This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McIntosh, P. B.
Right arrow Articles by Doorbar, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McIntosh, P. B.
Right arrow Articles by Doorbar, J.

 Previous Article  |  Next Article 

Journal of Virology, August 2008, p. 8196-8203, Vol. 82, No. 16
0022-538X/08/$08.00+0     doi:10.1128/JVI.00509-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Structural Analysis Reveals an Amyloid Form of the Human Papillomavirus Type 16 E1{wedge}E4 Protein and Provides a Molecular Basis for Its Accumulation{triangledown}

Pauline B. McIntosh,1 Stephen R. Martin,2 Deborah J. Jackson,1 Jameela Khan,1 Erin R. Isaacson,1 Lesley Calder,1 Kenneth Raj,1 Heather M. Griffin,1 Qian Wang,1 Peter Laskey,1 John F. Eccleston,2 and John Doorbar1*

Division of Virology,1 Division of Physical Biochemistry, MRC National Institute for Medical Research, London NW7 1AA, United Kingdom2

Received 7 March 2008/ Accepted 3 June 2008


arrow
ABSTRACT
 
The abundant human papillomavirus (HPV) type 16 E4 protein exists as two distinct structural forms in differentiating epithelial cells. Monomeric full-length 16E1{wedge}E4 contains a limited tertiary fold constrained by the N and C termini. N-terminal deletions facilitate the assembly of E1{wedge}E4 into amyloid-like fibrils, which bind to thioflavin T. The C-terminal region is highly amyloidogenic, and its deletion abolishes amyloid staining and prevents E1{wedge}E4 accumulation. Amyloid-imaging probes can detect 16E1{wedge}E4 in biopsy material, as well as 18E1{wedge}E4 and 33E1{wedge}E4 in monolayer cells, indicating structural conservation. Our results suggest a role for fibril formation in facilitating the accumulation of E1{wedge}E4 during HPV infection.


arrow
TEXT
 
Cervical cancer and its precursors usually arise from areas of low-grade disease caused by high-risk human papillomavirus (HPV) types such as HPV-16 (5, 32). The HPV E4 protein (also known as E1{wedge}E4) is expressed from an E1{wedge}E4 spliced mRNA prior to the assembly of infectious virions and accumulates to very high levels in cells supporting productive infection (10, 11). Abundant E1{wedge}E4 is seen in lesions caused by many different HPV types (10, 21, 26), suggesting that E1{wedge}E4 accumulation plays an important role in the virus life cycle.

Although E1{wedge}E4 function is not yet fully understood, its association with keratin filaments is thought to compromise the mechanical integrity of the cell during the late stages of infection (8, 33). Keratin binding is not, however, essential for E1{wedge}E4 accumulation, as keratin-binding mutants still accumulate to wild-type levels (24). Rather, it appears that E1{wedge}E4 accumulation depends on sequences at the C terminus of the protein. To understand 16E1{wedge}E4 function at the molecular level, and in particular to explain its ability to accumulate within the infected cell, we have carried out a structural analysis of the protein. We have found that monomeric 16E1{wedge}E4 does not have a stable tertiary fold but does contain elements of secondary structure that are key to its assembly into fibrils following N-terminal truncation. This work not only enhances our understanding of the HPV-16 E1{wedge}E4 protein but also contributes a viral protein to the growing family of proteins capable of using the amyloid fold (4, 13).

Two distinct forms of 16E1{wedge}E4 are expressed during productive infection. Although the E1{wedge}E4 protein of HPV-1 is known to be N terminally cleaved during natural infection (2, 9), it is not yet clear whether HPV-16 E1{wedge}E4 is similarly modified. To identify the E1{wedge}E4 species present in productively HPV-16-infected squamous epithelium, NIKS cells harboring the full HPV-16 genome were grown in raft culture (24), harvested at day 11 postlifting, and lysed in phosphate-buffered saline containing 6% sodium dodecyl sulfate (SDS) and 1.5 M urea. The E1{wedge}E4 species were characterized using the 16E1{wedge}E4 antibodies TVG402, which binds an epitope between amino acids 29 and 40 (33), and anti-N-term (10), which binds only to species which retain an intact N-terminal epitope (amino acids 1 to 14). TVG402 detected two E1{wedge}E4 species (Fig. 1A). Anti-N-term did not detect the smaller species, suggesting that it lacks the N-terminal epitope.


Figure 1
View larger version (34K):
[in this window]
[in a new window]

  		FIG. 1. An N-terminally truncated form of 16E1{wedge}E4 is observed in tissue culture and in HPV-16-induced cervical intraepithelial lesions. (A) Western blot analysis of the 16E1{wedge}E4 species observed in extracts of COS-7 cells at 24, 48, and 72 h after transfection with MV11 16E1{wedge}E4 and in extracts of an HPV-16 raft. Full-length 16E1{wedge}E4 (band 1) is detected by both TVG402 and anti-N-term, whereas the lower-molecular-mass species (band 2) is detected by only TVG402, indicating that it is N terminally truncated. (B) Immunofluorescence images of E1{wedge}E4 (top panel) in an HPV-16-induced CIN tissue section (magnification, x100, confocal) and of E1{wedge}E4 (bottom panel) in an HPV-16 raft (magnification, x20). A 4',6'-diamidino-2-phenylindole (DAPI) (blue) nuclear stain is included in both images.

In COS-7 monolayer cells transfected with MV11.16E1{wedge}E4 (7) and harvested at 24, 48, or 72 h posttransfection, two 16E1{wedge}E4 species comparable in size to those detected in the HPV-16 raft culture were apparent by Western blotting (Fig. 1A). Again, the lower-molecular-mass species (which accumulated over time) was detected by TVG402 and not by anti-N-term.

