The RNA stability regulator HuR regulates L1 protein expression in vivo in differentiating cervical epithelial cells

Virology 383 (2009) 142–149

Contents lists available at ScienceDirect

Virology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y v i r o

The RNA stability regulator HuR regulates L1 protein expression in vivo in differentiating cervical epithelial cells S.A. Cumming, T. Chuen-Im 1, J. Zhang, S.V. Graham ⁎ Faculty of Biomedical and Life Sciences, Division of Infection and Immunity, University of Glasgow, 120 University Place, Glasgow G12 8TA, Scotland, UK

a r t i c l e

i n f o

Article history: Received 6 August 2008 Returned to author for revision 5 September 2008 Accepted 1 October 2008 Available online 4 November 2008 Keywords: HPV16 Late gene expression Late regulatory element HuR Epithelial differentiation

a b s t r a c t Human papillomavirus (HPV) L1 and L2 capsid protein expression is restricted to the granular layer of infected, stratified epithelia and is regulated at least partly at post-transcriptional levels. For HPV16, a 79 nt late regulatory element (LRE) is involved in this control. Using W12 cells as a model for HPV16-infected differentiating cervical epithelial cells we show that HuR, a key cellular protein that controls mRNA stability, binds the LRE most efficiently in nuclear and cytoplasmic extracts of differentiated cells. Further, HuR binds the 3′ U-rich portion of the LRE directly in vitro. Overexpression of HuR in undifferentiated W12 cells results in an increase in L1 mRNA and protein levels while siRNA knock-down of HuR in differentiated W12 cells depletes L1 expression. In differentiated cervical epithelial cells HuR may bind and stabilise L1 mRNAs aiding translation of L1 protein. © 2008 Elsevier Inc. All rights reserved.

Introduction Human papillomaviruses (HPVs) infect epithelial cells causing benign lesion or warts, which in some rare cases can progress to malignancy. HPVs can be classified according to whether they infect cutaneous or mucosal epithelia and whether such infection poses a “low” or “high risk” of tumour formation. The most important subset medically are the anogenital infective mucosal “high risk” HPVs types (e.g. HPV16, 18, 31, 33, 45) that cause lesions, particularly on the cervix that, upon persistent infection, may progress to cancer. HPV16 is the most prevalent “high risk” anogenital type being found in around 60% of cervical lesions (zur Hausen, 2002). HPV16 has a circular double stranded 7.9 kb DNA genome that exists in the infected cell nucleus as multiple episomal copies. The genome can be divided into an early coding region, encoding proteins E1, E2, E1^E4, E5, E6 and E7, and a late coding region containing the L1 and L2 open reading frames that encode the capsid proteins. The early and late regions are separated by the long control region (LCR) that contains at its 3′ end, replication and transcription control elements and at its 5′ end, the late 3′ untranslated region (Longworth and

Abbreviations: HPV, human papillomavirus; LRE, late regulatory element; UTR, untranslated region; LCR, long control region; ARE, AU-rich element; EMSA, electrophoretic mobility shift assay. ⁎ Corresponding author. Pathological Sciences, Jarrett Building, Institute of Comparative Medicine, Garscube Estate, Glasgow G61 1QH, UK. Fax: +44 141 330 5602. E-mail address: [email protected] (S.V. Graham). 1 Present address: Department of Microbiology, Faculty of Science, Silpakorn University, Nakorn Pathom, 73000, Thailand. 0042-6822/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2008.10.003

Laimins, 2004) that contains two active late polyadenylation signals (Milligan et al., 2007). The papillomavirus life cycle is intimately linked to epithelial differentiation. The virus infects the dividing cells of the basal layer and the virus genome can persist in this layer (Longworth and Laimins, 2004). Differentiation of virus-infected basal layer cells results in a productive infection. Viral genome amplification occurs in the spinous epithelial layer but production of the virus capsid proteins is restricted to cells of the granular layer (Peh et al., 2002). Despite this, virus late transcripts can be detected in lower epithelial layers (Stoler et al., 1989, 1992; Beyer-Finkler et al., 1990). However, at least for HPV16, these RNAs are unprocessed (Milligan et al., 2007) and probably targeted for nuclear degradation. Factors expressed in the differentiated epithelial host cells must induce appropriate expression of the capsid proteins at a post-transcriptional level. Viral RNA elements that bind cellular proteins have been proposed to be involved in post-transcriptional regulation of capsid protein expression for a number of papillomaviruses. Indeed, such elements may be widespread among the papillomaviruses (Zhao et al., 2007). Late 3′ UTR regulatory elements have been identified for BPV-1 (Furth and Baker, 1991; Furth et al., 1994) and HPV-1 (Sokolowski et al., 1999) that regulate polyadenylation (Gunderson et al., 1997) and translation (Wiklund et al., 2002) respectively. Several elements have been identified for HPV16. All appear to repress gene expression in HeLa cells. One is present in the 5′ end and another at the 3′ end of the L2 open reading frame. Both these elements have resisted precise delineation but they may regulate mRNA stability and translation (Öberg et al., 2003). Indeed, they bind hnRNP K and poly(rC)-binding proteins 1 and 2, two proteins known

S.A. Cumming et al. / Virology 383 (2009) 142–149

to regulate RNA stability and translation (Collier et al., 1998). Another important region is located in the first 514 nucleotides of the L1 open reading frame (Collier et al., 2002). The dominant L1 element is within the first 129 nucleotides. There is evidence that this element acts as an exonic sequence silencer (ESS) by binding the splicing regulator hnRNP A1 to repress use of the 3′ splice site at the 5′ end of the L1 open reading frame (Zhao et al., 2004). Finally, the 79 nt HPV16 late regulatory element (LRE), which spans 27 nts at the 3′ end of the L1 open reading frame and extends 52 nts into the late 3′ UTR (Fig. 1) has a 5′ portion containing four weak 5′ splice sites that binds a U1 snRNP-like complex and a 3′ portion that is GU-rich (Cumming et al., 2003). In the nucleus, the 3′ portion binds the splicing factor U2AF (Dietrich-Goetz et al., 1997) and this may recruit the SR protein, SF2/ASF giving rise to formation of an early splicinglike complex (McPhillips et al., 2004). There is no detectable splicing of late transcripts within the 3′ UTR so this complex may be involved in regulating RNA processing and stability of the RNAs (McPhillips et al., 2004). The element may also bind CUGBP to regulate posttranscriptional repression (Goraczniak and Gunderson, 2008). Pre-

