Splicing thermotolerance maintains pre-mRNA transcripts in the splicing pathway during severe heat shock

EXPERIMENTAL

CELL

RESEARCH

202,

233-242 (1992)

ermotolerance Maintains Prethe Splicing Pathway during Seve ROBERT A. CORELL'ANDROBERTKGRCBS~ Department

of Biology and Molecular

Genetics Center, Dartmouth

Thermotolerance, the ability of cells and organisms to withstand severe elevated temperatures after brief exposure to mild elevated temperatures, has been studied in numerous laboratories. Survival thermotolerante is defined as the increase in cell or organism survival at severe elevated temperatures after a pretreatment at mild elevated temperatures. This study examines splicing thermotolerance in Drosophila melanogaster, the ability to splice pre-mRNAs made at the severe temperature (38°C) after a brief pretreatment at a milder temperature (35°C). It is probably one of a number of mechanisms by which cells adapt to heat shock. These experiments demonstrate that pre-mRNAs synthesized at the severe temperatures in splicing thermotolerant cells, although protected in splicing-competent are not actually processed to mature complexes, mRNAs until the cells are returned to their normal temperature. We have also studied the kinetics of acquisition and loss of splicing thermotolerance. As little as 10 min of pretreatment at 35°C was sufficient to provide full splicing thermotolerance to a 30-min severe heat shock of 38°C. Pretreatments of less than 10 min provide partial splicing thermotolerance for a 30-min severe heat shock. Full splicing thermotolerance activity egins to decay about 4 h after the cessation of the 35°C incubation and is completely lost by 8 h after the pretreatment. The kinetics experiments of pre-mRNAs synthesized during the 38°C treatment in splicing thermotolerant cells indicate that one or more splicing thermotolerance factors are synthesized during the 35°C pretreatment which interact with pre-mRNA-containing complexes to keep them in a splicing-competent state. These kinetic experiments also indicate that in cells which are partially splicing thermotolerant, the pre-mRNAs synthesized early during the 38°C incubation are protected, whereas those synthesized late are not. In the absence of splicing thermotolerant factors, the pre-mRNA-containing complexes leave the normal splicing pathway and are allowed to exit to the 0 1992 Academic Press, Inc. cytoplasm.

College, Hanover,

New Hampshire

03755

involved in splicing [l-6] (for review see [:I, 7]), a number of the early steps in splicing, such as those that commit the pre-mRNA to the splicing pathway, remain to be elucidated. A working model [S] ho1 s that shortly after an RNA is first transcribed, a dec ion is made about whether or not the RNA is g exit the nucleus immediately. According to this retention of the RNA in the nucleus is in mos dictated by the ability of the RNA to interact with the splicing machinery. If the RNA transcript is not r ing machinery, it most likely will the cytoplasm. Evidence for this model comes from three different sets of experiments. First, when a no cell is severely heat shocked, the pre-m duced during the heat shock is not spl nor does it ever reenter the splicing pathway [9]. This unspliced pre-mRNA exits the nucleus where it can be translated into an aberrant protein [IO]. Theoretically, the pre-mRNA exits the nucleus in this situation because the splicing machinery is stress [ll, 121. Second, experiments in yea interaction between the p paratus is needed for n mRNA [13]. If the highly quence is disrupted, unspliced 8 exit the nucleus qttenice that recogand are translated. If the U2 nizes the branch point seque sate for pre-mRNA mutations, retained in the nucleus After intron removal the nucleus to the eytspPasm The third piece of evidence comes from experiments ial t%r the retroviwith the AIDS virus, HIV. Bt is ess rus to maintain and the unsplice makes a number of proteins th.at ar retroviruses [P4, 151, one of wbi interacts with a sequence in the the Rev Responsive Element to promote the appear-

INTRODUCTION While a great deal of information has been gathered about many of the components and some of the steps

1 Present address: Seattle Biomedical Research Institute, son Street, Seattle, WA 98109-1651. 2 To whom reprint requests should be addressed.

233 All

Copyright 0 1992 rights of reproduction

4 Nicker-

00144827/92 $5.00 by Academic Press, Inc. in any form reserved.

