J. Cell Set. 79, 247-257 (1985) 247Printed in Great Britain © The Company of Biologists Limited 1985

WATER ORDERING DURING THE CELL CYCLE:

NUCLEAR MAGNETIC RESONANCE STUDIES OF

THE SEA-URCHIN EGG

S. ZIMMERMAN*Division of Natural Sciences, Glendon College, York University, Toronto,Ontario M4N 3M6, Canada

A. M. ZIMMERMANDepartment of Zoology, University of Toronto, Toronto, Ontario M5S 1A1, Canada

G.D.FULLERTON

Department of Radiology, The University of Texas Health Science Center at SanAntonio, San Antonio, Texas, U.SA.

R. F. LUDUENADepartment of Biochemistry, The University of Texas Health Science Center at SanAntonio, San Antonio, Texas, U.SA.

AND I. L. CAMERONDepartment of Cellular and Structural Biology, The University of Texas Health ScienceCenter at San Antonio, San Antonio, Texas, U.SA.

SUMMARY

Nuclear magnetic resonance was used to measure spin-lattice water proton relaxation times (T\)during the first cell cycle in sea-urchin zygotes of packed Strongylocentrotus purpuratus. Followinginsemination there was a 90% increase in the T\ value. The increase in T\ at fertilization could beaccounted for by the accumulation of extracellular fluid between the egg surface and thefertilization envelope. The T\ value then remained without change during the first cell cycle, exceptat metaphase when there was a significant 13% decrease. The lowered T\ values measured atmetaphase were not related to a change in the water content of the packed cells, which remainedfairly constant throughout the cell cycle. High hydrostatic pressure, low temperature andcolchicine (agents that depolymerize mitotic apparatus microrubules) did not affect the T\ values infertilized eggs. Treatment in vitro of a microtubule protein preparation with low temperature andcolchicine resulted in an increased T\, which accompanied the depolymerization of microtubuleprotein. Since depolymerization of the microtubules associated with the mitotic apparatus by highpressure, colchicine or low temperature does not alter the T\ of water protons in the cell, it isproposed that the increased state of ordered water molecules at metaphase is maintained by non-microtubular factor(s) of the metaphase egg.

INTRODUCTION

Nuclear magnetic resonance (n.m.r.) spectroscopy is a non-invasive method thatpermits study of the physical properties of water molecules in cells, in particular theircapacity for motional freedom. Measurements of the spin-lattice relaxation time (T{)

•Author for correspondence.

Key words: cell cycle, nuclear magnetic resonance, sea-urchin eggs.

248 5. Zimmerman and others

reflect the freedom of water after perturbations by an electromagnetic pulse. LongerT\ values are associated with freedom of movement and less organized water,whereas shorter T\ values are a reflection of less freedom of movement and a moreorganized state of the water molecules. There has been extensive scientific studyof the physical properties of water in biological systems (see reviews by Ling,1962; Drost-Hansen & Clegg, 1979). Beall and coworkers (Beall, Hazlewood & Rao,1976; Beall, Chang & Hazlewood, 19786; Beall, 1980; Beall, Brinkley, Chang &Hazlewood, 1982) have investigated the state of water in HeLa, Chinese hamsterovary (CHO) and breast cancer cells. Their findings indicate changes in the motionalfreedom of water associated with the cell cycle and with the extent of assembly ofmicro tubules in normal and in cancer cells. In this study, we investigated waterordering during the cell cycle of sea-urchin eggs. On the basis of past reports weanticipated changes in T\ at the time of assembly of microtubules in the mitoticapparatus. A preliminary report of this study was published elsewhere (Zimmermanetal. 1984).