143

viously we demonstrated that the HPV16 LRE could also bind HuR (Koffa et al., 2000). HuR is an RNA-binding protein that binds U- or AU-rich elements (AREs). It is a member of the ubiquitous Hu/ELAV protein family whose members regulate mRNA stability and translation and is predominantly nuclear but can shuttle between the nucleus and cytoplasm (Keene, 1999). All previous HPV16 RNA/protein binding studies were in HeLa cell extracts. Although HeLa cells are cervical epithelial cells they are highly transformed (they express high levels of the HPV18 oncoproteins E6 and E7) and they cannot differentiate to cells that will support the HPV life cycle. HeLa cells may not provide an appropriate background to investigate the role of cellular proteins in regulation of the HPV life cycle. In contrast, W12 cells (Stanley et al., 1989) are immortalised but not transformed, differentiation-competent HPV16 genome-containing cervical epithelial cells. These cells can support the HPV16 life cycle in monolayer and organotypic raft culture (Milligan et al., 2007). Using this model system, we show that HuR/LRE complex formation is most efficient in the nucleus and cytoplasm of differentiated W12 cells. Further, the 3′ U-rich portion of the element binds purified HuR directly with high affinity. Finally, transient transfection of undifferentiated W12 cells with an expression construct for HuR results in production of stable L1 mRNA and production of L1 protein. Conversely, siRNA knock-down of HuR expression in differentiated W12 cells results in abrogation of expression of L1 protein. This indicates that HuR may contribute to regulation of L1 protein production in an epithelial differentiation stage-specific manner. Results HuR binds the HPV16 late regulatory element (LRE) in a cervical epithelial differentiation-dependent manner

Fig. 1. HuR binds the HPV16 LRE most efficiently in differentiated W12 cell extracts. (A) The LRE nucleotide sequence. Diamonds, 4 weak 5′ splice sites (Cumming et al. 2003); underlined sequence, the 3′ U-rich region. The vertical line indicates the arbitrary division of the 5′ and 3′ portions of the LRE. The L1 stop codon is in bold type. (B) EMSA with [32P]-labelled in vitro-transcribed LRE RNA and nuclear (NE) and cytoplasmic (CE) extracts from undifferentiated and differentiated W12 cells. In tracks 3, 5, 8, and 10 the reaction mix was pre-incubated with HuR antibody 16A5 that causes complex depletion. The probe is indicated with an arrow. The arrowhead indicates probe resistant to denaturation at 70 °C. Stars indicate bands depleted in the presence of antiHuR antibody. (C) Western blots demonstrating fractionation (U2AF is a nuclear protein) and differentiation (involucrin is a cytoplasmic protein expressed in differentiated keratinocytes) of W12 cells and abundance of HuR (antibody 16A5) in the various fractions. CE, cytoplasmic extract; NE, nuclear extract, UW12, undifferentiated W12 cell extracts; DW12, differentiated W12 cell extracts.

Binding of HuR to the LRE was investigated previously by our group in HeLa cell nuclear extracts (Koffa et al., 2000). HeLa cells are highly transformed (they express high levels of the HPV18 E6 and E7 oncoproteins) and have lost the capacity to differentiate so they are unable to support the HPV life cycle. Moreover, recently HuR has been demonstrated to be overexpressed in many tumours and cancer cell lines (López de Silanes et al., 2005) such as HeLa cells. In order to investigate HuR/LRE binding during epithelial differentiation of nontransformed epithelial cells that better mimic the biochemical environment of the cells HPV16 infects we used extracts from undifferentiated and differentiated W12 cervical epithelial cells. Fig. 1B shows an electrophoretic mobility shift assay (EMSA) demonstrating that few protein complexes were formed on the LRE in undifferentiated W12 nuclear extracts (Fig. 1B, tracks 2 and 3). In track 1 which shows probe alone, the band of reduced mobility is due to secondary structure formation (Cumming et al., 2003) that is resistant to heat denaturation. The bands of similar mobility in tracks containing either nuclear or cytoplasmic extract may be a mix of undenatured probe and retarded denatured probe. There is little difference in the abundance and mobility of this band in the various tracks. However, a significant increase in binding of a high molecular weight complex was observed when undifferentiated W12 cytoplasmic extracts were used (track 4). This confirms what we found previously with HeLa cytoplasmic extracts (Koffa et al., 2000). Depletion of this complex (starred) was observed when the reaction mix contained anti-HuR antibody 16A5 (Fig. 1B track 5). In contrast, in differentiated W12 cells, there was clear formation of a high molecular weight complex in both nuclear (track 7) and cytoplasmic (track 9) extracts, and the largest amount of complex formed with cytoplasmic extract. Indeed this complex had retarded mobility compared to the complex observed with nuclear extract and so may have a different composition. These complexes were significantly depleted when the binding reactions were carried out in the

144

S.A. Cumming et al. / Virology 383 (2009) 142–149

presence of anti-HuR antibody 16A5 (starred in tracks 8 and 10). Fig. 1C shows that nuclear and cytoplasmic fractions from the undifferentiated and differentiated W12 cells were significantly enriched in proteins specific to each compartment. U2AF65, a nuclear protein, was found largely in the nuclear fraction and involucrin, a cytoplasmic filament protein, was found mostly in the cytoplasm. Levels of involucrin, a marker of differentiated keratinocytes, increased in the differentiated W12 cell cytoplasmic fraction as expected. HuR levels did not change significantly upon differentiation and HuR was found mainly in the nuclear fraction, as expected for a shuttling protein. To examine HuR cellular and tissue distribution W12 cells were grown on organotypic raft culture and raft tissues cross sections were subject to immunostaining using anti-HuR antibody 19F12. Fig. 2 shows that the distribution of HuR throughout the W12 epithelium was similar to the splicing factor U2AF65 but may be more abundant overall especially in differentiated epithelial cells. However, while U2AF65 was found to be mainly nuclear as expected HuR could be detected in both the nucleus and the cytoplasm. No antibody controls showed no brown staining (data not shown). This indicates a significant capacity for HuR shuttling in fully differentiation-competent W12 cells. The western blot in Fig. 1C does not show a very significant amount of HuR in the cytoplasm of undifferentiated and differentiated W12 cells. However it is difficult to equate staining intensity from immunohistochemistry to steady state levels of protein in cells detected by western blotting and the levels of HuR in the densely-stained nuclei relative to the cytoplasm of the cells in Fig. 2B may be high.