The intracellular distribution of both full-length and N-terminally truncated 16E1{wedge}E4 was probed by immunofluorescence in formalin-fixed, paraffin-embedded HPV-16-induced cervical intraepithelial neoplasia (CIN) tissue and in raft sections (Fig. 1B). As has been observed previously, these two antibodies give different staining patterns (10). The N-terminally truncated form of the protein appears more diffusely distributed in the cell, extending beyond the full-length protein to the periphery of the cytoplasm. These intracellular differences can also be observed in E1{wedge}E4-expressing cells in raft culture (Fig. 1B, bottom panel), but in contrast to HPV-1, where the full-length species is lost in the upper epithelial layers (2, 6, 9), both 16E1{wedge}E4 forms were found to persist to the epithelial surface (Fig. 1B).

The full-length 16E1{wedge}E4 protein contains elements of ordered secondary structure. To investigate the structure of 16E1{wedge}E4, the full-length protein was cloned into the pET-DUET (NcoI, NotI) and pET-28b (BamHI, XhoI) vectors so as to express N-terminally and C-terminally histidine-tagged proteins. After expression of the proteins in Escherichia coli BL21/DE3 (Novagen), the cells were lysed in 8 M urea, 50 mM phosphate, 10 mM imidazole, and 100 mM NaCl (pH 8.0). 16E1{wedge}E4 protein was purified from inclusion bodies on cobalt TALON (BD Biosciences) beads and eluted at pH 4.0. Refolding was carried out by dilution of the protein to 20 µM with refolding buffer (100 mM NaCl, 50 mM phosphate, 2 mM β-mercaptoethanol, 5 mM EDTA), followed by dialysis in refolding buffer at 4°C.

Purity and molecular mass were assessed by SDS-polyacrylamide gel electrophoresis, N-terminal sequencing, and mass spectrometry. Following refolding, 16E1{wedge}E4 was found to exist primarily as a monomer, with a molecular mass, determined by sedimentation equilibrium centrifugation, close to that predicted from the amino acid sequence (11,139 Da). Further analysis by sedimentation velocity centrifugation revealed a major species of 12 kDa and a minor species of 25 kDa, which corresponds to the dimer form (Fig. 2A). As reported previously, monomeric 16E1{wedge}E4 was found to assemble over time into higher-order multimers up to the size of hexamers that can be visualized by SDS-polyacrylamide gel electrophoresis in the absence of reducing agents. Electron microscopy (EM) studies, however, revealed that these higher-order species precipitated into aggregates (Fig. 2A) that were not amenable to structural analysis. Since the aggregation of full-length 16E1{wedge}E4 stimulated the emergence of a heterologous mixture of multimeric species, our structural characterization focused on monomeric 16E1{wedge}E4 with a view to gaining a first insight into its multimerization.


Figure 2
View larger version (19K):
[in this window]
[in a new window]

  		FIG. 2. Purification and biophysical characterization of 16E1{wedge}E4. (A) The sedimentation velocity [c(s)] profile illustrates that 16E1{wedge}E4 is primarily monomeric with a small proportion in a dimeric state (left). Storage of 16E1{wedge}E4 results in the formation of aggregates here observed by negative staining under EM (right) (inset magnified x2). OD, optical density. (B) The far-UV CD spectrum of 16E1{wedge}E4 reveals elements of both alpha helix and beta strand (left); however, the near-UV CD spectrum indicates a limited tertiary structure (right). {Delta}{varepsilon}mrw, CD extinction coefficient calculated using mean residue weight. (C) The predicted secondary structure of 16E1{wedge}E4 (blue indicates beta strand, and red indicates helical regions) with confidence levels (0, low; 9, high) assigned by PSIPRED indicates that the beta-stranded region of 16E1{wedge}E4 is located at the C terminus and the helical region is N terminal, as indicated in the proposed model.

The secondary-structure content of 16E1{wedge}E4 was estimated by spectral decomposition of the native far-UV circular-dichroism (CD) spectrum (30) to contain 11% {alpha} helices, 32% β strands, and 15% turns (Fig. 2B). Notably, there is also a considerable amount of random coil (42%). The predicted secondary structure (www.psipred.net) (16) of 16E1{wedge}E4 (Fig. 2C) indicates a helical character for 16 amino acids, which are predominantly N terminal; however, apart from K14 and L15, the confidence levels are low. A beta-stranded C-terminal region encompassing 18 amino acids (T68 to H79 and T85 to L90) is predicted with high confidence. These data suggest a secondary-structure model of 16E1{wedge}E4 with a short region of {alpha} helix around K14 and L15 and a region of β strand encompassing T68 to H79 and T85 to L90 (Fig. 2C). Interestingly, in addition to being highly conserved among alpha group HPV E1{wedge}E4 proteins, both the C- and N-terminal regions of 16E1{wedge}E4 play key roles in the association of 16E1{wedge}E4 with keratin and 16E1{wedge}E4-mediated disruption of keratin dynamics.

To obtain a more detailed insight into the structure of 16E1{wedge}E4, a range of biophysical approaches were used. The near-UV CD spectrum of 16E1{wedge}E4 (Fig. 2B), with a band extending from 255 to 300 nm, is clearly dominated by the single tyrosine (Y10), indicating that this region of the protein is in a folded conformation. The absence of any significant tryptophan contribution indicates that the tryptophan residues (W20, W34, and W67) may be exposed on the surface of the protein, indicating that the protein is largely disordered. The position of the histidine tag did not influence the CD spectra. One-dimensional 1H nuclear magnetic resonance analysis of the full-length 16E1{wedge}E4 protein revealed a limited dispersion of the chemical shifts, reflecting little structural stability and a high degree of backbone mobility (data not shown). These results are in line with a recent study which analyzed alpha group HPV protein sequences by using predictors of intrinsic disorder and found that E1{wedge}E4 falls into the category of intrinsically disordered proteins (31).