MATERIALS AND METHODS

OrganismsSea urchins, Strongylocentrvtus purpuratus, were purchased from Pacific BioMarine Lab-

oratories, Venice, California. Eggs were obtained by intracoelomic injection of 0-53M-KC1. Theshed eggs were washed three times with 250 ml artificial sea water (Instant Ocean) before use.Sperm were obtained from excised testes and stored at 4°C. Diluted sperm suspensions werefreshly prepared before use. Only those batches of eggs in which at teast 98% of the eggs showedfertilization membranes within 5 min of insemination were used for experimental purposes. Afterfertilization the zygotes were placed in sea water at 18 °C, at which temperature cells go throughcytokinesis 90-110 min after insemination. Studies by Fry (1936) and Harvey (1956) indicate thatthe time of appearance of mitotic events is fairly constant during the first cleavage cycle. Thus it ispossible to relate cleavage cycles from various batches of eggs. On the basis of previous experience,the division times for different batches of eggs were standardized at 100 min.

Nuclear magnetic resonance (n.m.r.) measurements

A 400-500 (A sample of concentrated cells was placed in an n.m.r. capillary tube and uniformlypacked by centrifugal force (272^for 3 min at room temperature). It was determined by packed cellvolume 'haematocrit' measurements that unfertilized and fertilized sea-urchin eggs attain uniformpacking volume in the capillary tube after 2 min of centrifugation (at 272 g), after which thefertilization envelopes of adjacent eggs were packed into a hexagonal pattern, indicating that little ifany space remained outside the fertilization envelope. There was no further measurable change incellular packing after longer periods of centrifugation. Pulsed proton n.m.r. relaxation measure-ments were made on separate equal samples of packed eggs at different times in the first cell cycleusing a Praxis model II instrument (San Antonio). This instrument has a 0-25 tesla permanentmagnet, sample coil and RF pulser tuned to 10-7 MHz. The instrument is interfaced with amicrocomputer for fast data acquisition and has built in data analysis software. The T\ or spin-lattice relaxation time was measured using the saturation recovery pulse sequence, 90°—T-90°. Theresultant analysis of 30 free-induction-decay peak heights with a sequence of increasing interpulsedelay times yields the Tl decay curve presented in Fig. 1. Such uni-exponential decay curvespermitted the determination of the relaxation time from the least-squares fit to all 30 data points.

Water content and cell diameter measurements

Following the n.m.r. measurements the individual samples were removed from the n.m.r. tube,weighed in pre-tared weighing pans and then dehydrated in a vacuum oven at 90 °C over a period of

n.tn.r. studies of sea-urchin eggs 249

several days until a stable weight was achieved. The difference between the initial wet weight of thesamples and the final dry weight was used to determine the percentage of water in the samples. Acalibrated ocular micrometer was used to measure the diameter of the egg proper, both before andafter fertilization. The diameter of the raised fertilization envelope after fertilization was alsomeasured. The volume of the egg proper as well as the volume of the space between the egg and theraised fertilization envelope could be calculated from the diameter measurements because of thespherical shapes.

Pressure experimentsn.m.r. capillary tubes containing 400-500/11 of packed cells or experimental solution

(microtubule protein solution) were filled with paraffin oil and covered with a soft plastic cap. Then.m.r. tubes were placed in a stainless steel pressure chamber. The pressure chamber wasconnected to a hydrostatic pressure pump (Zimmerman, 1970). Hydrostatic pressures up tol-6xl04lbf in"2 (equivalent to l - l O x ^ k N m " 2 ) were applied with an Amico pressure pump atthe rate of 5 X103 lbf in"2 (3 -45 X104 kN m~2) per stroke at a temperature of 23 CC. Decompressionwas achieved almost instantaneously by means of a needle valve. About 60 s were required from thetime of decompression to the start of the n.m.r. measurements.

TubulinMicrotubule protein was prepared from rat brain by a cycle of assembly and disassembly using

the method of Fellous, Francon, Lennon & Nunez (1977). The microtubule protein is obtainedin 100mM-2-(iV-morpholino)ethanesulphonic acid, pH6-4, 1 mM-ethyleneglycolbis(/3-amino-ethylether)-N, W-tetraacetic acid, 0-1 mM-ethylenediaminetetraacetic acid, 0-5 mM-MgClz, 1 mM-GTP and 1 mM-^-mercaptoethanol. The isolated protein contains about 65-70% tubulin andabout 20% microtubule-associated proteins. The microtubule protein is temperature-sensitive. At0°C the protein is clear and in a fluid state. At room temperature it is cloudy and gel-like(polymerized). The polymerization reaction can be reversed by lowering the temperature to 0°C.