234

CORELL

ante of full-length unspliced HIV mRNAs in the cytoplasm [16]. Rev can only act on pre-mRNAs that are poorly spliced [17] and therefore are retained in the nucleus for extended periods after transcription. Lindquist [9, 18-201 has found that in Drosophila melanogaster the expression of one of the major heat shock proteins, hsp70, is maximum at 38°C but is not detectable at 25°C. Hsp83 protein is expressed at low levels at 25°C but expression increases as the temperature increases, reaching a maximum at about 35°C [18,19,21]. Above this temperature, synthesis of hsp83 protein actually declines. In contrast, it was found that the level of hsp83 transcription was maximal at 38°C. The discrepancy arises from the fact that at 38°C the single intervening sequence of the hsp83 pre-mRNA is not removed and therefore a significant fraction of the hsp83 transcripts accumulates as unspliced pre-mRNA. If the cells were prestressed at 35°C and returned to 25°C for 3 h before being stressed at 38°C splicing occurs normally after the cells are returned to 25°C. This phenomenon in which the process of mRNA splicing is protected against severe stress is referred to as splicing thermotolerance. For clarity, in this paper we define all thermotolerance phenomena that involve protection against cell death as survival thermotolerance, and those phenomena involving protection of the splicing mechanism as splicing thermotolerance. If cells are incubated in the presence of cycloheximide during the prestress at 35°C and during the 3-h recovery at 25°C splicing thermotolerance at 38°C does not occur [33]. This indicates that there is at least one protein factor produced during the pretreatment at 35°C that is critical to preserving splicing under stress conditions. The disruption of splicing by severe stress has also been shown to occur in trypanosomes [22], soybeans [23], and Caenorhabditis elegans [24]. Bond has shown that in cell-free splicing systems, the interference of splicing in extracts from heat-shocked HeLa cells is correlated with a disruption of U2 and U4/5/6 snRNP particle structures [ 111. The experiments in this paper further define the phenomenon of splicing thermotolerance and allow us to suggest a possible mechanism for splicing thermotolerante. We demonstrate that a pretreatment of D. melanogaster cells for 10 min at 35°C is sufficient to provide complete splicing thermotolerance to a subsequent 30min 38°C incubation. Splicing thermotolerance is transient in nature and is rapidly induced. A 30-min 35°C treatment provides complete splicing thermotolerance to a subsequent 30-min 38°C heat shock for a period of about 4-5 h after the initial 35°C treatment, but this protection decays and is totally gone by 8 h after the 35°C treatment. We show that pre-mRNAs produced at 38°C are not spliced, even in thermotolerant cells. In order to splice the pre-mRNAs synthesized at 38°C in splicing thermotolerant cells, a return to 25°C is re-

AND

GROSS

quired after the 38°C. As splicing thermotolerance decays (between 4 and 8 h), those pre-mRNAs synthesized at the beginning of the 38°C incubation are more likely to be spliced than those synthesized at the end of the 38°C incubation. MATERIALS

AND

METHODS

Cell culture. Schneider’s Line 2 D. melanogaster cells [25] were grown in Schneider’s Drosophila medium from GIBCO and supplemented with 0.1 vol of heat-inactivated fetal bovine serum from HyClone. Cells were grown in Falcon tissue culture flasks at 25°C in an air incubator. Heat treatment of cells. All heat treatments of the cells were performed in airtight tissue culture flasks under water in a temperatureregulated water bath (+O.O5”C). The flasks were kept completely submerged for the entire heat treatment. Accurate temperature was determined by an ASTM thermometer (ERTCO Cat. No. 3199, tO.l”C). All recoveries from heat treatment were carried out in a 25°C air incubator. RNA extraction. Whole-cell RNA was extracted by a modification of the guanidine isothiocyanate method [26]. Briefly, cells were harvested, pelleted, and washed once with 1X DPBS (1X DPBS is 0.14 M NaCl, 2.7 mM KCl, 8.1 mM Na,HPO,, 1.5 mA4 KH,PO,). The cells were then lysed in 3 ml of GITC solution (GITC solution is 6 A4 guanidine isothiocyanate, 5 mM sodium citrate, pH 7.0, and 0.1 M fl-mercaptoethanol). The solution was then layered on 1.5 ml of 5.7 M CsCl in 0.1 A4 EDTA (pH 7.5). The RNA was pelleted in a swinging bucket rotor (Sorvall Model No. AH650) at 35,000 rpm (114,OOOg) for 18 h at 20°C. RNA gels. 10 micrograms of whole-cell RNA was dried in a Savant Speed Vat condenser (Model No. RH40-11) under vacuum. The RNA was then resuspended in 20 ~1 of denaturing solution [denaturing solution is 10 ~1 of formamide (Bethesda Research Labs), 2 ~1 of 10X Mops buffer (1X Mops buffer is 0.02 M 3-(N-morpholino)propanesulfonic acid (Sigma), pH 7,8 mM sodium acetate, and 1 mMEDTA], 3.4 ~1 of formaldehyde (Fisher), and 4.6 ~1 of H,O. The samples were denatured by a 5-min incubation at 65°C. The samples were briefly quenched on ice; loading dye was added and samples were loaded on the gel. The gel was 1% agarose (Bethesda Research Labs) in 1X Mops buffer and 6% formaldehyde. The running buffer was 1X Mops buffer and 6.5% formaldehyde. RNA blotting. The RNA was transferred to a nylon/nitrocellulose filter (Nylon 66 from Schleicher & Schuell, Cat. No. 41-40880) by the capillary action of 20x SSC (20x SSC is 3 M sodium chloride, 0.3 M sodium citrate, pH 7) being drawn up through the gel by blotting paper placed on top of the nitrocellulose overnight. The 20x SSC is drawn up from a buffer reservoir containing enough buffer to allow transfer to occur for at least 18 h. Large-scale preparation of plasmid DNA. Large quantities of plasmid DNA were made by the following modifications of established procedures [27, 281. Two microliters of bacterial stock was used to seed 10 ml of L-broth [L-broth is 10 g/liter Bacto-Tryptone (Difco), 5 g/liter Bacto Yeast extract (Difco), and 5 g/liter NaCl (Fischer), pH 71 and 20 pg/ml of the appropriate antibiotic. The bacteria were grown for at least 16 h with shaking at 37°C. This culture was used to seed 750 ml of plasmid L-broth [plasmid L-broth is 6 g/liter casamino acids, 100 ml/liter 10X M9 (M9 is Na,HPO,, 58 g/liter; KH,POI, 30 g/liter; NaCl, 5 g/liter; NH,Cl 10 g/liter), 1 ml/liter 1 M MgSO,, 5 ml/liter of 40% (v/v) glucose, 1 ml/liter of 10 mg/ml thiamine, 4 ml/liter of 25 mg/liter thymidine]. The plasmid L-broth culture was grown with shaking at 37°C until the A,,, = 0.8. Two milliliters of 75 mg/ml chloramphenicol was then added and shaking at 37°C was continued for an additional 16 h. The bacterial cells were then pelleted, resuspended in 50 ml of lyso-