The structure of the full-length 16E1{wedge}E4 monomer is dominated by a limited tertiary fold. Since the near-UV CD spectrum indicated that Y10 is restrained and adjacent amino acids are predicted to be helical, tryptophan fluorescence studies were carried out to assess if the protein does, indeed, have a fold. The 16E1{wedge}E4 fluorescence emission spectrum is dominated by a maximum at 352 nm, which indicates that the tryptophan residues (W20, W34, and W67) are solvent exposed. The accessibility of the tryptophan residues was further probed by acrylamide quenching (Fig. 3A). The effective quenching constant, Ksv, of 16E1{wedge}E4 was calculated to be 10 M–1. The Ksv of fully exposed tryptophan is in the range of 18 to 22 M–1, indicating that not all of the tryptophan residues are fully solvent exposed. To establish if, indeed, 16E1{wedge}E4 is folded in such a way as to limit the mobility of the tyrosine, two different fluorescence resonance energy transfer (FRET) studies were carried out with mutant proteins produced by the QuikChange site-directed mutagenesis protocol (Stratagene).


Figure 3
View larger version (40K):
[in this window]
[in a new window]

  		FIG. 3. Monomeric 16E1{wedge}E4 has a limited tertiary structure which, in the absence of N-terminal residues, forms amyloid-like fibrils; fibril formation is dependent on C-terminal amino acids. (A) Stern-Voler analyses of the acrylamide quenching of 16E1{wedge}E4 reveals that not all of the tryptophan residues are fully solvent exposed (left). Fluorescence excitation spectra were recorded to assess the contributions of W20 (green), W34 (blue), and W67 (red), observed at 280 nm, to the fluorescence spectrum of the MIANS label on C61. Background MIANS fluorescence is shown in black (middle). Excitation polarization spectra were recorded for the W34F (green) and W20F (red) mutant proteins to monitor tryptophan homotransfer for the W20/W64 pair and the W34/W64 pair, respectively (right). F and FO are the fluorescence in the presence and absence of acrylamide, respectively; au, arbitrary units. (B) Proposed arrangement of the 16E1{wedge}E4 polypeptide based on the tryptophan fluorescence data in which the polypeptide is folded so as to bring W20, W34, W67, and C61 into close proximity. The C-terminal region could occupy a number of orientations, as indicated by the letters A, B, and C; however, a less solvent-exposed orientation such as arrangement A may favor the highly hydrophobic C terminus. (C) Electron micrograph (inset magnified x2) of fibers assembled from N-terminally truncated E1{wedge}E4 (16E4{Delta}2-5) (left). Thioflavin T fluorescence at 485 nm, indicative of amyloid-like fibril structures, is observed in the presence of these fibers and fibers formed from 16E4 amino acids 66 to 92 (16E4 66-92) but not in the presence of 16E1{wedge}E4. Positive (amyloid-β) and negative (BSA) control samples are also included. (D) Far-UV CD spectrum of the C-terminal 26-amino-acid peptide of 16E1{wedge}E4 (16E4 amino acids 66 to 92) indicating a beta-stranded structure (left). The electron micrograph (right) shows that this peptide also assembles into fibrillar structures, as indicated by the thioflavin T binding data (inset magnified x2). {Delta}{varepsilon}mrw, CD extinction coefficient calculated using mean residue weight.

In the first experiment, FRET was used to assess the contributions of W20, W34, and W67 to the excitation spectrum of a MIANS (Molecular Probes) acceptor molecule attached to C61. W20F/W34F/C62A, W20F/W67F/C62A, and W34F/W67F/C62A mutant proteins were used to assess the relative contributions of W67, W34, and W20, respectively, to the excitation spectrum of the MIANS label. C61 was labeled with a MIANS label by reacting the protein with a fivefold excess of the probe and separating the labeled protein from the unlabeled probe on a PD10 column. A maximum was observed at 280 nm in the difference spectra of all samples, indicating that FRET takes place between the MIANS label and all of the tryptophan residues (Fig. 3A). A linear arrangement of the polypeptide chain would not facilitate this energy transfer. The fluorescence transfer is markedly lower for the W34F/W67F/C62A mutant protein, indicating that W20 is farther from C61 than either W34 or W67.

In the second experiment, excitation polarization spectra were used to study homo-FRET between the tryptophan residues in the W20/W67 and W34/W67 pairs with W34F and W20F single-tryptophan mutant proteins, respectively (Fig. 3A). Since homotransfer contributes to depolarization of the tryptophan emission spectrum more significantly upon excitation at 295 nm than at 310 nm, the 310/295-nm polarization ratio was used to provide an indication of the extent of energy transfer (22). The 310/295-nm polarization ratio was 2.06 for the W20/W67 pair and 2.03 for the W34/W67 pair. These data indicate that FRET occurred between both pairs of tryptophan residues and that W20 and W34 are at similar distances from W67, further supporting the idea of a limited fold and ruling out an extended conformation. These data are consistent with a simple arrangement of the polypeptide chain in which the elements of secondary structure are in close proximity (Fig. 3B), hence restraining Y10 and placing W20, W34, W67, and C61 in an arrangement which concurs with the fluorescence transfer data.

