DEVELOPMENTAL

BIOLOGY

150,363-3’71 (1992)

An Octamer Element Is Required for the Expression of the Alpha H2B Histone Gene during the Early Development of the Sea Urchin JEFFREYBELL,BHARAT

R. CHAR,ANDROBMAXSON~

Department of Biochemistry and Kenneth R Norris Hospital and Institute, University of Southern School of Medicine, 14.41Eastlake Avenue, Los Angeles, California 90033

California

Accepted January 14, 1992 Early (alpha) histone genes are one of several histone gene families in the sea urchin genome. They are expressed at high levels in blastula-stage embryos and are inactivated by the early gastrula stage. By microinjecting mutant early H2B genes into sea urchin zygotes and monitoring their transcriptional activity in blastula- and gastrula-stage embryos, we sought to identify the c®ulatory elements responsible for this dramatic change in early H2B gene activity. We found that deletion of DNA 5’ of -71 and 3’ of +591 did not affect the timing or magnitude of early H2B gene expression. Neither was early H2B gene expression affected by the replacement of sequences downstream of -36 with the corresponding region of the Ll late H2B gene, expressed after the peak transcription of the early H2B gene. Further deletion of early H2B promoter sequences from -71 to -56, removing a conserved octamer element, resulted in nearcomplete inactivation of the early H2B gene in both blastula- and gastrula-stage embryos. Also inactivating early H2B gene expression were an internal deletion of the octamer element and a base substitution mutation that altered its sequence. This base substitution mutation also caused a parallel reduction in the ability of the octamer element to bind a factor present in nuclear extracts of sea urchin blastulae. These data strongly suggest that the proper expression of the early H2B gene in cleavage- and blastula-stage embryos depends on the octamer element and a factor with which it interaCtS.

0 1992 Academic

Press, Inc.

Lai et al., 1988; McMahon et al., 1985; Vitelli et al, 1988). Vitelli et al. (1988) showed that when the entire early histone repeat unit of Psammechinus miliaris is injected into sea urchin zygotes, the H2A gene is expressed with a normal temporal profile. We demonstrated that 572 bp of 5’ flanking sequence and 465 bp of 3’ flanking sequence are sufficient for faithful expression of the early H2B gene of S. purpuratus (Colin et ab, 1988). DiLiberto et al. (1989) found that transcription of an S.prpuratus early H3 gene depends on several DNA elements in the 5’ promoter region, including a consensus CCAAT site. To more fully understand the mechanism that controls the timing of early histone gene expression during embryogenesis, and to determine whether distinct or common mechanisms regulate different early histone genes, we have analyzed the early H2B promoter of S. purpuratus. Using microinjection of histone genes into zygotes of the sea urchin Lytechinus p&us as an assay for &s-acting regulatory elements, we show that normal expression of the early H2B gene requires only 71 bp of 5 flanking sequence. This sequence contains a consensus octamer element, known to be involved in the expression and cell-cycle regulation of mammalian H2B histone genes among others (LaBella et al, 1988). Mutations in the early H2B octamer element greatly reduce expres-

INTRODUCTION

The genetic programs underlying metazoan development must be controlled in part by sequential changes in the transcription of specific genes. Our aim is to understand such transcriptional timing mechanisms. As a model, we use the “early” histone gene family of the sea urchin, Strong~locentrotus prpuratus. Organized in tandem arrays and repeated several hundredfold in the genome, early histone genes are one of several differentially regulated sea urchin histone gene families (reviewed in Hentschel and Birnstiel, 1981; Kedes, 1979; Maxson et al, 1983). They are active only during oogenesis and early embryogenesis. Their transcription reaches a sharp maximum in the rapidly cleaving, early blastula-stage embryo, coincident with the peak demand for histone synthesis (Maxson and Wilt, 1981, 1982; Weinberg et al., 1983). Although studied for two decades, the molecular mechanisms that bring about this dramatic rise and fall of early histone gene transcription are still not well understood. We and others have begun to use gene transfer technology to examine such mechanisms (Colin, 1986; ‘To whom correspondence dressed.

and reprint

requests should be ad363

001%1606/92 $3.00 Copyright All rights

0 1992 by Academic Press, Inc. of reproduction in any form reserved.

DEVELOPMENTALBIOLOGY V0~~~~150.1992

364

sion of the early H2B gene, as well as the ability of the octamer element to bind a protein present in sea urchin blastula nuclear extracts. MATERIALS

AND

METHODS

Preparation of Mutants of the S. purpuratus Early H2B Gene The 5’ deletion mutants of the early H2B gene designated A-78 (Fig. 1) A-59, and A-29 (Fig. 3A) were made by introducing Hind111 restriction sites into the A-500 subclone of pCO2 (Fig. 1; Overton and Weinberg, 1978) by the method of Norris et al. (1983). An internal deletion removing the octamer element (-84 to -59) of the early H2B gene (Fig. 3A) was prepared by introducing a Hind111 site at position -84 and using standard cloning techniques to join the DNA 5’of the Hind111 mutation at position -84 with the DNA 3’ of the Hind111 mutation at position -59. To generate A-100, A-71, and the “act mut” deletion/base substitution mutant (Fig. 3A), we used the polymerase chain reaction (PCR). The 5’ amplimers for the A-100 and A-71 mutants were oligonucleotides whose 5’ termini were at positions -100 and -71, respectively. These oligonucleotides, whose sequences are given below, were used in PCRreactions with a common 3’ amplimer spanning positions +572 to +591 (3’ of the early H2B gene). The “act mut” construct was prepared by using a mutant 5’ amplimer spanning the octamer element (Fig. 3A) together with the +572 to +5913’ amplimer. The net result was a 5’ deletion to position -80 and a replacement of all eight nucleotides composing the octamer element. The chimeric early-late H2B gene (E-L mutant, Fig. 4) was created by replacing the DNA sequence 3’ of the BamHI site at position -41 of the early H2B subclone A-500 (Fig. 1) with the corresponding DNA sequence 3’ of the BamHI site at -42 of the Ll late H2B subclone pSpLlH2B 5’ A-491,3’ A + 535 (45). A-100 GATCCTAAGACCAATGAAAG A-71 GGCTCATTTGCATACGG A-59 GGACCGCAGCATACGG Ott Mut CGAGACCGAGGAGACGGGTACGCAGGACCGCAGCATACGG +572- to +591 AAGCTTGATTATATCTTAAT