SPLICING

THERMOTOLERANCE

zyme buffer [lysozyme buffer is 50 mM glucose, 10 mM EDTA, 25 mM Tris-HCl, pH 8, and 5 mg/ml lysozyme was added just before use], and incubated at 37°C for 4 min. One hundred milliliters of 0.2 N NaOH, 1% SDS was added and the solution was mixed and incubated at 4°C for 5 min. Then 75 ml of AC-/K+ solution was added and incubated on ice for 30 min [AC-/K+ is 60 ml 5 M KOAc, 30 ml glacial acetic acid, and 10 ml H,O]. The solution was then centrifuged in a Sorvall GS3 rotor at 8000 rpm for 30 min. An equal volume of isopropanol was then added to the supernatant and incubated at room temperature for 2 h or more. The precipitate was then pelleted in a Sorvall GS3 rotor at 5000 rpm for 15 min, and the pellet was resuspended in 15 ml of 10 mMTris-HCl, 1 mMEDTA, pH 8. Next, 20 ~1 of RNase A (10 mg/ml) was added to the solution and incubated at 37°C for 15 min. The solution was then phenol/chloroform extracted, and the remaining DNA was ethanol precipitated. The DNA was pelleted in a GSA rotor at 6000 rpm for 20 min. The pellet was resuspended in 26 ml of 50 mM Tris-HCI, 1 mM EDTA, pH 8.0, and 29.025 g of CsCl was added. One milliliter of 5 mg/ml ethidium bromide was added and the solution was spun at 45,000 rpm in a Sorvall TV850 rotor for 16 h. The rotor was stopped without braking. The plasmid (lower) band was then removed from the centrifuge tube and the ethidium bromide was extracted with isoamyl alcohol. The DNA was then ethanol precipitated. RNA was removed from this plasmid pellet by centrifuging the DNA through a 1 M NaCl, 10 mM Tris-HCl, 1 mM EDTA, pH 8, cushion at 40,000 rpm for 6 h at room temperature in a Sorvall AH650 rotor. Hybridization and DNA probes. For DNARNA filter hybridization studies, purified plasmid DNA was linearized and labeled with 32P by a modification of the oligo labeling method [29,30]. The DNA in water in a total volume of 37 ~1 was first denatured in a boiling water bath for 7 min. Then after the DNA had been rapidly cooled to room temperature, 10 ~1 of OLB buffer [OLB is a mixture of solutions A:B:C at 2:5:3 ratio. Solution A is 1 ml Solution 0, 18 ~1 P-mercaptoethanol. Solution 0 is 1.25 M Tris-HCl (pH 8.0), 0.125 M MgCl,. Solution B is 2 M Hepes, pH 6.6. Solution C is hexadeoxyribonucleotides, 90 OD units/ml], 2 ~1 BSA (10 pg/ph Sigma), 1 ~1 of Klenow (New England Biolabs, 5 units/jd), and 5 pl of [cu-32P]dCTP (Amersham, a3000 Wmmol) were added to the DNA. The reaction was incubated at room temperature overnight. The reaction was stopped by addition of 200 ~1 of 10 mM Tris-HCI (pH 8), 1 mM EDTA. Unincorporated radioactivity was removed on Sephadex G-50. The DNA probe was then denatured in a boiling water bath for 7 min and quenched on ice for a minimum of 5 min before hybridization. The filter was prepared for hybridization by soaking for a minimum of 6 h at 45°C in 50% formamide, 5X SSC, 50 mMNa,HPO, (pH 7.0), 0.2% SDS, and 100 pg/ml highly polymerized yeast tRNA (CalBiothem Cat. No. 56712). The DNA probe was then added to the filter and hybridization was conducted at 45°C for at least 16 h. Nonspecific binding of the DNA probe to the filter was eliminated by four lo-min washes in 2~ SSC, 0.1% SDS at 45”C, followed by two 20-min washes in 0.2X SSC, 0.1% SDS at 60°C. Removal of specific hybridization from filters. The first radioactive probe was removed from the filter by washing the filter two times in 0.01X SSC, 0.03% SDS at 85°C for 15 min. Removal of the original probe was monitored by autoradiography.