As illustrated in Fig. 3B, it is difficult to precisely establish the orientation of the C-terminal beta-stranded region, which may be extended or in an antiparallel beta sheet conformation. Both the C- and N-terminal regions harbor clusters of hydrophobic amino acids, and close association of these elements of secondary structure, particularly in arrangement A (Fig. 3B), would minimize their solvent exposure. Our results suggest a structural model of the full-length 16E1{wedge}E4 protein in which regions of secondary structure close to the N and C termini are partially restrained, forming a limited fold in this largely disordered protein. The formation of oligomeric species as observed for 16E1{wedge}E4 both in vivo and in vitro is a phenomenon commonly observed for intrinsically disordered proteins (14).

N-terminal deletion results in the assembly of 16E1{wedge}E4 into ordered amyloid-like fibrils. Having shown that the 16E1{wedge}E4 protein can be modified by N-terminal deletion, we looked at how this might affect the structural integrity of the full-length protein described above. To examine this, we used two 16E1{wedge}E4 N-terminal mutant proteins, 16E4{Delta}2-5 and 16E4{Delta}12-16 (28). Since the precise nature of the cleavage event that leads to N-terminal loss in differentiating epithelial cells has not yet been determined, we took 16E4{Delta}2-5 to represent a minimally truncated form of the protein. 16E4{Delta}12-16 is deficient in keratin binding and has been used in previous functional studies. When expressed in COS-7 cells, 16E4{Delta}2-5, like the full-length protein, was found to be cytoplasmic (with TVG405) and, as expected, could not be detected by anti-N-term. A similar distribution and staining profile was apparent with 16E4{Delta}12-16.

To examine the structure of the N-terminal deletion-containing forms of 16E1{wedge}E4, these constructs were amplified and cloned into pET-28b (BamHI, XhoI) to produce the N-terminally truncated, C-terminally histidine-tagged 16E4{Delta}2-5 and 16E4{Delta}12-16 proteins. Protein purity was assessed to be similar to that seen with the full-length 16E1{wedge}E4 protein. By far-UV CD analysis, the secondary structure of 16E4{Delta}2-5 was found to comprise 10% {alpha} helices, 32% β strands, and 15% turns, which differs little from that found in the full-length protein. In contrast to full-length 16E1{wedge}E4, however, 16E4{Delta}2-5 was found to rapidly assemble into multimeric structures upon refolding and to exist as a wide range of high-molecular-mass species when examined by analytical ultracentrifugation. By EM, 16E4{Delta}2-5 was found to differ significantly from the disordered aggregates that form slowly from full-length 16E1{wedge}E4 and appeared as flexible unbranched fibrils following negative staining (1% sodium silicotungstate, pH 7) (Fig. 3C). These fibrils were typically 100 to 300 nm long and 7 nm wide and had a ribbon-like morphology characteristic of amyloid fibers. As illustrated in Fig. 3C, the fibers were not amorphous but were evenly dispersed on the EM grid, suggesting that ordered aggregation takes place on refolding. The architecture of the 16E1{wedge}E4 fibrils was further probed with thioflavin T, which is known to bind specifically to the crossed beta pleated sheet structure of amyloid fibrils, producing a fluorescence emission maximum at 485 nm when excited at 442 nm (Fig. 3C). Both positive and negative controls in the form of amyloid-β and bovine serum albumin (BSA) were included in the assay. The thioflavin T fluorescence spectra indicate that N-terminally truncated 16E1{wedge}E4 (16E4{Delta}2-5), but not the full-length protein, assembles into amyloid-like fibers. These data also indicate the importance of the N-terminal region in constraining the monomeric structure of 16E1{wedge}E4.

Interestingly, the 16E4{Delta}12-16 mutant protein that lacks the leucine cluster region necessary for keratin association could also assemble into amyloid-like fibrils, as determined by EM and thioflavin T binding (data not shown). It is probable that deletion of this region disrupts the monomeric structure by displacement of more N-terminal residues. Displacement of this region, as occurs during keratin binding (8) or mitochondrial association (27), may also trigger the assembly of 16E1{wedge}E4 into fibers, as is seen following N-terminal truncation.

Since our structural model indicated a beta-stranded C-terminal region, we used a 26-amino-acid peptide corresponding to the highly conserved C terminus of 16E1{wedge}E4 (16E4 amino acids 66 to 92) in order to assess its contribution to the formation of beta-strand-based 16E1{wedge}E4 fibrils. Analysis of the peptide was carried out in 10% acetic acid. This peptide had a strong beta-strand character (32%) (Fig. 3D) and spontaneously formed fibers (Fig. 3D). These peptide fibers also bound thioflavin T (Fig. 3C), indicating that they, like the N-terminal mutant forms, are amyloid like in nature. The peptide fibers were, however, less ordered and more aggregated than those observed with 16E4{Delta}2-5, indicating that although C-terminal amino acids are key to the formation of E1{wedge}E4 fibers, other residues contribute to their overall organization. The C terminus of 16E1{wedge}E4 combines high beta sheet and beta aggregation propensity with high hydrophobicity and a low net charge, characteristics that are compatible with the nucleation of 16E1{wedge}E4 into beta-strand-based fibrils and other multimeric species (33). In common with other amyloidogenic proteins (12), the intrinsic disorder that facilitates 16E1{wedge}E4 fiber formation necessitates purification under denaturing conditions and refolding in order to obtain soluble protein for structural analysis.

