DEVELOPMENTAL

BIOLOGY

Synthesis

148,138-146 (1991)

of Tubulin during Early Postgastrula Development of Artemia: lsotubulin Generation and Translational Regulation CARRIE M. LANGDON,'PARVANEHRAFIEE,~ANDTHOMAS H.MACRAE Depa.rtment of Biology, Dalhwsie University, Halifax, Nwua Scotia, Canada B2H J+Jl Accepted July 29, 1991

Isotubulin diversity and the synthesis of tubulin were examined during development of the brine shrimp, Atiemia. It was found, by Northern and dot-blot analyses, that Artemia possess constant amounts of one size class of mRNA each for Q- and @-tubulin during the first 24 hr of postgaslrula development. Two-dimensional gel electrophoresis and fluorography, following the in vitro translation of developmentally staged poly(A)+ mRNA, yielded one a- and one P-tubulin. Clearly, the isotubulin diversity seen on Coomassie blue-stained two-dimensional gels of Artemia tubulin is not generated by differential gene transcription during postgastrula growth, nor is development accompanied by synthesis of novel isotubulins resolvable by the methods employed. Characterization of polysomal poly(A)+ mRNA, and of proteins synthesized in viva, indicated very little tubulin was synthesized in Artemia as they developed from gastrula to first instar larvae. The results suggest control of tubulin synthesis in Artemia by a mechanism that restricts binding of the message to ribosomes. Of general significance, it appears that a complex metazoan animal is able to undergo extensive growth with limited tubulin synthesis and in the absence of differential expression of tubulin genes. Moreover, the capacity of microtubules to assume changing and/or increased functions associated with cellular development is seemingly not dependent on the synthesis of new tubulin isoforms. 8 1991 Academic Press. 1nc. INTRODIJCTION

Tubulin, a heterodimeric protein, is generally composed of cy- and /‘3-tubulins which are polypeptides of approximately 450 amino acids. Not only do o(- and @-tubulin share sequence similarity, but both polypeptides are well conserved across phylogenetic boundaries with variability restricted to specific regions of the molecules (Joshi and Cleveland, 1990; MacRae and Langdon, 1989). Tubulin will polymerize into microtuhules, cylindrical organelles 25 nm in diameter, which function in cell division and differentiation, maintenance of intracellular architecture, cell motility, and vesicle transport (Vallee and Shpetner, 1990; Cooper et al, 1990; Kelly, 1990; Matthews et al, 1990; Lai et aZ., 1988). Most eukaryotes exhibit multiple LY- and p-tubulin genes, and tubulin pseudogenes may be present. Characteristically, tubulin gene expression is spatially and temporally regulated, with mouse, chicken, and Drosophila tubulin genes particularly well examined from this perspective (Joshi and Cleveland, 1990; MacRae and Langdon, 1989). Evidence of differential tubulin gene expression led Fulton and Simpson (1976) to propose that tu-

i Current address: Rm 4H41, Department of Biochemistry, Health Sciences Center, McMaster University, Hamilton, Ontario, Canada L8N 325. * Current address: University of California Service, Veterans Administration Medical Center, 4150 Clement Street (151M2), San Francisco, CA 94121. 0012-1606/91 $3.00 Copyright All rights

0 1991 by Academic Press, Inc. of reproduction in any form reserved.

bulin isoforms are functionally specialized. The activity of a microtubule would thus be determined by its isotubulin composition. On the other hand, it has been suggested that isotubulins are functionally equivalent, and the presence of multiple tubulin genes permits transcriptional regulation of tubulin synthesis (Raff et al., 1987; Raff, 1984), presumably via upstream regulatory sequences and/or other cell-specific factors. In this scheme, regions of sequence variability do not indicate functional specificity but are regions where divergence, resulting from genetic drift, is tolerated. Regions of conserved sequence are those essential to functioning of tubulin (MacRae and Langdon, 1989; Lopata and Cleveland, 1987; Raff et al, 198’7; Raff, 1984), such as binding sites for GTP and microtubule-associated proteins and areas of cu/p tubulin interaction. The latter proposal is supported by copolymerization of divergent or chimeric tubulins with endogenous cellular tubulins and their integration into all microtubules within cells (Gu et ab, 1988; Joshi et al, 1987; Lewis et al, 1987) or by functional interchangeability of tubulin isotypes (May, 1989). Tubulin synthesis is controlled by translational regulation as well as by differential gene transcription. An autoregulatory mechanism involving a cotranslational binding of soluble tubulin to the amino terminal tetrapeptide of nascent &tubulin as it exits the ribosome has been demonstrated in cultured cells (Cleveland, 1989; Gay et ab, 1989; Yen et al., 1988a,b). Evidence for autoregulation of tuhulin synthesis has been provided hy Gong and Brandhorst (1988a,b) using sea urchin embryos, 138

