Posttranscriptional Lipopolysaccharide Regulation of the Lysozyme Gene at Processing of the Primary Transcript in Myelomonocytic HD11 Cells This information is current as of September 26, 2021. Ralph Goethe and Loc Phi-van J Immunol 1998; 160:4970-4978; ; http://www.jimmunol.org/content/160/10/4970 Downloaded from

References This article cites 52 articles, 28 of which you can access for free at: http://www.jimmunol.org/content/160/10/4970.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 26, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 1998 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Posttranscriptional Lipopolysaccharide Regulation of the Lysozyme Gene at Processing of the Primary Transcript in Myelomonocytic HD11 Cells1

Ralph Goethe2 and Loc Phi-van3

Lysozyme is increasingly expressed in macrophages in inflammatory response to bacterial LPS. In this study, we investigated the mechanisms that control expression of the lysozyme gene in myelomonocytic HD11 cells activated by LPS. Nuclear run-on transcription assays showed that LPS caused a 15-fold increase in the transcription rate of the lysozyme gene. However, Northern analyses with lysozyme cDNA and intron sequences revealed that the LPS-induced increase in nuclear lysozyme transcripts greatly exceeded the increase in transcription rate. Furthermore, nuclear lysozyme transcripts in untreated cells with a t1/2 of <10 min were more unstable than those accumulated in LPS-activated cells. We suggested, therefore, that the increased lysozyme expres- Downloaded from sion following LPS treatment was largely due to a nuclear stabilization of the primary transcript. Interestingly, the increase in stability of the lysozyme primary transcript was accompanied by changes in nuclear processing including an increase in poly(A) tail length, which gradually shortened after entering the cytoplasm. The long lysozyme poly(A) tail, however, did not result in any increase in polysomal recruitment for translation or in stability of the cytoplasmic lysozyme mRNA. The Journal of Immunology, 1998, 160: 4970–4978. http://www.jimmunol.org/ ysozyme is an antibacterial protein produced mainly in sozyme gene. We show that the lysozyme gene in myelomonocytic the chicken oviduct and in macrophages. In the tubular HD11 cells was regulated at both the transcriptional and posttran- L gland cells of the oviduct, the expression of lysozyme is scriptional levels in inflammatory response to LPS. Interestingly, dependent on steroid hormones (1). During the differentiation of the posttranscriptional regulation mechanism is a multistep process macrophages, the lysozyme gene is continuously activated from a that involves changes in the transcript stability leading to a nuclear low level of expression in precursors such as myeloblasts to a high accumulation of unspliced and incompletely spliced transcripts, level of expression in mature macrophages. The lysozyme gene changes in poly(A) tail length, and subsequent cytoplasmic thus is a well-characterized marker for the myeloid lineage (2–5). poly(A) shortening of the completely spliced lysozyme RNA tran- Bacterial endotoxins such as LPS, in the acute phase of bacterial scripts. The altered processing of the lysozyme primary transcript by guest on September 26, 2021 infection, activate immunologic and inflammatory responses, par- involving the increase in poly(A) tail length may be associated ticularly in cells of immunologic systems including B and T lym- with an enhanced nuclear lysozyme RNA stability. phocytes and macrophages (6). Macrophages, in response to LPS, increasingly express a variety of cytokines and antibacterial pro- teins including lysozyme (7, 8). We have previously shown that Materials and Methods the LPS activation of the lysozyme gene in myelomonocytic cells Recombinant DNA plasmids was transcriptionally regulated by the myeloid-specific transcrip- The plasmid pcEPCAT5 contains the CAT gene under transcriptional con- 4 tion factor C/EBP␤ in interaction with the far upstream Ϫ6.1-kb trol of the lysozyme promoter and the Ϫ6.1-kb lysozyme enhancer (9). The lysozyme enhancer (9, 10). plasmid ptkNeo contains the neomycin resistance gene controlled by the In the present study, we have employed the chicken my- thymidine kinase promoter (Ϫ109/ϩ51) from the herpes simplex virus elomonocytic cell line HD11 (11) to investigate the mechanisms (12). Plasmids pBSlysI1 and pBSlysI2 were constructed by insertion of a HindIII fragment containing lysozyme intron 1 sequence and an XbaI frag- involved in the regulation of LPS-activated expression of the ly- ment containing intron 2 sequence into the HindIII site or XbaI site of the pBluescript SKϩ (Stratagene, Heidelberg, Germany), respectively. These DNA fragments resulted from subcloned DNA probes of ␭lys30 (13).

Institut fu¨r Tierzucht und Tierverhalten (FAL), Celle, Germany Received for publication August 11, 1997. Accepted for publication January 26, 1998. Cell lines and cell culture The costs of publication of this article were defrayed in part by the payment of page HD11 cells of an established chicken myelomonocytic cell line, trans- charges. This article must therefore be hereby marked advertisement in accordance formed by the v-myc encoding retrovirus MC29 (11), were maintained in with 18 U.S.C. Section 1734 solely to indicate this fact. Iscove’s modified Dulbecco’s medium, supplemented with 8% FCS, 2% 1 This work was supported by grants to L.P. from the Deutsche chicken serum, 100 U/ml penicillin, and 100 ␮g/ml streptomycin, at 37°C Forschungsgemeinschaft. in humidified 95% air, 5% C02. The clone pc5 contains one copy of the 2 Present address: Institut fu¨r Mikrobiologie und Tierseuchen, Bischofsholer Damm CAT gene controlled by the lysozyme promoter and the Ϫ6.1-kb lysozyme 15, 30173 Hannover, Germany. enhancer (9). This clone was established from HD11 cells by stable cal- 3 Address correspondence and reprint requests to Dr. L. Phi-van, Institut fu¨r Tierzucht cium phosphate cotransfection performed as described previously (12) us- und Tierverhalten (FAL), Do¨rnbergstr. 25–27, 29223 Celle, Germany. E-mail ad- ing 20 ␮g pcEPCAT5 and 2 ␮g ptkNeo followed by the selection for dress: [email protected] resistance to G418 (500 ␮g/ml). The copy number of the integrated CAT 4 Abbreviations used in this paper: C/EBP␤, CCAAT/enhancer binding protein ␤; kb, gene was determined by quantitative Southern blotting of genomic DNA kilobase(s); CAT, chloramphenicol acetyltransferase; GAPDH, glyceraldehyde-3- cut with HindIII and SacI and hybridization to a 539-bp BamHI-HindIII phosphate dehydrogenase; TEN, 40 mM Tris-HCl (pH 7.5)/1 mM EDTA/150 mM fragment containing the Ϫ6.1-kb enhancer (14) as described NaCl. previously (15).