Colocalization of 16E1{wedge}E4 with amyloid binding probes in vivo. Thioflavin T and Congo red are commonly used in the detection of amyloid structures both in vitro and in vivo (34). Here we found that although thioflavin T bound to 16E4{Delta}2-5 fibrils in vitro (Fig. 3C), it exhibited only weak fluorescence colocalization with 16E1{wedge}E4 expressed in tissue culture. BTA-1, also commonly referred to as Pittsburgh compound B, is a direct derivative of thioflavin T and has been shown to detect a number of amyloidogenic proteins in vitro (17, 35) and in vivo (15, 18, 20) with a greater affinity than thioflavin T. COS-7 cells transfected with pMV11.16E1{wedge}E4 were stained with a 16E1{wedge}E4 antibody in conjunction with BTA-1 (10 mM BTA-1 [Sigma] in 50% ethanol for 30 min at room temperature, followed by a 5-s wash in distilled water). Amyloid aggregates were detected only in cells expressing 16E1{wedge}E4, and in these cells, 16E1{wedge}E4 and BTA-1 colocalization was clearly observed (Fig. 4A). We concluded that this enhanced fluorescence was due to the much higher affinity of BTA-1 for the beta-strand-based amyloid-like 16E1{wedge}E4 structure and is consistent with the presence of 16E1{wedge}E4 amyloid-like fibers in tissue culture.


Figure 4
View larger version (46K):
[in this window]
[in a new window]

  		FIG. 4. Detection of the E1{wedge}E4 protein with the amyloid binding probes BTA-1 and NIAD-4 is consistent with the presence of E1{wedge}E4 amyloid-like fibrils in vivo and indicates that fibril formation is key to the accumulation of E1{wedge}E4. (A) Immunofluorescence images illustrating that 16E1{wedge}E4 expressed in COS-7 cells is readily detected with BTA-1 48 h after transfection. Specificity of NIAD-4 for 16E1{wedge}E4 amyloid is shown in vitro, where NIAD-4 exhibits a fluorescence maximum at 600 nm, indicative of its association with β-strand-based fibrils, with both 16E4{Delta}2-5 (blue) and amyloid-β(1-42) (turquoise) but not with 16E1{wedge}E4 (black) or BSA (yellow), NIAD-4 alone is also shown (green). Specificity of NIAD-4 for 16E1{wedge}E4 amyloid is shown in vitro, where NIAD-4 exhibits a fluorescence maximum at 600 nm, indicative of its association with β-strand-based fibrils, with both 16E4{Delta}2-5 (blue) and amyloid-β(1-42) (turquoise) but not with 16E1{wedge}E4 (back) or BSA (yellow); NIAD-4 alone is also shown (green). The immunofluorescence colocalization of NIAD-4 and 16E1{wedge}E4 in 16E1{wedge}E4-transfected COS-7 cells illustrates that NIAD-4 is specific for E1{wedge}E4 in tissue culture. Immunofluorescence images (magnification, x10) of an HPV-16 CIN lesion reveals NIAD-4 staining within the 16E1{wedge}E4-positive regions; a dashed white line has been superimposed on each panel to indicate the limits of the upper cornified layer of the lesion. A higher-power image of a single 16E1{wedge}E4-positive cell is shown in the bottom panel. au, arbitrary units. (B) Alignment of the E1{wedge}E4 protein sequences of a number of high-risk alpha group viruses illustrates considerable sequence homology in the C-terminal region (shown here); amino acids are shaded according to percent conservation. In the context of the above amino acid alignment, the average and standard deviation values for the beta-aggregation propensity (TANGO) of each amino acid position calculated for the E1{wedge}E4 proteins of HPV-16, -18, -33, and -45 illustrate the high amyloidogenic potential of this conserved C-terminal region. COS-7 cells infected with HPV-18 and HPV-33 E1{wedge}E4 were double stained for E1{wedge}E4 with a polyclonal (HPV-18 or -33) antibody and for amyloid with BTA-1; the colocalization observed indicates that both 18E1{wedge}E4 and 33E1{wedge}E4 form amyloid-like structures. (C) The intracellular aggregates observed on expression of the keratin binding mutant, E4{Delta}12-16, are readily detected in COS-7 cells with BTA-1, confirming its assembly into amyloid-like fibrils. In contrast, the C-terminal mutant (16E4{Delta}86-92) is unable to form amyloid-like structures and is not detected by BTA-1. (D) Western blot showing the relative distribution of mutant and wild-type E1{wedge}E4 proteins between the soluble and insoluble fractions. 16E4{Delta}86-92 exists as an exclusively soluble protein and does not, like the wild-type and 16E4{Delta}2-5 mutant proteins, accumulate in the insoluble fraction. (E) Structural model of the intracellular assembly of 16E1{wedge}E4 into amyloid-like fibers following N-terminal truncation during epithelial differentiation. Previous data suggested the existence of hexameric 16E1{wedge}E4 structures, which may be assembly intermediates. 16E4{Delta}2-5 and 16E4{Delta}12-16, which have lost N-terminal sequences and are predicted to behave like N-terminally truncated 16E4, accumulate in the cell and stain with the amyloid imaging probe BTA-1. The C-terminal E4 mutant protein 16E4{Delta}86-92 retains weak keratin binding ability but does not accumulate in the cell and does not stain with amyloid imaging probes.