LANGDON, RAFIEE, AND MACRAE

making it likely this mechanism exists in many different organisms. The brine shrimp, Artemia, is a crustacean of the class Branchiopoda. Under favorable environmental conditions Artemia young are released from the female as free-swimming nauplii, but in high salinity and low oxygen the young are shed as dormant gastrula or cysts. Inside the cyst, which has a shell impermeable to most substances other than water or gases, the gastrula is in a cryptobiotic state (Persoone et ab, 1980; Hentschel and Tata, 1976). Dormancy is terminated upon incubation in a well-aerated, aqueous solution at an appropriate salt concentration and temperature. Development is rapid, with emergence of a membrane-enclosed prenauplius from the cyst, followed by hatching or release of a freeswimming nauplius (instar I larva), all in 15-20 hr (Go et al, 1990; Rafiee et al., 1986a). Development to an instar I nauplius is achieved in the virtual absence of mitosis (Olson and Clegg, 1978; Nakanishi et al., 1962), indicating that only cell rearrangement and differentiation are required for early postgastrula development. The use of Artemia thus provides an interesting opportunity to study early development of a complex multicellular organism separately from the events of cell division. As revealed by Coomassie staining of two-dimensional gels, Artemia contain three cy-and two P-tubulins which are unchanged during development to instar I larvae (Rafiee et al., 1986b). Artemia appear to have a tubulin gene family of limited complexity, with single size classes of (Y- and P-tubulin mRNAs, each of which yields one tubulin isoform (Langdon et al., 1990). Posttranslational acetylation accounts for a portion of the a-tubulin diversity seen in Artemia (Langdon et al., 1990). This study was undertaken to further examine the generation of tubulin diversity and the synthesis of tubulin during development of Artemia. Of most significance is the demonstration that there is a constant amount of tubulin mRNA in the developing postgastrula organism and it is sparingly utilized, suggesting a translational regulation of tubulin synthesis. In addition, within the limits of the techniques employed, modification of transcription and translation did not lead to the synthesis of novel tubulins during development, nor did it increase the diversity of the tubulin population in Artemia. The organism thus appeared able to undergo complex developmental events over an extended time period in the near absence of tubulin synthesis and without production of specific tubulin isotypes. MATERIALS

Hydration

and Incubation

AND

METHODS

of Artemia

Dormant Artemia cysts (Sanders Brine Shrimp Co., Ogden, UT) were hydrated, collected by suction in a

Postgastrula Synth,esis c$ Tubdin

139

Buchner funnel, rinsed several times with cold distilled water, and either used without incubation (termed 0 hr) or 20 g (wet weight) of cysts were incubated in 1 liter of hatch medium (Warner et ah, 1979) at 28°C in the dark on a rotary shaker at 250 rpm for 15 or 24 hr. All chemicals, unless otherwise stated, were obtained from BDH Inc. (Dartmouth, N.S., Canada) or Sigma Chemical Co. (St. Louis, MO) and were of analytical quality or better. Preparation

of Cell-Free Homogenates and Tubulins

Artemia cell-free homogenates and Artemia tubulin were prepared as described by MacRae and Ludueiia (1984). Protein concentrations were determined by the method of Lowry et al. (1951). Electrophoresis

of Proteins

One-dimensional sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis, isoelectric focusing, and two-dimensional gel electrophoresis were as previously described by MacRae and Luduefia (1984). Drosophila

Tub&in

Gene Clones

Cloned Drosophila a- and P-tubulin genes were contained in plasmids pDmTa1 (Kalfayan and Wensink, 1981; Theurkauf et al, 1986) and DTB2 (Natzle and McCarthy, 1984), respectively. Their use in the analysis of Artemia DNA and RNA has been described (Langdon et ab, 1990). Dr. P. C. Wensink (Department of Biochemistry, Brandeis University, Waltham, MA) supplied pDmTa1 and Dr. J. E. Natzle (Department of Biochemistry and Biophysics, University of California, San Francisco, CA) supplied DTBB. Preparation

of Artemia

RNA

Total RNA and the poly(A)+ mRNA fraction of total RNA were prepared from Artemia developed for 0, 15, and 24 hr as described by Langdon et al. (1990). Preparation of polysomal RNA was the same as described for total RNA up to production of the 12,000g supernatant. At this stage, the 12,000g supernatant was divided equally into two centrifuge tubes, underlayed with 5 ml per tube of TEMN buffer containing 30% RNAase-free sucrose (Sigma Chemical Co.) and centrifuged at 113,000gin a Beckman SW28 swinging bucket rotor for 4 hr at 4°C. The supernatant was discarded and each pellet was resuspended in 5 ml of sterile TE buffer on ice. TE buffer contained 10 mMTris and 1 mMEDTA at pH 8.0. The contents of the two tubes were pooled, adjusted to 6 Mguanidine-HCl, incubated at 65°C for 20 min with frequent swirling, and quickly cooled to room temperature on ice. The polysomal RNA preparation was then extracted and precipitated with ethanol as for total

140

DEVELOPMENTALBIOLOGYV0~~~~148.1991

RNA preparations. Poly(A)+ polysomal mRNA was prepared by oligo(dT) cellulose chromatography (Langdon et al., 1990). Characterization

of Artemia

RNA

Preparations of Artemia RNA were electrophoresed on 1.5% agarose minigels prepared in 0.02 M 3-(N-morpholino)propanesulfonic acid (Mops) (Research Organits, Cleveland, OH), 5 mM sodium acetate, and 1 mM EDTA, pH 7.2, containing 6% formaldehyde, blotted to nitrocellulose by the northern procedure, and probed with 32P-labeled Drosophila tubulin genes (Langdon et aZ.,1990). For quantization by dot-blot analysis the RNA was prepared as for electrophoresis except that the samples were heated for 1 hr at 5O”C, then chilled on ice, and diluted 1:l with RNA gel buffer. The twofold serial dilutions were initiated with samples containing 2.5 pg of RNA which were reduced to 0.04 pg of RNA in the final dot. The membranes were air dried for 1 hr and baked at 80°C for 2-3 hr. Prehybridization and hybridization were as for the Northern blots. In Vitro Translation

of Artemia

RNA

In vitro translation of Artemia RNA, taxol-induced coassembly of radioactively labeled and unlabeled Artemia tubulin to permit analysis of tubulin synthesized in vitro, and fluorography were previously described (Langdon et al., 1990). Taxol was the kind gift of Dr. M. Suffness (NCI, Bethesda, MD). For the results shown in this paper, tubulin synthesized in vitro was purified by coassembly with nonlabeled tubulin from Artemiadeveloped 15 hr. Coassembly with tubulin purified from developmental stages corresponding to those from which the mRNA was prepared did not alter the results. Incorporation of [35S]methionine into in vitro-translated proteins was monitored by using a method suggested by the manufacturer of the rabbit reticulocyte lysate.