ESULTS

Detection of Splicing Thermotolerance 38°C Incubation at 25°C

Requires a Post

When Yost and Lindquist [9] conducted their experiments on splicing thermotolerance, they reported on a fixed set of experimental conditions. To make the cells splicing thermotolerant, the cells were first subjected to

STABILIZES

PRE-mRNA

235

a 35°C heat shock for 0.5 h. The cells were then recovered for 3 h at their normal growth temperature Of 25’C before they were challenged at 38°C fOr 0.5 h. The extent of lnRNA splicing was not tested at .&is time but instead the cells were returned to 25°C for 1 h before mRNA processing was investiga’ In the experiments presented b we have examined splicing under a wide range of co tisns, providing us splicing thermotolerwas extracted for “C, there was accumulation of pre-mRNA and no e ence Of splicing thermotolerance (Fig. 1, lane 9). wever, if the hour at 25°C after the 38°C severe st s was included, there was no detectable pre-mRNA (Fig. 1, lane 11). The &our at 25°C after the 38°C treat required for the cells to display splici ante, could reflect the need for either an a 38°C incubation time or a lower temperature. To address this question, the amount of time the cells were incubated at 38°C was varied from 9.5 2 h. The 25°C (post 38°C) treatment was varied from result of this experiment is shown in Fig. 1. In the absence Of a return to 25°C there is always an accumu tion of pre-mRNA (see Fig. 1, lanes 3, 4, 6, and 9). general, the longer the the accumulation Of pre and 9). However, if ceils accumula of pre-mRNA pare lanes 4 and 5, lanes 6 an and Eanes 9 and 10). This is best seen in lanes 9 through 12 where the cel%s were beat shocked at 38°C for 0.5 h (lane 9) and then returned to 25°C fOr 0.5, 1, or I.5 h (Banes 19-E). It length of incubati clearly can be seen that as t at 25°C increases, the level a~c~rn~~~at~d pre decreases. This does not occur in cells that kave no retreated at 35°C (our Splicing

Thermotolerance

Activity

1s ~~~~~1~ Induced

Yost and Lindquist [9] returned ells to 25°C for 3 h after the initial pretreatment at before challenging them at 38°C and measuring splicing thermotslerante. In order to investigate the effect of the length the recovery time between the 35°C pretreatment. a the 38°C severe heat shock, a time course experiment was performed by varying the recovery time between the pretreatment and the severe heat shock from 0 to 24 h. The results of this experiment are shcrwn in Fig. 2. After a 0.5-h eatment, the cells were immediately fully splicing motolerant to a 0.5-b treatment at 38°C (Fig. 2, lane 4). The cells remain fully splicing thermotolerant for at least 4 h after the ~~‘etreatmen~ at 35°C (lanes 4-8). owever, by 5 h after the pretreat-

236

CORELL

AND

A -

-I

l n

2

= extract RNA

time at38”

38"

A

ment, the cells begin to show an accumulation of premRNA by the end of the 38°C treatment (lane 9). The amount of pre-mRNA that accumulates in these cells is still substantially less than that in cells that were not pretreated (lane 3). The amount of pre-mRNA that accumulates due to incubation at 38°C increases as the post 35°C time increases from 5 h (lane 9) to 8 h (lane 12). By 8 h after the pretreatment, the amount of premRNA that accumulates is the same as that in cells that were not pretreated.

Treatment

j-035"

1

11hour. 250 l

38"

2 hr

1.5 hr

Less than 10 min at 35°C Is Required to Achieve Full Splicing Thermotolerance to a 0.5-h Incubation at 38°C

6

7 8 9 10 11 3 hours

12 -)hr

e I

1

2

I

I

1

0

B UJI

.-I

0.5 3

2

3

4

time

5

6

7

I

I I (Ifours) 5

8

910

1

I

6

I

GROSS

8

1112

-

prwnRNA

-

mRNA

2.4 1.4-

FIG. 1. Analysis of time and temperature requirements for splicing thermotolerance. (A) A pictorial representation of the treatments of the cells before RNA was extracted. Lane 1,35”C for 0.5 h, lane 2, 38°C for 0.5 h followed by 1 h at 25°C; lanes 3 through 12 were all treated at 35°C for 0.5 h and returned to 25°C for 3 h before subsequent treatment. Subsequent treatments were: lane 3, 38°C for 2 h; lane 4,38”C for 1.5 h; lane 5,38”C for 1 h followed by 25°C for 0.5 h; lane 6,38” for 1 h; lane 7,38”C for 1 h followed by 25°C for 0.5 h; lane 8,38”C for 1 h followed by 25°C for 1 h; lane 9,38”C for 0.5 h; lane 10, 38°C for 0.5 h followed by 25°C for 0.5 h; lane 11, 38’C for 0.5 h followed by 25°C for 1 h; lane 12, 38°C for 0.5 h followed by 25°C for 1.5 h. (B) Northern gel analysis-The extracted RNA was fraction-

