Mar 25, 1987 - At this time, aphidicolin or hydroxyurea was added to the cells to block DNA synthesis, and the kinetics of decay of the endogenous mouse ...
The EMBO Journal vol.6 no.6 pp.1825-1831, 1987
Sequences controlling histone H4 mRNA abundance
Olga Capasso', Giselle C.Bleecker and Nathaniel Heintz Laboratory of Biochemistry and Molecular Biology, The Rockefeller University, 1230 York Avenue, New York, NY 10021, USA 'Present address: Laboratorio di Biologia Cellulare, Consiglio Nazionale delle Richerche, via G.Romagnosi 18/A, 00196 Roma, Italy Communicated by C.Baglioni
The post-transcriptional regulation of histone mRNA abundance is manifest both by accumulation of histone mRNA during the S phase, and by the rapid degradation of mature histone mRNA following the inhibition of DNA synthesis. We have constructed a comprehensive series of substitution mutants within a human H4 histone gene, introduced them into the mouse L cell genome, and analyzed their effects on the post-transcriptional control of the H4 mRNA. Our results demonstrate that most of the H4 mRNA is dispensable for proper regulation of histone mRNA abundance. However, recognition of the 3' terminus of the mature H4 mRNA is critically important for regulating its cytoplasmic half-life. Thus, this region of the mRNA functions both in the nucleus as a signal for proper processing of the mRNA terminus, and in the cytoplasm as an essential element in the control of mRNA stability. Key words: Histone H4 mRNA/post-transcriptional control and accumulation/cytoplasmic half-life/cell cycle
Introduction The accumulation of replication variant histone mRNAs during the S phase of the cell cycle has been thoroughly studied in a variety of eukaryotic cells (see review by Schumperli, 1986). It is now well documented that the increase in histone mRNA during S phase involves both transcriptional and posttranscriptional control (Heintz et al., 1983; Sittman et al., 1983). Recent in vivo (Capasso and Heintz, 1985; Artishevsky et al., 1985; Osley et al., 1986) and in vitro (Heintz and Roeder, 1984; Hanly et al., 1985; Dailey et al., 1986) evidence indicates that transcription of histone genes during cell cycle is mediated by soluble trans-acting transcriptional regulatory factors. In contrast, it is not evident whether the regulation of processing, nuclear cytoplasmic transport or mature mRNA stability is responsible for the post-transcriptional component of histone mRNA accumulation during DNA synthesis. The original demonstration that mature cytoplasmic histone mRNA accumulates preferentially during S phase (Robbins and Borun, 1967) and is rapidly degraded after the inhibition of DNA synthesis (Gallwitz, 1975) has been reproduced in many laboratories using a variety of techniques (for review see Schumperli, 1986). The facts that this response can be elicited by any agent which can effectively block DNA synthesis, that is specific for histone mRNA and that is rapidly reversible upon the resumption of DNA synthesis suggest that the regulation of mature mRNA stability may be critically important in the specific accumulation of histone mRNA during the S phase of the cell cycle. Recent (a IRL Press Limited, Oxford, England
data (Ross and Kobs, 1986; Ross et al., 1986) indicate that the
mechanism for degradation of histone mRNA following a block in DNA synthesis involves degradation by a 3' exonuclease. However, no direct evidence supporting a role for this mechanism in the normal turnover of histone mRNA as cells exit S phase has been obtained. That histone mRNA must be correctly processed at its 3' terminus for its temporal regulation during the cell cycle has been demonstrated (Luscher et al., 1985). Mutant histone genes which are not correctly processed within the nucleus are present at equivalent levels in both GI arrested and exponentially growing cell populations. Furthermore, an 80 nucleotide fragment surrounding the 3' processing sites of a mouse histone H4 gene can confer growth regulation to a chimeric gene when re-introduced into cells by transfection (Stauber et al., 1986). Although these studies are consistent with the notion that the regulation of maturation at the 3' end of histone mRNA is responsible for its increased abundance in the cytoplasm during S phase, they do not rule out models in which the 3' end of the mRNA is involved in other post-transcriptional regulatory events. In an attempt to define further the processes responsible for post-transcriptional control of histone mRNA abundance, we have constructed a comprehensive series of substitution mutants within a human histone H4 mRNA and analyzed their effect both on the rapid degradation of histone mRNA after blocking DNA synthesis and on accumulation during the cell cycle. Our results indicate that recognition of the 3' terminus of mature histone mRNA is critical for its regulation in response to DNA synthesis inhibitors, and that this may be important for its proper accumulation during the cell cycle.