A. The homogenates were centrifuged as previously described for the preparation of Artemia cell-free extracts (MacRae and Ludueiia, 1984). Homogenates of dormant gastrula were prepared by incubating 100 mg of hydrated cysts in [14C]sodium bicarbonate-containing hatch medium for 5 min, followed by filtration and processing as just described. The amount of 14Cincorporated into TCA-precipitable material was determined. Labeled proteins were characterized by electrophoresis on 10% SDS-polyacrylamide gels followed by fluorography. Tubulin, induced to assemble by addition of 10 PM taxol to cell-free supernatants, was collected by centrifugation through 15% sucrose cushions and analyzed by electrophoresis and fluorography as for cell-free homogenates.

RESULTS PuriJcation

of Artemia

RNA

Poly(A)+ mRNA was prepared from total and polysoma1 RNA, at 0,15, and 24 hr of development, by chromatography on oligo(dT) cellulose. Consistent yields were obtained for all RNA preparations with the amount of RNA per gram of Artemia less in organisms developed for 24 hr than those developed for 0 or 15 hr. This variation may reflect the different physical nature of the organisms at the stages of development under comparison. There was also a smaller percentage of the total RNA present as polysomal poly(A)+ mRNA in encysted (0 hr) organisms than in those which had undergone development. Tub&n

mRNA during Artemia

Development

Five micrograms of poly(A)+ mRNA from Artemia after 0, 15, and 24 hr of development was applied to individual lanes of 1.5% agarose gels containing 6% formaldehyde. Following electrophoresis, transfer to nitrocellulose, and hybridization with 32P-labeled L?roIn Vivo Labeling of Artemia Proteins sophila tubulin gene clones contained in plasmids Artemia were labeled in vivo by placing 100 mg of pDmTa1 and DTBB, one size class of tubulin mRNA of hydrated cysts into 2 ml of hatch medium lacking unla- approximately 1.9 kb was revealed upon autoradiograbeled sodium bicarbonate and supplemented with 175 phy (Fig. 1). We had previously reported the size of the &i of [14C]bicarbonate from a 2 mCi/ml stock solution tubulin mRNA as 1.8 kb (Langdon et ah, 1990). The sigobtained from New England Nuclear, Du Pont Canada nal intensity of the a-tubulin mRNA bands, as well as Inc. (Mississauga, Ontario). The Artemia were incu- their size, was similar for the poly(A)+ mRNA from orbated at 28°C in the dark at 250 rpm for 15 or 24 hr, ganisms at each stage of development, with the same collected by filtration on 100~pm Nitex mesh (B. and being true for fl-tubulin mRNA. Stronger hybridization S. H. Thompson, Montreal, Quebec), and homogenized bands were obtained with the a-tubulin probe, pDmTa1, in 0.5 ml of Pipes buffer containing 4 M glycerol and than with the /3-tubulin probe, although both probes laproteolytic enzyme inhibitors including 5 pg of leupep- beled equally well during nick translation procedures. tin, 5 pg of soybean trypsin inhibitor, 10 pg of phenyl- Hybridization of the 32P-labeled plasmids to poly(A)RNA yielded very faint, 1.9-kb bands on northern blots, methylsulfonyl fluoride (PMSF), and 5 pg of pepstatin

LANGDON, RAFIEE, AND MACRAE

141

Synthesis of Tubulin

31231231

P’

kb

Postgastrula

23

2.4l.91.4-

0.3-

123 FIG. 1. Northern blot analysis of Atiemia poly(A)+ mRNA. Poly(A)+ mRNA from organisms developed 0 hr (lane l), 15 hr (lane 2), and 24 hr (lane 3) was electrophoresed on 1.5% denaturing agarose gels, transferred to nitrocellulose, and hybridized with 3ZP-labeled plasmids pDmTa1 (o) and DTB2 (p). Five micrograms of RNA was loaded per lane. A single band of approximately 1.9 kb was seen for both Artemia 01-and P-tubulin mRNA. Size markers in kb are shown on the left side of the autoradiogram and the origin of the lanes is indicated by 0.

demonstrating that the poly(A)- fraction contained little tubulin mRNA (not shown). To assess more accurately the quantity of tubulin mRNA, dot blots of serially diluted Artemia poly(A)+ mRNA were hybridized to 32P-labeled Droso&iZa tubulin gene clones in plasmids pDmTol1 and DTBB. Visual inspection indicated that similar signal intensities were obtained for RNA preparations from organisms at each stage of development examined, with both the a!- and the P-tubulin gene probes (Fig. 2). That Artemia contain equivalent amounts of cu-tubulin mRNA at each stage of

0

(Y

15 24

FIG. 2. Dot-blot analysis of Artemia tubulin mRNA. Poly(A)’ mRNA from organisms developed 0, 15, and 24 hr (as labeled) was serially diluted, dotted onto Gene-Screen Plus, and hybridized to 32Plabeled cloned Drosophila OI- and fl-tubulin genes contained in plasmids pDmTcv1 (a) and DTB2 (p), respectively. The first undiluted dot contained 2.5 fig of RNA.