RESULTS

Nuclear magnetic resonance spectroscopy was used to monitor the freedom ofmotion of water molecules in sea-urchin eggs, 5. purpuratus. Individual samples ofthe sea-urchin eggs were uniformly packed into n.m.r. glass sample tubes and spin-lattice relaxation times (T{) were measured in unfertilized eggs and fertilized eggs atspecific stages following insemination, i.e. syngamy, interphase, metaphase anddivision. Cells removed from n.m.r. tubes after cell packing showed a divisionschedule like that of the control cells. This observation indicates that packing of theeggs had no adverse effect on their progression through the cell cycle. Since T\ valuesrelate to mobility of water molecules, these measurements may be influenced bymacromolecular structural changes as well as changes in water content of cells.Therefore, it was necessary to determine the water content of the sea-urchin eggs atall the developmental stages studied, since changes in water content could alter T\values. Although the percentage hydration was lower in the unfertilized eggs, nostatistically significant differences were found in the water content of unfertilizedeggs compared with fertilized eggs, throughout the first cell cycle (Fig. 2B).

T I values at fertilization and during the first cell cycle

In general, the mean water proton relaxation time of unfertilized sea-urchin eggs(520 ms) was significantly shorter than that of fertilized sea-urchin eggs at each stage

250 S. Zimmerman and others

of the cell cycle measured. There was a 90% increase in mean T\ (991 ms) 15 minafter insemination (syngamy), which remained fairly constant throughout the cellcycle except at metaphase. The mean diameter of 24 unfertilized eggs proper was76±0-5/im, yielding a calculated volume of 2-30X105 jun3, which remained thesame after fertilization. The mean diameter of the raised fertilization envelope was100-8 ± 0-4/im, yielding a calculated volume of space between the egg and fer-tilization envelope of 3-O6xlOs/«n3. At metaphase, the mean T\ value was 896ms;this represents a 13% decrease from the interphase value (1024 ms). The shorter T{

value at metaphase represented a significant change from the other T\ values duringthe cell cycle. At cell division,T\ returned to the pre-metaphase value (Fig. 2A). Themore rapid water relaxation time at metaphase was not due to a change in watercontent at this mitotic stage but may relate to assembly of the mitotic apparatus. Inparticular, we thought that polymerization of tubulin into spindle microtubulesassociated with the mitotic apparatus might be responsible for the decrease in waterproton relaxation time. To test this hypothesis, sea-urchin embryos were treatedwith various spindle microtubule depolymerizing agents (hydrostatic pressure, lowtemperature and colchicine) and T\ values were measured.

Influence of hydrostatic pressure on T\ values

Fertilized eggs, at specific cell cycle stages were uniformly packed in n.m.r.capillary tubes and treated with hydrostatic pressure of 10000 lbf in"2 for 1 min at

3000

2000

1000

- r300 600

Tune (ms)900

Fig. 1. A representative T\ decay curve of sea-urchin eggs at metaphase. A longitudinalproton (Ti) relaxation decay curve obtained using a 90°-r-90° pulsed sequence on thePraxis model II pulsed n.m.r. analyser with a 0-25 tesla permanent magnet sample coiland RF pulser tuned to 10-7MHz. The amplitude intercept is 3258; Tu 722ms;correlation coefficient, 0-997.

n.m.r. studies of sea-urchin eggs 251

1100

f 90°

700

500

100

I* 90

80

- /

i

7 Unfertilized

20 40 60 80

B

100

^Unfertilized

Syngamy Metaphasei

Divisioni

20 40 60 80Time after insemination (min)

100

Fig. 2. Water proton T\ and the % of water in eggs of the sea urchin, 5. purpuratus, areshown at various stages of the cell cycle. Each measurement was made on a separatefreshly packed sample of eggs. The Tx at metaphase is statistically lower than the other Tx

values in the fertilized eggs. The water content of the fertilized and unfertilized eggs didnot change at each of the stages.