HuR binds the 3′ portion of the LRE directly The 3′ portion of the HPV16 LRE contains a class III AU-rich element (ARE) so we investigated whether it bound HuR directly. GST–HuR was expressed in bacteria, purified then incubated with radiolabelled full length LRE probe (Fig. 1A) in an electrophoretic gel mobility shift assay (EMSA) (Fig. 3A). The LRE probe has the capacity to form heat-resistant secondary structure so the probe alone track (Fig. 3A, track 1) shows a main band representing denatured probe and a less intense lower mobility band representing non-denatured probe. Increasing concentrations of GST–HuR caused increased formation of LRE RNA/protein complexes (efficient binding was observed with 260 pM GST–HuR; track 3) while GST alone did not bind the LRE (track 5). As noted for Fig. 1B, the smallest LRE-HuR complex may have a similar mobility to the nondenatured LRE probe that was not complexed with protein (note the increased abundance of the starred band). Binding was specific because nonspecific competitor, pBluescript polylinker RNA, did not interrupt complex formation when incubated with the probe at 4 and 8-fold molar excess (tracks 6 and 7). Cold competitor LRE RNA was able to compete efficiently the binding of HuR to the LRE (tracks 8–10). EMSA and UV crosslinking studies were used to determine which portion of the LRE binds HuR. GST–HuR/RNA complexes were formed when radiolabelled full length LRE and 3′ LRE probes were used (Fig. 3B tracks 2 and 8) but not when a 5′ LRE probe was used (Fig. 3B track 5). An antisense LRE probe did not bind GST–HuR indicating binding was specific for sense LRE RNA (Fig. 3B track 11). Addition of HuR antibody 19F12 caused a supershift of the complexes formed (tracks 3 and 9) confirming that these were due to HuR binding. This is different to the depletion of HuR complexes observed in Fig. 1B where antibody 16A5 was used. These monoclonal antibodies recognise different epitopes and this likely accounts for the difference in response in probing RNA/protein interactions. Fig. 3C shows the results of a UV crosslinking experiment confirming that GST–HuR binds the 3′ U-rich portion of the LRE (track 3) rather than the 5′ portion (track 2). Fig. 3D shows a competition UV crosslinking experiment. 32P-labelled LRE RNA was incubated with GST–HuR in the presence of no cold probe (tracks 1 and 2), or in the presence of cold 3′ LRE probe or cold 5′ LRE probe. Cold 3′ LRE probe competed efficiently with the 32P-labelled full length LRE RNA for binding HuR. There was some competition for binding of HuR using cold 5′ LRE probe but a significant amount of LRE RNA was still complexed to GST–HuR. There is a short U-rich region within the 5′ LRE probe that may bind HuR inefficiently leading to some inefficient competition in UV crosslinking. Experiments were performed in duplicate and duplicate tracks are shown. Site directed mutagenesis studies, changing 5 nucleotides at a time, across the LRE sequence, showed no depletion in binding of a band at around 40 kDa that we identified as HuR in UV cross linking studies (data not shown). This indicates that HuR may require a large region within the 3′ LRE (30 nts in length) for most efficient binding. HuR has been shown to bind cooperatively to ARE RNAs of at least 18 nucleotides in length (Fialcowitz-White et al., 2007). Oligomerisation of HuR on the LRE template can be observed in Fig. 3A tracks 3 and 4. Overexpression of HuR in undifferentiated W12 cells induces expression of L1 protein

Fig. 2. HuR is present in the nucleus and cytoplasm of differentiating W12 cells. Immunostaining of 4 μm paraffin-embedded sections of organotypic raft cultures of W12 cells. (A) Staining with antibody MC3 against U2AF65, a nuclear RNA processing factor. (B) Staining with antibody 16A5 against HuR. Slides were counterstained with hematoxylin.

To determine whether HuR has any regulatory activity via LRE binding on stability of LRE-containing RNAs, HeLa cells were transfected with pcDNA3.1, as a vector control or a mammalian expression vector for HuR, pcDNA3HuR and with luciferase reporter gene constructs containing the HPV16 late 3′ UTR with or without the LRE and reporter gene expression assayed. No significant change in levels of luciferase was observed upon HuR expression (data not

S.A. Cumming et al. / Virology 383 (2009) 142–149

shown). However, EMSA data indicated that HuR does not have significant capacity to bind the LRE in HeLa cell cytoplasmic extracts where it may have its effects on stability (data not shown). As HuR is clearly able to bind the element in W12 nuclear and cytoplasmic extracts we tested HuR control of endogenous L1 expression in these cells. In undifferentiated W12 cells, LRE-containing RNAs are not exported from the nucleus so L1 protein is not produced (Koffa et al., 2000; Milligan et al., 2007). We tested whether overexpression of HuR in undifferentiated W12 cells could induce expression of L1 protein. Undifferentiated W12 cells were transfected with pcDNA3HuR or with the vector construct with no insert, pcDNA3.1. 16 h following transfection half of the cells were differentiated by suspension culture in methylcellulose for 48 h. The remaining cells were kept at low density in F-medium with low concentrations of calcium to inhibit differentiation. RNA and protein was isolated from the undifferentiated and differentiated W12 cells and northern and western blotting performed. Figs. 4A and B tracks 1 show that L1 mRNA and L1 protein respectively was not detected in undifferentiated W12 cells transfected with the control vector,

145

pcDNA3.1. However, in cells transfected with pcDNA3HuR (tracks 2) significant levels of L1 mRNA and L1 protein could be detected. For L1 protein this was around 5-fold over background upon quantification with respect to levels of GAPDH (Fig. 4C). A smear representing L1-containg RNAs was observed in Fig. 4A, track2 because the probe hybridises to the central portion of thirteen L1-containing, polycistronic mRNAs (Milligan et al., 2007). In differentiated W12 cells L1 mRNA and protein was detected in both the vector and pcDNA3HuR-transfected cells as expected and little significant increase was observed in cells overexpressing HuR (data not shown). HuR overexpression caused a similar level of increase in HuR protein concentrations in both cell populations (data not shown). This indicates that HuR may have an involvement in regulation of L1 expression and perhaps other LRE-containing virus late mRNAs. siRNA depletion of HuR in differentiated W12 cells results in reduced levels of L1 protein Next HuR levels in W12 cells were depleted by treatment with a pool of HuR siRNAs. Undifferentiated W12 cells were transfected with control scrambled siRNAs or HuR siRNAs. 16 h after transfection the cells were differentiated by growth in methylcellulose for 48 h. The level of knock down of the protein was around 50%. Greater levels of knock down gave poor yields of transfected cells. This may be because HuR is known to be required for cell signalling during differentiation and has a broad antiapoptotic function (Abdelmohsen et al., 2007). L1 mRNA was detected by semi-quantitative RT-PCR (Fig. 4D). GAPDH and L1 mRNAs were simultaneously amplified from cDNAs reverse transcribed from cells transfected with control scrambled siRNA (tracks 1 and 2) and test HuR siRNAs (tracks 3 and 4). Reverse transcriptase was omitted from the cDNA synthesis step (tracks 1 and 3) as a control for DNA contamination. Less L1 mRNA was present in differentiated W12 cells transfected with HuR siRNAs (track 4), than cells transfected with control scrambled siRNAs and similarly differentiated (track 2). L1 protein was detected by western blotting of lysates of W12 cells transfected with the control siRNAs or the HuR siRNAs followed by differentiation in methylcellulose (Fig. 4E). In accordance with the mRNA data, there were reduced levels of L1 protein in differentiated W12 cells transfected with the HuR siRNAs (Fig. 4E track 2) compared to cells transfected with control siRNAs (track 1). L1 mRNA and protein were