FIG. 3. In vitro translation of Artemia mRNA. Poly(A)’ mRNA from organisms developed 0 hr (lane l), 15 hr (lane 2), and 24 hr (lane 3) was translated in vitro in rabbit reticulocyte lysate. The products were either separated immediately on a 10% SDS-polyacrylamide gel (A and B) or coassembled with unlabeled 15-hr Artemia tubulin (C and D) prior to electrophoresis. A and C, Coomassie blue-stained gel; B and D, corresponding fluorograms. Molecular weight markers X10m3 are shown on the left side of the figure. The positions of the a- and @-tubulins are indicated. TUB, tubulin.

development was confirmed by liquid scintillation counting of the hybridization spots and analysis of the autoradiograms using a Bio-Rad Model 620 video densitometer with lD-Analyst software (not shown). The same result was obtained for P-tubulin mRNA. In Vitro Translation of Artemia Poly(A)+ mRNA Artemia poly(A)+ mRNA from organisms developed 0, 15, and 24 hr was translated in a rabbit reticulocyte lysate system, with 0-hr mRNA yielding the least amount of acid-precipitable material under standardized conditions. Fluorograms of samples from in vitro translation mixtures revealed that [%]methioninelabeled polypeptides, over a molecular weight range of 17,000-80,000, were synthesized (Figs. 3A and 3B). As the tubulin synthesized in vitro could not be resolved on either one- (Fig. 3B) or two-dimensional (not shown) gels, it was purified by taxol-induced copolymerization with nonradiolabeled Artemia tubulin, followed by centrifugation through 15% sucrose cushions. Coomassiestained gels of coassembled samples revealed only aand P-tubulin (Fig. 3C), a result also seen on fluorograms, regardless of the developmental stage from which the mRNA was prepared (Fig. 3D). A small amount of degraded or prematurely terminated tubulin purified with the intact tubulin, as evidenced by the trailing seen below the P-tubulin band on the fluorograms. Tubulin svnthesized in vitro and subsequently nurified by coassembly was analyzed on two:dimensIonal gels. Location of the tubulin spots was determined by Coomassie blue staining of gels (Fig. 4A). Single (Y-and fl-tubulin spots were seen in corresponding locations on

142

DEVELOPMENTALBIOLOGY V0~~~~148,1991 PH-

kb l‘“2 9%

SDS +

3

61

23456

4.4-

2.41.9IA-

15

0

24

FIG. 4. Two-dimensional analysis of Artemia tubulin translated in vitro. Five micrograms of poly(A)+ mRNA from organisms developed 0, 15, and 24 hr (as labeled) was translated in vitro with [sr’S]methionine in rabbit reticulocyte lysate, coassembled with unlabeled 15 hr Artemia tubulin, and then analyzed in one direction by isoelectric focusing (pH) and in the other direction by SDS-polyacrylamide gel electrophoresis (SDS). A, Coomassie blue-stained gels; B, corresponding fluorograms. The arrows in B point to an unidentified translation product. Only the tubulin region of the gels is shown as there were no other spots on the fluorograms.

fluorograms of the two-dimensional Coomassie bluestained gels (Fig. 4B). Even when the exposure times used for the fluorograms were extended or reduced, only single (Y- and ,&tubulins were resolved in the in vitro translation products of poly(A)+ mRNA. Thus, as determined by the two-dimensional gel system employed, Artemia synthesize only one isotype of tubulin from each of the single size classes of ct!-and P-tubulin mRNA present in the organism. Analysis of Tub&in Development

mRNA

FIG. 5. Comparison of Artemia o(- and fl-tubulin mRNA in total and polysomal poly(A)+ mRNA. Five micrograms of total poly(A)’ mRNA was applied to lanes numbered 1, 3, and 5 while 5 Kg of polysomal poly(A)+ mRNA was applied to lanes numbered 2,4, and 6. The RNA was prepared from organisms developed 0 hr (lanes 1 and 2), 15 hr (lanes 3 and 4), and 24 hr (lanes 5 and 6). The RNA was electrophoresed on 1.5% denaturing agarose gels, transferred to nitrocellulose, and hybridized with %P-labeled pDmTol1 (A) and DTB2 (B). Size markers in kb are shown on the left side of the figure.

lent conditions, the polysomal mRNA from cysts (0 hr development) produced the least amount of acid-precipitable material while 15 hr mRNA produced the most. Many different polypeptides were synthesized in vitro using polysomal poly(A)+ mRNA from 15- and 24-hr organisms, but mRNA from cysts yielded a smaller number of polypeptides (Figs. 6A and 6B). Electrophoresis of samples after coassembly with unlabeled Artemia tubulin indicated synthesis of almost no tubulin with polyso-

Use during Artemia

To determine if tubulin mRNA is translated during the first 24 hr of postgastrula development, polysomal poly(A)+ mRNA was isolated and analyzed on northern blots in parallel with total poly(A)+ mRNA (Figs. 5A and 5B). The autoradiograms showed that the polysoma1 poly(A)+ mRNA, in comparison to the total poly(A)+ mRNA, contained only a small amount of message capable of hybridizing to the cloned Drosophila tubulin genes. This result was true for RNA preparations from organisms developed for 0, 15, or 24 hr. To reveal faint bands on blots of polysomal poly(A)+ mRNA it was necessary to overexpose the autoradiogram and even this failed to yield bands when DTBZ was used as the probe (not shown). This could be expected as DTB2 tended to hybridize to Artemia tubulin mRNA less strongly than did pDmTa1 (see results in Figs. 1 and 5). In order to determine if functional message was present and to see if a species of tubulin mRNA incapable of hybridizing to the cloned Drosophila tubulin genes existed on polysomes, the polysomal poly(A)+ mRNA was translated in rabbit reticulocyte lysate. Under equiva-

FIG. 6. 1~ vitro translation of Artemia polysomal poly(A)+ mRNA. Polysomal poly(A)+ mRNA from organisms developed 0 hr (lane l), 15 hr (lane 2), or 24 hr (lane 3) was translated in vitro in rabbit reticulocyte lysate. The translation products either were applied directly to 10% SDS-polyacrylamide gels (A and B) or were coassembled with unlabeled tubulin from Atiemia developed 15 hr and then resolved by electrophoresis (C and D). A and C, Coomassie blue-stained gels; B and D, corresponding fluorograms. Molecular weight markers are on the left side of the figure and the positions of LY-and P-tubulin are indicated. TUB, tubulin.

