Demonstration of antibodies against human papillomavirus type-11 E6 and L2 proteins in patients with recurrent respiratory papillomatosis

AURlS

.\lASUS

LARYNX IhTtRNA-IoNAL IOUANAL OF OR, 6, HNI Auris

Nasus

Larynx

24 (1997)

185-191

Demonstration of antibodies against human papillomavirus type-l 1 E6 and L2 proteins in patients with recurrent respiratory papillomatosis Atsushi Sameshima a, Toshinobu Fujiyoshi b,*, Sukjai Pholampaisathit ‘, Masato Ushikai a, Motoko Kono a, Soontorn Antarasena d, Katsunori Fukuda a, Shigeru Furuta ‘, Shunro Sonoda b, Masaru Ohyama a a Departmeni of Otoluryngology. Faculty of‘ IvJedicine, Kagoshima University, 8-35- 1 Sakuragaoka, Kagoshima 890, .Japan b Department of Virology, Faculty of Medicine. Kagoshima Unitrersity, 8-35-l Sakuragaoka, Kugoshima 890, Japan ’ Department of’ Virology, Ministry of Public Health, Amphur Huang, Nonthahuri. Thailand ’ Department of Otolaryngology, Rajcithi Hospital. Phyathai. Bangkok 10400, Thailand Accepted

16 October

1996

Abstract Recurrent respiratory papillomatosis (RRP) is highly prevalent in Thailand, with the human papillomavirus (HPV) type-l 1 being the most widespread. In this study, we isolated the HPV type-l 1 (HPV-11) genome from subjects with RRP and subcloned the E6 and L2 open reading frames (ORFs) with the expression vectors pEX1 and pEX3. The recombinant E6//?-gal and L2//?-gal fusion proteins were expressed in E. co/i. Using the recombinant proteins, we demonstrated the presence of antibodies against HPV-11 E6 and L2 in RRP patients by Western blot analysis. The prevalence of seropositivity for HPV-11 E6 and L2 were 5% (l/20) and 10% (2/20), respectively. Although RRP is caused by infection on the mucosal surface, it appears that an immune response occurs against viral proteins expressed in the epithelial lesions. 0 1997 ELsevier Science Ireland Ltd. Keywords:

Otolaryngology;

Recurrent

respiratory

papillomatosis;

Human

papillomavirus

type-l 1

1. Introduction

* Corresponding 992 658164.

author. Tel.: + 81 992 755283: fax: + 81

Recurrent respiratory papillomatosis (RRP) is a benign epithelial neoplasm of laryngeal mucosa, particularly involving the true vocal cords [1,2]. Clinically, RRP develops prior to puberty [3] and

0385-8146/97/$37.00 C 1997 Elsevier Science Ireland Ltd. All rights reserved PII SO385-8146(96)00000-X

186

A. Sameshima

Table 1 Summary Patient

of serological no.

SROOl SR005 SR006 SR007 SR009 SROlO SROll SR014 SR015 SR016 SR017 SR018 RAJOOS RAJ007 RAJOl7 RAJ019 CM002 CM004 CM006 CM007

Sex

F M F M M F F M F F F M M F F F M M F F

studies

for anti-HPV-11

Age at onset

5 4 6 3 5 4 5 3 4 4 3 4 2 2 1 -a -a -a -a -a

et al. /Auris

Xasus

Larynx

24 (1997)

185-191

antibodies HPV

DNA

(PCR-E6)

11 11 11 11 11 11 11 11 11 II 11 11 11 11 II 11 11 11 11 11

Anti-HPV-11

Ab (x 100)

Anti-D-gal

E6

L2

-

-

>16384

+ -

-

64 >I6384 >I6384 >I6384 4096 >I6384 4096 16 384 >I6384 16 384 4096 >16384 >16384 1024 64 1024 >I6384 > 16 384 >I6384

+ -

-

+ -

Ab

a Age not available.