Fig. 3. GST–HuR binds the 3′ portion of the LRE with high affinity. (A) Specific binding of GST–HuR to [32P]-labelled in vitro-transcribed full length LRE RNA. Track 1, no protein; track 2, 130 pM GST–HuR (40 ng); track 3, 260 pM GST–HuR (80 ng); track 4, 520 pM GST HuR (160 ng); track 5 GST alone. Tracks 6 and 7 EMSA reaction as in track 3 but with 4 and 8-fold molar excess of cold pBluescript polylinker RNA (65 nts), a nonspecific competitor; tracks 8–10 EMSA reaction as in track 3 but with 1, 2 and 4-fold molar excess cold LRE RNA as a specific competitor. LRE RNA probe runs as a fully denatured RNA and a band with retarded mobility due to incomplete denaturation. GST–HuR/RNA complexes are indicated by a line to the left of the autoradiograph. A line to the right hand side indicates probe. A possible GST–HuR complex that migrates with undenatured probe is indicated with an asterisk. (B) EMSA with [32P]-labelled in vitro-transcribed LRE-derived RNAs showing that GST–HuR (50 ng) can form a complex specifically with the 3′ U-rich region of HPV16 LRE RNA. GST–HuR/RNA complexes are indicated by a line to the left of the autoradiograph. Radiolabelled free probes are indicated with a line to the right hand side of the autoradiograph. LRE, full length LRE probe; 5′ LRE, nts 7127– 7174 of the HPV16 genome; 3′ LRE, nts 7175–7206 of the HPV16 genome; αsense LRE, in vitro transcribed full length antisense LRE sequence. Tracks 3, 6, 9 and 12 show supershift of HuR/LRE RNA complexes using antibody 19F12. (C) [32P]-labelled in vitro-transcribed LRE-derived RNAs UV crosslinked to GST–HuR protein (50 ng). LRE, full length LRE RNA probe; 5′, RNA probe transcribed from the 5′ portion of the LRE sequence, nts 7127–7174; 3′ RNA probe transcribed from the 5′ portion of the LRE sequence, nts 7175–7206; αs, antisense RNA probe transcribed from full length LRE. (D) [32P]-labelled in vitrotranscribed LRE-derived RNAs UV crosslinked to GST–HuR protein (50 ng) in the presence of no cold specific competitor (tracks 1 and 2), 3′ LRE RNA probe (as above) cold specific competitor (tracks 3 and 4) or 5′ LRE RNA probe (tracks 5 and 6). Experiments were performed in duplicate and shown electrophoresed on the same gel.

146

S.A. Cumming et al. / Virology 383 (2009) 142–149

Fig. 4. HuR regulates expression of L1 protein in W12 cells (A) Northern blot of mRNA isolated from W12 cells transiently transfected with (1) vector control, pcDNA3.1 and (2) HuR expressing plasmid, pcDNA3HuR and probed with [32P]-labelled, random primed insert from pHPV16-RL1 containing sequences 6151 to 6819 of the L1 coding region of the HPV16 genome. The blots were stripped and reprobed with an actin probe. (B) Western blots of total protein lysates from W12 cells transfected as above with (1) pcDNA3.1 and (2) pcDNA3HuR and probed with L1 antibody, HuR antibody 19F12 to show endogenous levels of HuR (1) and successful transfection with pcDNA3HuR (2) and GAPDH antibody to show similar levels of protein lysates fractionated in each track. UW12, undifferentiated W12 cell extracts; DW12, differentiated W12 cell extracts. (C) Quantification of data from three separate experiments to show a 5-fold increase in L1 protein levels (relative to levels of GAPDH as a loading control) upon transfection of the HuR expression construct. (D) Semiquantitative RT-PCR showing relative levels of GADPH and L1 mRNAs in cells expressing HuR siRNA. 1. Control scrambled siRNA, no reverse transcriptase. 2. Control scrambled siRNA, with reverse transcriptase. 3. HuR siRNA, no reverse transcriptase. 4. HuR siRNA, with reverse transcriptase. (E) Western blots of total protein lysates from W12 cells transfected with control scrambled siRNA and HuR siRNA. Antibodies were used as in (B). (F) Quantification of a representative experiment (three separate experiments were carried out) showing an approximate 2-fold reduction in L1 protein levels in response to a 50% reduction in levels of endogenous HuR. A similar amount of protein was electrophoresed in each track in the western blots in B and D but different exposure times are shown.

undetectable in undifferentiated W12 cells as expected (Milligan et al., 2007) and this was also the case when the cells were transfected with HuR siRNAs (data not shown). Discussion Papillomavirus late gene expression is controlled post-transcriptionally (Graham, 2006). Although the capsid proteins are expressed exclusively in granular layer cells of infected epithelia, RNAs encoding these proteins can be detected in less differentiated cells (Beyer-Finkler et al., 1990; Stoler et al., 1989). Previous data has indicated that HPV16 late gene expression may be regulated via the LRE present in the late 3′ UTR at the level of mRNA stability (Kennedy et al., 1991; Tan et al., 1995). Tan et al. (1995) showed that in HeLa cells transfected with a retrovirus-based expression vector containing the HPV16 L1 gene, but including only the first 25 nucleotides of the LRE, L1 mRNAs were present in the cytoplasm but L1 protein was not made. This suggested that L1 mRNAs were unable to be translated either because they were unstable or because they were unable to reach the polysomes. However, the expression vector also contained the 5′ end of the HIV1 genome, including the LTR so these HIV-specific signal sequences that are known to have post-transcriptional effects may have contributed to the regulation observed. Kennedy et al., (1991) demonstrated that in vitro transcribed 445 nt RNAs containing the LRE and the HPV16 late 3′ UTR were very unstable in a HeLa in vitro decay assay. In contrast, a 227 nt RNA at the 3′ end of the 445 nt fragment, that does not contain the LRE, was considerably more stable. More recently, we demonstrated that although HPV16 late transcripts are readily detected in total RNA from both undifferentiated and