LANGDON, RAFIEE, AND MACRAE

ma1 poly(A)+ mRNA from 0- and 24-hr organisms and a small amount of tubulin with RNA from 15-hr organisms (Figs. 6C and 6D), even though the translation of polysomal mRNA, except for the 0-hr sample, was as good as for total mRNA. A direct comparison of the relative amounts of tubulin synthesized in vitro from total and polysomal poly(A)+ mRNA was made by parallel analysis of translation mixtures containing equal amounts of these different mRNAs. Assuming equal efficiency of coassembly in the taxol-driven assembly reactions, a much larger amount of tubulin was synthesized from the total poly(A)+ mRNA than from the polysomal fraction (not shown). The low level of tubulin synthesized in vitro from polysomal poly(A)+ mRNA supported the finding that very little tubulin mRNA was detectable by Northern blot analysis of the same RNA. Brine shrimp appear to utilize only a small amount of the total tubulin mRNA available to the organism during the first 24 hr of postgastrula development. In Vivo Labeling of Artemia

Tub&in

To examine the synthesis of Artemia tubulin, and as an adjunct to the analysis of polysomal mRNA, Artemia were labeled in vivo by incubation in hatch medium containing [14C]sodium bicarbonate in place of unlabeled sodium bicarbonate. Cell-free homogenates were prepared and the amount of ‘*C in acid-precipitable material therein was determined. Homogenates from organisms developed 24 hr exhibited the greatest incorporation of 14Cinto acid-precipitable material with relatively little background as shown for homogenates prepared from cysts (0 hr). Twenty-five micrograms of protein from Artemia developed for 0,15, and 24 hr in the presence of [14C]bicarbonate was electrophoresed on a 10% SDS-polyacrylamide gel. The resulting Coomassie blue-stained gel and its corresponding fluorogram are shown in Figs. 7A and 7B. Extraction of undeveloped cysts (0 hr), which were only exposed to [14C]bicarbonate for 5 min, yielded a homogenate with no labeled proteins. However, cell-free homogenates from Artemia exposed to [14C]bicarbonate for either 15 or 24 hr contained many different labeled proteins. To determine if tubulin was synthesized in viva, microtubule formation was induced in cell-free extracts by the addition of taxol. The microtubules were collected by centrifugation through sucrose cushions and electrophoresed. The Coomassie blue-stained gel and corresponding fluorogram are shown in Figs. 7C and 7D. No tubulin, either labeled or unlabeled, was detected in 0-hr samples. We are uncertain why microtubules failed to form from tubulin present in 0-hr supernatants (Fig. 7C, lane l), but the short exposure to [‘“Cl-

Postgastrula Synthesis of Tub&n

143

FIG. 7. ‘%-labeling of Arfemia proteins ~TLviva. Artemia were incubated in the presence of [‘%]sodium bicarbonate for 0 hr (lane l), 15 hr (lane 2), and 24 hr (lane 3) prior to preparation of cell-free homogenates. Twenty-five micrograms of total protein was run on lo’% SDSpolyacrylamide gels (A and B). Tubulin in 200 fig total protein was induced to assemble with taxol and centrifuged through 20% sucrose cushions, followed by electrophoresis (C and D). A and C, Coomassie blue-stained gels; B and D, corresponding fluorograms. Fluorograms were exposed for 18 days. Molecular weight markers are on the left side of the figure.

bicarbonate leading to the near absence of labeled polypeptides in homogenates at this stage (Fig. 7B) and the very low level of 14C-labeled acid precipitable material preclude the possibility that tubulin was synthesized at 0 hr. Fifteen- and 24-hr samples contained one labeled band of both a- and /3-tubulin. Although the bands appear strong, it was necessary to assemble tubulin in 200 mg of cell-free supernatant protein and to expose the fluorogram for 18 days in order to achieve this result. Little 14Chas been incorporated into tubulin during the first 24 hr of postgastrula development. This is more readily apparent when it is realized that many of the stronger bands in 7B occur in locations for which there is no visible Coomassie-stained band in 7A while 7C has obvious tubulin bands. It is unlikely that weak labeling of tubulin in relation to other proteins is due to differences in amino acid compositions, rather than to rates of synthesis, since the 14Cof the sodium bicarbonate would have been incorporated into a wide range of amino acids, with aspartic acid and alanine, both of which are found in tubulin, as the heaviest labeled (Clegg, 1976). Moreover, if only a limited number of different amino acids are labeled, Artemia would not be expected to yield such a large number of proteins on fluorograms for it would suggest that many proteins in the organism exhibit an unusual amino acid composition. In an attempt to elucidate their isotope composition, tubulins labeled in vivo and purified by taxol-induced polymerization were analyzed on two-dimensional gels. The small amount of tubulin present, combined with background arising due to the long exposure time needed to produce visible LY- and P-tubulin spots on fluorograms, made it impossible to determine