The previous experiment showed that as splicing thermotolerance activity declines, intermediate amounts of pre-mRNA accumulate when the cells are exposed to a severe 38°C heat stress. These cells are referred to as partially splicing thermotolerant. Another way of making the cells partially thermotolerant is to shorten the length of time the cells are pretreated at 35°C [9]. In their experiments, Yost and Lindquist [9] used a recovery time of 3 h between the 35°C pretreatment and the 38°C severe heat shock. Their results showed that heat pulses at 35°C for as short as 5 min could reduce the amount of accumulating pre-mRNA during the subsequent 0.5-h 38°C heat shock, and a pretreatment for 15 min virtually eliminated any accumulation of premRNA. In order to better characterize the time response and to see if the time period in between the pretreatment and the 38°C severe heat shock is critical, a more comprehensive experiment was conducted. The cells were pretreated at 35°C for 0,2.5,5,10,15,20,25,30,40, or 50 min followed by 0.25 h at 25”C, 0.5 h at 38°C and 1 h at 25”C, at which time RNAs were extracted for analysis. The results of this experiment are shown in Fig. 3. As seen in lane 4, a pretreatment as short as 2.5 min can substantially reduce the amount of pre-mRNA that accumulates during the 38°C incubation (densitometric scans indicate about a 45% reduction, data not shown). A pretreatment of 10 min virtually eliminates any premRNA that accumulates as a result of an incubation at 38°C (lane 6). These results show that splicing thermotolerance is rapidly induced. Splicing

Thermotolerance

Activity

Is Limiting

When splicing thermotolerance activity is less than maximal (e.g., 5-7 h after 35°C pretreatment or with

ated on a 1% agarose-formaldehyde gel. The RNA was next transferred to nitrocellulose and hybridized with a DNA probe consisting of a portion of the hsp83 gene containing both intron and exon portions.

SPLICING

THERMOTOLERANCE

STABILIZES

PRE-mRNA

237

Treatment

b B 25°C control 3.5” 2

3

25”

Kb

1

2

3

4

5

6

7

8

9101112

13

FIG. 2. Analysis of kinetics of loss of splicing thermotolerance. (A) A pictorial representation of the treatments of the cells before RNA was extracted. Lane 1,25”C; lane 2,35”C for 0.5 h; lane 3,38”C for 0.5 h followed by 1 h at 25’C; lanes 4 through 13 were all treated at 35°C for 0.5 h and returned to 25°C for varying lengths of time before they were subsequently incubated at 38°C for 0.5 b followedby 25°C for P h. The varying lengths of time in between the 35 and the 38°C treatments were: lane 4,0 h; lane 5, I h; lane 6,2 h; lane 7,3 h; lane 8: 4 h; lane 9,5 h; lane 10,6 h; lane 11,7 h; iane 12,8 h; lane 13,24 h. (B) Northern gel analysis-The extracted RNA was fractionated on a 1% agarose-formaldehyde gel. The RNA was next transferred to nitrocellulose and hybridized with a DNA probe consisting of a portion of the hsp83 gene containing both intron and exon portions. The data in this part of the figure are taken from two different Northern gels that best represenr. the results seen from several experiments.

238

CORELL

AND

GROSS

A 1 2

25°C

l = extract RNA

-35”

Time at 35°C 8” -

3

1 hour

0 min

4

2.5 min

5

5 min

6

10 mill

7

15 min

8

20 min

9

25 min

10

30 min

11

0

40 min 50 min

12

B

Kb

123456

7

8

9 10

11 12

pre-mRNA

EIRNA 2.4 -

FIG. 3. Analysis of kinetics of acquisition of splicing thermotolerance. (A) A pictorial representation of the treatments of the cells before RNA was extracted. Lane 1, 25°C; lane 2, 35°C for 0.5 h; lane 3, 38°C for 0.5 h followed by 1 h at 25°C; lanes 4 through 12 were all treated at 35°C for varying lengths of time and returned to 25°C for 15 min before they were subsequently incubated at 38°C for 0.5 h followed by 25°C for 1 h. The varying lengths of time at 35°C were: lane 4,2.5 min; lane 5,5 min; lane 6,lO min; lane 7,15 min; lane 8,20 min; lane 9,25 min; lane lo,30 min; lane 11,40 min; lane 12,50 min. (B) Northern gel analysis-The extracted RNA was fractionated on a 1% agarose-formaldehyde gel. The RNA was next transferred to nitrocellulose and hybridized with a DNA probe consisting of a portion of the hsp83 gene containing both intron and exon portions.