causes serious sequelae, airway obstruction, frequent recurrences and a tendency to spread throughout the respiratory tract. RRP often requires repeated surgical procedures, tracheostomy or endoscopic excisions [ 1 - 51. We have previously demonstrated an extraordinarily high prevalence of RRP in Thai children [6]. The rate of RRP per 100000 children (ages O-14 years) in Thailand is 2.8, five to ten times higher than that of other countries [7710]. We also have previously shown HPV DNA in the lesions of Thai subjects with RRP. HPV-11 was isolated from 21 to 25 patients (84%) and HPV-6 from 1 of 25 (4%) [ll]. In this study, we have isolated HPV-11 DNA of viral isolates from a Thai RRP patient and expressed the E6 and L2 genes in Escherichia coli as p-galactosidase (p-gal) fusion proteins (E6//?-gal and L2/B-gal). These bacterially derived papilloma viral gene products were used to examine sera from RRP patients for antibodies to papillomavirus type-11 antigens (E6 and L2) by Western

blot analysis. We demonstrated antibodies against HPV-11 E6 and L2, indicating an immune response in some patients.

2. Materials

and methods

2.1. Subjects Twenty patients (8 males and 12 females) with RRP, who were admitted to Rajvithi Hospital, Siriraj Hospital, Bangkok, and University Hospital, Faculty of Medicine, Chiang Mai IJniversity, Thailand, were the subjects of this study. The average age of patients was 3.7 years (range l-6 years, Table 1). Sera and biopsy specimens were collected from patients at the time of endoscopic operations and stored at - 80°C until .assay. The presence of HPV-11 DNA in the biopsy specimens was determined by PCR and dot blot hybridization in a previous study [l 11.

A. Sameshima

et ul. /Auris

Nasus Laryr~s

24 (1997)

185-191

187

TPG-1 r”\

Fig. I. Cloning strategy for expression lac-Z, Esctwichia coli /I-galactosidase

of the HPV-I gene.

I E6 and L2 gene products.

2.2. Cloning strategy for expression of the HP V- 11 E6 and L2 gene products HPV-11 genomic DNA was cloned (SR006, see Table 1) using the EMBL3 vector and a BcrnrHI cut of the entire HPV-11 DNA (TPR-1) was obtained as described previously [l 11. Fig. 1 shows the cloning strategies for the expression of the E6 and L2 proteins. Briefly, an AvaII-PsfI fragment (nt 33-572) of the E6 gene region and a HindIII-XbaI fragment (nt 4557-5885) of the L2 gene region from TPG-1 were subcloned into the SmaI-PstI site of pEX1 [16,17] and the SmuIXbaI site of pEX3 [16,17], respectively. To verify the proper orientation and correct vector-insert junction, the SucI-X&I fragment of pEX-L2 were partially sequenced by Sanger’s dideoxy method P81 using the AmpliTaqT” sequencing kit (Takara, Kyoto, Japan). E. coli Pop2186 harboring HPV expression vectors were precultured overnight in LB medium

TPG-1,

Thai papilloma

gene-l

[I I]; PR, promotor:

containing ampicillin at 30°C. The overnight culture was diluted 1:100 and incubated at 310°Cfor 3 h. The cultures were then incubated at 40°C for 3 h to induce fusion proteins by a temperature shift [ 11,121.The cells were pelleted by centrifugation, suspended in 1 x sodium dodecylsulfate (SDS) gel loading buffer (50 mM Tris-Cl pH 6.8, 100 mM dithiothreitol, 2% SDS, 0.1% bromopheno1 blue, and 20% glycerol) and boiled for 3 min prior to being separated by 6% polyacrylamide gel electrophoresis (PAGE) [ 191. The proteins were stained with Coomassie brilliant blue. 2.3. Western blot analysis und udsorption of anti p -gul untibodies E. coli extracts harboring pEX3, pEX.-E6 or pEX-L2 were subjected to electrophoresis on 6% polyacrylamide gels containing 10% SDS. Following electrophoresis, the recombinant proteins were electrotransferred onto an immobilon-P mem-