23 °C. This pressure and duration of treatment are known to depolymerize spindlemicrotubules (Zimmerman & Marsland, 1964; Zimmerman, 1970; Salmon, 19756).Following decompression, Tj values were determined. There were no statisticallysignificant differences between the Tx values for control and pressure-treated cells atsyngamy, interphase, metaphase or division (Table 1). Pressure-treated cells wereremoved from the n.m.r. tubes and observed. The division schedule of these cellswas delayed (cf. Zimmerman & Silberman, 1965).

Low-temperature effects on Ti

Cells at metaphase were placed at 0°C in an ice-water bath for 20min; the cellswere returned to room temperature following which Tx measurements were made.Low temperature is known to depolymerize spindle microtubules (Inoud, 1964).Measurements of mean T\ values for low temperature-treated cells (800 ± 32) andcontrol cells (781 ± 23) showed no statistically significant differences.

252 5. Zimmerman and others

Effects ofcolchicine on T]

Fertilized eggs at early metaphase were placed into 2-5xlO~4M-colchicine for40-54 min. This concentration of colchicine depolymerizes spindle microtubules (cf.Sauaia & Mazia, 1961; Zimmerman & Zimmerman, 1967). The mean 7^ value forthe colchicine-treated cells (863 ± 62) was not statistically different from that fornon-treated control cells (781 ± 23). The water content of colchicine-treated eggs atmetaphase was 88-00 ± 1-25%; this value was similar to that for metaphase controlcells (89-25 ±1-39%).

Effects of physical and chemical agents on Ti in vitro tubulin studies

To test further whether polymerized and depolymerized states of tubulin reflectdifferences in water proton T\ values, similar depolymerization agents were used instudies in vitro with microtubule protein samples prepared according to the methodof Fellous et al. (1977).

The polymerized state of the microtubule protein preparation (17-13 mgml"1) inthe presence of GTP and Mg2"1" is temperature-dependent. At 23 °C, microtubulepreparations appeared cloudy and gel-like (Fig. 3), indicating the polymerized formof tubulin. Glass beads placed on the surface of this material did not fall through thesolution at unit gravity. When this microtubule protein sample was put in an ice bathat 0°C, it became clear and fluid-like (depolymerized) within 5 min. Glass beadsplaced on the surface of this preparation fell rapidly through the solution. Thecorrelation between temperature and the state of polymerization of the microtubuleprotein preparation was further borne out by n.m.r. measurements of the spin-latticerelaxation time for water molecules in this material. The mean T\ value at 23 °C was

Table 1. Pulsed proton Ti values (ms ± s.E.M.) of fertilized sea-urchin eggs duringthe first cell cycle and immediately after exposure to high hydrostatic pressure (1 min

at 10000 Ibf in'2)Time afterfertilization(min)

155375

100Row mean

Stage of thecell cycle

SyngamyInterphaseMetaphaseDivision

Results of a two-way analysis of variance

n

8595

test:

Stage of cell cycleTreatment conditionInteraction

Treatment conditions

No pressure

991 ±991024 ± 35896±60

1043 ±109975

F value10-7511-9320-444

n High pressure

8 1027 ± 775 1004 ±689 933 ± 455 1089 ±104

1003

P value< 0-001*

Not significantNot significant

Columnmean

10091014914*

1066

•Results of Student-Newman-Keul's multiple-range test showed that the value at metaphasewas significantly different from all other column mean values at P< 0-0023; none of the othercolumn mean values were significantly different.

n.m.r. studies of sea-urchin eggs 253

Fig. 3. Photographs showing polymerized (A) and non-polymerized (B) microtubuleprotein preparation (1-7%). The opacity of the polymerized sample (at 23 °C) is evidentwhen compared with the non-polymerized (0°C) sample.

1316 ± 16-9ms, whereas at 0°C it was 1511 ±5-9ms; these values are statisticallydifferent (P< 0-01). The increase in relaxation time reflects the greater freedom ofmobility of water molecules that exists in the depolymerized state.