differentiated W12 cells, polyadenylated mRNA is detected only in differentiated W12 cells (Milligan et al., 2007). Previous studies indicated that these RNAs were confined to the nucleus (Koffa et al., 2000). This suggests that although late genes are transcribed in infected undifferentiated epithelial cells, these RNAs are not polyadenylated and are unstable. HuR binds to AU-rich elements (AREs) in the 3′ UTR of mRNAs that are unstable. There are three classes of AREs. Class I contain the classical pentamer AUUUA, class II have a nonamer UUAUUUA(U/A)(U/ A) and class III lack an AUUUA motif but are U-rich. Overexpression of HuR appears to stabilise such RNAs by inhibiting their degradation. For example, human vascular endothelial growth factor (VEGF) has a class III ARE in its 3′ UTR that binds HuR with high affinity. In hypoxia, HuR overexpression efficiently stabilises VEGF mRNA (Levy et al., 1998). In addition to a role in controlling RNA stability, HuR can regulate alternative splicing (Zhu et al., 2006) and translation (Kawai et al., 2006). HuR is a nucleo-cytoplasmic shuttling protein and evidence suggests that it binds target RNAs in the nucleus, accompanies them to the cytoplasm and promotes their stability on polysomes (Fan and Steitz, 1998; Peng et al., 1998). Using EMSA and UV crosslinking studies we have demonstrated binding of HuR to the HPV16 LRE in W12 cell extracts in a differentiation stage-specific manner. UV crosslinking studies showed that HuR is one of the main proteins that can interact with LRE RNA (data not shown). In EMSA, purified GST–HuR binds with high affinity to the LRE and analysis of LRE deletion mutants revealed that binding was predominantly to the U-rich 3′ portion. The 3′ portion is 30 nts in length, 12 nts longer than a model minimum high affinity binding site for HuR (Fialcowitz-White et al., 2007). Site-directed mutagenesis did not reveal a shorter portion

S.A. Cumming et al. / Virology 383 (2009) 142–149

that was capable of binding HuR efficiently. Multimerisation of HuR upon the entire LRE was observed and to a lesser extent upon its 3′ portion (Figs. 3A and B). The 3′ LRE resembles a class III ARE as there are no AUUUA motifs in the element. These results contribute to the list of RNA sequences to which HuR can bind. EMSA revealed that in in vitro cell extracts HuR can bind the LRE in the cytoplasm of undifferentiated W12 cells but binds particularly efficiently in both the nucleus and cytoplasm of differentiated W12 cells. The RNA/protein complex detected with cytoplasmic extracts of differentiated W12 cells appears to comprise largely HuR as it is severely depleted upon inclusion of 16A5 anti-HuR antibody (this antibody does not cause a supershift) in the EMSA reaction mix. We suggest that in differentiated epithelial cells HuR is necessary for export and stability of LRE-containing mRNAs to the cytoplasm. In undifferentiated epithelial cells LRE-containing late mRNAs do not reach the cytoplasm (Koffa et al., 2000) and are unstable (Milligan et al., 2007) perhaps because HuR is not available for LRE binding as the element binds a splicing-related complex (Cumming et al., 2003). Alternatively, nuclear HuR may be sequestered in these cells by binding to other proteins or RNAs. When differentiation takes place the affinity of the splicing-like complex for the LRE may decrease perhaps due to phosphorylation of the SF2/ASF component (McPhillips et al., 2004) or due to increased availability of HuR to the LRE. The LRE-containing mRNAs begin to be efficiently polyadenylated and exported to the cytoplasm accompanied by HuR in a stabilising complex, perhaps in partnership with other proteins. One such protein might be hnRNP A1 that we have shown binds the central portion of the LRE in W12 cells and whose levels are increased in differentiated epithelial cells (Chuen-Im et al., 2008). hnRNP A1 could help stabilise the late transcripts leading to efficient translation. The hypothesis that HuR is involved in regulating HPV16 late gene expression was tested by overexpressing the protein in undifferentiated and differentiated W12 cells and assaying expression of the late mRNAs and L1 protein. Differentiation was by the methylcellulose method that allows rapid differentiation of transfected W12 cells during the expected time course (up to 48 h) of expression of the transiently transfected vector. There was no obvious effect on expression of late mRNAs and L1 protein in differentiated W12 cells. It is possible that overexpression has little effect on the nuclear and cytoplasmic pools of HuR available to LRE-containing late RNAs in these cells. In contrast in undifferentiated cells increased levels of HuR resulted in a clear increase in L1 mRNA expression mirrored by the appearance of detectable levels of L1 protein. Increasing levels of HuR in undifferentiated W12 cells may tip the balance of what protein complexes can bind the LRE allowing access of HuR to the LRE on at least some late transcripts in competition with the splicing-related complex that usually binds. This could result in alleviation of the proposed splice complex inhibition of polyadenylation (Graham, 2008) leading to stabilisation of LRE-containing mRNAs and efficient translation in the cytoplasm. In contrast, siRNA knock down of HuR in differentiated W12 cells depleted levels of L1 mRNA and protein. This lends support to the hypothesis that HuR is a positive regulator of L1 expression during differentiation of HPV16-infected epithelia. CUGBP1 has recently been shown also to bind the 3′ LRE and is proposed to synergise with U1 snRNP bound to the 5′ portion. However, functional analysis of CUGBP1 has only been performed in HeLa cells thus far (Goraczniak and Gunderson, 2008). Our present study reports the first in vivo functional study of an HPV16 LREbinding protein in a good model system for the virus life cycle. HuR appears to be a major regulator of HPV16 late gene expression during differentiation of infected epithelial cells and may contribute to repression of capsid protein expression in undifferentiated epithelial cells. Our data may also suggest that HuR acts as a positive regulator of capsid protein production in differentiated epithelial cells. Discovery of such cellular factors that control late protein production may open the way for design of novel antiviral therapies. Small molecule