Copyright © 1998 by The American Association of Immunologists 0022-1767/98/$02.00 The Journal of Immunology 4971

CAT assay Northern blotting. Blots were sequentially hybridized to nick-translated lysozyme cDNA and GAPDH cDNA as described above. pc5 cells were grown to a density of 5 ϫ 106 cells per 8.5-cm plate. For stimulation with LPS, cells were maintained in Iscove’s modified Dulbec- Run-on transcription assay co’s medium with 0.5% FCS for 48 h before they were activated with 5 ␮g/ml LPS from Salmonella typhimurium (Sigma, Deisenhofen, Germany) Nuclear run-on transcription assays were performed according to the pro- ϫ 7 for further 24 h. For extract preparation, cells were washed twice in PBS, cedure described by Greenberg and Ziff (22). Briefly, 1.5 10 cells were scraped by a rubber policeman in TEN containing 40 mM Tris-HCl (pH suspended in 1 ml of a Nonidet P40 lysis buffer containing 10 mM Tris- 7.5), 1 mM EDTA, and 150 mM NaCl, and pelleted in an Eppendorf HCl (pH 7.5), 10 mM NaCl, 3 mM MgCl2, 0.5% (v/v) Nonidet P40 and centrifuge for 10 s. The cell pellet was suspended in 0.25 M Tris-HCl (pH incubated at 4°C for 5 min. Nuclei were harvested by centrifugation at ϫ 7.8), lysed by three cycles of freeze-thawing, and cleared by centrifugation 500 g for 5 min at 4°C and then washed once with 2 ml of the Nonidet ␮ at 10,000 ϫ g for 10 min. Protein concentrations of cell extracts were P40 lysis buffer. After centrifugation, nuclei were suspended in 100 lof determined by the method of Bradfort (Bio-Rad, Mu¨nchen, Germany). a storage buffer containing 50 mM Tris-HCl (pH 8.3), 5 mM MgCl2, 0.1 Ϫ CAT assays were performed with 50 ␮g protein of each cell extract as mM EDTA, and 40% (v/v) glycerol and frozen at 80°C. For run-on previously described (12). transcription, nuclei were thawed, immediately mixed with an equal vol- ume of a run-on transcription buffer containing 10 mM Tris-HCl (pH 8.0), ␮ RNA isolation 5 mM MgCl2, 300 mM KCl, 0.5 mM each ATP, CTP, GTP, and 100 Ci 32 ϩ [␣- P]UTP (800 Ci/mmol) (DuPont NEN), and incubated for 30 min at Poly(A) RNA preparation was performed according to the method de- 30°C. RNA was isolated by treatment with 200 U of RNase-free DNase for ϫ 7 scribed by Rahmdorf et al. (16). Briefly, 1 10 cells were washed twice 15 min at 30°C, followed by incubation with 800 ␮g/ml proteinase K for with PBS, lysed in SSTE containing 10 mM NaCl, 0.5% (w/v) SDS, 20 45 min at 37°C, extraction with phenol-chloroform-isoamyl alcohol, and mM Tris-HCl (pH 7.5), 10 mM EDTA, and homogenized with a Janke & two cycles of precipitation, first with isopropanol and then with ethanol.

Kunkel Ultra-Turrax (Staufen, Germany). Following an incubation of the Downloaded from ␮ ϩ For binding on nylon membranes, isolated DNA fragments containing the lysate with 300 g/ml proteinase K for 30 min at 37°C, poly(A) RNA chicken lysozyme cDNA (18), the CAT gene (23), the c-rel cDNA (24), the was prepared by binding to oligo(dT) cellulose (Boehringer, Mannheim, ϩ GAPDH cDNA (19), or chicken genomic DNA were denatured with 0.2 N Germany) in the presence of 0.5 M NaCl. The bound poly(A) RNA was NaOH for 10 min at 37°C and slot-blotted in the presence of 0.125 ϫ SSC washed four times with a solution containing 10 mM Tris-HCl (pH 7.5), ϩ onto nylon membranes (Appligene) using a Hybri-Slot Manifold (Life 0.3 M NaCl, 5 mM EDTA, and 0.1% (w/v) SDS, and then poly(A) RNA Technologies). For analysis of the synthesized RNAs, the baked filters was eluated with deionized water. ϩ were hybridized to the purified labeled RNAs according to the method To isolate nuclear and cytoplasmic poly(A) RNAs, cells were washed developed by Church and Gilbert (20). twice in PBS and lysed in a cell lysis buffer containing 10 mM Tris (pH http://www.jimmunol.org/

7.5), 10 mM NaCl, 3 mM MgCl2, 0.5% (v/v) Nonidet P-40, 25 U/ml Polysome preparation RNasin (Promega, Heidelberg, Germany). Following incubation at 4°C for 5 min, nuclei were separated from the cytoplasm by centrifugation of the Polysome preparation was conducted as described previously (25) with ϫ 8 cell lysate at 10,000 ϫ g for 90 s, then washed twice with cell lysis buffer. some modifications. Briefly, 1 10 cells were washed twice with PBS, Cytoplasmic and nuclear poly(A)ϩ RNAs were then isolated by binding harvested, and lysed in 3 ml of a polysome buffer (PB) containing 20 mM ␮ with oligo(dT) cellulose as described above. Tris-HCl (pH 7.4), 10 mM MgCl2, 300 mM KCl, 4 g/ml polyvinyl sul- Total cellular RNA was isolated by using a RNeasy kit from Qiagen fate, 0.5% (v/v) Nonidet P40, 2 mM DTT, 25 U/ml RNasin (Promega), and ␮ (Hilden, Germany). To eliminate traces of genomic DNA, RNA samples 10 g/ml cycloheximide. Following centrifugation of the homogenate at ϫ (50 ␮g) were incubated with 15 U of RNase-free DNase in the presence of 12,000 g for 15 min, the postmitochondrial supernatant was layered over 15 U of RNasin (Promega) in 100 ␮l of a buffer containing 10 mM Tris- a 6-ml cushion of 35% (w/v) sucrose in PB, and the polysomes were col- lected by centrifugation at 150,000 ϫ g for2hat4°CinaBeckman SW40 by guest on September 26, 2021 HCl (pH 8.0), 50 mM KCl, and 1.5 mM MgCl2 at 37°C for 30 min. The ϩ samples were extracted twice with Roti-phenol (Roth, Karlsruhe, Ger- Ti rotor (Munich, Germany). Poly(A) RNAs from the polysome fraction many) and precipitated with ethanol, and RNA pellets were dissolved in and the postpolysomal supernatant were isolated and then subjected to 100 ␮l of deionized water. Northern blotting as described above. Northern analysis Immunoblotting 7 For Northern analysis, RNAs were denatured by 0.5 M glyoxal and 27% Cells were grown to a density of 10 cells per 8.5-cm plate and maintained (v/v) DMSO, electrophoretically fractionated through 1.4 to 1.8% agarose in Iscove’s modified Dulbecco’s medium with 0.5% FCS for 48 h. Fol- lowing washing twice with PBS, cells were incubated in 3 ml of the same gels containing 10 mM NaH2PO4 (pH 6.9), Southern-transferred to nylon membranes (Appligene, Heidelberg, Germany) (17), and immobilized by medium with or without LPS for the indicated times. At various time baking at 80°C for 30 min. Plasmid DNA containing the chicken lysozyme points, cell culture medium was transferred into a new tube, and cells were cDNA (18), intron 1 (pBSlysI1), intron 2 (pBSlysI2), or glyceraldehyde- washed twice with PBS, scraped in 0.8 ml TEN, harvested by centrifuga- 3-phosphate dehydrogenase (GAPDH) cDNA (19) was labeled with tion in an Eppendorf centrifuge for 15 s, and lysed in 250 ␮l of 0.25 M [␣-32P]dCTP (3000 Ci/mmol) (DuPont NEN, Bad Homburg, Germany) by Tris-HCl (pH 7.8) by four freeze-thaw cycles. Cell lysates were cleared by using a nick translation kit from Life Technologies (Eggenstein, Germany). centrifugation at 10,000 ϫ g for 10 min. The blots were prehybridized, hybridized, washed according to the method For immunoblotting, 2 ␮l of each cell lysate and 10 ␮l of each cell described by Church and Gilbert (20), and exposed to x-ray films with culture medium were separated on 12% SDS-polyacrylamide gels and elec- intensifying screens at Ϫ80°C. To quantify hybridization signals, autora- troblotted onto nitrocellulose membranes. Following blocking in NTT (150 diograms were scanned with an imaging densitometer from Bio-Rad. Ly- mM NaCl, 30 mM Tris-HCl (pH 8.0), and 0.05% Tween 20) containing sozyme mRNA levels were normalized for GAPDH on the same blot. 0.5% gelatin for 30 min at room temperature, the blots were incubated with a rabbit anti-lysozyme antiserum at a 1:1000 dilution in NTT for1hat Reverse transcription-polymerase chain reaction room temperature. After three 5-min washes in NTT, the blots were incu- bated with an alkaline phosphatase-conjugated anti-rabbit Ab (Dianova, RT-PCR of total RNA was performed by using a Titan One tube RT-PCR Hamburg, Germany) diluted 1:5000 in NTT. Following washing twice in kit from Boehringer Mannheim. Briefly, the cDNA reaction was conducted NTT, bound alkaline phosphatase was visualized by incubating the blots with 0.5 or 1 ␮g of total RNA at 50°C for 30 min followed by a PCR with with 100 ␮g/ml nitro blue tetrazolium, 50 ␮g/ml 5-bromo-4-chloro-3- 43 cycles (each 1 min at 94°C and 1 min at 60°C) using the primer pair indolylphosphate in 100 mM Tris-HCl (pH 9.5), 50 mM NaCl, and consisting of P1 (5Ј-ACGACACTGGCAACATGAGG-3Ј) and P2 (5Ј-AT 5 mM MgCl . TCCAACATCACGCAGACC-3Ј). PCR products were fractionated on 2% 2 agarose gels at 75 V for 3 h, stained with ethidium bromide, and quantified using an ethidium bromide gel documentation system from Bio-Rad. Results LPS induction of an accumulation of lysozyme pre-mRNAs in Polyadenylation analysis with RNase H myelomonocytic HD11 cells ϩ To remove poly(A) tails, 4-␮g poly(A) RNA samples were hybridized to Treatment with LPS caused increased expression of various genes ␮ 0.6 g of oligo(dT)12–18 (Pharmacia, Freiburg, Germany), and digested with RNase H as described by Kleene et al. (21). The samples were ex- in macrophages. Therefore, we examined the effect of LPS on tracted twice with Roti-Phenol (Roth) and precipitated with ethanol. RNA expression of the lysozyme gene in myelomonocytic HD11 cells. pellets were dissolved in 4 ␮l of deionized water and then subjected to Cells were cultured in the presence of LPS from S. typhimurium at 4972 LPS REGULATION OF THE LYSOZYME GENE