Colchicine, at concentrations of 2-5XlCT4M and 2-5xlO~6M, was added to themicrotubule protein preparation at 0 °C and the preparation was allowed to warm to23 °C. The-change in temperature from 0°C to 23 °C occurred within 5min. Themean Tx values in the presence of 2-5xlO~4M and 2-5xlO~6M-colchicine were1539 ±18-3 and 1456 ± 16-8 ms, respectively (Fig. 4). These 7\ values for thecolchicine preparations were statistically different (P<0-01) from that of controlmicrotubule protein at 23°C (1316± 16-9ms) (Fig. 4). Moreover, the T\ values forthe two colchicine-treated preparations were statistically different from one another(P<0-01). At thermal equilibrium (23 °C), 10-15 min after the preparations wereremoved from 0°C, the difference between T\ values for colchicine and controlsamples exceeded 100 ms.

Another depolymerizing treatment applied to microtubule protein preparationswas hydrostatic pressure of 160001bf in"2 for 30 min at 23 °C. n.m.r. measurementsof Tx were initiated 60 s after decompression. The mean Tj was 1348 ± 6-4 ms, whichwas not statistically different from the mean T\ of non-pressurized microtubuleprotein at 23 °C (Fig. 4). Since there was the possibility of a rapid reversal, from thedepolymerized to the polymerized state, upon decompression, glass beads wereplaced on the surface of the polymerized microtubules before compression.Observations of these experiments revealed that some of the glass beads had moveddown about 0-5 cm below the surface of the material, which may indicate a partialdepolymerization effect.

254 5. Zimmerman and others

1600 -

1200

Treatment —

Temperature 0°C

16000lbfin"2 colchitine colchicine

23°C 23°C 23°C 23°C

Microtubule protein

Fig. 4. Effect of low temperature, hydrostatic pressure and colchicine treatment on theT\ of microtubule protein preparations (Yl^XZugmT^). At 0°C the microtubulepreparation was clear and in a fluid state (depolymerized). When the tubulin preparationwas warmed to room temperature it became cloudy and changed to a gel state(polymerized). This polymerized state is reflected in a shortening of the 7\. Microtubulepreparations at 23 °C subjected to pressures of 160001bf in"2 for 30min showed nochange in T\. Tubulin preparations at 23 CC treated with colchicine at concentrations of2-SxlO~4M and 10~6M showed increases in T\ that were directly proportional to thecolchicine concentration.

DISCUSSION

What accounts for the 90 % increase in the mean water proton relaxation time uponfertilization? It seems reasonable to think that the longer T\ value for fertilized eggsmight be due to the increased fluid in the space between the egg membrane and theraised fertilization envelope, which was not removed by packing of the fertilizedeggs. Assuming a model of fast proton exchange between the egg and this space in thepacked eggs we can evaluate this suggestion using the following equation:

l / T i = / . x l / T I a + / b X l / 7 ' I b >

where l/Tj is the relaxation rate of the packed fertilized eggs, / a is the volumefraction of the space between the egg membrane and the fertilization envelope (0-57),l/^ia is the relaxation rate of the fluid in this space (assuming it is sea water, thisvalue is 0-37), / b is the volume fraction of the egg proper (0-43), l/Tib is the

n.m.r. studies of sea-urchin eggs 255

relaxation rate of the egg proper (because we could not measure this value in thefertilized egg we have assumed it is the same or similar to the value of the unfertilizedegg, 0-0019). Solving for T\ we obtain 988ms as the expected 7\ value. Thisexpected value compares with an observed measured value of 991 ms for the newlyfertilized eggs. Thus the increase in relaxation time at fertilization can be accountedfor by the increase in the fluid in the space between the egg surface and the raisedfertilization envelope.

These studies show a significant decrease in the T\ values for water molecules infertilized sea-urchin eggs during metaphase compared with other stages of the cellcycle. The lower T\ values measured at metaphase were not related to a change inwater content of the fertilized eggs, which remained fairly constant throughout thecell cycle. One possibility for the reduced Tx value seen at metaphase may be relatedto the assembly of a mitotic apparatus, whose macromolecular structure may cause agreater ordering of water in the cell. The mitotic apparatus, formed at metaphase, isan extensive structure that comprises approximately 12% of the total cell protein(Mazia & Roslansky, 1956). Thus water molecules may become increasinglyassociated with this macromolecular structure, thereby reducing their freedom ofmotion. Other possibilities for the change in T\ at metaphase also exist.