SPLICING

THERMOTOLERANCE

less than 10 min of 35°C pretreatment), cells treated at 38°C accumulate intermediate amounts of pre-mRNA. There are at least two models to explain these results. The first is that the factors responsible for splicing thermotolerance provide a general protection of the splicing machinery. When splicing thermotolerance factors are limiting, the cells are only partially splicing thermotolerant and the splicing machinery can only operate at a fraction of the normal splicing rate. At this reduced efficiency the splicing machinery can only process precursors at a reduced rate; i.e., 30,50, or 70% of the normal rate. In the second model, splicing thermotolerance factors are again limiting when the cells are partially splicing thermotolerant. However, in this model the splicing thermotolerance factors, instead of protecting the overall efficiency of the splicing process, protect individual spliceosomes or spliceosome components. In this case, once the supply of splicing thermotolerance factors is exhausted, newly synthesized premRNAs will not be spliced because newly formed spliceosomes can not be protected. The first model would predict that all partially splicing thermotolerant cells would begin to accumulate premRNA immediately at 38°C but they would accumulate pre-mRNAs at different rates based on the efficiency at which the splicing machinery is operating (i.e., the extent of splicing thermotolerance). On the other hand, the second model would predict that cells containing increasing levels of splicing thermotolerance factors would not begin to accumulate pre-mRNAs until increasingly longer times at 38°C. Further, the second model would predict that once the cells begin to accumulate pre-mRNA, the pre-mRNAs would accumulate at the same rate. To evaluate the models, a compound time course experiment was performed. Briefly, cells were pretreated at 35°C for 0.5 h followed by recovery at 25°C for 0,2,4, 6, or 8 h. After the various recovery periods, the cells were incubated at 38°C for 0, PO, 20, 30, 40, or 50 min following which they were incubated for 1 h at 25°C and the RNA was extracted. The results of the RNA blot analyses are seen in Fig. 4. As shown, the time required to begin accumulation of pre-mRNA is different in the sets of cells recovered for different lengths of time. In cells that have abundant splicing thermotoierance activity, i.e., 0 or 2 h after the pretreatment, the accumulation of pre-mRNA during the 38°C severe heat shock does not occur during the 50-min severe heat shock. However, as the time ofrecovery at 25°C increases, the time at 38°C required for preNA accumulation decreases, until 8 h of recovery, when the accumulation of pre-mRNA begins in the same amount of time as that in the cells which had not been pretreated. In Fig. 4, the line defined by the squares represents the accumulation of pre-mRNA in cells that were pre-

STABILIZES

239

PRE-mRNA

19-24

x-30

...-. 31-36

8 Hours

B

FIG. 4. An.alysis of compound time course of splicing thermotolerance. (A) A pictorial representation of the treatments of the cells before RNA was extracted. Times at 38°C are as indicated in B. After treatment at 38°C all cells were returned to 25°C for 1 h before the RNA was extracted for analysis. Lanes 1-6 were not pretreated at 35°C before 38°C treatment. Lanes 7-12 were pretreated at 35°C for 0.5 h and immediately transferred to 38°C. Lanes 13-18 were pretreated at 35°C for 0.5 h and returned to 25°C for 2 h before treatment at 38°C. Lanes 19-24 were pretreated at 35°C for 0.5 h and returned to 25°C for 4 h before treatment at 38’“C. Lanes 25-30 were pretreated at 35°C for 0.5 h and returned-to 25’C for 6 h before treatment at 38°C. Lanes 31-36 were pretreated at 35°C for 0.5 h and returned to 25°C for 8 h before treatment at, 38°C. (S) After exposure to X-ray film, the amount of Imp83 pre-mRNA in each lane was quantitated by scanning with a densitometer. The hsp83 gene probe was then removed from the nitrocellulose, the filter was reprobed with an 18s ribosomal RNA gene probe, and the resulting autoradiograph was scanned. The amount of hybridization to the rDNA probe was used to normalize the amount of pre-mRNA accumulation in each set of lanes.

240

CORELL AND GROSS

treated and incubated for 8 h at 25°C before the 38°C incubation-a time by which splicing thermotolerance has disappeared. Note that the lag period before the onset of pre-mRNA accumulation decreases as the cells become less splicing thermotolerant, but the rate of accumulation of pre-mRNA is approximately the same once pre-mRNA accumulation starts. These results are most consistent with the second model. DISCUSSION

Pre-mRNAs Made at 38°C in Splicing Thermotolerant Cells Are Not Spliced until the Cells Are Returned to 25°C In our experiments and those of Yost and Lindquist [9], when the RNA was not extracted for analysis until after a l-h 25°C incubation following the 38°C heat shock, but directly extracted after the 38°C incubation (i.e., no post 38°C incubation), no splicing thermotolerante can be detected. This indicates either that the cells might require a return to the 25°C temperature before they could process the pre-mRNA made at 38°C or possibly that the cells could require the extra time to process the pre-mRNA. As the results in Fig. 1 clearly show, the cells require the return to a lower temperature before processing the pre-mRNA. When the length of the incubation at 38°C is increased, the amount of pre-mRNA accumulating increases, roughly in proportion to the length of the incubation. However, as soon as the cells are returned to 25°C for a period as short as 0.5 h, the amount of detectable pre-mRNA decreased. These experiments clearly show that in Drosophila cells splicing, even in splicing thermotolerant cells, does not occur at severe temperatures. These results are consistent with the hypothesis that splicing thermotolerance factors bind to the premRNP, the splicing snRNPs, or both to stabilize a complex that holds the pre-mRNAs in a location or configuration so that when the severe stress is removed, the pre-mRNA can proceed through the splicing pathway. Pre-mRNAs that are made in the absence of splicing thermotolerance factors are not kept in the splicing pathway. Instead, these pre-mRNAs bypass the splicing pathway and exit the nucleus to the cytoplasm where they can be translated [lo]. Splicing Thermotolerance Transient in Nature

Is Rapidly

Induced and

The results of the time course experiments shown in Figs. 2 and 3 show that acquisition of full splicing thermotolerance activity is rapid and that the activity is transient in nature. Previous results had shown that less than 15 min at 35°C were required to induce full splicing thermotolerance for a 0.5-h incubation at 38°C