FIGURE 3. LPS stimulates expression of a stably integrated CAT gene controlled by the lysozyme promoter and the Ϫ6.1-kb lysozyme enhancer in pc5 cells. CAT assays were performed as described in Materials and FIGURE 1. LPS stimulates the accumulation of lysozyme pre-mRNAs Methods. in HD11 cells. Poly(A)ϩ RNAs isolated from cells activated with LPS for Downloaded from the indicated time periods were electrophoretically fractionated on a 1.4% agarose gel, Southern blotted, and sequentially hybridized to 32P-labeled moter and the Ϫ6.1-kb lysozyme enhancer in HD11 cells (9). To lysozyme cDNA and GAPDH cDNA; the blot was autoradiographed for 3 days and 1 day, respectively. RNA sizes were estimated using RNA stan- investigate the transcriptional control of the lysozyme gene ex- dards from Life Technologies. The long arrow indicates the position of the pression, the rate of its transcription in untreated and in LPS-ac- mature lysozyme mRNA. tivated HD11 cells and pc5 cells was determined in a run-on tran- scription assay. pc5 of the HD11 cell line harbors a stably http://www.jimmunol.org/ integrated CAT gene controlled by the lysozyme promoter and the 5 ␮g/ml for 10, 20, 30, and 60 min, and the levels of accumulated Ϫ6.1-kb lysozyme enhancer. Nuclei isolated from untreated cells lysozyme poly(A)ϩ RNA were determined by Northern analysis and from cells activated with LPS for 30 min, 1 h, and 5 h were using the full-length lysozyme cDNA (18). Figure 1 shows that in incubated for 30 min with [␣-32P]UTP to elongate nascent RNA contrast to control HD11 cells that expressed weakly the lysozyme transcripts. The 32P-labeled RNA was isolated and then hybridized gene, HD11 cells activated by LPS exhibited a marked increase in to immobilized DNA fragments specific to the lysozyme, c-rel, the level of lysozyme RNA transcripts (33- and 255-fold after 30 and GAPDH gene. The isolated 32P-labeled RNA from pc5 was hybridized to the CAT gene. Figure 2 shows that the basal tran- and 60 min of LPS treatment, respectively). In addition to the by guest on September 26, 2021 mature lysozyme mRNA, three lysozyme RNA transcripts were scription of the lysozyme gene in untreated HD11 cells was low, seen, the largest transcript at 3.9 kb appearing as early as 20 min but markedly increased ϳ15-fold after exposure to LPS for 30 after addition of LPS; the other two, at 2.1 and 0.8 kb, were sub- min; this increase remained nearly constant during the indicated sequently observed within 30 and 60 min of LPS stimulation, re- time periods. The c-rel gene was thought to be stimulated by LPS. spectively. In contrast, the mature lysozyme mRNA appeared to be Here, the rate of transcription determined under the same condi- constant during the incubation time periods. The results demon- tions was clearly increased (6-fold) after stimulation with LPS. strated that LPS stimulated an accumulation of lysozyme pre- The transcription of the GAPDH gene, on the other hand, was not mRNAs in HD11 cells. influenced by LPS, consistent with the steady state levels of GAPDH mRNA in HD11 cells. pc5 exhibited a 17-fold increased LPS stimulates the transcription of the lysozyme gene CAT activity in response to LPS (Fig. 3), and this increase was We have previously shown that LPS activated expression of a clearly due to a 15-fold enhanced rate of transcription of the stably transiently transfected CAT gene controlled by the lysozyme pro- integrated CAT gene and thus consistent with our previous results obtained from transient transfection experiments (9). Taken to- gether, these data show that the lysozyme gene was transcription- ally activated by LPS.