Microinjection of Histone Genes into L. pi&us Zygotes and Detection of H2B Transcripts and Genes in Injected Embryos Early and late H2B histone genes, at a concentration of 50-100 mg/ml, were injected into L. J&&.-S one-cell zygotes as described in Colin et al. (1988) and Zhao et al. (1990,199l). Their cognate mRNAs were detected by an RNase protection assay (Colin et al, 1988; Zhao et aL, 1990,199l). H2B DNA in injected embryos was assayed

by slot blot (Flytzanis et ab, 1987; Zhao et aL, 1990). The details of these procedures are given in the legend to Fig. 1.

Gel Retardation and Methyl&on Interference Footprint Assays Nuclear protein was extracted from S. purpuratus blastulae as described by Calzone et al. (1988). A 58-bp BamHI-DdeI (-99 to -41) fragment containing the octamer-binding site was purified by gel electrophoresis and used as a probe in gel retardation assays as well as in the methylation interference footprint. When used in gel mobility shift assays, the fragment was radiolabeled at its 3’ ends by using T4 DNA polymerase and [%P]dATP. In Figs. 5A and 5C probes were generated by PCR. For wild-type probe, a 5’ amplimer spanning -71 to -55 (5’-GGCTCATTTGCATACGG-3’) was used with the common 3’ amplimer that spanned +572 to i-591. In the case of the octamer mutant probe, the “act mut” amplimer was used in conjunction with the common 3’ amplimer. The PCR fragments were radiolabeled at their 5’ termini by using polynucleotide kinase and pP]ATP and digested with BumHI, and the appropriate fragments were purified by polyacrylamide gel electrophoresis. Gel retardation assays were performed essentially as described in Zhao et al. (1990) and Fried and Crothers (1981) except that nuclear extracts were preincubated for 10 min with 5 pg of sheared salmon sperm DNA, in addition to the other components of the binding reaction, before the radiolabeled probe was added. Competitions were performed as described in Zhao et al. (1990) using the double-stranded oligonucleotides B-GATCCTAAGACCAATGAAAG-3’ as the CCAAT competitor and 5’-GGCTCATTTGCATACGG-3’ as the Ott competitor. Methylation interference footprint assays were performed as described in Hendrickson and Schlief (1985). Sea urchin SpOct protein was prepared by transcription/translation in vitro. A 3.5-kb cDNA bearing the entire SpOct protein-coding sequence was introduced into pBluescript. T3 RNA polymerase was used to transcribe linearized (HindIII) pBluescript-SpOct, and the resultant SpOct mRNA was capped and translated in vitro in a reticulocyte lysate following the manufacturer’s instructions (Ambion Inc.). RESULTS

As a first step toward locating the &s-acting control regions of the S. purpuratus early H2B histone gene, we carried out a deletion analysis in which we pared increasing lengths of DNA from the early H2B gene in the

BJZLL,CHAR,ANDMAXSON kb I

I

H2A Hlndlll

psp102

I

I

Hl Pstl

EC&I

,

I

WI

B I

H4

365

Sea Urchin H&one Gene Regulation

H2B Hhal

I H3

I

G

stage of dev

I M-

Endog.

RNA

4

S. p”rp.

RNA

4

S. purp.

DNA

-

A-2300

A-500

A-78 FIG. 1. Effect of gross deletions of the early histone repeat unit on the timing of expression of the early H2B gene. A 6.5-kb Hind111 fragment bearing one copy of the S purpuratus early histone repeat unit (pCO2; Overton and Weinberg, 1978) was digested with the indicated restriction endonucleases to generate the fragments shown in the left panel. These fragments were injected into approximately 60 L p&us one-cell zygotes (Materials and Methods). The zygotes were cultured until they reached the blastula (B) or gastrula (G) stage of development, when they were lysed and their nucleic acids extracted (Materials and Methods). Early H2B transcripts were detected by RNase protection analysis performed on 2/3 of the embryo lysate. The probe, a 560-nt RNA spanning the 5’ terminus of the early H2B gene (Colin et al, 1988), protects a 149-nt segment of the 5! purpuratus early H2B transcript (right panel, “S. purp. RNA”) and a 60-nt segment of the endogenous L pi&us early H2B transcript (“Endog. RNA”). Early H2B DNA was detected in the remaining l/3 of the embryo lysate by slot blot analysis using the same probe (“S. purp. DNA”). As the endogenous signal was identical in all experiments it is shown only with the pCO2 construct. A quantitative analysis of these data is presented in Table 1.