[9]. These results show that the time needed for induction of splicing thermotolerance at 35°C can be further reduced to less than 10 min as shown in Fig. 3 (extending the results of Yost and Lindquist [9]). In addition, the acquisition of splicing thermotolerance is not an allor-none phenomenon; increasing levels of splicing thermotolerance are obtained with increasing incubation times at 35°C up to about 10 min when complete splicing thermotolerance (to a 30-min 38°C incubation) is achieved. Interestingly, in trypanosomes, which have a 5’ transspliced leader added to each message, severe heat stress inhibits trans-splicing activity in genes that are expressed at normal temperatures [22], but does not affect the trans-splicing of the hsp70 mRNA [4].

In Partially Splicing Thermotolerant Cells, Pre-mRNAs Made Early during Severe Heat Stress Are Protected, but Those Made Later Are Not As shown in Fig. 4, the time to the onset of pre-mRNA accumulation during the 38°C incubation decreases as the interval between the 35 and the 38’C incubations increases. By 8 h the onset time for pre-mRNA accumulation at 38°C is approximately the same as that in the cells that have not been pretreated. Once pre-mRNA accumulation begins, the rate of accumulation during further incubation at 38” is approximately the same in all cells whether they had been pretreated at 35°C or not. These results strongly support the hypothesis that under partial splicing thermotolerance conditions, those pre-mRNA made early in heat shock are most likely to be protected in a splicing-competent state during the course of the 38°C heat stress. The pre-mRNAs made later during a 38°C heat stress (after the splicing thermotolerance activity has been depleted) are not protected. Another model consistent with these results is that in cells that are partially splicing thermotolerant, as the amount of pre-mRNA that has been made and needs to be protected exceeds the amount of splicing thermotolerance activity, those pre-mRNAs that are newly made displace those that were made first from the protected group. The pre-mRNAs that were made early in the 38°C treatment now leave the nucleus for the cytoplasm. Definitive proof of either of these models awaits a pulse-chase study.

A Model for Splicing

Thermotolerance

The processing of pre-mRNA to mRNA takes place in the nucleus. The lack of splicing at severe temperatures

SPLICING

THERMOTOLERANCE

is consistent with a loss of integrity of splicing snRNPs and an inability of these snRNPs to interact with the pre-mRNAs to form a functional spliceosomes [ll, 121. However, snRNPs from thermotolerant cells exposed to severe thermal stress retain their stability (Bond, personal communication; [El)* The results of Yost and Lindquist [9] using Sl nuclease analysis on pre-mRNAs synthesized at 38°C showed that these RNAs are blocked from splicing in nonthermotolerant cells before the first splicing step. These pre-mRNAs can be transported from the nucleus to the cytoplasm and are translated into protein [lo]. The size of the pre-mRNAs detected in our Northern analyses is consistent with pre-mRNAs that have not undergone the first step of splicing. This is true in premRNAs detected both in nonthermotolerant cells and in thermotolerant cells that have not been allowed to incubate at 25°C following the 38°C severe heat shock. This indicates that pre-mRNAs transcribed at 38°C in splicing thermotolerant cells are not spliced during the severe heat shock but instead are held in a splicingcompetent state that can be processed after the severe stress is removed. Bond [II] showed that when nonthermotolerant cells are exposed to severe heat stress the structur NPs is disrupted. However, in splicing ther cells that are exposed to a subsequent severe stress the snRNPs remain intact (Bond, communication). pe se a heat treatment (35°C) is needed to induce splicing thermotolerance, and the acquisition of splicing thermotolerance is prevented in the presence of cycloheximide ([9] and our data) or puromycin (our data, not shown), it is reasonable to suspect the involvement of one or mare of the beat shock proteins in splicing thermotolerance. 0f all of the heat shock proteins, only hsp70 has been shown definitively to have a nuclear localization following beat shock [31]. Furthermore, proteins related to the beat shock proteins, particularly those in the hsp70 gene family, collectively called the chaperonins, have been shown to have a function related to proper assembly of multicomponent complexes [32,33] and to be involved in renaturation of denatured proteins after heat shock [34]. We suggest that hsp703 facilitates splicing thermotolerance by stabilizing premRNPs, snRNPs, or snRNP components during the 38°C beat shock. The end of the 38°C stress leads to the

3 Although some studies have indicated that hsp70 appears to be locaiized in the nucPeolus following heat shock, this is not inconsistent with our model. The majority of RNA processing that takes place in the cell is rRNA processing, and if the flouresence studies are examined carefuliy there is evidence of a punctate pattern throughout the nucleus. This pattern is consistant with the data seen for the distribution of the spliceosomal snRNPs [35,36]. However, the signal is overwhelmed by the signal from the nucleolus.