144

DEVELOPMENTALBIOMGY V0~~~~148,1991

1987; Lewis et al, 1987; Bond et aZ.,1986). Furthermore, coassembly selects for complete or nearly complete tubulins and provides, in concert with comigration on gels, another means to identify tubulins synthesized in vitro. DISCUSSION When separated in two dimensions, one spot for each of Previous work has indicated that Artemia contain a CY-and P-tubulin was obtained at all stages of developtubulin gene family of limited complexity (Langdon et ment examined. Single (Y- and P-tubulin spots were al,, 1990), a finding noteworthy in light of the number of shown by Grosfeld and Littauer (1976) when Acrtemia tubulin genes in most other metazoan organisms (Joshi proteins synthesized in vitro were resolved on two-diand Cleveland, 1990; MacRae and Langdon, 1989; Sulli- mensional gels. This earlier work was based on analysis of unfractionated translation products using mRNA van, 1988). In spite of the restricted number of tubulin genes in Artemia, the organism possesses at least five from only one stage of development, and the identificaisotubulins as determined by Coomassie blue staining of tion of electrophoretic spots as tubulin was based solely two-dimensional gels, and these do not appear to change on gel migration patterns. The possibility that the sinduring development (Rafiee et al., 198613).Since these gle CY-and P-tubulin spots we observed are composed of observations had interesting implications regarding the more than one electrophoretically identical tubulin arisutilization of tubulin during Artemia development, in ing from two or more a- and/or P-tubulin mRNAs of the addition to the generation of isotubulin diversity and same size, as is the case for Physarum (Birkett et al., the regulation of tubulin synthesis, we undertook the 1985), cannot be ruled out. The simplest interpretation of these data, however, is that single tubulins arise from studies described in this and the accompanying paper. Northern blot analysis of total Artemia poly(A)+ translation of single mRNAs, and this situation does not mRNA, which is composed of the free and ribosome- change during the first 24 hr of postgastrula developbound messages, revealed a single size class mRNA of ment. In order to determine if the tubulin message was be1.9 kb for both CY-and P-tubulin. The size and abundance of the tubulin mRNA were constant for RNA prepared ing utilized, polysomal poly(A)+ mRNA was isolated from organisms developed 0,15, and 24 hr, with the tu- from Artemia after 0,15, and 24 hr of postgastrula develbulin mRNA found almost exclusively in the poly(A)+ vs opment. The mRNA was then analyzed on Northern blots and by translation in vitro. Dormant Artemia gasthe poly(A)) fractions. The size of the Artemia tubulin mRNA is within the range of sizes for tubulin mRNAs trula contain, in relation to developing organisms, a refrom other organisms and is sufficiently long to encode duced amount of polysomes, presumably left over from those present in the embryo prior to desiccation or in a complete IZ- or /3-tubulin. Although the Northern blot analysis indicated the place to allow rapid resumption of protein synthesis presence of only one CY-and P-tubulin mRNA, it was once development resumes (Grosfeld and Littauer, 1975; possible that multiple messages of the same size existed Clegg and Golub, 1969). Accordingly, the percentage of for both 01-and P-tubulin and that these messages var- total RNA present as polysomal poly(A)+ mRNA was ied during development. To test this possibility, total less than half of that present at the other stages and poly(A)+ mRNA from 0-, 15-, and 24-hr organisms was this RNA was much less actively translated. Of most translated in vitro in a rabbit reticulocyte lysate system. interest, however, was that a very small portion of the tubulin mRNA was incorporated into polysomes, even In accordance with findings from other laboratories (Slegers and Aerden, 1989; James and Tata, 1980), after the organisms had developed for 24 hr. This result poly(A)+ mRNA from dormant gastrula (0 hr) trans- was observed by Northern blotting of mRNA, in vitro lated less well than mRNA from organisms developed 15 translation, and analysis of proteins synthesized in vivo. or 24 hr. Generally, however, Artemia poly(A)+ mRNA A somewhat larger amount of tubulin appears to be translated well in vitro, yielding a large number of dif- translated in vitro from 15-hr polysomal poly(A)+ mRNA than from 0- or 24-hr preparations. It is possible ferent polypeptides. that more tubulin mRNA is present on polysomes from To resolve the Artemia tubulins synthesized in vitro, they were purified by taxol-induced coassembly with organisms developed 15 hr, but better translation of 15Artemia tubulin and centrifugation through sucrose hr poly(A)+ mRNA may account for the heavier tubulin bands. cushions, followed by two-dimensional gel electrophoreThe combined evidence indicates that Artemia consis. Coassembly as a method to purify tubulin synthetain two pools of tubulin mRNA during the first 24 hr of sized in vitro was used by Lai et al. (1979). It is unlikely that use of this procedure excludes isoforms as diver- postgastrula development, a small translated polysomal gent tubulins are known to copolymerize (Baker et al., pool and a larger, nontranslated cytoplasmic pool. Why 1990; Prescott et al., 1989; Gu et al., 1988; Hussey et al., Artemia maintain a surplus of untranslated tubulin accurately the number of isotubulins

vivo.

synthesized in

LANGDON,RAFIEE, AND MACRAE

mRNA and how its utilization is regulated are unclear. A similar situation exists in mouse, where myoblast fusion during muscle differentiation is accompanied by an increase in the message for Mn4, a divergent a-tubulin (Lewis and Cowan, 1988), but only low levels of the Mo14 tubulin are found. Lewis and Cowan proposed that Ma4 mRNA is subject to translational or post-translational regulation. Such a mechanism could involve binding of tubulin to its mRNA, as occurs when Escherichia coli ribosomal proteins bind specifically to their mRNAs and prevent translation (Dean and Nomura, 1980). Evidence for autoregulatory control of tubulin synthesis has been shown previously (Yen et al., 1988a,b; Gong and Brandhorst, 1988a,b) although there was no indication that the tubulin message was prevented from binding to ribosomes. Clearly, differential gene expression does not account for the diversity of Artemia tubulin isoforms seen on Coomassie blue-stained two-dimensional gels, nor, as a corollary, is there any indication of differential tubulin gene transcription during Artemia development. The data support the conclusion that metazoan animals do not require a complex array of isotubulins for normal cell function. Moreover, the organism undergoes extensive postgastrula growth seemingly with relatively little need for tubulin synthesis or for the generation of new isotubulins to accommodate the anticipated increase and/or change in cell function during development. This situation pertains even in the presence of mRNA sufficient to permit more tubulin synthesis than is apparent from in vitro translation of polysomal poly(A)+ mRNA and in &JO labeling of proteins. It will be extremely interesting to determine the mechanism that regulates tubulin mRNA expression in Artemia and to see if it occurs in other organisms or for other cytoskeletal proteins. We thank Dr. P. C. Wensink and Dr. J. C. Natzle for the generous gifts of cloned Drosophila tubulin genes. Taxol was a gift from Dr. M. Suffness. The work was supported by a Natural Sciences and Engineering Research Council of Canada operating grant to T.H.M. REFERENCES BAKER, H. N., ROTHWELL,S. W., GRASSER,W. A., WALLIS, K. T., and MURPHY,D. B. (1990). Copolymerization of two distinct tubulin isotypes during microtubule assembly in dro. J. Cell Biol. 110,97-104. BIRKETT, C. R., FOSTER,K. E., JOHNSON,L., and GULL, K. (1985). Use of monoclonal antibodies to analyse the expression of a multi-tubulin family. FEBS Letf. 187, 211-218. BOND,J. F., FRIDOVICH-KEIL, J. L., PILLUS, L., MULLIGAN, R. C., and SOLOMON,F. (1986). A chicken-yeast chimeric P-tubulin protein is incorporated into mouse microtubules in viva. Cell 44,461-468. CLEGG,J. S. (1976). Interrelationships between water and cellular metabolism in Artemiu cysts. V. ‘%O, incorporation. J. Cell. PhysioL 89,369-380.