In paraffin-embedded tissue sections of HPV-16-induced CIN, we found that the amyloid imaging probe NIAD-4 (Nomadics Inc., Life Sciences) offered the greatest sensitivity. NIAD-4 has been used to detect amyloid-β protein both in vivo and in vitro, where binding is accompanied by an enhancement of the fluorescence emission and a shift in the excitation spectrum (25). Here we have examined the in vitro binding of NIAD-4 to both 16E1{wedge}E4 and the N-terminal mutant protein 16E4{Delta}2-5 (Fig. 4A). Samples were analyzed in phosphate-buffered saline with 4 µM NIAD-4 and 16 µM protein. An enhancement of fluorescence at 600 nm (accompanied by a shift in the excitation maximum) was observed in the presence of 16E4{Delta}2-5 and amyloid-β (positive control) but not with 16E1{wedge}E4 or BSA (negative control). In addition, we have shown by immunofluorescence analysis that NIAD-4 can also specifically detect 16E1{wedge}E4 in transfected COS-7 cells (Fig. 4A). These data indicate that NIAD-4 specifically binds to the amyloid-like fibrils formed by 16E4{Delta}2-5. Tissue sections were stained with TVG405, followed by incubation with NIAD-4 (final concentration of 10 mM in 10% dimethyl sulfoxide-90% propylene glycol) for 45 min at room temperature. NIAD-4 staining was confined to pockets in the lesion where E1{wedge}E4 was found (Fig. 4A), with colocalization within the cell being the most apparent at high-power magnification (Fig. 4A, lower panels). Differences in the intracellular patterns seen with NIAD-4 and TVG405 may indicate variations in the multimeric state of 16E1{wedge}E4 or differences in the ability of antibodies and probes to access their respective binding sites.

Colocalization of the high-sensitivity amyloid imaging probes BTA-1 and NIAD-4 with 16E1{wedge}E4, shown here, is consistent with the presence of 16E1{wedge}E4 amyloid both in HPV-16-induced CIN lesions and when 16E1{wedge}E4 is expressed in tissue culture.

A comparison of the E1{wedge}E4 sequences of the most prevalent, high-risk, alpha group viruses (23) reveals a high degree of sequence homology particularly at the C and N termini. In addition, these proteins also display a strong C-terminal beta aggregation propensity (Fig. 4B). Preliminary work here indicates that E1{wedge}E4 proteins of HPV-18 and -33 can also be detected with BTA-1 (Fig. 4B), raising the possibility that detection of E1{wedge}E4 from multiple HPV types could be facilitated with amyloid imaging probes. These data also indicate that the E1{wedge}E4 proteins of HPV-18 and -33 may have overall structural properties similar to those defined here for 16E1{wedge}E4. The HPV alpha group has more than 40 members, containing a large group of homologous E1{wedge}E4 proteins that could potentially provide a valuable resource to further our understanding of the sequence determinants and mechanisms of amyloid fiber formation.

Accumulation of 16E1{wedge}E4 correlates with its ability to assemble into amyloid-like fibrils. The accumulation of 16E1{wedge}E4 in the upper layers of the epithelium, and in the insoluble compartment of the cell (8, 29, 33), has previously been attributed to its association and multimerization on the keratin intermediate-filament network. Our observation that 16E1{wedge}E4 N-terminal loss can stimulate fiber formation raises the possibility that, like the accumulation of other amyloidogenic proteins within the cell, that of 16E1{wedge}E4 may be linked to its ability to assemble into amyloid-like fibers. When expressed in COS-7 cells, both 16E4{Delta}12-16 and 16E4{Delta}2-5 could be visualized with the BTA-1 imaging probe (Fig. 4C), in agreement with the biophysical analysis outlined above. In addition, both of these mutant proteins (which are defective for keratin binding [33]) were found to accumulate to levels similar to those of the wild-type protein and were found primarily in the insoluble compartment of the cell (Fig. 4D). When expressed in cells, 16E1{wedge}E4 is processed to a mixture of full-length and truncated proteins; hence, it accumulates via a combination of keratin binding and N-terminal truncation. 16E1{wedge}E4 could also be detected with BTA-1 (Fig. 4A). The accumulation of the keratin-binding mutant form (16E4{Delta}12-16), however, is due to its ability to form amyloid-like fibrils. Wild-type and mutant 16E1{wedge}E4 proteins were expressed and fractionated as previously described (33). By contrast, 16E4{Delta}86-92, which lacks C-terminal sequences and which associates only poorly with keratin filaments (33), did not accumulate in the cells to high levels and was localized primarily in the soluble compartment (Fig. 4D). This is in agreement with our in vitro data, which show 16E4{Delta}86-92 to be monomeric when expressed in E. coli (data not shown). 16E4{Delta}86-92 was readily apparent with TVG405 and anti-N-term but could not be detected with amyloid imaging probes (Fig. 4C). These results indicate an important role for E1{wedge}E4 fiber formation in regulating its abundance.

Taken together, our results suggest that the in vivo accumulation of 16E1{wedge}E4 is linked to its ability to assemble into amyloid-like fibers. Fiber formation is facilitated by N-terminal truncation of 16E1{wedge}E4 but may also be triggered by the association of 16E1{wedge}E4 with keratin, both of which would free the C terminus of the protein for self-association (Fig. 4E). In addition to its role in keratin binding, it appears that the N terminus of 16E1{wedge}E4 is also required to constrain the structure of the protein in order to regulate its function in the cell.

HPV infection is known to induce abnormalities of the cornified cell envelope, with expression of 11E1{wedge}E4 in differentiated keratinocytes being associated with morphological cellular changes (3). The identification of an amyloid-like form of 16E1{wedge}E4 goes some way toward explaining its abundance during productive infection and gives new insight into the way in which E1{wedge}E4 may affect cellular integrity, as intracellular amyloid is associated with the disruption of a range of cellular functions (1, 19).


arrow
ACKNOWLEDGMENTS
 
We are grateful to John Skehel and Jonathan Stoye in the Division of Virology for supporting this work at NIMR.