STABILIZES

241

PRE-m

components from the splicing thermotolerance factors (or co~lfo~matio~a~ change of the splicing complex) wbic , in turn, allows pre-mRNAs to be processes. When splicing thermotolerant cells are exposed to a severe heat stress, splicing t Fame factors stabilizes the snRNPs (and/or interact with pre-mRNPs th vere stress. The interaction of s mg thermctolerance formation of an acfactors with snRNPs prevents tive spliceosome, but allows mation of a splicing rice the severe stress is removed, competent complex. the splicing thermotoierance factors dissociate from the snRNPs allowing the formation of active sphceosomes, and RNA processing proceeds. xpesiments are under way to test this model. This work was supported in part by Public Heaith Service Grant GM36057 to R.H.G. and by Dartmouth College BRSG funds. We thank Linda Friedlander, Rebecca Eustance, Jessica Riordan, Beth Sawin, Phii Rice, and Wynne Chen for their many hei-pful suggestions during the course of the work.

1. 2. 3. 4.

Grabowski, 3455353. Hernandez,

P. J., Seller, S. R., and Sharp, P. A. (1985) Cell 42, N., and Keller,

W. (1983) Cell 35,89999.

Krainer, A. R., Maniatis, T., Ruskin, B., and Green, M. R. (‘1984) Cell 36,993-1005. Muhich, M. L., and Boothroyd, J. C. (1989) J. Aiol. Chem. 264, 7107-7110.

5.

Padgett, R. A., Konarska, M. M., Grabowski: I?. J., Hardy, and Sharp, P. A. (1984) Science 225,898-903.

6.

Ruskin, B., Krainer, Cell 38, 317-331.

7.

Padgett, R. A., Grabowski, P. J., Konarska, M. M., Seiler, S., and Sharp, P. A. (1986) Annu. Ret;. &o&em. $6, 1119-1150.

A. R., Maniatis,

S. F.,

T., and Green, M. R. (1984)

8. 9. 10. 11.

Blobei, 6. (1985) hoc. NC&. Acad. Sci. USA 88,8527X3529. Yost, H. J., and Lindquist, S. (1986) Ce’eEl45, 185-193. Yost, H. J., and Lindquist, S. (1988) Science 24:2, 1544-1548. Bond, U. (1988) E2MBO J. 7, 35093518.

12.

Wright-Sandor, L. G., Reichlin, M., and Tobin, S. L. (1989) J: Cell Biol. 108, 2007-2016. Legrain, P., and Rosbash, M. (1989) CelE 67, 573-583.

13. 14.

Gallo, R. C., Wong-Staal, F., Montagnier, L., Haseltine, and Yoshida, M. (1988) Nature 333, 504.

15.

Ratner, L., Haseltine, W. A., Patarca, R., Livak, K. J., Starcicb, B., Josephs, S. F., Doran, E. R., Rafalski, J. A., Whitehorn, E. A., Baumeister, K., Ivanoff, L., Petteway, S. R., Jr., Pearson, M. L., Lautenberger, J. A., Papas, T. S., Ghrayeb, J., Chang, N. T., Gallo, R. C., and Wong-Staal, F. (1985) Nature 318, 277-284. Emerman, M., Vazeux, R., and Peden, K. (1989) Cell 87, i1551165. Chang, D. D., and Sharp, P. A. (1989) &El s99 789-795. Lindquist, S. (i980) Dev. Biol. 17, 463-469.

16. 17. 18.

W. A.,

242 19. 20. 21. 22.

CORELL Lindquist, S. (1981) Nature 294, 311-314. McGarry, T. J., and Lindquist, S. (1985) Cell 42, 903-911. Hightower, L. E., and White, F. P. (1981) J. Cell Physiol. 108, 261-275. Muhich, M. L., and Boothroyd, J. C. (1988) Mol. Cell. Biol. 8,

3837-3846. 23. Czarnecka, E., Nagao, R. T., Key, J. L., and Gurley, W. B. (1988) Mol. Cell. Bio2. 8, 1113-1122. Kay, R. J., Russnak, R. H., Jones, D., Mathia, C., and Candido, E. P. M. (1987) Nucleic Acids Res. 15,3723-3735. 25. Schneider, I. (1972) J. Embryol. Exp. Morphol. 27, 353-365. 26. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, 1st ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 27. Birnboim, H. C. (1983) Methods Enzymol. 100, 243-255. 24.

Received December 26, 1991 Revised version received June 3, 1992

AND

GROSS

28. Radloff, 29.

R., Bauer, W., and Vinograd, J. (1967) Proc. Natl. Acad. Sci. USA 57, 1514-1521. Feinberg, A. P., and B. Vogelstein (1983) Anal. Biochem. 132,

6-13. 30. Feinberg, A. P., and Vogelstein, B. (1984) Anal. Biochem. 13'7, 266-267. 31. Velazquez, J. M., and Lindquist, S. (1984) Cell 36, 655-662. 32. Ellis, R. J., and Hemmingsen, S. M. (1989) TIBS 14,339-342. 33. Flynn, G. C., Chappell, T. G., and Rothman, J. E. (1989) Science 245,385-390. 34. Skowyra, D., Georgopoulos, C., and Zylicz, M. (1990) Cell 62, 939-944.

35. Potashkin, 36.

J. A., Derby, R. J., and Spector, D. L. (1990) Mol. Cell. Biol. 10, 3524-3534. Spector, D. L. (1990), Proc. Natl. Acad. Sci. USA 87, 147-151.