CLEGG,J. S., and GOLUB,A. L. (1969). Protein synthesis in Artemia

Postgustrula

Synthesis

of Tubalin

145

salina embryos. II. Resumption of RNA and protein synthesis upon cessation of dormancy in the encysted gastrula. Dev. Biol. 19,178 200. CLEVELAND,D. W. (1989). Use of DNA transfection to analyze gene regulation and function in animal cells: Dissection of the tubulin multigene family. Cell Motil. CytoskeL 14, 147-155. COOPER,M. S., CORNELL-BELL,A. H., CHERNJAVSKY,A., DANI, J. W., and SMITH, S. J. (1990). Tubulovesicular processes emerge from trans-Golgi cisternae, extend along microtubules, and interlink adjacent trans-Golgi elements into a reticulum. Cell 61, 135-145. DEAN, D., and NOMURA,M. (1980). Feedback regulation of ribosomal protein gene expression in E. coli. Proc. Natl. Acad. Sci., USA 77, 3590-3594. FULTON,C., and SIMPSON,P. A. (1976). Selective synthesis and utilization of flagella tubulin: The multi-tubulin hypothesis. In “Cell Motility” (R. Goldman, T. Polland, and J. Rosenbaum, Eds.), pp. 987-1005. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. GAY, D. A., SISODIA,S. S., and CLEVELAND,D. W. (1989). Autoregulatory control of /3-tubulin mRNA stability is linked to translation elongation. Proc. Natl. Acad. Sci. USA 86, 5763-5767. GO,E. C., PANDEY,A. S., and MACRAE, T. H. (1990). Effect of inorganic mercury on the emergence and hatching of the brine shrimp Artemia franciscana. Mar. Biol. 107, 93-102. GONG, Z., and BRANDHORST,B. P. (1988a). Stabilization of tubulin mRNA by inhibition of protein synthesis in sea urchin embryos. Mol. Cell. BioL 8, 3518-3525. GONG,Z., and BRANDHORST,B. (1988b). Autogenous regulation of tubulin synthesis via RNA stability during sea urchin embryogenesis. Development 102,31-43. GROSFELD,H., and LITTAUER, U. Z. (1976). The translation in vitro of mRNA from developing cysts of Artemia salina. Ewr. J Biochem. 70, 589-599.

GROSFELD,H., and LI~AUER, U. Z. (1975). Cryptic form of mRNA in dormant Atiemiu salina cysts. Biochem. Biophys. Res. Commw~. 67, 176-181. Gu, W., LEWIS, S. A., and COWAN,N. J. (1988). Generation of antisera that discriminate among mammalian n-tubulins: Introduction of specialized isotopes into cultured cells results in their coassembly without disruption of normal microtubule function. J. Cell BioZ. 106, 2011-2022. HENTSCHEL,C. C., and TATA, J. R. (1976). The molecular embryology of the brine shrimp. Trends Biochem. Sci. 1,97-100. HUSSEY,P. J., TRAAS, J. A., GULL, K., and LLOYD, C. W. (1987). Isolation of cytoskeletons from synchronized plant cells: The interphase microtubule array utilizes multiple tubulin isotopes. J. Cell Sci. 88, 225-230. JAMES, T. C., and TATA, J. R. (1980). Messenger RNA during early embryogenesis in Artemia salinn: Altered translatability and sequence complexity. Diferentiatiw 16, 9-21. JOSHI,H. C., YEN, T. J., and CLEVELAND,D. W. (1987).In ~ivocoassembly of a divergent B-tubulin subunit (C/j’s) into microtubules of different function. J. Cell BioZ. 105, 2179-2190. JOSHI, H. C., and CLEVELAND,D. W. (1990). Diversity among tubulin subunits: Toward what functional end? Cell Motil. Cytoskel. 16,159163. KALFAYAN, L., and WENSINK,P. C. (1981). oc-Tubulin genes of Drosophila. Cell 24, 97-106. KELLY, R. B. (1990). Microtubules, membrane traffic, and cell organization. CeZZ61, 5-7. LAI, E. Y., WALSH, C., WARDELL, D., and FULTON, C. (1979). Programmed appearance of translatable Aagellar tubulin mRNA during cell differentiation in Naegleria. Cell 17,867-878. LAI, E. Y., REMILLARD, S. P., and FULTON, C. (1988). The n-tubulin

146

DEVELOPMENTALBIOLOGY

gene family expressed during cell differentiation in Nuegleria

pu-

beri. J. Cell Biol. 106, 2035-2046.