Results Construction of mutant histone H4 genes The altered histone genes employed in this study were constructed by recombing appropriate fragments derived from a series of 5' and 3' Bal3l deletion mutants described by Hanly et al. (1985). In each case, the deleted H4 mRNA sequence was replaced by approximately 30 nucleotides from the M13mplO polylinker. Nucleotide sequencing of these recombinants revealed that many of them do not preserve the histone H4 translation frame: several of the 5'-most substitution mutants utilize out of frame stop codons within the coding region of this gene, whereas three of the 3'-most constructs are translated beyond the normal H4 translation stop codon. As detailed below, to control for the effects of these frame shifts on histone mRNA accumulation, additional point mutations were introduced into these constructs which result in stop codons in all reading frames very close to the normal UGA codon. This was accomplished using oligonucleotide directed mutagenesis (see Sive et al., 1986) to alter the carboxyterminal portion of the H4 coding sequence from 5' GGU UUC GGU GGU UGA GCG 3' to 5' GGU UUA GGU GAU UGA GGG 3'.
1825
O.Capasso, G.C.Bleecker and N.Heintz +390
+348
+37
+1
i
i
CAP
UGA
AUG
75-160
8-27
25-58 -
I
119-160
5&
201 -279
218 -279
1 158-206 i
i~~~~~~~I
I
265-299
338-390
Fig. 1. Map showing histone H4 mRNA sequences deleted in constructs employed in this study. The line drawing at the top of this figure represents the mature pHu4a human histone H4 mRNA. The numbers indicate the mRNA cap site (+ 1), the translation initiation codon (+37), the translation stop codon (+348) and the mature 3' terminus (+390) of the miRNA. Each deletion mutant is described by the solid lines below the mature mRNA, with associated numbers referring to the nucleotide positions of the end points of each deletion.
As shown in Figure 1, the constructions we have analyzed span most of the human histone H4 mRNA. With the exception of
deletion 338 -390, none of these altered genes disrupt sequences previously shown to be required for histone H4 mRNA 3' processing (Birnstiel et al., 1984). Thus, these constructs are designed to address specifically whether mRNA sequences other than those required for mRNA maturation in the nucleus are required for post-transcriptional control of histone mRNA abundance. Phenotypes of cell lines carrying the altered histone H4 genes We have previously demonstrated that the wild-type human histone H4 gene is normally regulated when stably introduced into the mouse L cell genome (Capasso and Heintz, 1985). Therefore, to analyze the constructions described above, we have established stable cell lines carrying each gene by cotransfection with the neomycin resistance gene and selection in G418 (Southern and Berg, 1982). We have made at least three transfections for each gene and analyzed pools of colonies containing an average of 10-15 copies per haploid genome. In each case, the transformed cells were synchronized, released into S phase, and histone mRNA allowed to accumulate for 3 h. At this time, aphidicolin or hydroxyurea was added to the cells to block DNA synthesis, and the kinetics of decay of the endogenous mouse histone mRNA and the introduced human H4 mRNA measured using the SI nuclease mapping procedure (Berk and Sharp, 1977). In all cases the efficacy of the DNA synthesis inhibitor employed was monitored by measuring the incorporation of 3H deoxycytidine into DNA (Heintz et al., 1983). Using this protocol, we have previously shown that this human histone H4 mRNA decays with a 10-15 min half-life. Figure 2A presents the data from analysis of all of the transfected cell lines in which the introduced human histone gene is normally post-transcriptionally regulated. In these cell lines, both the accumulation and decay of the human mRNA parallel that of the endogenous mouse