147

inhibitors of HuR have recently been identified (Meisner et al., 2007) that may prove therapeutically useful in future. Materials and methods Cell culture 20863 (W12E) (Jeon et al., 1995) is a subclone of the W12 cell line (Stanley et al., 1989) and contains 50–100 episomal copies of the HPV16 genome. For monolayer culture, W12E cells were co-cultured with mitomycin C-treated J2 3T3 fibroblast feeder cells at a ratio of 1:5, seeding W12E cells at 2 × 105 cells/100 mm dish (Jeon et al., 1995). The W12E line was grown without passaging for up to 5 days for undifferentiated cells, and 10 days for differentiated cells (McPhillips et al., 2004) (Milligan et al., 2007). HeLa cells were cultured in DMEM supplemented with 10% foetal bovine serum and 2 mM L-glutamine (Invitrogen). All cells were grown at 37 °C in 5% CO2. Transient transfection Undifferentiated W12 cells were grown at low density in Fmedium in the presence of J2 3T3 cell feeder layers. After 4 days 3T3 cells were removed by trypsinisation and W12 cells were transfected with pcDNA3.1 or pcDNA3HuR using Lipofectamine 2000 (Invitrogen) at 0.5 μg DNA per 2 × 105 cells per 35 mm well. 16 h after transfection cells were differentiated for 48 h in methylcellulose (Ruesch et al., 1998). For siRNA knock-down HuR siRNA pools or scrambled control siRNA pools (0.5 μg each siRNA pool (Santa Cruz)) were transfected using the siRNA transfection reagent provided into 2 × 105 W12 cells in 35 mm wells and cells were differentiated in methylcellulose as above. Total cell, nuclear and cytoplasmic protein extraction Prior to preparation of W12 extracts feeder, cells were removed by treatment with 0.1% trypsin–0.5 mM EDTA. For total cell extracts, cells were washed twice in PBS and lysed directly in 2 × protein loading buffer [63 mM Tris–HCl, pH 6.8, 1% (wt/vol) SDS, 10% glycerol, 5% (vol/ vol) β-mercaptoethanol, 0.004% (wt/vol) bromophenol blue] followed by sonication until viscosity was decreased. Nuclear and cytoplasmic extracts were prepared as described (Dignam et al., 1983). Western blotting Proteins were separated on 12% SDS-PAGE before electroblotting onto polyvinylidene fluoride (PVDF) membranes (Amersham). The membranes were blocked in 5% (w/v) dried milk powder in phosphate-buffered saline (PBS) overnight at 4 °C. Primary antibodies were diluted in PBS with 5% (w/v) dried milk powder, 0.05% (v/v) Tween-20 (PBS-T) and incubated with the membranes for 1 h at room temperature with shaking. The anti-HuR antibodies 19F12 (Santa Cruz) and 16A5 (gift from Henry Furneaux; Levy et al., 1998) were used at a dilution of 1/500, antibody MC3 against U2AF65 (gift from M. Carmo-Fonseca) was used at a dilution of 1/100, monoclonal antibodies against involucrin (SY5; Sigma) and GAPDH (6CS; Biodesign International) were used at a dilution of 1/1000. L1 antibody (Pharmingen) was used at a dilution of 1/250. After washing with PBST, the membranes were incubated with an anti-mouse secondary antibody-horseradish peroxidase (Sigma) at a dilution of 1/1000 for 1 h at room temperature. Following washing in PBS-T, antibody– protein complexes were visualised using ECL reagents (Amersham) according to the manufacturer's instructions. Electrophoretic mobility shift assay (EMSA) and UV crosslinking Primer sequences used to generate riboprobe templates by PCR amplification of pCATPE445 were as described (Cumming et al., 2003).

148

S.A. Cumming et al. / Virology 383 (2009) 142–149

PCR products were purified on 6% polyacrylamide gels. Riboprobes were synthesised with the Stratagene RNA in vitro transcription kit with addition of 25 μCi [α-32P]UTP (800 mCi/mmol; NEN) following the manufacturer's protocol and purified from 5% denaturing polyacrylamide gels. EMSAs were carried out as described (Cumming and Graham, 2005). Specific competitor RNA was in vitro transcribed with the addition of unlabelled UTP substituting for [α-32P] UTP. Nonspecific competitor RNA (65 nts) was transcribed from pBluescript KS (+) (Stratagene) polylinker linearised with EcoRI. Supershift EMSA experiments were carried out by preincubation of nuclear or cytoplasmic extracts with 1 μg HuR antibody 19F12 for 15 min on ice. Depletion reactions were carried out by preincubating extracts with 1 μg 16A5 HuR antibody for 15 min on ice before addition of the probe. The EMSA reactions were electrophoresed on a 5% non-denaturing gel. UV crosslinking was carried out as described (Cumming and Graham, 2005) using a Stratalinker (Stratagene) at a setting of 250 mJ. For competition experiments, cold probe was added prior to addition of 50 ng GST–HuR. Reactions were carried out at room temperature for 15 min. followed by RNase A digestion (20 μg per reaction) at 37 °C for 30 min. Labelled proteins were fractionated on 12% SDS-PAGE.

istry was performed using the ABC Elite Kit (Vector laboratories) according to the manufacturer's instructions. For antigen retrieval, the tissue was heated in citrate buffer at pH 6.0 for 10 min in a microwave. Sections were blocked in PBS-20% foetal calf serum for 1 h before incubating with primary antibody for 1 h at 4 °C. Monoclonal antibodies MC3 against U2AF65 and 16A5 against HuR were used at concentrations of 1/10 and 1/5 respectively. Diaminobenzidine tetrahydrochloride (DAB) (Vector Laboratories was used as the chromogen and tissues were counterstained with hematoxylin.