188

A. Sameshima

et al. /Auris

brane (Millipore, Bedford, MA) via a semi-dry transfer system [20]. The membrane then was blocked with 5% non-fat dry milk, 0.1% sodium azide in phosphate-buffered saline (PBS)-0.1% Nonidet P-40 at 4°C overnight [21]. Blocking was followed by several washes with PBS NonidetP40, and the dried membranes cut into 5 mm wide strips. Patient sera were incubated with milk buffer containing E. coli-pEX3 lysates overnight to adsorb anti-p-gal antibodies. Milk buffer was prepared from induced E. co/i-pEX3 cultures, that had been pelleted, resuspended in 4% SDS, and sonicated. The final concentration of SDS was reduced to 0.14”/0 by adding PBS Nonidet P-40 containing 5% nonfat dry milk. Pretreated sera were diluted 1:lOO and incubated with strips for 30 min. The strips were washed with PBS Nonidet P-40 several times, then incubated with biotinylated goat antihuman IgM and IgG as a second antibody, and detected by the ABC method using a Vectastatin ABC kit (Vector Lab, CA).

3. Results

Nasus Lnr~nx

24 (1997)

185-191

4 L24Sgal dEWa-gal 116ZKD

+ B -gal

Fig. 2. Expression of recombinant E6/p-gal and L2/,!I-gal fusion proteins. Cell lysates of E. coli pop2186 (lane 2), pop2186 transformed with pEX-3 (lane 3), pEX-E6 (lane 4) and pEX-L2 (lane 5) following 6”/0 SDS-PAGE: and staining with Coomassie brilliant blue. /I-gal, E6:/J-gal and L2//I-gal proteins were produced by heat induction (lanes 3. 4 and 5). The position of marker proteins and their molecular weights in kilodaltons are shown in lane 1.

antibody titers were higher than the mean of 21 normal Japanese (14 males, 7 females, 1:2173. P < 0.0001). Because cross-reactivity of anti-p-gal 1

3.1. E.xpression of HP V- 11 E6 and L2 $usion proteins

S.M. preadsorption

E. coli pop2186 harboring pEX-E6 and pEX-L2 produced recombinant fusion proteins EG//j’-gal and L2/p-gal, respectively, following heat induction (Fig. 2). The molecular weight of each recombinant protein on SDS-PAGE gel (B-gel: 116 kD, E6//?-gal: 134 kD, and L2ip-gal: 165 kD) correlates to the predicted values calculated from sequence data (E6: 18 kD, L2: 44 kD). 3.2. Adsorption

of anti-p-gal

200KD t

SROlO -+-+

3

4

5

SROll

rL2/f3-gal

-4 Efi/#bgal

antibodies

We tested for the presence of antibody against HPV-11 E6 and L2 in the sera of 20 RRP patients by Western blot of strips prepared as described in Section 2. Fig. 3 shows the presence of anti-o-gal antibody that cross-reacts with E6/p-gal and L2/ a-gal. All 20 RRP subjects had anti-/j-gal antibody. The titers of anti-/?-gal antibody ranged from 1:64 to greater than 1.16384 (Table 1). The

2

116.25, KD

4 p-gal

Fig. 3. Demonstration of anti-p-gal antibody binding to recombinant fusion proteins by Western blot. f?. di lysates harboring pEX-3. pEX-E6 and pEX-L2 were separated by 6% SDS-PAGE, then transferred onto an immobilon-P membrane. Western blot strips were incubated with untreated patient sera (lanes 2 and 4) and preadsorbed patient sera (lanes 3 and 5). Lane 1 shows size marker.

A. Same&ma 12 S.M. preadsorption

3 SR005 -+-+

4

et al. :‘Auris

5

RAJO17

Nasus Laryn.u

24 (1997)

185-191

189

were positive for anti-p-gal antibody. The rate of seropositivity of anti HPV-11 proteins was 0% (O/20) among the healthy Japanese control.