This work was funded by the UK Medical Research Council.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Division of Virology, MRC National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, United Kingdom. Phone: 44 (0)20 8816 2623. Fax: 44 (0)20 8906 4477. E-mail: jdoorba{at}nimr.mrc.ac.uk Back

{triangledown} Published ahead of print on 18 June 2008. Back


arrow
REFERENCES
 
    1
  1. Almeida, C. G., R. H. Takahashi, and G. K. Gouras. 2006. Beta-amyloid accumulation impairs multivesicular body sorting by inhibiting the ubiquitin-proteasome system. J. Neurosci. 26:4277-4288.[Abstract/Free Full Text]
    2. 2
  2. Breitburd, F., O. Croissant, and G. Orth. 1987. Expression of human papillomavirus type 1 E4 gene products in warts. Cancer Cells 5:115-122.[CrossRef]
    3. 3
  3. Bryan, J. T., and D. R. Brown. 2000. Association of the human papillomavirus type 11 E1{wedge}E4 protein with cornified cell envelopes derived from infected genital epithelium. Virology 277:262-269.[CrossRef][Medline]
    4. 4
  4. Chiti, F., and C. M. Dobson. 2006. Protein misfolding, functional amyloid, and human disease. Annu. Rev. Biochem. 75:333-366.[CrossRef][Medline]
    5. 5
  5. Doorbar, J. 2006. Molecular biology of human papillomavirus infection and cervical cancer. Clin. Sci. (London) 110:525-541.
    6. 6
  6. Doorbar, J., I. Coneron, and P. H. Gallimore. 1989. Sequence divergence yet conserved physical characteristics among the E4 proteins of cutaneous human papillomaviruses. Virology 172:51-62.[CrossRef][Medline]
    7. 7
  7. Doorbar, J., R. C. Elston, S. Napthine, K. Raj, E. Medcalf, D. Jackson, N. Coleman, H. M. Griffin, P. Masterson, S. Stacey, Y. Mengistu, and J. Dunlop. 2000. The E1{wedge}E4 protein of human papillomavirus type 16 associates with a putative RNA helicase through sequences in its C terminus. J. Virol. 74:10081-10095.[Abstract/Free Full Text]
    8. 8
  8. Doorbar, J., S. Ely, J. Sterling, C. McLean, and L. Crawford. 1991. Specific interaction between HPV-16 E1{wedge}E4 and cytokeratins results in collapse of the epithelial cell intermediate filament network. Nature 352:824-827.[CrossRef][Medline]
    9. 9
  9. Doorbar, J., H. S. Evans, I. Coneron, L. V. Crawford, and P. H. Gallimore. 1988. Analysis of HPV-1 E4 gene expression using epitope-defined antibodies. EMBO J. 7:825-833.[Medline]
    10. 10
  10. Doorbar, J., C. Foo, N. Coleman, L. Medcalf, O. Hartley, T. Prospero, S. Napthine, J. Sterling, G. Winter, and H. Griffin. 1997. Characterization of events during the late stages of HPV16 infection in vivo using high-affinity synthetic Fabs to E4. Virology 238:40-52.[CrossRef][Medline]
    11. 11
  11. Doorbar, J., A. Parton, K. Hartley, L. Banks, T. Crook, M. Stanley, and L. Crawford. 1990. Detection of novel splicing patterns in a HPV16-containing keratinocyte cell line. Virology 178:254-262.[CrossRef][Medline]
    12. 12
  12. Fowler, D. M., A. V. Koulov, C. Alory-Jost, M. S. Marks, W. E. Balch, and J. W. Kelly. 2006. Functional amyloid formation within mammalian tissue. PLoS Biol. 4:e6.[CrossRef][Medline]
    13. 13
  13. Fowler, D. M., A. V. Koulov, W. E. Balch, and J. W. Kelly. 2007. Functional amyloid—from bacteria to humans. Trends Biochem. Sci. 32:217-224.[CrossRef][Medline]
    14. 14
  14. Frieden, C. 2007. Protein aggregation processes: in search of the mechanism. Protein Sci. 16:2334-2344.[CrossRef][Medline]
    15. 15
  15. Ishikawa, K., K. Doh-ura, Y. Kudo, N. Nishida, I. Murakami-Kubo, Y. Ando, T. Sawada, and T. Iwaki. 2004. Amyloid imaging probes are useful for detection of prion plaques and treatment of transmissible spongiform encephalopathies. J. Gen. Virol. 85:1785-1790.[Abstract/Free Full Text]
    16. 16
  16. Jones, D. T. 1999. Protein secondary structure prediction based on position-specific scoring matrices. J. Mol. Biol. 292:195-202.[CrossRef][Medline]
    17. 17
  17. Klunk, W. E., Y. Wang, G. F. Huang, M. L. Debnath, D. P. Holt, and C. A. Mathis. 2001. Uncharged thioflavin-T derivatives bind to amyloid-beta protein with high affinity and readily enter the brain. Life Sci. 69:1471-1484.[CrossRef][Medline]
    18. 18
  18. Klunk, W. E., Y. Wang, G. F. Huang, M. L. Debnath, D. P. Holt, L. Shao, R. L. Hamilton, M. D. Ikonomovic, S. T. DeKosky, and C. A. Mathis. 2003. The binding of 2-(4'-methylaminophenyl)benzothiazole to postmortem brain homogenates is dominated by the amyloid component. J. Neurosci. 23:2086-2092.[Abstract/Free Full Text]
    19. 19
  19. LaFerla, F. M., K. N. Green, and S. Oddo. 2007. Intracellular amyloid-beta in Alzheimer's disease. Nat. Rev. Neurosci. 8:499-509.[CrossRef][Medline]
    20. 20
  20. Matsuzaki, M., T. Hasegawa, A. Takeda, A. Kikuchi, K. Furukawa, Y. Kato, and Y. Itoyama. 2004. Histochemical features of stress-induced aggregates in alpha-synuclein overexpressing cells. Brain Res. 1004:83-90.[CrossRef][Medline]