GST–HuR purification

Abdelmohsen, K., Lal, A., Kim, H.H., Gorospe, M., 2007. Posttranscriptional orchestration of an anti-apoptotic program by HuR. Cell Cycle 6, 1288–1292. Beyer-Finkler, E., Stoler, M.H., Girardi, F., Pfister, H.J., 1990. Cell differentiation-related gene expression of human papillomavirus 33. Med. Microbiol. Immunol. 179, 185–192. Chuen-Im, T., Zhang, J., Milligan, S.G., McPhillips, M.G., Graham, S.V., 2008. The alternative splicing factor hnRNP A1 is up-regulated during virus-infected epithelial cell differentiation and binds the human papillomavirus type 16 late regulatory element. Virus Res. 131, 189–198. Collier, B., Goobar, L., Sokolowski, M., Schwartz, S., 1998. Translational inhibition in vitro of human papillomavirus type 16 L2 mRNA mediated through interaction with heterogeneous ribonucleoprotein K and poly (rC)-binding proteins 1 and 2. J. Biol. Chem. 273, 22648–22656. Collier, B., Oberg, D., Zhao, X., Schwartz, S., 2002. Specific inactivation of inhibitory sequences in the 5′ end of the human papillomavirus type 16 L1 open reading frame results in production of high levels of L1 protein in human epithelial cells. J. Virol. 76, 2739–2752. Cumming, S.A., Graham, S.V., 2005. Analysis of regulatory motifs within HPV transcripts. In: Davy, C., Doorbar, J. (Eds.), Human Papillomaviruses: Methods and Protocols. Humana Press, Totowa, New Jersey, pp. 291–315. Cumming, S.A., McPhillips, M.G., Veerapraditsin, T., Milligan, S.G., Graham, S.V., 2003. Activity of the human papillomavirus type 16 late negative regulatory element is partly due to four weak consensus 5′ splice sites that bind a U1 snRNP-like complex. J. Virol. 77, 5167–5177. Dietrich-Goetz, W., Kennedy, I.M., Levins, B., Stanley, M.A., Clements, J.B., 1997. A cellular 65 kDa protein recognizes the negative regulatory element of human papillomavirus late mRNA. Proc. Natl. Acad. Sci. U. S. A. 94, 163–168. Dignam, J.D., Lebovitz, R.M., Roeder, R.G., 1983. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475–1489. Fan, X.C., Steitz, J.A., 1998. Overexpression of HuR, a nuclear-cytoplasmic shuttling protein, increases the in vivo stability of ARE-containing mRNAs. EMBO J. 17, 3448–3460. Fialcowitz-White, E.J., Brewer, B.Y., Ballin, J.D., Willis, C.D., Toth, E.A., Wilson, G.M., 2007. Specific protein domains mediate cooperative assembly of HuR oligomers on AUrich mRNA-destabilizing sequences. J. Biol. Chem. 283, 20948–20959. Furth, P.A., Baker, C.C., 1991. An element in the bovine papillomavirus late 3' untranslated region reduces polyadenylated cytoplasmic RNA levels. J. Virol. 65, 5806–5812. Furth, P.A., Choe, W-T., Rex, J.H., Byrne, J.C., Baker, C.C., 1994. Sequences homologous to 5' splice sites are required for inhibitory activity of papillomavirus late 3' untranslated regions. Mol. Cell. Biol. 14, 5278–5289. Goraczniak, R., Gunderson, S.I., 2008. The regulatory element in the 3′-untranslated region of human papillomavirus 16 inhibits expression by binding CUG-binding protein 1. J. Biol. Chem. 283, 2286–2296. Graham, S.V., 2006. Late events in the life cycle of human papillomaviruses. In: Campo, M.S. (Ed.), Papillomavirus Research: From Natural History to Vaccines and Beyond. Caister Academic Press, Wymondham, Norfolk, pp. 193–212. Graham, S.V., 2008. Papillomavirus 3′UTR regulatory elements. Front. Biosci. 13, 5646–5663. Gunderson, S.I., Polycarpou-Schwarz, M., Mattaj, I.W., 1998. U1 snRNP inhibits premRNA polyadenylation through a direct interaction between U1 70K and poly(A) polymerase. Mol. Cell 1, 255–264. Jeon, S., Allen-Hoffman, B.L., Lambert, P.F., 1995. Integration of human papillomavirus type 16 into the human genome correlates with a selective growth advantage of cells. J. Virol. 69, 2989–2997. Kawai, T., Lal, A., Yang, X., Galban, S., Mazan-Mamczarz, K., Gorospe, M., 2006. Translational control of cytochrome c by RNA-binding proteins TIA-1 and HuR. Mol. Cell. Biol. 26, 3295–3307.

GST-tagged HuR expression plasmid (pGEX-HuR; gift from Joan Steitz, Yale University, New Haven, USA) was transformed into BL-21 cells. Logarithmic phase cells were induced with 1 mM IPTG at 37 °C for 3 h, then pelleted. The pellet was resuspended in 1 ml lysozyme (10 mg/ml) in PBS and incubated on ice for 15 min before incubating with 100 μl Triton X-100 for another 15 min. The supernatant was collected by centrifugation at 9500 rpm and mixed with 1 ml of 50% glutathione sepharose (Amersham) for 1 h at 4 °C before washing with PBS. The protein was eluted in 5 mM reduced glutathione in 50 mM Tris pH 8.0 three times and the concentration of purified HuR was determined by Bradford assay. RNA preparation and Northern blotting For preparation of RNA from W12 cells 3T3 feeder layer cells were removed by trypsinisation. W12 cells were washed with PBS then scraped into Trizol reagent (Invitrogen) and total RNA was prepared according to the manufacturer's instructions. Polyadenylated RNA was selected by hybridisation to oligo(dT) cellulose using an Oligotex mRNA purification kit (Qiagen) according to the manufacturer's instructions. 0.5 μg polyadenylated RNA per track was fractionated on a 1.2% agarose, 2.2M formaldehyde gel in MOPS buffer and transferred to Hybond-N (Amersham) as described (Sambrook et al., 1989). Blots were hybridised in 5xSSC, 50% formamide at 42 °C overnight and washed to 0.1 × SSC at 65 °C. Blots were dried and exposed to X-ray film overnight, or longer. Reverse transcriptase PCR (RT-PCR) was carried out using an Invitrogen First Strand Superscript III cDNA synthesis kit according to the manufacturer's instructions followed by simultaneous PCR amplification of L1 and GAPDH cDNAs using L1 Forward primer (5′CCTACACCTAGTGGTTCTATGGTTACC3′) and L1 Reverse primer (5′GGAAACTGATCTAGGTCTGCAG3′) with GAPDH Forward primer (5′TCCACCACCCTGTTGCTGTA3′) and GAPDH Reverse primer (5′ACCACAGTCCATGCCATCAC3′) and Taq polymerase (Invitrogen). 35 cycles were performed of 94 °C for 15 s. 55 °C for 30 s and 72 °C for 1 min. Immunohistochemistry of raft tissue Organotypic raft cultures were performed exactly as described (Meyers and Laimins, 1994). Following growth at the liquid/air interface for 14 days, raft tissues were fixed in 10% neutral buffered formalin overnight and paraffin-embedded. 4 μm sections were cut and fixed on poly-L-lysine (sigma)-coated slides. Immunocytochem-