LANGDON,C. M., BAGSHAW,J. C., and MACRAE, T. H. (1990). Tubulin isoforms in the brine shrimp, Artemia: Primary gene products and their posttranslational modification. Eur. .I Cell Biol. 52, 17-26. LEWIS, S. A., and COWAN,N. J. (1988). Complex regulation and functional versatility of mammalian a- and /I-tubulin isotopes during the differentiation of testis and muscle cells. J. Cell Biol. 106, 20232033. LEWIS, S. A., Gu, W., and COWAN,N. J. (1987). Free intermingling of mammalian P-tubulin isotopes among functionally distinct microtubules. Cell 49, 539-549. LOPATA,M. A., and CLEVELAND,D. W. (1987). 1~ nivo microtubules are copolymers of available fl-tubulin isotopes: Localization of each of six vertebrate P-tubulin isotopes using polyclonal antibodies elicited by synthetic peptide antigens. J. Cell Biol. 105, 1707-1720. LOWRY,0. H., ROSENBROUGH, N. J., FARR, A. L., and RANDALL, R. J. (1951). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193,265-275. MACRAE, T. H., and LUDUE~~A,R. F. (1984). Developmental and comparative aspects of brine shrimp tubulin. Biochern. J. 219, 137-148. MACRAE, T. H., and LANGDON,C. M. (1989). Tubulin synthesis, structure, and function: What are the relationships? B&hem. Cell Biol. 67,770-790. MA’ITHEWS,K. A., MILLER, D. F. B., and KAUFMAN, T. C. (1990). Functional implications of the unusual spatial distribution of a minor a-tubulin isotope in Drosophila: A common thread among chordotonal ligaments, developing muscle, and testis cyst cells. Dev. Biol. 137,171-183. MAY, G. S. (1989). The highly divergent fl-tubulins of Aspergillus nidw Za)zsare functionally interchangeable. J. Cell Biol. 109, 2267-2274. NAKANISHI, Y. H., IWASAKI, T., OKIGAKA, T., and KATO, H. (1962). Cytological studies of Artemia sa2ina. I. Embryonic development without cell multiplication after the blastula stage in encysted dry eggs. Annot. 2001. Jpn. 35, 223-228.

NATZLE,J. E., and MCCARTHY,B. J. (1984). Regulation of Drosophila r~and P-tubulin genes during development. Dea Biol. 104,187-198. OLSON,C. S., and CLEGG,J. S. (1978). Cell division during the development of Artemia salina. Wilhelm Rwxk Arch. Dev. Biol. 184, l-14. PERSOONE,G., SORGELOOS, P., ROELS,O., and JASPERS,E. (1980). “The Brine Shrimp Artemia.” Vols. l-3. Universa Press, Wetteren, Belgium. PRESCOTT,A. R., FOSTER,K. E., WARN, R. M., and GULL, K. (1989). Incorporation of tubulin from an evolutionary diverse source, Phy-

VOLUME1481991 sarum

polycephalum,

into the microtubules of a mammalian cell. J.

Cell Sci. 92, 595-605.

RAFF, E. C., DIAZ, H. B., HOYLE, H. D., HUTCHENS,J. A., KIMBLE, M., RAFF, R. A., RUDOLPH,J. A., and SUBLER,M. A. (1987). The origins of multiple gene families: Are there both functional and regulatory constraints? A “Development as an Evolutionary Process” (R. A. Raff and E. C. Raff, Eds.), pp. 203-238. A. R. Liss Inc., New York. RaFF, E. C. (1984). Genetics of microtubule systems. J. Cell Biol. 99, l-10. RAFIEE, P., MATTHEWS,C. O., BAGSHAW,J. C., and MACRAE, T. H. (1986a). Reversible arrest of Artemia development by cadmium. Can. J. Zool. 64, 1633-1641. RAFIEE, P., MACKINLAY, S. A., and MACRAE, T. H. (1986b). Taxol-induced assembly and characterization of microtubule proteins from developing brine shrimp (Artemia). B&hem. Cell Biol. 64,238-249. SAMBROOK,J., FRITSCH, E. F., and MANIATIS, T. (1989). “Molecular Cloning A Laboratory Manual,” 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. SLEGERS,H., and AERDEN,M. (1989). Messenger ribonucleoproteins of cryptobiotic gastrulae of Artemia: Mechanisms of activation and repression of non-polysomal messenger ribonucleoproteins. In “Cell and Molecular Biology of Artemia Development” (A. H. Warner, T. H. MacRae, and J. C. Bagshaw, Eds.), pp. 259-266. Plenum, New York. SULLIVAN, K. F. (1988). Structure and utilization of tubulin isotypes. Ann. Rev. Cell Biol. 4, 687-716. THEURKAUF,W. E., BAUM, H., Bo, J., and WENSINK,P. C. (1986). Tismelanrr sue-specific and constitutive a-tubulin genes of Drosophila ouster code for structurally distinct proteins. Proc. N&l. Acad. Sci. USA 83, 8477-8481. VALLEE, R. B., and SHPETNER,H. S. (1990). Motor proteins of cytoplasmic microtubules. Annu. Rev. Biochem. 59,909-932. WARNER,A. H., MACRAE, T. H., and WAHBA, A. J. (1979). The use of Artemia saliva for developmental studies: Preparation of embryos, tRNA, ribosomes and initiation factor 2. In “Methods in Enzymology” (K. Moldaue and L. Grossman, Eds.), Vol. 60, pp. 298-311. Academic Press, San Diego. YEN, T. J., GAY, D. A., PACHTER,J. S., and CLEVELAND,D. W. (1988a). Autoregulated changes in stability of polyribosome-bound fl-tubulin mRNAs are specified by the first 13 translated amino acids. Mol. Cell. Bid. 8, 1224-1235. YEN, T. J., MACHLIN, P. S., and CLEVELAND,D. W. (1988b). Autoregulated instability of fl-tubulin mRNAs by recognition of the nascent amino terminus of p-tubulin. Nature (Lm~don) 334,580-585.