6.5kb genomic repeat unit (Cohn et aL, 1976; Overton and Weinberg, 1978). We microinjected the various deletion mutants shown in Fig. 1 into L. p&us zygotes and monitored the amount of early H2B mRNA at the blastula and gastrula stages of development by an RNase protection assay (Colin et ah, 1988; Zhao et cd, 1990). This assay distinguishes between S. purpuratus and L. pi&s early H2B transcripts. The endogenous (L, p&&s) transcripts protect several RNA probe fragments of approximately 60 nt, while the transcript of the injected S. purpuratus early H2B gene protects a fragment of 149 nt (Colin et a2l, 1988). We also measured the amount of injected early H2B DNA in blastula- and gastrula-stage embryos by slot blot analysis (Flytzanis et al, 1987). In general, the expression of injected genes varied little as a function of the amount of injected DNA in the embryo. This is probably because the injected DNA is usually at or above the level required to saturate the endogenous transcriptional machinery (J. Bell and R. Maxson, unpublished observations). That the injected early H2B DNA is saturating means that the early H2B mRNA signal, independent of the early H2B DNA, should indicate the transcriptional activity of the injected early H2B gene. It is difficult to be certain, however, that the

amount of injected DNA is saturating for each mutant gene and each batch of embryos. Because of this uncertainty, we have selected experiments in which the amounts of such DNAs are similar, enabling us to compare accurately the transcriptional activity of different early H2B DNAs. We note also that the timing of early H2B inactivation did not vary with the amount of injected DNA, even at subsaturating amounts (data not shown). Data presented in Fig. 1, and in a quantitative form in Table 1, show that when we injected the early H2B gene as a part of the entire early histone repeat unit, it was expressed at a high level in blastula-stage embryos and at a much lower level in gastrula-stage embryos (Fig. 1; Table 1, pCO2). This profile of expression closely resembles that of the endogenous early H2B gene (Fig. 1). Further deletion to EcoRI sites 3 kb 5’and 0.5 kb 3’of the early H2B gene did not change its expression in blastula- and gastrula-stage embryos (Fig. 1, pSplOZ; Table 1). Neither did successive deletion of DNA located 5’ of positions -2300, -1250, -500, and -78 have a significant effect on the timing of early H2B expression or the quantity of early H2B transcripts (Fig. 1; Table 1). These results extend our previous finding that only 572

366

DEVELOPMENTAL BIOLOGY

TABLE 1 THE EFFECT OF GROSS DELETIONS ON LEVELS OF EARLY H2B HrsTONE GENE EXPRESSION IN BLASTLJLA-STAGE AND GASTRLJLA-STAGE EMBRYOS Blastula Injected DNA

Gastrula

RNA/embryo DNA/embryo RNA/embryo DNA/embryo (mol. X 10V5) (mol. X 10-6) (mol. X 10w5) (mol. X 10T6)

pco2 psp102 A-2300 A-1250 A-500 A-78

15.4 11.4 15.8 11.9 10.8 15.7

18.6 10.1 7.64 1.80 1.95 21.2

0.60 0.51 0.80 0.37 0.48 1.98

14.1 10.8 12.4 3.49 17.1 42.4

Note. To determine the absolute numbers of early H2B mRNA and early H2B DNA molecules in injected embryos, we scanned the autoradiograms shown in Fig. 1 with a densitometer. We compared the resultant densities with those of early H2B RNA and early H2B DNA standards on the same gel or blot and calculated the quantities of early H2B mRNA and DNA represented by these densities as described by Zhao et al (1990). The amounts of early H2B mRNA and DNA are given as molecules per embryo.

bp of 5’ flanking sequence and 465 bp of 3’ flanking sequence are necessary for faithful expression of the S. purpuratus early H2B gene after microinjection into L. pictw zygotes (Colin et al, 1988). Comparisons of nucleotide sequences of early H2B genes of three sea urchin species revealed that several elements located within 80 bp 5’ of the TATA box are closely conserved (Fig. 2). The most striking of these are a CCAAT element at position -87, an octamer element at -57, and the sequence CGCAGC at -47. Both the early H2B and the Ll late H2B genes possess CCAAT and octamer elements, although the CCAAT elements are in opposite orientations. Sequence elements at -100 and -47 are unique to the early H2B genes. The behavior of the -78 deletion (Fig. 1) shows that despite the conservation of the -100 and CCAAT elements among early H2B genes of different sea urchin species, they are not essen-

CCA+T

Early S.purp

con Ll

VOLUME 150,1992

tial for the expression or regulation of the injected early H2B gene during early development. To locate the early H2B 5’ regulatory elements more exactly, and to determine whether the octamer and the -47 (“early specific”) elements are functionally important, we prepared a series of 5’ deletions whose breakpoints ranged between -100 and -29 (Fig. 3A). We tested the activity of these mutant DNAs by microinjection into L. pictus zygotes as described above. A mutant bearing a 5’ deletion to -87, like the A-78 mutant, was expressed at wild-type levels in blastula- and gastrulastage embryos (Fig. 3B; data not shown). Further deletion to position -59, removing the conserved octamer element, resulted in a dramatic reduction in the activity of the early H2B gene in blastula-stage embryos. A mutant DNA with a still greater deletion to -29 was also inactive (Fig. 3B). These data support the view that the octamer element or a closely associated sequence is important for high-level expression of the early H2B gene in blastula-stage embryos. As a further test of the role of the octamer element in the expression and regulation of the early H2B gene, we compared the transcriptional activity of a 5’ deletion to position -71,5 bp upstream of the octamer element, with that of a similar deletion mutant (to -80) bearing, in addition, a base substitution mutation in the octamer element (Fig. 3A, “act mut”). Microinjection of these mutant H2B genes revealed that the A-71 mutant was expressed at a high level in blastula-stage embryos and at a much lower level in gastrula-stage embryos (Fig. 3C). The behavior of this mutant is similar to that of a A-100 control H2B gene, shown previously to be indistinguishable from a wild-type control (A-500; data not shown). In contrast, the base substitution octamer mutant was expressed at much lower levels than either the A-71 or the A-100 mutants. However, the expression of the octamer base substitution mutant still declined between the blastula and the gastrula stages (Fig. 3C). These results show that at least some regulatory capa-