Evidence for an increase in lysozyme transcript stability in LPS-activated cells Using this run-on transcription assay under the same conditions, we have previously shown that LPS induced an 3.5-fold increase in C/EBP␤ transcription rate that reflected the 3.2-fold increase in ␤ steady state level of C/EBP mRNA with a t1/2 of 1 h (10). Thus, the run-on transcription used in our experiments reflected in vivo gene transcription. However, the 15-fold increased lysozyme gene FIGURE 2. LPS stimulates the rate of lysozyme transcription. Nuclei transcription induced by LPS cannot fully account for the LPS- were isolated from untreated cells (Ϫ) or from cells activated with LPS for induced lysozyme gene expression. This induction exhibited a 33- 30 min, 1 h, and 5 h and incubated in a run-on transcription assay with [␣-32P]UTP for 30 min to elongate RNA transcripts as described in Ma- fold and a 255-fold increase over the basal steady state level of terials and Methods. RNAs were isolated, and equal amounts of radio- lysozyme mRNA after 30 and 60 min of LPS treatment, respec- laleled RNAs were hybridized to cDNA fragments specific to lysozyme, tively (Fig. 1). In fact, the LPS-induced increase in lysozyme tran- c-rel, GAPDH, and CAT and separated to chicken genomic DNA immo- scripts exceeds the increase in transcription rate even more if the bilized by slot blotting on nylon membranes. levels of LPS-induced pre-mRNAs are compared directly with The Journal of Immunology 4973

of lysozyme transcripts. Thus, not only the increase in lysozyme transcription, but also the nuclear stabilization of LPS-induced ly- sozyme primary transcript accounted for the large accumulation of lysozyme mRNA in LPS-activated cells.

Delayed, sequential splicing of the lysozyme primary transcript in LPS-activated HD11 cells The chicken lysozyme gene contains three intron sequences (1, 2, and 3) of about 1270, 1810, and 79 bp, respectively (see Refs. 3, FIGURE 4. Analysis of nuclear lysozyme transcript stability. Untreated 13, and 26, and Fig. 6). To determine whether the three larger cells and cells activated with LPS for 30 min were treated with actinomycin lysozyme transcripts at 3.9, 2.1, and 0.8 kb were unspliced or D (Act. D) for 10, 20, 30, and 60 min. Total RNAs were isolated and ϩ treated with RNase-free DNase to eliminate genomic DNA. RT-PCR was partially spliced lysozyme RNA, Northern blots of poly(A) performed with 1 ␮g total RNA from untreated cells or 0.5 ␮g RNA from RNAs from LPS-activated cells were hybridized to lysozyme LPS-activated cells. PCR products, resolved on a 2% agarose gel, were cDNA and intron sequences of the lysozyme gene. As shown in stained with ethidium bromide, and the intensity of the 248-bp band was Figures 5 and 6, hybridization to intron 1 revealed two transcripts quantified by densitometric scanning. DNA marker fragments were a at 3.9 and 2.1 kb, whereas hybridization to intron 2 showed only 100-bp DNA ladder. the largest transcript at 3.9 kb. This 3.9-kb band should represent

the primary transcript because it had a size expected from the Downloaded from published sequence of the lysozyme gene (13, 26). Fractionating those of newly synthesized lysozyme pre-mRNAs in untreated ϩ poly(A) RNA from cells activated by LPS for 30 min on a 1.8% cells during the same time period. Since the lysozyme mRNA was agarose gel revealed that the RNA band at 2.1 kb consisted of two stable with a t of ϳ9 h (see below), and the discrepancy between 1/2 transcripts (see Fig. 8, lanes 3 and 4). Because intron 3 contains the LPS-induced increases in the lysozyme gene transcription rate only 79 bases, it is most likely that the upper band at 2.1 kb con- and in the levels of lysozyme transcripts was already observed

tained introns 1 and 3, whereas the lower band carried only intron http://www.jimmunol.org/ early within 60 min of LPS treatment, we suggested a nuclear 1. Interestingly, we never found transcripts containing both intron co-/posttranscriptional process affecting the lysozyme transcript 2 and 3, suggesting that intron 1 cannot be spliced in the presence stability. of intron 2. Thus, these results demonstrated a splicing pathway in To evaluate this suggestion, we employed RT-PCR to estimate which the intervening sequences, in contrast to most intron se- the half-life of nuclear lysozyme transcripts in untreated and in quences, were sequentially and stepwise unusually slowly spliced LPS-activated HD11 cells after inhibiting transcription with acti- from the primary transcript. The hybridization pattern presented in nomycin D. First-strand lysozyme cDNA, synthesized by reverse Figure 5 shows clearly that intron 2 was the first and intron 1 was transcription using primer P2, antisense to the 5Ј end of lysozyme the last to be removed from the lysozyme primary transcript.

intron 1, was derived only from lysozyme intron 1-containing by guest on September 26, 2021 RNA, but not from mRNA. PCR performed with this cDNA and two primers, P1, sense to the 5Ј end of exon 1, and P2, should yield LPS induces increases in poly(A) tail length of the lysozyme a diagnostic 248-bp DNA fragment for nuclear lysozyme tran- RNA transcripts scripts. In contrast to the Northern analysis data (Fig. 1), nuclear The hybridization pattern presented in Figures 1 and 5 shows lysozyme transcripts were detected by RT-PCR not only in LPS- that the LPS-induced lysozyme RNA transcript at 0.8 kb was activated cells, but also in control untreated cells. The levels of completely spliced, but was still larger than the mature ly- transcripts in untreated cells, however, decreased rapidly in the sozyme mRNA from untreated HD11 cells, indicating a further presence of actinomycin D, in contrast to those detected in LPS- step in the posttranscriptional processing of the LPS-induced activated cells (Fig. 4). A densitometric scanning of the results lysozyme RNA transcripts. Therefore, this step was investi- Ͻ revealed a t1/2 of 10 min for nuclear lysozyme transcripts in gated after inhibiting transcription with actinomycin D, an in- ϳ untreated cells and a t1/2 of 30 min for those in LPS-activated hibitor of RNA polymerases. Before the actinomycin D treat- cells. These results indicated that LPS alters the rate of degradation ment, cells were activated with LPS for 30 min. As shown in

FIGURE 5. The splicing pathway of the lysozyme primary transcript in LPS- activated HD11 cells. Poly(A)ϩ RNAs were isolated from cells activated with LPS for 0, 0.5, 2, and 6 h, fractionated on 1.4% agarose gels, and Southern blotted onto nylon membranes. The blots were hybridized to lysozyme cDNA, intron 1, and intron 2 and were subsequently stripped and rehybridized to GAPDH cDNA. The arrow indicates the position of the mature lysozyme mRNA. 4974 LPS REGULATION OF THE LYSOZYME GENE

FIGURE 6. Schematic representation of the splicing pathway of lysozyme primary transcript. Downloaded from

