Laboratory diagnosis of latent human papillomavirus infection

679

DIAGN MICROBIOLINFECTDIS 1992;15:679-683

Laboratory Diagnosis of Latent Human Papillomavirus Infection Patricia McNicol, Fernando Guijon, Robert Brunham, Michael Gray, and Maria Paraskevas

The etiologic association of human papillomavirus (HPV) with uterine cervical cancer has prompted the need for improved laboratory diagnosis of this virus. The application of conventional hybridization technology, including filter in situ hybridization (FISH) and Southern-blot analysis, has revealed that the detection and typing of the virus is inconsistent between sequential specimens from the same individual. To determine whether the polymerase chain reaction (PCR) can be used to provide a more accurate assessment of infection status, two exfoliated cervical cell specimens obtained sequentially from a c o -

hort of 30 women without clinical evidence of cervical abnormalities were analyzed in parallel by FISH and PCR at 6month intervals. Neither of the procedures provided consistent findings with two sequential specimens suggesting that multiple analyses are necessary to assess infection accurately. However, PCR was less subjective in interpretation and demonstrated greater specificity than did FISH. With the increased sensitivity inherent to PCR, our findings indicated that PCR is more likely to identify latent HPV infection with a single specimen.

INTRODUCTION

infection status become a useful prognostic indicator of cervical disease risk, then diagnostic procedures must overcome these variables. The application of polymerase chain reaction (PCR) (Saiki et al., 1985) to detection of HPV (Shibata et al., 1988) affords us the opportunity to determine w het her PCR facilitates consistent diagnosis of HPV infection as compared with a conventional screening procedure such as FISH.

Ongoing evaluation of w o m e n with cervical h um an papillomavirus (HPV) infection by filter in situ hybridization (FISH) revealed inconsistencies in HPV detection and typing between sequential exfoliated cell specimens (McNicol, et al., 1990). This phenomenon has been doc um ent ed by other investigators using FISH (Reeves et al., 1989a) and Southern-blot analysis (Rando et al., 1989; Schneider, et al., 1988). The variability in HPV detection and typing observed between serial specimens using these procedures could result from sampling error; different numbers of HPV-infected cells are collected, and/or different regions of the cervix are sampled. Fluctuation in virus genome copy n u m b e r per infected cell may be another confounding factor. Should HPV From the Cadham ProvincialLaboratory(P.M.), Winnipeg; and Departments of Medical Microbiology(P.M., R.B., M.G.), Obstetrics Gynecologyand ReproductiveSciences (F.G.), and Pathology (M.P.), Universityof Manitoba, Winnipeg, Manitoba, Canada. Address reprint requests to Dr. P. McNicol, Cadham Provincial Laboratory, 750 William Avenue, P.O. Box 8450, Winnipeg, Manitoba R3C 3Y1, Canada. Received 7 October 1991; revised and accepted 14 February 1992. © 1992 Elsevier Science Publishing Co., Inc. 655 Avenue of the Americas, New York, NY 10010 0732-8893/92/$5.00

MATERIALS A N D M E T H O D S Patient P o p u l a t i o n For this investigation, a cohort of 30 w o m e n with at least two successive normal Papanicolaou smears were assessed for HPV infection by parallel FISH and PCR analysis of exfoliated cervical cells. The w o m e n were reevaluated by these two procedures after 6 months and also u n d e r w e n t a colposcopic examination and second Papanicolaou smear to confirm that the cervix remained normal. These w o m e n were considered to be at risk for HPV infection because of specific socioeconomic and behavioral characteristics including a low level of education and income and multiple sex partners, although partners did not change during the course of the investigation.