4. Discussion

Fig. 4. Demonstration of anti-E6 and L2 antibody recognizing the HPV-11 E6 and L2 fusion proteins on Western blot. Patient sera were used at a 1:lOO dilution either untreated (lanes 2 and 4) or preadsorbed with milk buffer containing E. co/i-pEX3 lysate (lane 3 and lane 5). Lane 1 shows size marker.

antibody interferes with E6/p-gal and L2:‘P-gal fusion protein detection, patients sera was pretreated with milk buffer containing E. coli-pEX3 lysate to adsorb anti-b-gal antibody (Fig. 3, lanes 3 and 5). 3.3. Demonstration

of anti-E6

and L2 antibodies

Patient sera were tested at a 1:lOO dilution either untreated or preadsorbed with milk buffer containing E. coli-pEX3 lysate. As shown in Fig. 4, preadsorbed sera reacted only with the E6:/3gal or L2:P-gal fusion protein (lanes 3 and 5), though untreated patient sera reacted with P-gal. E6/P-gal and L2i/?-gal proteins (lanes 2 and 4). None of the 20 control Japanese samples reacted with E6/b-gal and L2/P-gal proteins after adsorption of anti P-gal antibodies (data not shown). The rate of seropositivity for antibodies against b-gal, E6!p-gal and L2/p-gal in 20 RR‘P patients is shown in Table 1 and summarized in Table 2. One patient was positive for anti-E6 antibody, and two patients were positive for anti-L2 antibody at 5 and lo%, respectively, while all subjects

Our previous study has revealed that HPV-11 DNA was detected in 84% of Thai subjects with RRP [ 111. Thus, HPV-11 seems to be the predominant HPV type in Thai RRP cases. In this report, we detail humoral immune responses against the E6 and L2 proteins of HPV-11 in RRP patients in Thailand using recombinant fusion proteins. These recombinant fusion proteins were used in Western blot analyses to detect antibo’dies. Although the problem of antibodies directed against E. coli /?-gal proteins is formidable, it. can be overcome by extensive preadsorption of sera with milk buffer containing a large amount of bacterial lysate as described [2 l-241. All patients had anti-p-gal antibodies. While 100% of normal Japanese individuals also have anti-p-gal antibodies, the anti-b-gal antibody titers of Thai RRP cases were higher than those of normal Japanese (P < 0.0001). Thus, adsorption of P-gal antibodies was very important to avoid false-positive reactions. We demonstrated anti-E6 and L2 antibodies among some Thai RRP patients. The E6 protein is expressed in the early stages of HPV infection and is not included in mature virions [I 3,25,26]. The presence of anti-E6 antibody indicates that HPV-infected cells of the respiratory lepithelial cells are recognized by the host immune system. While, anti-E6 antibodies have been demonstrated among patients with cervical neoplasia [26-291, this is the first report demonstrating anti-E6 antibodies among RRP patients. Bonnez et al. [14] have reported antibody responses to HPV-11 in RRP patients using a whole HPV-11 virion-based ELISA. Forty-seven percent of patients’ sera had reactivity above the cut-off value. HPV-11 virions were obtained from condyloma acuminata instead of RRP lesions. HPV-11 virions contain two viral capsid proteins, Ll and L2, but not early gene products [13,25,26]. Whole virion ELISA systems could detect the antibodies

190 Table 2 Prevalence Examined

A. Samrshima

of anti-E6

or anti-L2

cases

antibodies Anti-E6

20

et al. / Auris

among

Thai

RRP

Ab( +)

,Vasus Larynx

Ab( +)