OCT

E specific

BarnHI

AG-CCAATgAAAGGaTCGA-----gaccgaGGCT--CATTTGCATACGgAcCGCAGC---atacGGAtCCGgCCCCGc-TG I IIIIII I llll II II I I II I II I I I I IIIIII I tGAttggTGAAAG---CtggtgcgctcCCGAtcCT--tATTTGCATACcgtGGcCgGtacgtTttGGATCC--CggCGttTGaTAT~

TATA

II

II

TATAAA IIIlll

FIG. 2. Evolutionary conservation of nucleotide sequences within the proximal promoter of the sea urchin early H2B gene. We compared the proximal promoter sequences of early H2B histone genes of the sea urchins S purpuratus (Sures et aZ.,1978), Stronsylocentrotus drobuchien.sis (Busslinger et d, 1982), and Psammechinus miltiti (Schaffner et al, 1978) using the Genalign program of the Intelligenetics Suite. Conserved elements are boxed. “Early con” is the consensus sequence derived from the comparison of the three early H2B sequences. The sequence of the X purpuratus Ll late H2B proximal promoter is shown below (Maxson et a& 1987).

BELL, CHAR,

-500 I

..../ _........”

sea Urchin H&one

AND MAXSON

+500

0 I

I

I

I

I

+looo I

... . .“” .“’ ,~._.....’).,,.... ,(.,..“‘~“’ ,...... “” ,._.....

---. .%.._._..,,,(,, -“‘----~- . ......_._._..,,,,,

CCAAT -100 t CTGCGACGCCTAAG

OCT

-87 CCAATG

367

Gene Regulation,

-71

E specific -59

BarnHI

TATA

-41

-29

GGATCGAGACCGAGGCT -Ott Mut

AG#+zEq+

C B

DNA ini.

S. purp.

RNA-b

S. purp. DNA+

A-87A-59A-29

DNA inj.

Stage of dev.

A-71 BG

OOtmut BG

A-l 00 BG

D DNA Inj.

S. purp.

RNA-b

S. purp. RNA-t

S. purp.

DNA-b

S. purp.

DNA+

FIG. 3. Effect of 5’, internal, and base substitution mutations in the proximal early HZB promoter on the timing and magnitude of early H2B expression. We created a series of 5’ deletions, a base substitution mutation, and an internal deletion (shown in A) in the proximal promoter of the S purpratus early H2B gene (Materials and Methods). The breakpoints of the 5’deletions are indicated by the downward arrows; the base substitution (“act mut”) and internal deletion (A-84 to -59) mutations are shown as bold lines with the mutated bases indicated. Boxed DNA sequences are consensus elements. We injected these mutant genes into L @ctus zygotes and monitored their activity in blastula-stage embryos (B and D) or in both blastula- and gastrula-stage embryos (C) by RNase protection as described in the legend to Fig. 1. A-87 (B), A-100 (C), and A-500 (D) all serve as wild-type controls (Fig. 1).

bility remained even in largely inactive H2B genes with disabled octamer elements. We next asked whether the octamer element is necessary for early HZB expression in the context of 5’70 bp of 5’ flanking sequence. We prepared a mutant H2B gene bearing an internal deletion of DNA sequences -34 to -59, but retaining upstream sequences to position -572. Figure 3D shows that this mutant H2B gene is transcribed at a much lower level than a wild-type control (A-500); therefore, sequences between the octamer element and -500 are unable to drive expression of the early H2B gene when the octamer is deleted. We have shown that the octamer element is essential for the high-level expression of the early H2B gene in blastula-stage embryos and that sequences upstream of the octamer are dispensable. Does the octamer element participate in the inactivation of the early H2B gene in gastrula-stage embryos? As a first test of this proposition, we prepared a chimeric H2B gene (-500 E-L, Fig. 4A) by fusing the 5’ promoter of the early H2B gene,

including the octamer element, to the body of the Ll late H2B gene. The fusion breakpoint was a conserved BumHI site 11 bp upstream of the TATA box in the late H2B gene and 10 bp upstream of the TATA box in the early H2B gene. This substitution of the late H2B proximal promoter with that of the early H2B results in the replacement of the two octamer sites in the late promoter (Maxson et a& 1987) with the single octamer site of the early promoter. We injected this chimeric construct into zygotes and tested its expression at the blastula and gastrula stages by RNase protection. As a control, we monitored the expression of a mutant Ll late H2B gene lacking its 3’ enhancer element (Ll H2B A-491 mutant; Fig. 4). The expression of the E-L mutant declined approximately twofold between the blastula and the gastrula stages while expression of the control Ll H2B gene increased (Fig. 4B). Since both mutants share the Ll late H2B mRNA coding sequence, this difference in timing is not due to mRNA half-life, but is caused by a difference in transcription, We note, however, that

368

DEVELOPMENTAL BIOLOGY A

-500 I

BamHI Ll HZB

0 I BamHI

I

+500 I

I

+lOOO I

BglII

B

DNA inj.