The pattern of Tj values during the cell cycle in sea-urchin eggs is differentfrom that of HeLa and CHO cells (Beall et al. 1976; Beall, Bohmfalk, Fuller &Hazlewood, 1978a; Beall, 1980). n.m.r. measurements during the cell cycle insynchronized HeLa and CHO cells showed the longest T\ times during mitosis andthe shortest Tj times in the S phase. The percentage hydration of these cells washigher in mitosis and lower in 5 phase. Nevertheless, Beall (1980) states that thedifferences in T\ in mitosis and 5 phase in HeLa cells cannot be attributed entirely tothe change in water content since T\ values increased from 5 to Gi with no change inwater content occurring at that time. Beall and coworkers (1976) relate thedifferences in T\ values to the chromatin condensation cycle of the cell and proposeda greater binding of water to diffuse chromatin than to chromatin in condensedchromosomes. The relationship of water-ordered structure to specific cell cyclestages in the sea urchin is the opposite to that in HeLa and CHO cells. In sea-urchineggs chromatin may have less involvement in water ordering since there is a largecytoplasmic to nuclear ratio during the cleavage period, as opposed to a larger nuclearto cytoplasmic ratio in HeLa cells (Beall, 1980).

Beall et al. (1982) suggested that the polymerization and depolymerization oftubulin in cells is correlated with the behaviour of cell water. They reported increasesin T\ values in human breast cancer cells with less microtubule material anddecreases in T\ in cells with more microtubule material. In addition, they found thatT\ increased in HeLa and CHO cells treated with the depolymerizing agent,colcemid. Thus we reasoned that the microtubule complex associated with themitotic apparatus, in sea-urchin eggs at metaphase, might play an important role inthe decrease in Tx observed at this cell-cycle stage. However, sea-urchin cells treatedwith depolymerizing agents such as hydrostatic pressure, low temperature andcolchicine did not show any significant changes in the motional freedom of water

256 S. Zimmerman and others

molecules. When microtubule protein preparations were treated in vitro with thesedepolymerizing agents, the Tl values increased significantly, except in the case ofhydrostatic pressure, where there was no change in T\ measurements. The absenceof a measurable change in T\ as a result of pressure treatment was unexpected, sincehydrostatic pressure is known to depolymerize microtubules in vivo (Tilney,Hiramoto & Marsland, 1966; Kennedy & Zimmerman, 1970) andm vitro (Salmon,1975a). Since pressure effects are rapidly reversible, the microtubule proteinsolution may have repolymerized before n.m.r. measurements could be made.

The in vitro experiments clearly show that microtubules can influence the orderingof water. Nevertheless, the depolymerization of microtubules by high pressure, lowtemperature and colchicine, in sea-urchin eggs, was not reflected in the T\ valuesof the water molecules in these cells. Thus the significant shortening of the waterproton T\ seen at the metaphase stage of the cell cycle is not due to decreased watercontent of the cell nor to the known increase in microtubule assembly. We proposethat the ordered state of water molecules at metaphase is maintained by othermacromolecular structures of the metaphase cell, which are not disrupted bydepolymerization of the microtubules.

This work was supported in part by grants from the USPHS (no. CA36372) to I.L.C., from theNSERC to A.M.Z., by Glendon College Research grants to S.Z., and a USPHS grant (no.CA26376) to R.F.L. This work waa carried out in the Department of Cellular and StructuralBiology, The University of Texas Health Science Center at San Antonio while S.Z. and A.M.Z.were visiting scientists. The authors thank Loree Cameron for her support and Veena Prasad forher skilled technical assistance.

REFERENCES

BEALL, P. T. (1980). Water-macromolecular interactions during the cell cycle. In Nuclear-Cytoplasmic Interactions in the Cell Cycle (ed. G. L. Whitson), pp. 223-247. New York:Academic Press.

BEALL, P. T., BOHMFALK, J., FULLER, G.& HAZLEWOOD, C. F. (1978a). Effects of temperature onwater microtubule interactions in a purified solution and in Wl-38 cells. J. Cell Biol. 79, 281a.