histone mRNAs. Examples of the SI nuclease mapping data which were used to generate the data in Figure 2A are shown for the cell lines containing the wild type human H4 gene (Figure 2B) and a mutant H4 construct in which nucleotides 8-27 have been substituted (Figure 2C). It is noteworthy that in some cell lines (Figure 2B) the relative increase of the introduced mRNA is not as large as that of the resident histone muRNA. We attribute this to the dosage compensation phenomenon we have previously reported (Capasso and Heintz, 1985). In these cases, however, quantitation of the S1 mapping data reveals that the rate of decay of the introduced H4 mRNA that has accumulated during the 3 h post release is similar to that of the resident mouse histone H4 mRNA (Figure 2). In several 1826
the degradation of the endogenous mouse histone H3 mRNA was also analyzed (Figure 2D) and shown to occur with the same kinetics as the histone H4 mRNAs. Two of the constructs which fall into this category are of particular interest. Firstly, the mutant 8-27 removes almost entirely 5' non-coding region of this histone H4 mRNA, but is still normally regulated. Thus, if the 5' terminus of the mRNA is important for post-transcriptional regulation, those elements responsible must reside within the first eight base pairs. Secondly, the mutant 25-58 deletes the histone H4 AUG codon translation initiation codon and, as a result, can only code for a 17 amino acid out of frame polypeptide between positions 119 and 170 of the altered mRNA. Since this mRNA is regulated normally, translation of histone H4 from the substrate mRNA is not required for its degradation. Furthermore, since the open reading frame for translation of this mRNA is sufficient for occupancy by only a single ribosome, formation of a normal polysome is not essential for the post-transcriptional control of histone mRNA abundance. As shown in Figure 3, an entirely different result was obtained when we analyzed cell lines containing the 200-279, 218-279 or 265 -299 substitution mutants. In this case, the altered human H4 mRNAs do not accumulate during the first 3 h of S phase and they are not degraded after blocking DNA synthesis with aphidicolin (Figure 3A,B). As expected, the endogenous mouse histone mRNAs in these cell lines are regulated normally (Figure 3A,C,D). To ensure that the result shown in Figure 3 did not simply reflect our use of a 5' end-labelled SI nuclease probe, that the 3' terminus of the altered H4 mRNA is properly processed, and to measure the endogenous mouse H4 mRNA, we repeated the experiments using 3' labelled SI nuclease probes. As shown in Figure 3C, the 218 -279 H4 mRNA is processed to the expected size and, in agreement with the results of Figure 3B, is not subject to the normal post-transcriptional control mechanisms regulating the abundance of the mouse histone mRNA. These results indicate, therefore, that correct processing of the 3' terminus of histone mRNA is not sufficient for its accumulation upon release of the cells into S phase or for its rapid degradation following the inhibition of DNA synthesis. Accumulation of the 218-279 H4 mRNA during the cell cycle The apparent loss of post-transcriptional control over the 200279, 218-279 and 265 -299 mRNAs observed in the short-term experiments described above led us to investigate the behaviour of these mRNAs throughout the cell cycle. The synchronization protocol employed is identical to that previously utilized (Capasso and Heintz, 1985) to study several cell lines containing the wild type human H4 mRNA in mouse L cells. Exponentially growing cells are first treated with indomethacin for 24 h, leading to the cases,
Histone H4 mRNA abundance
A A
I
100h+
IN
75 +
B
0
+
i I i
+
+
3 10 20 50
0
z II (-I
d-
50t
U
eecU -o_{
./
_M
eess
U
T
-mouLse
25
01L
J
-
-
5c
0
0
3
D
4
+
C
0
3
+
+
1 0 20 30
+
50
D
D . S)
.. ":
+ a.