680

P. McNicol et al.

Specimen Analysis Exfoliated cervical cells were collected for parallel FISH and PCR analysis by the combined method of swabbing the endocervical canal and scraping the squamocolumnar junction. The cells were suspended in 5 ml of phosphate-buffered saline (PBS) and held at -20°C until parallel analysis by FISH and PCR. Prior to FISH analysis, a 50-p3 aliquot of cell suspension estimated to contain - 2 . 5 x 105 cells was removed from each specimen by using a Gilson Microman pipette (Mandel Scientific) with disposable tip and piston. These aliquots were analyzed by PCR as described below. The remaining cell suspension was analyzed by FISH as previously described (McNicol et al., 1989). For the PCR reaction, DNA was isolated from each of the 50-~l-specimen aliquots as previously described (McNicol and Dodd, 1990) using positive displacement pipettes with disposable tips and pistons throughout the procedure. In addition, two physically separated laboratories were used for specimen preparation and PCR setup and for analysis of PCR products to prevent problems with contamination. Oligonucleotide primers (Table 1)--targeting sequences within the human HLA class-II locus (Erlich and Bugawan, 1990)--were used to amplify 250 ng of DNA from each specimen under conditions described below to insure that amplifiable cellular DNA was present. The remainder of each specimen was TABLE 1

analyzed for HPV DNA by using oligonucleotide primers (Table 1) targeting type-specific sequences within the E6 gene of HPV 6, 11, 16, and 18 for the amplification of 189, 230, 240, and 160 base pairs, respectively. Two sets of primer pairs for HPV 11 and 16 or HPV 16 and 18 were added to a final concentration of I p,M in duplicate 100-p,1 reactions. PCR reactions contained 50 mM KC1, 10 mM TrisCI, pH 8.3, 1.5 mM MgC12,200 ~M each of dATP, dGTP, dCTP, and TFP, and -750 ng of cellular DNA. The concentration of Taq polymerase was 2.5 U. Thirty-five cycles of amplifications were carried out beginning with DNA denaturation at 94°C for 1 min and primer annealing at 55°C for I rain, followed by primer extension at 65°C for 30 sec and then at 72°C for 2 rain. Negative controls included reactions lacking DNA template and reactions with human DNA lacking HPV target sequences. Positive control reactions contained 55 pg of HPV genomic DNA. Amplification products were precipitated and analyzed as previously described (McNicol and Dodd, 1990). The molecular weight of amplification products was determined by comparison to Hae III restriction fragments from pBR322 plasmid DNA after agarose gel electrophoresis. HPV type specificity of products was confirmed by sequential hybridization of Southern blots to oligonucleotide probes (Table 1), 5'-endlabeled with 32p. Blots were prehybridized for 3 hr and then hybridized for 1 hr at 45°C as previously described. Membranes were washed at 45°C in 6 ×

Sequence and Location of Oligonucleotide Primers and Probes

Primer or Probe

Sequence (5'-3')

Genome location (bp)

HPV16 Primer Primer Probe

GATGGGAATCCATATGCTGTA TCGACCGGTCCACCGACCCCT GCCACTGTGTCCTGAAGAAAAGC

E6 E6 E6

269-289 488-508 427-449

HPV 18 Primer Primer Probe

CAGTATACCCCATGCTGCATGCC CGGTTTCTGGCACCGCAGGCACC CAGACTCTGTGTATGGAGACAC

E6 E6 E6

278-300 417-437 349-370

HPV 6 Primer Primer Probe

GCTGGATATGCAACAACAGTTG CATGCATGTTGTCCAGCAGTG GCTACCTGTGTCACAAACCG

E6 E6 E6

348-369 514-535 412-432

HPV 11 Primer Primer Probe

GCAGCGTGTGCCTGTTGCTTA AAGCAACGACCCTTCCACTGG CCTGTGTCACAAGCCGTTGTG

E6 E6 E6

285-305 494-514 415-435

HLA-DQ Primer GH26 Primer GH27

GTGCTGCAGGTGTAAACTTGTACCAG CACGGATCCGGTAGCAGCGGTAGAGTTG

HPV, human papillomavirus; and bp, base pairs.

HLA class II DQ locus

Latent HPV

TABLE 2

681

HPV Detection and Typing by FISH Between Sequential Specimens

A 123

4

56

7

8

First Specimen Second Specimen

No HPV

No HPV 6/11 16/18 6/11/16/18

5 1 1 2

3 0 0 0

0 0 0 0

2 1 2 3

10 2 3 5

Total

9

3

0

8

20

6/11 16/18 6/11/16/18 Total

B1

2 3

4

5

6

7

8

HPV, human papillomavirus; and FISH, filter in situ hybridization. ..~-24obp

SSC-0.1% SDS for 10 min and then in 2 x SSC-0.1% SDS for 10 min. Autoradiography was carried out for 6-8 hr with intensifying screens.