Stage of dev.

C U Ll A-491 E-L S

G

S

G

A-491 HhaI

E H2B

I

VOLUME 150,1992

BamHI

BamHI

S. purp.

RNA=

S. purp.

DNA-W

A-500

E-L H2B

FIG. 4. Temporal regulation of the chimeric H2B gene formed by fusion of early and late H2B promoters. (A) Schematic map of injected H2B genes. We fused a portion of the early H2B proximal promoter to the Ll late H2B gene at a conserved BarnHI site located 11 bp 5’ of the early H2B TATA box and 10 bp 5’ of the late H2B TATA box (Materials and Methods). A 500-bp HA&BamHI fragment of EH2B A-500 was inserted into the Ll H2B mutant A-491 in place of the BarnHI promoter fragment, thus creating the E-L H2B mutant. (B) RNase protection assays of injected H2B gene expression in embryos. We injected Ll H2B A-491 and E-L H2B into L pi&us zygotes and monitored their expression at the blastula (B) and gastrula (G) stages of development as described in the legend to Fig. 1 using a probe specific for the Ll H2B message (Colin et aa, 1988). Two protected bands are seen due to alternate start sites for transcription of the Ll H2B message. Lane C, analysis of 1 pg of whole cell RNA from gastrula-stage S. purpuratusembryos. Lane U, analysis of 1 pg of whole cell RNA from uninjected control L pi&a embryos.

transcripts of the injected E-L chimeric H2B gene declined less between the blastula and the gastrula stages (twofold) than did those of intact early H2B genes (loto 20-fold; compare Figs. 1 and 4), implying that upstream sequences, including the octamer element, are not sufficient for the full inactivation of the early H2B gene. Rather, such upstream sequences are responsible for approximately 10-20s of the decline in early H2B gene activity between the blastula and the gastrula stages. Several proteins that bind the octamer element have been characterized in mammalian cells (Sen and Baltimore, 1986; Barberis et aL, 1987; Sturm et aZ., 1987; Rosner et al, 1990; reviewed in Rosenfeld, 1991 and Ruvkun and Finney, 1991); and Calzone et al, (1988) identified an octamer-binding activity in extracts of sea urchin blastulae. If an octamer-binding protein is essential for the high level of early H2B gene expression in blastula-stage embryos, then we would expect that nuclear extracts from sea urchin blastulae should contain such an activity. Also, mutations in the early H2B octamer element should cause parallel reductions in the ability of this element to bind protein and in the transcriptional activity of the early H2B gene. We used gel shift and methylation interference assays to examine interactions between embryonic nuclear proteins and the octamer site of the early H2B gene. We incubated an extract of blastula-stage sea urchin embryonic nuclei (Calzone et aZ,, 1988) with a radiolabeled DNA probe spanning the octamer site of the early H2B gene. A gel shift assay revealed a closely grouped set of protein DNA complexes (Fig. 5A) which were greatly reduced in amount by competition with a 17-bp oligonucleotide bearing the octamer site, but not by competition with a 20-bp oligonucleotide bearing the early H2B CCAAT element. These complexes are thus the result of specific protein-DNA interactions, mostly involving the octamer site (Fig. 5A). Three further observations con-

firmed these inferences. First, substitution of 13 bp of the octamer element greatly reduced its ability to bind nuclear protein (Fig. 5A). Second, a methylation interference footprint showed that methylation of a G residue located at position -62 within the octamer site (see octamer sequence, Fig. 3A) interfered with complex formation (Fig. 5B). No other footprints were observed over the 58-bp probe fragment. Finally, a cloned sea urchin octamer-binding factor, SpOct, bound avidly to the wild-type early H2B promoter fragment, but weakly to a fragment bearing a mutation in the octamer site (Fig. 5C). The SpOct protein, obtained by transcription/ translation of a cDNA clone, is a member of the POU II protein family (B. Char, J. Bell, and R. Maxson, in preparation). This family includes Ott-1 and Ott-2, both known to bind the octamer element (He et cd., 1989; LeBowitz et cd, 1989). Our results demonstrate that early H2B promoter fragments specifically detect octamer-binding proteins in sea urchin embryonic nuclear extracts. They demonstrate further that the reduced level of expression of injected early H2B gene bearing a base substitution in the octamer site (Fig. 3B) correlates well with the reduced binding of a cloned POU II-class protein to the mutant site. This correspondence of protein binding in vitro and gene function in wivo strengthens the conclusion that an octamer-binding protein is important for the high level expression of the early H2B gene in blastula-stage embryos. DISCUSSION

We have shown that the faithful expression of the early H2B gene in blastula- and gastrula-stage embryos requires only 71 bp of 5’ flanking sequence, including a canonical octamer element and an adjacent element unique to the early H2B gene. Mutations in the octamer element virtually eliminate expression of the early H2B

