 179—189.

SPERLING, K. & RAO, P. N. (1974). Mammalian cell fusion. V. Replication behaviour of hetero-chromatin as observed by premature chromosome condensation. Chromosoma 45, 121-131.

SUBTELNY, S. & BRADT, C. (1963). Cytological observations on the early developmental stagesof activated Rana pipens eggs receiving a transplanted blastula nucleus. J. Morph. 112, 45-59.

SUNKARA, P. S., WRIGHT, D. A. & RAO, P. N. (1979a). Mitotic factors from mammalian cellsinduce germinal vesicle breakdown and chromosome condensation in amphibian oocytes.Proc. natn. Acad. Sci. U.S.A. 76, 2799-2802.

SUNKARA, P. S., WRIGHT, D. A. & RAO, P. N. (19796). Mitotic factors from mammalian cells:a preliminary characterization.^, supramolec. Struct. 11, 189-195.

206 R. C. Adlakha and otliers

WASSERMAN, W. J. & MASUI, Y. (1976). A cytoplasmic factor promoting oocyte maturation:its extraction and preliminary characterization. Science, N.Y. 191, 1266-1268.

WOODLAND, H. R. (1974). Changes in the polysome content of developing Xenopus laevisembryos. Devi Biol. 40, 90-101.

WRAY, V. P., ELGIN, S. C. R. & WRAY, W. (1980). Proteins of metaphase chromosomes andinterphase chromatin. Nucl. Acids Res. 8, 4155-4163.

WRAY, W. & STUBBLEFIELD, E. (1970). A new method for the rapid isolation of chromosomes,mitotic apparatus, or nuclei from mammalian fibroblasts at near neutral pH. Exptl Cell Res.59, 469-478.

Wu, M. & GERHART, J. C. (1980). Partial purification and characterization of the maturation-promoting factor from eggs of Xenopus laevis. Devi Biol. 79, 465-477.

{Received 5 October 1981)