Figure 7, A and B, the lysozyme 0.8-kb RNA transcript ap- To examine the polysomal recruitment of lysozyme mRNA for peared first after 40 min of exposure to actinomycin D, and translation, postmitochondrial supernatants from LPS-activated intron 1 was more slowly removed than intron 2 and 3. Inter- HD11 cells were fractionated by centrifugation through 35% su- estingly, a continuous shortening of the 0.8-kb RNA transcript crose gradients. Polysomal mRNA from the pellet and nontrans- was observed throughout 15 h of actinomycin D addition. After lating mRNA from the supernatant were isolated and analyzed by http://www.jimmunol.org/ 15 h, the size of the 0.8-kb RNA transcript was similar to that Northern blotting. Northern blots probed to the lysozyme cDNA, of the mature lysozyme mRNA in untreated cells (Fig. 7B). presented in Figure 9, showed that all lysozyme mRNAs were Several studies have demonstrated posttranscriptional regulation associated with polysomes irrespective of their poly(A) tail length. involving changes in poly(A) tail length of completely spliced Thus, the polysomal recruitment of lysozyme mRNA for transla- RNAs (27–29). To determine whether the difference in size of the tion did not change after exposure to LPS. LPS-induced 0.8-kb RNA transcript and the mature lysozyme Next, to determine whether the increase in poly(A) tail length of mRNA may be due to a difference in their poly(A) tail length, we the lysozyme mRNA might be associated with the message sta- used RNase H and oligo(dT)12–18 to selectively digest the poly(A) ϩ bility, we measured the decay of lysozyme mRNA in the absence

tails. Poly(A) RNAs from untreated, from LPS-activated or from by guest on September 26, 2021 of transcription. HD11 cells were left untreated or activated with LPS- and actinomycin D-treated HD11 cells were hybridized to LPS and then incubated with 5 ␮g/ml of actinomycin D for 2 to oligo(dT)12–18 and subsequently digested with RNase H, which specifically cleaves RNA in RNA:DNA hybrids. The deadenylated 10 h, and the levels of lysozyme message were determined by RNAs were analyzed by Northern blotting and hybridization to the Northern analysis. Unfortunately, a longer than 10-h incubation lysozyme cDNA. As shown in Figure 8, when poly(A) tails were with actinomycin D at this concentration was very toxic to the cells completely digested, the size difference between the LPS-induced (data not shown). Actinomycin D at a concentration of 5 ␮g/ml has 0.8-kb RNA transcript and the mature lysozyme mRNA was elim- been previously shown to be able to inhibit transcription in HD11 inated, indicating that their poly(A) tail lengths were quite differ- cells (10) (see also Fig. 7). Figure 10 shows that the lysozyme ent. These data confirmed that LPS induced increases in the mRNA was found to be stable in untreated cells and in LPS-acti- ϳ poly(A) tail length of the lysozyme RNA transcripts and that the vated cells; panel B shows a t1/2 of 9 h for lysozyme mRNA in progressive shortening of LPS-induced lysozyme mRNA was due both untreated and LPS-activated HD11 cells. to a reduction in poly(A) tail length occurring in a 3Ј-to-5Ј direc- tion. Figure 7B shows further that the poly(A) shortening of LPS- induced lysozyme mRNA occurred slowly throughout 15 h after Nuclear localization of the unspliced and incompletely spliced actinomycin D addition. After this time period, the LPS-induced lysozyme RNA transcripts mRNA still contained a shortened but not completely removed Next, we determined the subcellular localization of the accumu- poly(A) tail, which corresponded to the mature, uninduced ly- lated lysozyme pre-mRNAs induced by LPS. Cytoplasmic and nu- sozyme mRNA. Longer incubation with actinomycin D, up to ϩ clear poly(A) RNAs were fractionated from LPS-activated acti- 24 h, did not cause a decrease in poly(A) tail length (data not nomycin D-treated HD11 cells; the subcellular distribution of the shown), indicating that the process of deadenylation could stop intron-containing lysozyme pre-mRNAs was analyzed by North- within 15 h of actinomycin D treatment. Interestingly, the poly(A) shortening was observed only from the LPS-induced mRNA, but ern blot hybridization to the lysozyme cDNA. A densitometric never from the uninduced lysozyme mRNA during this time. scanning of the autoradiogram presented in Figure 11 revealed that Posttranscriptional regulation of gene expression occurs fre- most of the intron-containing lysozyme transcripts at 3.9 kb and quently by changing the cytoplasmic message stability affecting 2.1 kb were, as expected, accumulated in the nucleus, whereas steady state levels of specific mRNAs (31, 32). Furthermore, the Ͼ90% of the 0.8-kb transcripts were located in the cytoplasm. This increase in poly(A) tail length can influence the translation rate of implied that the poly(A) tails of LPS-induced lysozyme mRNA specific mRNAs (33–36). Therefore, we examined whether the began to shorten following translocation into the cytoplasm, while increase in poly(A) tail length of lysozyme mRNA was accompa- the change in poly(A) tail length occurred, on the other hand, in the nied by changes in the translational efficiency or in the message nucleus. Rehybridizing the blot to GAPDH cDNA demonstrated stability. the cytoplasmic localization of the mature GAPDH mRNA. The Journal of Immunology 4975 Downloaded from

FIGURE 8. Analysis of poly(A) tails on lysozyme pre-mRNAs and ma- ture mRNA. Poly(A)ϩ RNAs from untreated cells (Ϫ) or from cells acti- vated with LPS for 30 min and subsequently treated with actinomycin D http://www.jimmunol.org/ (Act. D) for the indicated time periods were subjected to RNase H diges-

tion following hybridization to oligo(dT)12–18. The resulting deadenylated RNAs were analyzed by sequential Northern hybridization to lysozyme and GAPDH. The two arrows indicate two RNA bands after digestion with RNase H, and the long arrow indicates the position of the mature lysozyme mRNA.

Discussion by guest on September 26, 2021 LPS has been shown to activate expression of various genes. The activation is regulated in many cases at the transcriptional level by various transcription factors such as nuclear factor ␬B (NF-␬B) (37–39), activator protein 1 (AP-1) (40), or C/EBP␤ (9, 10, 41). Our previous study on the LPS activation of the lysozyme gene in FIGURE 7. Northern blot analysis of the poly(A) tail shortening of myelomonocytic HD11 cells indicated a transcriptional control of LPS-induced lysozyme mRNA. HD11 cells were activated with LPS for the lysozyme gene regulated by LPS, which contributed to an ac- 30 min and then treated with 5 ␮g/ml actinomycin D (Act. D) for the cumulation of the lysozyme mRNA (9). In this detailed study, we ϩ indicated time periods. Poly(A) RNA were isolated, and 4 ␮g were show that both transcriptional and posttranscriptional mechanisms electrophoretically fractionated on 1.8% agarose gels and blotted onto are involved in the regulation of the lysozyme gene in myelomono- nylon membranes. The blots were sequentially hybridized to lysozyme cytic cells in response to LPS. Analyses of nuclear and cytoplas- and GAPDH. The arrow indicates the position of the mature lysozyme mic RNAs demonstrated that the posttranscriptional regulation mRNA. was a multistep process including a transient accumulation of the pre-mRNAs with an intranuclear elongation of poly(A) tail length

LPS stimulates lysozyme synthesis in myelomonocytic HD11 cells To examine whether LPS-induced lysozyme mRNA is functional and translatable, we performed Western blot analysis for ly- sozyme. HD11 cells were left untreated or activated with LPS. After 1 to 9 h, cell culture media and cell lysates were subjected to Western blot analysis using an antiserum against chicken ly- sozyme. As shown in Figure 12, production and secretion of ly- FIGURE 9. Analysis of polysomal recruitment of lysozyme mRNA. sozyme were increased following stimulation with LPS. Since ly- HD11 cells were activated with LPS for 30 min or 6 h and then treated with or without actinomycin D (Act. D) for 3 and 15 h. Postmitochondrial su- sozyme was undetectable in control untreated cells, the maximal pernatants were fractionated by 35% sucrose gradients into polysomal frac- fold induction by LPS could not be determined. The increase in tions (P) and postpolysomal supernatants (S). Poly(A)ϩ RNA was isolated lysozyme production seemed to correlate temporally with the in- from each fraction and analyzed by Northern hybridization to lysozyme crease in lysozyme mRNA after LPS treatment. cDNA. 4976 LPS REGULATION OF THE LYSOZYME GENE Downloaded from