BELL, CHAR, AND MAXSON

Sea Urchin H&me

Gene Regulation

369

A Probe Extract

’ -

WI + +

+‘I

mut - +’

c Probe mRNA

WI

mut

TT

Complex [

FIG. 5. Binding of sea urchin nuclear protein to octamer element. (A) Autoradiogram of mobility shift gel. Sea urchin nuclear protein from blastula-stage embryos was incubated with probes containing either a wild-type (wt) or mutant (mut.) octamer consensus element (Materials and Methods). We incubated these fragments alone, with 2 Pg blastula-stage nuclear extract but without competitor DNA, or in the case of the wild-type probe with a 50-fold molar excess of either octamer-bearing (Ott) or CCAAT-bearing (Cat) competitor oligonucleotide. We electrophoresed these mixtures on a nondenaturing 5% acrylamide gel and visualized the free probe and protein-DNA complexes by autoradiography. A broad band representing octamer protein-DNA complexes is indicated by brackets. The single band below the bracketed complex, present in the control without nuclear extract, is an artifact of probe preparation. (B) Methylation interference footprint of octamer protein-DNA complex. A 58-bp DNA fragment bearing the octamer element (-99 to -41; Materials and Methods) was end-labeled, methylated with DMS, and incubated with 10 pg of blastula-stage nuclear protein (Materials and Methods). The resultant protein-DNA complexes were resolved by electrophoresis on a nondenaturing polyacrylamide gel. Bands corresponding to free probe (“Free”) or to octamer protein-DNA complexes (“Complex”) were excised. The labeled DNA was eluted, cleaved with piperidine, electrophoresed on a denaturing urea-8% polyacrylamide gel, and visualized by autoradiography. G and G + A reactions (Maxam and Gilbert, 1980), performed on the labeled probe, are shown as markers. We show only the sense strand footprint; no footprint was evident on the opposite strand. (C) Autoradiogram of mobility shift gel showing binding of a cloned sea urchin POU protein, SpOct, to wild-type or mutant octamer DNA probes. The probes were the same as in A. SpOct was prepared by in vitro transcription and translation as described under Materials and Methods. Two microliters of the lysate was incubated with wild-type or mutant octamer probes in the same manner as in A and electrophoresed on the same gel.

gene, demonstrating for the first time that an octamer element functions in early embryos. Consistent with a prominent role for the octamer element in early HZB gene expression, Wu and Simpson (1985) have observed a nuclease hypersensitive site spanning approximately 100 nt of the early H2B promoter with the octamer element at its center. This site is present in chromatin from early blastulae, in which the early H2B gene is

actively transcribed, but not in chromatin from late blastulae, in which this gene is inactive. That only 71 bp of 5’ flanking sequence are required for the faithful expression of the injected early H2B gene confirms our previous finding that the timing of early H2B expression does not depend on the integrity of the early histone repeat unit (Colin et al, 1988) and is consistent with the results of DiLiberto et al. (1989),

370

DEVELOPMENTALBIOLOGY

showing that the timing of early H3 gene expression requires only 97 bp of flanking sequence. A comparison of the nucleotide sequences involved in the expression of early H3 and early HZB genes reveals no similarities. Thus, despite the attractive simplicity of models proposing common regulators of the five histone genes in the early histone repeat unit, it appears that different mechanisms are responsible for the expression of the early H3 and early H2B genes. There is now ample documentation that the octamer element, ATGCAAAT, has a prominent role in the expression and regulation of a number of genes (reviewed in Scott et aZ., 1989). It mediates, for example, S-phasedependent transcription of mammalian H2B histone genes (LaBella et d, 1988) and cell type-specific transcription of immunoglobulin genes (Sen and Baltimore, 1986) and is required for the expression of certain viral promoters (e.g., Mackem and Roizman, 1982). This diversity of effects is likely due to a multiplicity of POU homeodomain proteins that bind the octamer site (Scholer et al, 1989), to post-translational modifications that alter the activity of such proteins (Kapiloff et cd, 1991; Roberts et al, 1991), and to largely uncharacterized accessory proteins with which POU homeodomain proteins interact (Kristie et aZ., 1989; Kristie and Sharp, 1990). Among the group of POU homeodomain proteins are Ott-1, regulating viral promoters and probably H2B histone genes (LaBella et al, 1988; Mackem and Roizman, 1982), Ott-2, responsible for cell-type-specific expression of immunoglobulin genes (Gerster et al, 1990; LeBowitz et al, 1988; Scheidereit et al, 1987; Staudt et al, 1986), and several others that are expressed in specific tissues and in early mammalian embryos (He et al, 1989; Rosner et aZ., 1990). We have shown recently that the sea urchin possesses at least two genes encoding POU homeodomain proteins (B. Char, J. Bell, and R. Maxson, in preparation). Designated SpOct and SpPOUl, these are members of the POU II and POU III families (He et al, 1989). Messenger RNA encoded by the SpOct gene is present at substantially higher levels in early embryos than in subsequent stages of development, making this gene a leading candidate for a role in the expression of early H2B histone genes (B. Char, J. Bell, and R. Maxson, in preparation). Extensive mutagenesis of the early H2B promoter has shown that the octamer element is essential for highlevel expression of the early H2B gene in cleavage-stage and blastula-stage embryos. The octamer element is insufficient, however, for the inactivation of the early H2B gene in late-stage embryos. Chimeric constructs in which an early H2B octamer element is placed upstream of a late H2B TATA element and coding sequence were expressed in gastrula-stage embryos at levels only about twofold below the level in blastula-stage embryos,