    21. 21
  21. Middleton, K., W. Peh, S. Southern, H. Griffin, K. Sotlar, T. Nakahara, A. El-Sherif, L. Morris, R. Seth, M. Hibma, D. Jenkins, P. Lambert, N. Coleman, and J. Doorbar. 2003. Organization of human papillomavirus productive cycle during neoplastic progression provides a basis for selection of diagnostic markers. J. Virol. 77:10186-10201.[Abstract/Free Full Text]
    22. 22
  22. Moens, P. D., M. K. Helms, and D. M. Jameson. 2004. Detection of tryptophan to tryptophan energy transfer in proteins. Protein J. 23:79-83.[CrossRef][Medline]
    23. 23
  23. Munoz, N., F. X. Bosch, X. Castellsague, M. Diaz, S. de Sanjose, D. Hammouda, K. V. Shah, and C. J. Meijer. 2004. Against which human papillomavirus types shall we vaccinate and screen? The international perspective. Int. J. Cancer 111:278-285.[CrossRef][Medline]
    24. 24
  24. Nakahara, T., W. L. Peh, J. Doorbar, D. Lee, and P. F. Lambert. 2005. Human papillomavirus type 16 E1{wedge}E4 contributes to multiple facets of the papillomavirus life cycle. J. Virol. 79:13150-13165.[Abstract/Free Full Text]
    25. 25
  25. Nesterov, E. E., J. Skoch, B. T. Hyman, W. E. Klunk, B. J. Bacskai, and T. M. Swager. 2005. In vivo optical imaging of amyloid aggregates in brain: design of fluorescent markers. Angew. Chem. Int. Ed. Engl. 44:5452-5456.[CrossRef]
    26. 26
  26. Peh, W. L., K. Middleton, N. Christensen, P. Nicholls, K. Egawa, K. Sotlar, J. Brandsma, A. Percival, J. Lewis, W. J. Liu, and J. Doorbar. 2002. Life cycle heterogeneity in animal models of human papillomavirus-associated disease. J. Virol. 76:10401-10416.[Abstract/Free Full Text]
    27. 27
  27. Raj, K., S. Berguerand, S. Southern, J. Doorbar, and P. Beard. 2004. E1{wedge}E4 protein of human papillomavirus type 16 associates with mitochondria. J. Virol. 78:7199-7207.[Abstract/Free Full Text]
    28. 28
  28. Roberts, S., I. Ashmole, L. J. Gibson, S. M. Rookes, G. J. Barton, and P. H. Gallimore. 1994. Mutational analysis of human papillomavirus E4 proteins: identification of structural features important in the formation of cytoplasmic E4/cytokeratin networks in epithelial cells. J. Virol. 68:6432-6445.[Abstract/Free Full Text]
    29. 29
  29. Roberts, S., I. Ashmole, S. Rookes, and P. Gallimore. 1997. Mutational analysis of the human papillomavirus type 16 E1{wedge}E4 protein shows that the C terminus is dispensable for keratin cytoskeleton association but is involved in inducing disruption of the keratin filaments. J. Virol. 71:3554-3562.[Abstract]
    30. 30
  30. Sreerama, N., and R. W. Woody. 2000. Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set. Anal. Biochem. 287:252-260.[CrossRef][Medline]
    31. 31
  31. Uversky, V. N., A. Roman, C. J. Oldfield, and A. K. Dunker. 2006. Protein intrinsic disorder and human papillomaviruses: increased amount of disorder in E6 and E7 oncoproteins from high risk HPVs. J. Proteome Res. 5:1829-1842.[CrossRef][Medline]
    32. 32
  32. Walboomers, J. M., M. V. Jacobs, M. M. Manos, F. X. Bosch, J. A. Kummer, K. V. Shah, P. J. Snijders, J. Peto, C. J. Meijer, and N. Munoz. 1999. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J. Pathol. 189:12-19.[CrossRef][Medline]
    33. 33
  33. Wang, Q., H. Griffin, S. Southern, D. Jackson, A. Martin, P. McIntosh, C. Davy, P. J. Masterson, P. A. Walker, P. Laskey, M. B. Omary, and J. Doorbar. 2004. Functional analysis of the human papillomavirus type 16 E1{wedge}E4 protein provides a mechanism for in vivo and in vitro keratin filament reorganization. J. Virol. 78:821-833.[Abstract/Free Full Text]
    34. 34
  34. Westermark, G. T., K. H. Johnson, and P. Westermark. 1999. Staining methods for identification of amyloid in tissue. Methods Enzymol. 309:3-25.[Medline]
    35. 35
  35. Ye, L., A. Velasco, G. Fraser, T. G. Beach, L. Sue, T. Osredkar, V. Libri, M. G. Spillantini, M. Goedert, and A. Lockhart. 2008. In vitro high affinity {alpha}-synuclein binding sites for the amyloid imaging agent PIB are not matched by binding to Lewy bodies in postmortem human brain. J. Neurochem. 105:1428-1437.[CrossRef][Medline]

Journal of Virology, August 2008, p. 8196-8203, Vol. 82, No. 16
0022-538X/08/$08.00+0     doi:10.1128/JVI.00509-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Wang, Q., Kennedy, A., Das, P., McIntosh, P. B., Howell, S. A., Isaacson, E. R., Hinz, S. A., Davy, C., Doorbar, J. (2009). Phosphorylation of the Human Papillomavirus Type 16 E1^E4 Protein at T57 by ERK Triggers a Structural Change That Enhances Keratin Binding and Protein Stability. J. Virol. 83: 3668-3683 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McIntosh, P. B.
Right arrow Articles by Doorbar, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McIntosh, P. B.
Right arrow Articles by Doorbar, J.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»