50
258
258 -
207 --
207ci
-moIuse H3
.
-mz0ouse H3 variant
105-10Fig. 2. Analysis of constructs showing 'wild type' regulation after inhibition of DNA synthesis. Cell lines carrying integrated copies of each mutant were synchronized, released into S phase and, after 3 h, treated with aphidicolin (Materials and methods). Total RNA was extracted at the G1/S boundary (time 0), the peak of S phase (time 3), and at various times after treatment with aphidicolin (+ 10, +20, +30, etc.). (A) Quantitation of SI nuclease mapping data showing accumulation and decay of several human histone H4 substitution mutants: (X) wild type; (0) 8-27; (0) 25-58; (OI) 75-160; (U) 119-160; (A) 158-206. (B) 3' S1 mapping data of total RNA isolated from L cell lines carrying the wild type human histone H4 gene. The probe employed was labelled at position 141 of the mRNA. WT indicates band resulting from full length protection of the introduced human histone H4 mRNA. Mouse indicates protection of the endogenous L cell mRNAs to the end of the highly conserved coding region. (C) SI nuclease mapping data from the L cell line containing the 8-27 altered human histone H4 gene. The SI probes and resulting protected bands are analogous to those shown in (B). (D) Si nuclease mapping of 8-27 RNA utilized in (C), but employing a 3' SI probe to measure the accumulation and decay of the endogenous mouse histone H3 mRNA.
arrest of cells in Gi phase (Bayer et al., 1979). The cells are then released for 2 h and, to obtain precisely synchronized cell populations, they are arrested at the G1/S boundary by a 12-h
treatment with a DNA synthesis inhibitor (aphidicolin or hydroxyurea). Release from the second block results in a very high synchronous cell population which traverses S phase in unison and 1827
-
O.Capasso, G.C.Bleecker and N.Heintz
100
I 50 j
-
\
+
+
+-
+
I 0
\ -I \\0 --
+
0 3 10 20 30 50 90
207 -
1
75 1
25
B
,_
0
r-
-
-
1- 218-279
105-
91-
0' +
0
3
+
1 0 20
+
+
30
50
90
D
C
+
0
154-
31lo
+
t me
+ + + + + 0 3 10 20 30 50 90
+
2050
* ~-flw
400p
"
-
75-
-moLuse H3 -mouLse
H3
vari ant
-218-279
a _ ) mnouse
Fig. 3. Analysis of constructs showing a 'mutant' phenotype for the introduced human histone H4 mRNA. (A) Quantitation of SI mapping data as in Figure 2A showing that the 200-279 (0), 218-279 (0) and 265 -299 (L]) mRNAs do not accumulate upon release into S phase and are not degraded after a block in DNA synthesis. (B) 5' SI nuclease mapping of total RNA from the 218-279 cell line isolated at the times indicated at the top of the figure. The probe employed was labelled at the Narl site at position 141 of the mRNA. (C) 3' SI mapping of RNA employed in (B). The probe utilized was 3' labelled at the Bste2 site at position 279 in the mRNA. (D) SI nuclease mapping of the endogenous mouse H3 mRNA (see Figure 2D) in the 218-279 RNA samples analyzed above (A,B,C). ,
remains quite synchronous in a second S phase approximately 14-18 h post-release. The experiment presented in Figure 4 shows the accumulation of both the endogenous mouse histone H4 mRNA and the 218279 human H4 mRNA during long-term synchronization protocol described above. The accumulation of the mouse H4 mRNA 1828
in these cells is as expected: it accumulates to significant levels only during the two S phases, in parallel with the incorporation of deoxycytidine into DNA. In contrast, the accumulation of the 218 -279 mRNA is not coupled to the rate of DNA synthesis: its accumulation peaks approximately 10 h after release from the indomethacin treatment, in the presence of the DNA synthesis
Histone H4 mRNA abundance
A
-218-2 79
B
1001
-16 -14