2 ( 10%)

against not only linear epitopes, but. also the conformational epitopes. We used a portion of the L2 proteins as recombinant B-gal/L2 fusion proteins to detect antibodies. Thus, anti-L2 antibody positivity in this study was lower than that previously described using ELISA. Tachezy et al. [1.5] have examined the presence of anti-L2 epitope antibodies among adults with RRP using synthetic peptides derived from common epitopes of the HPV-6 and - 11 L2 proteins. They showed higher positivity rates of anti-L2 antibodies using this synthetic peptide, again suggesting the ELISA system might have a higher sensitivity than the Western blot system used in our studies. Anti-E6 antibodies were also detected among HPV-related malignancy patients (‘Table 3). Muller et al. [27,28] demonstrated that 49% of carcinoma patients were positive for anti-E6 antibody using ELISA. Furthermore, Sasagawa et al. [26] demonstrated that anti-E6 antibody was positive in 23% of Japanese patients with cervical neoplasias. Thus, early gene products of HPV appeared to be recognized by host immune system among HPV related malignancy patients and also RRP patients. Because there are no major differences in the HPV associated with RRP in Thailand versus that

Patients

positivity

among

HPV-related

Acknowledgements We would like to thank Dr A. Hakura for his input and S. Katahira for her kind assistance. This work was supported by a Grant-in-Aid for Cancer Research and Overseas Scientific Research Survey (0142006) from the Japanese Ministry of Education, Science and Culture.

References [l]

[3] Researchers

Cervical neoplasias Cervical cancer

23

Sasagawa

Cervical neoplasias

3.5

49

Ab( +)

in other countries [l 11, the high prevalence of RRP in Thailand cannot be explained by virulence of these HPV strains. Some other cofactors could be involved in this prevalence of RRP in Thailand [I 11. Examination of the prlesence of antibodies against HPV proteins using recombinant protein assays could be a useful technique to determine the prevalence of HPV infection in the general population and could enhance understanding of modes of viral transmission. Further epidemiologic investigations that include environmental and genetic data are necessary.

malignancy

Positivities of anti-E6 Ab (%$

Anti-/!-gal 20 (IOO’XJ)

[Z] Table 3 Anti-E6 antibody patients

185-191

patients Anti-L2

1 (5%)

24 (1997)

[4] et al.

1261

[5]

Muller et al. ~27,281 Ghosh et al. [29]

[6]

Kashima HK, Shah K. Recurrent respiratory papillomatosis, clinical overview and managemenl. principles. Obst Gynecol Clinics N Am 1987;14:581-588. Kashima HK, Mounts P. Leventhal B et al. Sites of predilection in recurrent respiratory papillomatosis. Annu Otol Rhino] Laryngol 1993;102:580-583. Strong MS, Vaugman CW, Hearly GB, et al. Recurrent respiratory papillomatosis. Management with CO, laser. Annu Otol Rhino1 Laryngol 1976;85:5088516. Benjamin B, Parsons DS. Recurrent respiratory papillomatosis A 10 year study. J Laryngol Otolog 1988:10221028. Abramson AL, Steinberg BM, Winkler BD’. Laryngeal papillomatosis. Clinical. histopathological and molecular studies. Laryngoscope 1987;97:6788685. Antarasena S. Clinical study of juvenile laryngeal papilloma. Acta Otolaryngol Suppl (Stockh) 1988;458:163166.