VOLUME150,19%

This decline is severalfold below that usually observed for injected early H2B genes. Thus the octamer element contributes at most lo-20% of the down-regulation of the early H2B gene. What, then, is the mechanism that regulates the timing of early H2B expression? Despite extensive mutagenesis of the early H2B promoter and mRNA leader sequences, we have not identified a mutation that changes the time at which the early H2B gene is inactivated (Figs. 1-3; and J. Bell, and R. Maxson, unpublished). One intriguing possibility, now under investigation, is that regulatory sequences lie within the body of the early H2B gene. J.B. was supported by NC1 Training Grant 5-T32-CA09569-03 and a grant from the Wright Foundation. B.R.C. was partially supported by a fellowship from the California Foundation for Biochemical Research. This work was supported by NIH Grant HD18582 to R.M. and by funds from NC1 Institutional Grant 5P30-CA14089-14. We thank Dr. Michael Stallcup for a critical review of the manuscript, and we are grateful to Lynn Williams for the synthesis of oligonucleotides. REFERENCES Barberis, A., Superti-Furga, G., and Busslinger, M. (1987). Mutually exclusive interaction of the CCAAT-binding factor and of a displacement protein with overlapping sequences of a histone gene promoter. Cell 50,347-359. Busslinger, M., Rusconi, S., and Birnstiel, M. L. (1982). An unusual evolutionary behavior of a sea urchin histone gene cluster. EMBOJ. 1,27-33. Calzone, F. J., Theze, N., Thiebaud, P., Hill, R. L., Britten, R. J., and Davidson, E. H. (1988). The developmental appearance of factors that bind specifically to &-regulatory sequences of a gene expressed in the sea urchin embryo. Genes Dev. 2,1074-1088. Cohn, R. H., Lowry, J. C., and Kedes, L. H. (1976). Histone genes of the sea urchin S. pwpuratus cloned in E. co.&: Order, polarity and strandedness of the five histone-coding and spacer regions. Cell 9, 147-161. Colin, A. (1986). Rapid repetitive microinjection. Curr. Methods Cdl BioL 27,395-406. Colin, A. M., Catlin, T. L., Kidson, S. H., and Maxson, R. E. (1988). Closely linked early and late H2b histone genes are differentially expressed after microinjection into sea urchin zygotes. Proc NatL Aw.d Sci. USA 85,507-510. DiLiberto, M., Lai, Z-C., Fei, H., and Childs, G. (1989). Developmental control of promoter-specific factors responsible for the embryonic activation and inactivation of the sea urchin early histone H3 gene. Genes Lkv. 3,973-985. Flytzanis, C. N., Britten, R. J., and Davidson, E. H. (1987). Ontogenic activation of a fusion gene introduced into sea urchin eggs. Proc. NatL Ad 81% USA 84,151-155. Fried, M., and Crothers, D. (1981). Equilibria and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. NucEeic Acids Res. 9, 6505-6525. Gerster, T., Balmaceda, C., and Roeder, R. G. (1990). The cell-specific octamer transcription factor OTF-2 has two domains required for the activation of transcription. EMBO J. 9,1635-1643. He, X., Treaty, M. N., Simmons, D. M., Ingraham, H. A., Swanson, L. W., and Rosenfeld, M. G. (1989). Expression of a large family of POU-domain regulatory genes in mammalian brain development. Nature 340,35-42. Hendrickson, W., and Schlief, R. (1985). A dimer of AraC protein con-

BELL, CHAR,

AND MAXSON

Sea Urchin Histone Gene Regulation

tacts three adjacent major groove regions of the AraI DNA site. Proc Nati Acad Sci USA 82,3129-3133. Hentschel, C., and Birnstiel, M. L. (1981). The organization and expression of histone gene families. Cell 25,301-313. Kapiloff, M. S., Farkash, M., Wegner, M., and Rosenfeld, M. G. (1991). Variable effects of phosphorylation of Pit-l dictated by the DNA response elements. Science 253, ‘786-789. Kedes, L. H. (1979). Histone genes and histone messengers. Annu. Rev. Riochem 48,837-870. Kristie, T. M., LeBowitz, J. H., and Sharp, P. A. (1989). The octamerbinding proteins form multi-protein-DNA complexes with the HSV aTIF regulatory protein. EMBO J. 8,4229-4238. Kristie, T. M., and Sharp, P. A. (1990). Interactions of the Ott-1 POU subdomains with specific DNA sequences and with the HSV atransactivator protein. Genes Dev. 4,2383-2396. LaBella, F., Sive, H. L., Roeder, R. G., and Heintz, N. (1988). Cell-cycle regulation of a human histone H2B gene is mediated by the H2b subtype-specific consensus element. Genes Dev. 2,32-39. Lai, Z., Maxson, R., and Childs, G. (1988). Both basal and ontogenic promoter elements affect the timing and level of expression of a sea urchin Hl gene during early embryogenesis. Genes Dev. 2,1’73-183. LeBowitz, J. H., Clerc, R. G., Brenowitz, M., and Sharp, P. A. (1989). The Ott-2 protein binds cooperatively to adjacent octamer sites. Genes Dev. 3,1625-1638. LeBowitz, J. H., Kobayashi, T., Staudt, L., Baltimore, D., and Sharp, P. A. (1988). Octamer binding protein from B or HeLa cells stimulates transcription of the immunoglobulin heavy-chain promoter in vitro. Genes Dew. 2, 1227-1237. Mackem, S., and Roizman, B. (1982). Differentiation between alpha promoters and regulatory regions of herpes simplex virus 1: The functional domains and sequence of a movable alpha regulator. Proc. Natl. Acad Sci. USA 79,4917-4921. Maxam, A., and Gilbert, W. (1980). Sequencing end-labelled DNA with base-specific cleavages. In “Methods in Enzymology” (L. Grossman and K. Moldave, Eds.), Vol. 65, pp. 499-513. Academic Press, San Diego. Maxson, R., Mohun, T., Gormezano, G., Childs, G., and Kedes, L. (1983). Distinct organizations and patterns of expression of early and late histone gene sets in the sea urchin. Nature (London) 301, 120-125. Maxson, R. E., Mohun, T. J., Gormezano, G., and Kedes, L. H. (1987). Evolution of late H2A, H2B and H4 genes of the sea urchin Stwngylocentrotus purpuratus. Nucleic Acids Res. 15,10569-10582. Maxson, R. E., and Wilt, F. H. (1981). The rate of synthesis of histone mRNA during development of sea urchin embryos.Strongylocentro tuspurpuratus. Dev. Biol 83,380-386. Maxson, R. E., and Wilt, F. H. (1982). Accumulation of the early histone mRNAs during the development of Strongylowntrotus pqwuratus. Dev. Biol 94,435-440. McMahon, A., Flytzanis, C., Hough-Evans, B., Katula, K., Britten, R., and Davidson, E. (1985). Introduction of cloned DNA into sea urchin egg cytoplasm: Replication and persistence during Embryogenesis. Dev. BioL 108,420-430. Norris, K., Norris, F., Christiansen, L., and Fiil, N. (1983). Efficient site-directed mutagenesis by simultaneous use of two primers. Nucleic Acids Rex 11,5103-5112.