FIGURE 11. Subcellular localization of lysozyme pre-mRNAs. HD11 cells were activated with LPS for 30 min and then treated with actinomycin http://www.jimmunol.org/ D (Act. D) for the indicated time periods. Cytoplasmic and nuclear poly(A)ϩ RNAs were isolated and analyzed by Northern hybridization to lysozyme and GAPDH.

transcription. Furthermore, the splicing of intron 1 can occur after the splicing of intron 2 and 3 has been completed (Figs. 5 and 6). Thus, the order of the splicing did not occur in a linear sequence FIGURE 10. Analysis of cytoplasmic lysozyme mRNA stability. A, Un- from the 5Ј or 3Ј end of the primary transcript. As expected, all treated HD11 cells or LPS-activated cells were treated with actinomycin D intron-containing lysozyme pre-mRNAs were retained in the nu- by guest on September 26, 2021 (Act. D) for 2, 4, 6, 8, and 10 h. Total RNAs were isolated. 1/5 and 1/25 cleus. After the splicing had been completed, the lysozyme mRNA of RNA from untreated cells and LPS-activated cells, respectively, were analyzed by Northern hybridization to lysozyme cDNA. The blots were was quickly transported from the nucleus into the cytoplasm (Fig. exposed to x-ray films for 6 days and for 12 h, respectively. Subsequently, 11), indicating that splicing complexes prevent the export of un- the blots were stripped and rehybridized to GAPDH cDNA. Data shown spliced precursor RNAs from the nucleus. This direct competition are from a representative experiment. B, Stability of lysozyme mRNA in between the processes of splicing complex formation and nuclear untreated cells (ᮀ) and LPS-activated cells (f) at various times after ex- export has been reported by Legrain and Rosbash (43) and Chang posure to actinomycin D. Autoradiograms were analyzed by densitometric and Sharp (44). In fact, mutated introns that were not recognized scanning. For each curve, the values at 2, 4, 6, 8, and 10 h were expressed as percentage of the value at time zero. Data shown are from two inde- pendent experiments. and a subsequent cytoplasmic partial deadenylation of the com- pletely spliced mRNA. Primary transcripts of most eukaryotic genes have been shown to be spliced very quickly immediately after the transcription has been completed before leaving the nucleus to enter the cytoplasm. Recently, a number of studies have provided evidence that splicing can occur cotranscriptionally through binding of pre-mRNA splic- ing factors with the C-terminal domain of RNA polymerase II (for a review, see Ref. 42). In contrast, splicing of the lysozyme pri- mary transcript in LPS-activated HD11 cells was extremely slow, and the intron sequences 1, 2, and 3 have been shown to be com- pletely removed at 3 h after transcription. Interestingly, the ex- tremely slow splicing of the lysozyme gene primary transcript en- abled us to employ Northern analysis using intron sequences to FIGURE 12. LPS stimulates lysozyme production. HD11 cells were investigate its splicing pathway. In this pathway, the intron se- treated with and without LPS for the indicated time periods. Following quences carried in the primary transcript were sequentially spliced removal of the cell culture media, cells were disrupted, and cell lysates in the following order: intron 2 was the first and intron 1 was the were isolated. Two microliters of each cell lysate (A) and 10 ␮l of each cell last to be removed. Intron 2 and 3 were spliced out within 40 min, culture medium (B) were subjected to Western analysis using an antiserum whereas intron 1 was completely removed after 3 h following the against chicken lysozyme. The Journal of Immunology 4977 by the splicing machinery and not spliced were able to leave the ening for at least 15 h. It is possible that this process is a prereq- nucleus (43, 44). uisite to the lysozyme mRNA decay and a consequence of the There is evidence indicating that a great part of eukaryotic pri- poly(A) tail length increase induced by LPS. Perhaps the increased mary transcripts is quickly degraded after transcription, whereas poly(A) tail has to shorten to reach normal size before the normal only a minor part can undergo the nuclear processing to become degradation pathway of the lysozyme mRNA can occur. Thus, the mature mRNAs before leaving the nucleus (45). Posttranscrip- poly(A) tail shortening seems to be a control mechanism involved tional regulation affecting the stability of unspliced transcripts has in the regulation of the lysozyme message stability. already been suggested (46, 47). In the present study, we investi- Poly(A) tails have been shown to be involved in translational gated mechanisms involved in the accumulation of lysozyme pre- efficiency (53). Although the polysomal recruitment of the ly- mRNAs in the nucleus following LPS treatment. Our data dem- sozyme mRNA was shown to be independent of the poly(A) tail onstrating that LPS activated transcription of the lysozyme gene length, the possibility that long poly(A) tails are nevertheless more also indicated that an intranuclear stabilization of the lysozyme suited for efficient translation than short poly(A) tails can not be primary transcript accounted, in part, for the nuclear accumulation excluded. Directly determining the translation rate of the lysozyme of lysozyme transcripts, thus leading to a considerable increase in mRNA with different poly(A) tail lengths may better reveal dif- cytoplasmic lysozyme mRNA level. The increased stability of the ferences in their translatability. nuclear transcripts, however, was clearly associated with LPS-ac- In a recent study, Huang and Carmichael (54) have provided tivated transcription of the lysozyme gene and seemed to be caused evidence for a role of poly(A) tails in transport of RNA across the by the altered processing of lysozyme pre-mRNAs, because simul- nuclear membrane. The nucleocytoplasmic transport is energy de- Downloaded from taneous treatment with LPS and actinomycin D failed to activate pendent and saturable and thus a carrier-mediated process (55). accumulation of the lysozyme pre-mRNAs (data not shown). In- Perhaps this process can be facilitated by altered poly(A) tail terestingly, the splicing process of the lysozyme primary transcript length, particularly when a great amount of mRNA must be ex- occurred very slowly in LPS-activated cells. At present, we are not ported within a short time period, for example, following LPS able to detect this splicing pattern in untreated cells, although tran- treatment. scription of the lysozyme gene in these cells was clearly detectable In summary, our study demonstrates a posttranscriptional mul- by a run-on transcription assay. Indeed, Northern hybridization to tistep process of LPS-regulated lysozyme gene expression. The http://www.jimmunol.org/ the lysozyme intron sequences 1 and 2 did not reveal any signal exact mechanism by which the stability of the primary transcript even after very long autoradiography. Thus, it is possible that the was increased and the role of each step remain to be investigated duration of the splicing process, but not the splicing itself, may be in a further study. altered by LPS. Alternatively, the splicing process of the lysozyme primary transcript in untreated cells was the same in LPS-activated Acknowledgments cells, but the lysozyme primary transcript could not be detected We thank K. Zimmermann for skilful technical assistance, because of rapid degradation. D. Wulf for RT-PCR, Prof. G. Mieskes (Max-Planck-Institut fu¨r Bio-