BEALL, P. T., BRINKLEY, B. R., CHANG, D. C. & HAZLEWOOD, C. F. (1982). Microtubule

complexes correlated with growth rate and water proton relaxation times in human breast cancercells. Cancer Res. 42, 4124-4130.

BEALL, P. T., CHANG, D. C. & HAZLEWOOD, C. F. (19786). Structural changes in chromatinduring the HeLa cell cycle: effect on water n.m.r. relaxation times. In Biomolecular Structureand Function (ed. P. Agria), pp. 233-237. New York: Academic Press, Inc.

BEALL, P. T., HAZLEWOOD, C. F. & RAO, P. N. (1976). Nuclear magnetic resonance patterns ofintracellular water as a function of HeLa cell cycle. Science 192, 904—907.

DROST-HANSEN, W. & CLEGG, J., eds (1975). Cell Associated Water. New York: Academic Press.FELLOUS, A., FRANCON, J., LENNON, A. M..& NUNEZ, J. (1977). Microtubule assembly in vitro

purification of assembly-promoting factors. Eur.jf. Biochem. 78, 167-174.FRY, H. J. (1936). Studies of the mitotic figure. V. The time schedule of mitotic changes in

developing Arbacia eggs. Biol. Bull. mar. biol. Lab., Woods Hole 70, 89-99.HARVEY, E. B. (1956). The American Arbacia and Other Sea Urchins. Princeton, New Jersey:

Princeton University Press.iNOUfi, S. (1964). Organization and function of the mitotic spindle. In Primitive Motile Systems in

Cell Biology (ed. R. D. Allen & N. K. Kamiya), pp. 549-598. New York: Academic Press.KENNEDY, J. R. & ZIMMERMAN, A. M. (1970). The effects of high hydrostatic pressure on the

microtubulea of Tetrahymena pyriformis. J. Cell Biol. 47, 568-576.

n.m.r. studies of sea-urchin eggs 257

LING, G. N. (1962). Physical Theory of the Living State. Philadelphia, Pennsylvania: Blaisdell.MAZIA, D. & ROSLANSKY, J. D. (1956). The quantitative relations between total cell proteins and

the proteins of the mitotic apparatus. Protoplasma 46, 528—534.SALMON, E. D. (1975a). Pressure-induced depolymerization of brain microtubules in vitro.

Science 189, 884-886.SALMON, E. D. (19756). Pressure-induced depolymerization of spindle microtubules. I. Changes

in birefringence and spindle length..7. Cell Biol. 65, 603-614.SAUAIA, H. & MAZIA, D. (1961). Action of colchicine on the mitotic apparatus. Path. Biol. 9 (no.

5-6).TlLNEY, L., HlRAMOTO, Y. & MARSLAND, D. (1966). Studies on the microtubules inHeliozoa. I II .

A pressure analysis of the role of these structures in the formation and maintenance of theaxopodia of Actinosphaerium nucleofilum (Barrett). J. Cell Biol. 29, 77-95.

ZIMMERMAN, A. M. (1970). High pressure studies on synthesis in marine eggs. In High PressureEffects on Cellular Processes (ed. A. M. Zimmerman), pp. 235-257. New York: AcademicPress.

ZIMMERMAN, A. M. & MARSLAND, D. (1964). Cell division: Effects of pressure on the mitoticmechanisms of marine eggs (Arbacia punctulata). Expl Cell Res. 35, 293-302.

ZIMMERMAN, A. M. & SILBERMAN, L. (1965). Cell division: The effects of hydrostatic pressure onthe cleavage schedule in Arbacia punctulata. Expl Cell Res. 38, 454—464.

ZIMMERMAN, A. M. & ZIMMERMAN, S. (1967). Action of colcemid in sea urchin eggs. J. Cell Biol.34, 483-488.

ZIMMERMAN, S., ZIMMERMAN, A. M., FULLERTON, G. D., LUDUENA, R. F. & CAMERON, I. L.

(1984). Water ordering during the cell cycle of sea urchin eggs. J. Cell Biol. 99, 446a.

(Received 21 December 1984-Accepted, in revised form, 25 June 1985)