A. Samrshima

et al. / Auris

M, Antarasena S. Sonoda S. et al. Juvenile 171 Ohyama laryngeal papillomatosis in Thailand. In: Sasaki R and Aoki K. eds. Epidemiology and Prevention of Cancer, Nagoya, Japan: University Nagoya Press. 1990;53-59. Acta OtoPI Bomholt A. Juvenile laryngeal papillomatosis. laryngol (Stockh) 1988;105:367~371. P, Kashima HK. Association of human papillo191 Mount mavirus subtype and clinical course in respiratory papillomatosis. Laryngoscope 1984;94:28-33. A. Juvenile la[lOI Christensen PH, Jorgensen K, Grentved ryngeal papillomatosis of the larynx. Acta Otolaryngol Suppl (Stockh) 1984;412:37-39. IllI Ushikai M, Fujiyoshi T, Kono M. et al. Detection and cloning of human papillomavirus DNA associated with recurrent respiratory papillomatosis in Thailand. Jpn J Cancer Res 1994;85:699-703. u21 Sah KV, Kashima HK, Polk BF, et al. Rarity of caesarean delivery in cases of juvenile laryngeal papillomatosis. J Obstet Gynecol 1986;68:795-799. of papillo[I31 Crum CP, Barber S, Roche JK. Pathobiology mavirus-related cervical diseases. Prespects for immunodiagnosis. Clin Microbial 1991;4:270&285. HK, Leventhal B. et al. Antibody [I41 Bonnez W, Kashima response to human papillomavirus (HPV) Type 11 in children with juvenile-onset recurrent respiratory papillomatosis (RRP). Virology 1992;188:384-387. E, Ranst MV. et al. Antibody [ISI Tachezy R, Hamsikova response to a synthetic peptide derived from the human papillomavirus Type 6/l I L2 protein in recurrent respiratory papillomatosis. Correlation between sourhern blot hybridization, polymerase chain reaction. and serology. J Med Virol 1994;42:52-59. of a new family of 1161Stanley KK, Luzio JP. Construction high efficiency bacterial expression vectors. Identification of cDNA clones coding for human liver proteins. EMBO J 1984:3:1429%1434. H, Herz J, Bressan GM et al. Efficient con[I71 Haymerie struction of cDNA libraries in plasmid expression vectors using an adapter strategy. Nucleic Acids Res 1986;14. t181 Sanger F, Nicklen S. Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Nat1 Acad Sci USA 1977;74:5463-5467.

!Vasus Larynx

24 (1997)

185- 191

191

UK. Cleavage of structural proteins during the u91 Laemmli assembly of the head of bacteriophage T4. Nature 1970;227:680-685. transfer PO1 Burnette W. ‘Western blotting’: Electrophoretic of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein. Anal Biochem 1981;112:195-203. A et al. IdentificaPII Jenison SA, Firzlaff JM, Langenberg tion of immunoreactive antigens of human papillomavirus Type 6b by using Esrherichia co/i-expressed fusion proteins. J Virol 1988;62:2115-2123. P21 Tomita Y, Shirasawa H, Sekine H et al. Expression of the human papillomavirus Type 6b L2 open reading frame in Escherichiu r&i. L2-p-galactosidase fusion prol:eins and their antigenic properties. Virology 1987:158:8% 14. Y. Shirasawa H, Simizu B et al. Expression of ~231 Tomita human papillomavirus Types 6b and 16 Ll open reading frames in Eschuichia co/i. Detection of a 56OOO-dalton polypeptide containing genus-specific (common) antigens. J Virol 1987:61:2389-2394. 1241 Tomita Y, Fuse A, Sekine H et al. Human papillomavirus Type 6 and 11 E4 gene products in condyloma acuminata. J Gen Virol 1991;72:731-734. LP, Haugen TH. Transforming and regulatory 1251 Turek functions of bovine papillomavirus Type 1. In: Syrjanen K, Gissmann L and Koss LG. eds. Papillomaviruses and Human Disease. Berlin: Springer-Verlag. 1987;4#09-442. of PI Sasagawa T, Inoue M: Tanizawa 0 et al. Identification antibodies against human papillomavirus Type 16 E6 and E7 proteins in sera of patients with cervical ncoplasias. Jpn J Cancer Res 1992;83:705-713. M, Gansepohl H, Martynoff G et al. Identifica[271 Muller tion of seroreactive regions of the human papillomavirus Type-16 proteins E4, E6 and E7. J Gcn Virol 1990;71:2709~2717. WI Muller M, Viscidi RP. Sun Y et al. Antibodies to HPV-16 E6 and E7 proteins as markers for HPV-16 associated invasive cervical cancer. Virology 1992: 187:508- 5 14. re~291 Ghosh AK, Smith NK, Stacy SN et al. Serological sponses to HPV-16 in cervical dysplasia and neoplasia. Correlation of antibodies to E6 with cervical cancer. Int J Cancer 1993;53:591-596.