Interestingly, the increase in lysozyme transcript stability in physikalische Chemie, Go¨ttingen, Germany) for the anti-lysozyme anti- by guest on September 26, 2021 LPS-activated cells seemed to be associated with an increase in the serum, and Prof. G. F. Gerlach (University of Hannover, Hannover, Ger- poly(A) tail length of the lysozyme transcript. This process ap- many) for critical reading of the manuscript. peared to occur in the nucleus, in contrast to many developmental References systems of the mouse (36), Xenopus (33, 48), and Drosophila (49) in which preexisting short poly(A) tails of mRNAs were elongated 1. Palmiter, R. D. 1972. Regulation of protein synthesis in chick oviduct. J. Biol. Chem. 247:6450. in the cytoplasm. 2. Gordon, S., J. Todd, and Z. A. Cohn. 1974. In vitro synthesis and secretion of Nuclear mechanisms increasing poly(A) tail length of nuclear lysozyme by mononuclear phagocytes. J. Exp. Med. 139:1228. RNAs have already been described (27, 29). It has been proposed 3. Hauser, H., T. Graf, H. Beug, I. Greiser-Wilke, W. Lindenmaier, M. Grez, H. Land, K. Giesecke, and G. Schu¨tz. 1981. Structure of the lysozyme gene and that this process may be involved in the stability of cytoplasmic, expression in the oviduct and macrophages. In Haematology and Blood Trans- mature mRNAs (29, 50, 51). Surprisingly, Ford et al. (52) have fusion, Vol. 26. R. Neth, R. C. Gallo, T. Graf, K. Mannweiler, and K. Winkler, shown that poly(A) tails were able to stabilize RNA in an in vitro eds. Springer Berlin, pp. 175–178. 4. Cross, M., I. Mangelsdorf, A. Wedel, and R. Renkawitz. 1988. Mouse lysozyme RNA stability system only with nuclear extracts. Our results show M gene: isolation, characterization, and expression studies. Proc. Natl. Acad. Sci. that the cytoplasmic message stability was similar in both un- USA 85:6232. 5. Ja¨gle, U., A. M. Mu¨ller, H. Kohler, and C. Bonifer. 1997. Role of positive and treated and LPS-activated HD11 cells. Although any correlation negative cis-regulatory elements in the transcriptional activation of the lysozyme between an extended polyadenylation and a stabilization of pri- locus in developing macrophages of transgenic mice. J. Biol. Chem. 272:5871. mary transcripts has not been reported yet, we could not exclude 6. Hamilton, T. A., and D. O. Adams. 1987. Molecular mechanisms of signal trans- duction in macrophages. Immunol. Today 8:151. the possibility that the increase in poly(A) tail length of the ly- 7. Rietschel, E. T., and H. Brade. 1992. Bacterial endotoxins. Sci. Am. 267:54. sozyme transcript might result in an increased stability of the ly- 8. Cohn, Z. A., and E. Wiener. 1963. The particulate hydrolases of macrophages. II. sozyme pre-mRNAs, but not account for the stability of the cyto- Biochemical and morphological response to particle ingestion. J. Exp. Med. 118: 1009. plasmic mature mRNA, because of the following two 9. Goethe, R., and L. Phi-van. 1994. The far upstream chicken lysozyme enhancer considerations. First, the mature lysozyme mRNA was a stable at Ϫ6.1 kilobase, by interacting with NF-M, mediates lipopolysaccharide-in- ϳ duced expression of the chicken lysozyme gene in chicken myelomonocytic cells. message with a t1/2 of 9 h. Increasing its stability may result J. Biol. Chem. 269:31302. more likely in an extended half-life than in an elevated level of the 10. Goethe, R., and L. Phi-van. 1997. Evidence for an enhanced transcription-de- lysozyme message and an increased lysozyme production of mac- pendent de novo synthesis of C/EBP␤ in the LPS activation of the chicken ly- sozyme gene. J. Leukocyte Biol. 61:367. rophages in early response to LPS or at the beginning of a bacterial 11. Beug, H., A. Von Kirchbach, G. Do¨derlein, J.-F. Conscience, and T. Graf. 1979. infection. Second, the increased poly(A) tail length of lysozyme Chicken hematopoietic cells transformed by seven strains of defective avian leu- mRNA induced by LPS began to be gradually shortened immedi- kemia viruses display three distinct phenotypes of differentiation. Cell 18:375. 12. Phi-van, L., J. P. Von Kries, W. Ostertag, and W. H. Stra¨tling. 1990. The chicken ately after entering the cytoplasm. It has already been reported that lysozyme 5Ј matrix attachment region increases transcription from a heterologous the rate of mRNA decay is determined by the rate of poly(A) tail promoter in heterologous cells and dampens position effects on the expression of ␤ transfected genes. Mol. Cell. Biol. 10:2302. removal (30). Like the stable -globin mRNA (30), the LPS-in- 13. Lindenmaier, W., M. C. Nguyen-huu, R. Lurz, M. Stratmann, N. Blin, T. Wurz, duced lysozyme mRNA underwent a slow, gradual poly(A) short- H. J. Hauser, A. E. Sippel, and G. Schu¨tz. 1979. Arrangement of coding and 4978 LPS REGULATION OF THE LYSOZYME GENE

intervening sequences of chicken lysozyme gene. Proc. Natl. Acad. Sci. USA 34. Simon, R., J.-P. Tassan, and J. D. Richter. 1992. Translational control by poly(A) 76:6196. elongation during Xenopus development: differential repression and enhancement 14. Grewal, T., M. Theisen, U. Borgmeyer, T. Grussenmeyer, R. A. W. Rupp, by a novel cytoplasmic polyadenylation element. Genes Dev. 6:2580. A. Stief, F. Qian, A. Hecht, and A. E. Sippel. 1992. The Ϫ6.1-kilobase chicken 35. Dreyfuss, G., M. Hentze, and A. I. Lamond. 1996. From transcript to protein. Cell lysozyme enhancer is a multifactorial complex containing several cell-type-spe- 85:963. cific elements. Mol. Cell. Biol. 12:2339. 36. Vassalli, J. D., J. Huarte, D. Belin, P. Gubler, A. Vassalli, M. L. O’Connell, 15. Phi-van, L., and W. H. Stra¨tling. 1996. Dissection of the ability of the chicken I. A. Parton, R. J. Rickles, and S. Strickland. 1989. Regulated polyadenylation Ј lysozyme gene 5 matrix attachment region to stimulate transgene expression and controls mRNA translation during meiotic maturation of mouse oocytes. Genes to dampen position effects. Biochemistry 35:10735. Dev. 3:2163. 16. Rahmsdorf, H. J., A. Scho¨nthal, P. Angel, M. Litfin, U. Ru¨ther, and P. Herrlich. 37. Lenardo, M. J., and D. Baltimore. 1989. NF-␬B: a pleiotropic mediator of in- 1987. Posttranscriptional regulation of c-fos mRNA expression. Nucleic Acids ducible and tissue-specific gene control. Cell 58:227. Res. 15:1643. 17. Southern, E. M. 1975. Detection of specific sequences among DNA fragments 38. Oeth, P. A., G. C. N. Parry, C. Kunsch, P. Nantermet, C. A. Rosen, and separated by gel electrophoresis. J. Mol. Biol. 98:503. N. Mackman. 1994. Lipopolysaccharide induction of tissue factor gene expres- sion in monocytic cells is mediated by binding of c-rel/p65 heterodimers to a 18. Sippel, A. E., H. Land, W. Lindenmaier, M. C. Nguyen-huu, T. Wurtz, ␬ K. N. Timmis, K. Giesecke, and G. Schu¨tz. 1978. Cloning of chicken lysozyme B-like site. Mol. Cell. Biol. 14:3772. structural gene sequences synthesized in vitro. Nucleic Acids Res. 5:3275. 39. Phi van, L. 1996. Transcriptional activation of the chicken lysozyme gene by ␬ ␬ 19. Dugaiczyk, A., J. A. Haron, E. M. Stone, O. E. Dennison, K. N. Rothblum, and NF- Bp65 (relA) and c-rel, but not by NF- Bp50. Biochem. J. 313:39. R. Schwartz. 1983. Cloning and sequencing of a deoxyribonucleic acid copy of 40. Baeuerle, P. A. 1991. The inducible transcription activator NF-␬B: regulation by glyceraldehyde-3-phosphate dehydrogenase messenger ribonucleic acid isolated distinct protein subunits. Biochim. Biophys. Acta 1072:63. from chicken muscle. Biochemistry 22:1605. 41. Godambe, S. A., D. D. Chaplin, T. Takova, and C. J. Bellone. 1994. Upstream 20. Church, G. M., and W. Gilbert. 1984. Genomic sequencing. Proc. Natl. Acad. Sci. NFIL-6-like site located within a DNase I hypersensitivity region mediates LPS- USA 81:1991. induced transcription of the murine interleukin-1␤ gene. J. Immunol. 153:143. 21. Kleene, K. C., R. J. Distel, and N. B. Hecht. 1984. Translational regulation and 42. Steinmetz, E. J. 1997. Pre-mRNA processing and the CTD of RNA polymerase