Acknowledgments This work was funded by a grant from the BBSRC (grant reference 17/ G16909). We are grateful to Margaret Stanley and Paul Lambert for the gift of the W12 cells and their clones. We thank Joan Steitz for the gift of the HuR expression plasmid, Henry Furneaux for the anti-HuR antibody 16A5 and Maria Carmo-Fonseca for the MC3 antibody against U2AF65. References

S.A. Cumming et al. / Virology 383 (2009) 142–149 Keene, J.D., 1999. Why is Hu where? Shuttling of early-response-gene messenger RNA subsets. Proc. Natl. Acad. Sci. U. S. A. 96, 5–7. Kennedy, I.M., Haddow, J.K., Clements, J.B., 1991. A negative regulatory element in the human papillomavirus type 16 genome acts at the level of late mRNA stability. J. Virol. 65, 2093–2097. Koffa, M.D., Graham, S.V., Takagaki, Y., Manley, J.L., Clements, J.B., 2000. The human papillomavirus type 16 negative regulatory element interacts with three proteins that act at different posttranscriptional levels. Proc. Natl. Acad. Sci. U. S. A. 97, 4677–4682. Levy, N.S., Chung, S., Furneaux, H., Levy, A.P., 1998. Hypoxic stabilization of vascular endothelial growth factor mRNA by the RNA-binding protein HuR. J. Biol. Chem. 273, 6417–6423. Longworth, M.S., Laimins, L.A., 2004. Pathogenesis of human papillomaviruses in differentiating epithelia. Microbiol. Mol. Biol. Rev. 68, 362–372. López de Silanes, I., Lal, A., Gorospe, M., 2005. Post-transcriptional paths to malignancy. RNA Biol. 2, e11–e13. McPhillips, M.G., Veerapraditsin, T., Cumming, S.A., Karali, D., Milligan, S.G., Boner, W., Morgan, I.M., Graham, S.V., 2004. SF2/ASF binds the human papillomavirus type 16 late RNA control element and is regulated during epithelial differentiation. J. Virol. 78, 10598–10605. Meisner, N.-C., Hintersteiner, M., Mueller, K., Bauer, R., Seifert, J.-M., Naegeli, H.-U., Ottl, J., Oberer, L., Guenat, C., Moss, S., Harrer, N., Woisetschlaeger, M., Buehler, C., Uhl, V., Auer, M., 2007. Identification and mechanistic characterization of low-molecularweight inhibitors for HuR. Nat. Chem. Biol. 3, 508–515. Meyers, C., Laimins, L.A., 1994. In vitro systems for the study and propagation of human papillomaviruses. Curr. Top. Microbiol. Immunol. 186, 199–215. Milligan, S.G., Veerapraditsin, T., Ahamat, B., Mole, S., Graham, S.V., 2007. Analysis of novel human papillomavirus type 16 late mRNAs in differentiated W12 cervical epithelial cells. Virology 360, 172–181. Öberg, D., Collier, B., Zhao, X., Schwartz, S., 2003. Mutational inactivation of two distinct negative RNA elements in the human papillomavirus type 16 L2 coding region induces production of high levels of L2 in human cells. J. Virol. 77, 11674–11684. Peh, W.L., Middleton, K., Christensen, N., Nicholls, P., Egawa, K., Sotlar, K., Brandsma, J., Percival, A., Lewis, J., Liu, W.J., Doorbar, J., 2002. Life cycle heterogeneity in animal models of human papillomavirus-associated disease. J. Virol. 76, 10411–10416. Peng, S.S.-Y., Cehn, C.Y.A., Xu, N., Shyu, A.-B., 1998. RNA stabilization by the AU-rich element binding protein, HuR, an ELAV protein. EMBO J. 17, 3461–3470.

149

Ruesch, M.N., Stubenrauch, F., Laimins, L.A., 1998. Activation of papillomavirus late gene transcription and genome amplification upon differentiation in semisolid medium is coincident with expression of involucrin and transglutaminase but not keratin. J. Virol. 72, 5016–5024. Sambrook, J., Fritsch, E., Maniatis, T., 1989. “Molecular Cloning: A Laboratory Manual”. Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York. Sokolowski, M., Furneaux, H., Schwartz, S., 1999. The inhibitory activity of the AU-rich RNA element in the human papillomavirus type 1 late 3′ untranslated region correlates with its affinity for the elav-like HuR protein. J. Virol. 73, 1080–1091. Stanley, M.A., Browne, H.M., Appelby, M., Minson, A.C., 1989. Properties of a nontumourigenic human cervical keratinocyte cell line. Int. J. Cancer 43, 672–676. Stoler, M.H., Wolinsky, S.M., Whitbeck, A., Broker, T.R., Chow, L.T., 1989. Differentiationlinked human papillomavirus types 6 and 11 transcription in genital condylomata revealed by in situ hybridisation with message-specific RNA probes. Virology 172, 331–340. Stoler, M.H., Rhodes, C.R., Whitbeck, A., Wolinsky, S.M., Chow, L.T., Broker, T.R., 1992. Human papillomavirus type 16 and 18 gene expression in cervical neoplasia. Human Pathol. 23, 117–128. Tan, W., Felber, B.K., Zolotukhin, A.S., Pavlakis, G.N., Schwartz, S., 1995. Efficient expression of the human papillomavirus type 16 L1 protein in epithelial cells by using rev and the rev-response element of human immunodeficiency virus or the cis-acting transactivation element of simian retrovirus type 1. J. Virol. 69, 5607–5620. Wiklund, L., Sokolowski, M., Carlsson, A., Rush, M., Schwartz, S., 2002. Inhibition of translation by UAUUUAU and UAUUUUUAU motifs of the AU-rich instability element in the HPV-1 late 3′ untranslated region. J. Biol. Chem. 277, 40462–40471. Zhao, X., Rush, M., Schwartz, S., 2004. Identification of an hnRNP A1-dependent splicing silencer in the human papillomavirus type 16 L1 coding region that prevents premature expression of the late L1 gene. J. Virol. 78, 10888–10905. Zhao, X., Rush, M., Carlsson, A., Schwartz, S., 2007. The presence of inhibitory RNA elements in the late 3′-untranslated region is a conserved property of human papillomaviruses. Virus Res. 125, 135–144. Zhu, H., Hasman, R.A., Barron, V.A., Luo, G., Lou, H., 2006. A nuclear function of Hu proteins as neuron-specific alternative RNA processing regulators. Mol. Biol. Cell 17, 5101–5114. zur Hausen, H., 2002. Papillomaviruses and cancer: from basic studies to clinical application. Nat. Rev., Cancer 2, 342–350.