371

Overton, G. C., and Weinberg, E. S. (1978). Length and sequence heterogeneity of the histone gene repeat unit of the sea urchin, S purpm&s. CeU 14,247~257. Roberts, S. B., Segil, N., and Heintz, N. (1991). Differential phosphorylation of the transcription factor Octl during the cell cycle. Science 253,1022-1026. Rosenfeld, M. G. (1991). POU-domain transcription factors: pou-er-ful developmental regulators. Genes Deu. $897-907. Rosner, M. H., Vigano, M. A., Ozato, K., Timmons, P., Poirier, F., Rigby, P. W. J., and Staudt, L. M. (1990). A POU-domain transcription factor in early stem cells and germ cells of the embryo. Nature 345,686-692. Ruvkun, G., and Finney, M. (1991). Regulation of transcription and cell identity by POU domain proteins. Cell 64,475-478. Schaffner, W., Kunz, G., Daetwyler, H., Telford, J., Smith, H. O., and Birnstiel, M. L. (1978). Genes and spacers of cloned sea urchin histone DNA analyzed by sequencing. Cell 14,655-671. Scheidereit, C., Heguy, A., and Roeder, R. G. (1987). Identification and purification of a human lymphoid-specific octamer binding protein OTF-2 that activates transcription of an immunoglobulin promoter in vitro. Cell 51,783-793. Scholer, H. R., Hatzopoulos, A. K., Balling, R., Suzuki, N., and Gruss, P. (1989). A family of octamer-specific proteins present during mouse embryogenesis: Evidence for germline-specific expression of an Ott factor. EMBO J. 8,2543-2550. Scott, M. P., Tamkun, J. W., and Hartzell, G. W., III (1989). The structure and function of the homeodomain. B&him Biophys. Ada 989, 25-48. Sen, R., and Baltimore, D. (1986). Multiple factors interact with the immunoglobulin enhancer sequences. Cell 46,705-716. Staudt, L. M., Singh, H., Sen, R., Wirth, T., Sharp, P. A., and Baltimore, D. (1986). A lymphoid-specific protein binding to the octamer motif of immunoglobulin genes. Nature 323,640~643. Sturm, R., Baumruker, T., Franza, B. R., and Herr, W. (1987). A lOOkDa HeLa cell octamer binding protein (OBPlOO) interacts differently with two separate octamer-related sequences within the SV40 enhancer. Genes Deu. 1,1147-1160. Sures, I., Lowry, J., and Kedes, L. H. (1978). The DNA sequence of sea urchin X purpuratus: H2A, H2B and H3 histone coding and spacer regions. CeU 15,1033-1044. Vitelli, L., Kemler, I., Lauber, B., Birnstiel, M. L., and Busslinger, M. (1988). Developmental regulation of micro-injected histone genes in sea urchin embryos. Dev. Biol. 127,54-63. Weinberg, E. S., Hendricks, M. B., Hemminki, K., Kuwabara, P. E., and Farrelly, L. A. (1983). Timing and rates of synthesis of early histone mRNA in the embryo of Strcmg¢rotus purptwatus. Dev. Biol. 98,117. Wu, T. C., and Simpson, R. T. (1985). Transient alterations of the chromatin structure of sea urchin early histone genes during embryogenesis. Nucla’c Acids Res. 13,6187-6203. Zhao, A. Z., Colin, A. M., Bell, J., Baker, M., Char, B. R., and Maxson, R. (1990). Activation of a late H2B histone gene in blastula-stage sea urchin embryos by an unusual enhancer located 3’ of the gene. Mol. Cell Biol. 10,6’730-6741. Zhao, A. Z., Vansant, G., Bell, J., Humphreys, T., and Maxson, R. (1991). Activation of the Ll late H2B histone gene in blastula-stage sea urchin embryos by an Antennapedia-class homeoprotein. Meek Dev. 34,21-28.