deadenylation of a protamine mRNA during spermiogenesis in the mouse. Dev. II: the tail that wags the dog? Cell 89:491. Downloaded from Biol. 105:71. 43. Legrain, P., and M. Rosbash. 1989. Some cis- and trans-acting mutants for splic- 22. Greenberg, M. E., and E. B. Ziff. 1984. Stimulation of 3T3 cells induces tran- ing target pre-mRNA to the cytoplasm. Cell 57:573. scription of the c-fos proto-oncogene. Nature 311:433. 44. Chang, D. D., and P. A. Sharp. 1989. Regulation of HIV depends upon recog- 23. Luckow, B., and G. Schu¨tz. 1987. CAT constructions with multiple unique re- nition of splice sites. Cell 59:789. striction sites for the functional analysis of eukaryotic promoters and regulatory elements. Nucleic Acids Res. 15:5490. 45. Soeiro, R., M. Y. Vaughan, J. R. Warner, and J. E. Darnell. 1968. The turnover 24. Capobianco, A. J., D. L. Simmons, and T. D. Gilmore. 1990. Cloning and ex- of nuclear DNA-like RNA in HeLa cells. J. Cell Biol. 39:112. ␣ pression of a chicken c-rel cDNA: unlike p59v-rel, p68c-rel is a cytoplasmic pro- 46. Cohen, R. B., T. R. Boal, and B. Safer. 1990. Increased elF-2 expression in tein in chicken embryo fibroblasts. Oncogene 5:257. mitogen-activated primary T lymphocytes. EMBO J. 9:3831. http://www.jimmunol.org/ 25. Wormington, M. W. 1991. Preparation of synthetic mRNAs and analyses of 47. Kiledjian, M., and T. Kadesch. 1991. Post-transcriptional regulation of the human translational efficiency in microinjected Xenopus oocytes. Methods Cell Biol. liver/bone/kidney alkaline phosphatase gene. J. Biol. Chem. 266:4207. 36:167. 48. Paris, J., and J. D. Richter. 1990. Maturation-specific polyadenylation and trans- 26. Jung, A., A. E. Sippel, M. Grez, and G. Schu¨tz. 1980. Exons encode functional lational control: diversity of cytoplasm polyadenylation elements, influence of and structural units of chicken lysozyme. Proc. Natl. Acad. Sci. USA 77:5759. poly(A) tail size, and formation of stable polyadenylation complexes. Mol. Cell. 27. Shiels, B. R., W. Northemann, M. R. Gehring, and G. H. Fey. 1987. Modified Biol. 10:5634. ␣ nuclear processing of 1-acid glycoprotein RNA during inflammation. J. Biol. 49. Salles, F. J., M. E. Lieberfarb, C. Wreden, J. P. Gergen, and S. Strickland. 1994. Chem. 262:12826. Coordinate initiation of Drosophila development by regulated polyadenylation of 28. Carrazana, E. J., K. B. Pasieka, and J. A. Majoub. 1988. The vasopressin mRNA maternal messenger RNAs. Science 266:1996. poly(A) tail is unusually long and increases during stimulation of vasopressin 50. Huez, G., G. Marbaix, D. Gallwitz, E. Weinberg, R. Devos, E. Hubert, and Y gene expression in vivo. Mol. Cell. Biol. 8:2267. Cleuter. 1978. Functional stabilisation of HeLa cell histone messenger RNAs by guest on September 26, 2021 29. Murphy, D., K. Pardy, V. Seah, and D. Carter. 1992. Posttranscriptional regula- injected into Xenopus oocytes by 3Ј-OH polyadenylation. Nature 271:572. tion of rat growth hormone gene expression: increased message stability and 51. Zeevi, M., J. R. Nevins, and J. E. Darnell. 1982. Newly formed mRNA lacking nuclear polyadenylation accompany thyroid hormone depletion. Mol. Cell. Biol. polyadenylic acid enters the cytoplasm and the polyribosomes but has a short 12:2624. half-life in the absence of polyadenylic acid. Mol. Cell. Biol. 2:517. 30. Shyu, A.-B., J. G. Belasco, and M. E. Greenberg. 1991. Two distinct destabilizing 52. Ford, L. P., P. S. Bagga, and J. Wilusz. 1997. The poly(A) tail inhibits the elements in the c-fos message trigger deadenylation as a first step in rapid mRNA Ј Ј decay. Genes Dev. 5:221. assembly of a 3 -to-5 exonuclease in an in vitro RNA stability system. Mol. Cell. 31. Belasco, J. G., and G. Brawerman. 1993. Control of Messenger RNA stability. Biol. 17:398. Academic Press, San Diego. 53. Sachs, A. 1990. The role of poly(A) in the translation and stability of mRNA. 32. Sachs, A. B. 1993. Messenger RNA degradation in eukaryotes. Cell 74:413. Curr. Opin. Cell Biol. 2:1092. 33. McGrew, L. L., E. Dworkin-Rastl, M. B. Dworkin, and J. D. Richter. 1989. 54. Huang, Y., and G. G. Carmichael. 1996. Role of polyadenylation in nucleocy- Poly(A) elongation during Xenopus oocyte maturation is required for transla- toplasmic transport of mRNA. Mol. Cell. Biol. 16:1534. tional recruitment and is mediated by a short sequence element. Genes Dev. 55. Zasloff, M. 1983. tRNA transport from the nucleus in a eukaryotic cell: carrier- 3:803. mediated translocation process. Proc. Natl. Acad. Sci. USA 80:6436.