RESULTS Of the 30 women examined initially, 20 complied with our request to provide a second specimen after 6 months. Results of the first and second FISH and PCR analyses are compiled in Tables 2-5. A representative analysis of specimens by PCR is presented in Figure 1. The limit of sensitivity of FISH was 105 HPV genome copies; PCR could detect six HPV genome copies per specimen (data not shown). Both procedures gave variable results for the sequential specimens. The results of two analyses were in agreement for eight (40%) and nine (45%) of the 20 sequential specimens analyzed by FISH (Table 2) and PCR (Table 3), respectively. Discrepant findings included switching between positive and negative infection status as well as changes in HPV type. There was no discernable pattern to the variations observed. The overall prevalence of specific HPV types within the study cohort was similar for the first and second specimens analyzed by the same procedure. There was no significant difference in the prevalence of HPV infection as identified by FISH between the first and second specimens (p = 0.75). The FISH procedure identified - 50% (16 of 30, Table 4; and 10 of 20, Table 2) of the total population to be infected with HPV. As reported by other investigators (de Villiers et al., 1987; Reeves et al., 1989), FISH identified a high prevalence of mixed infections. As expected, PCR analysis detected a higher prevalence of HPV infection than did FISH, although this was evident for the first specimen only (p=0.012). Of the 30 women, 25 (83%) were positive for HPV at the first sampling (Table 4) and 11 (55%) of 20 were positive at the second sampling (Table 5). Once

C12

3

4

5

6

7

8

.~.23obp

FIGURE 1 Detection of human papiUomavirus (HPV) type 16 and 11 sequences in cervical cells by PCR. (A) Amplification products were resolved by agarose gel electrophoresis: lanes 1, 3, and 4, cervical specimens negative for HPV 16 and 11; lane 2, pBR322 marker DNA restricted with Hae III; lane 5, the expected 240 base pair (bp) fragment indicative of HPV 16 infection in a cervical specimen; lane 6, reaction without DNA template as a negative control; lane 7, the expected 230 bp fragment is observed in the HPV-11-DNA-positive control; and lane 8, the 240-bp fragment is visible in the HPV 16-DNA positive control. (B) An 18-hr autoradiograph from the blot of the gel shown in A, probed with the HPV-16-specific probe, shows a strong signal in lanes 5 and 8 only as expected. (C) The 18hr autoradiograph of same blot probed with the HPV-11specific probe shows a signal in lane 7, the HPV-11-positive control.

again, discrepant findings between the two sequential specimens included changes in HPV type and switching between positive and negative status. The most prevalent HPV type carried by the cohort was HPV 16. Of 20 women, eight (40%) providing two specimens were consistently positive for this type by PCR (Table 3), with up to 73% of the cohort demonstrating positivity for this type at least once (Table 4). PCR detected a lower prevalence of mixed infec-

P. McNicol et al.

682

TABLE 3

HPV Detection and Typing By Polymerase Chain Reaction Between Sequential Specimens First Specimen

Second Specimen

No HPV

16

18

16 + 18

6 + 18

Total

No HPV 16 18 6+16 6+11

2 0 0 0 0

3 6 0 1 0

0 0 1 0 1

3 1 1 0 0

1 0 0 0 0

9 7 2 1 1

Total

2

10

2

5

1

20

HPV, human papillomavirus.

tions with seven (23%) observed for the 30 first specimens (Table 4) and two (10%) observed for the 20 second specimens (Table 5). W h e n the results obtained by FISH and PCR were c o m p a r e d for the first (Table 4) and second specimens (Table 5), there was a g r e e m e n t for six (20%) of 30 specimens a n d 12 (60%) of 20 specimens, respectively. The poorest agreem e n t with PCR was o b s e r v e d for the FISH HPV 6/11 c o m b i n e d genomic probe. Of the first 15 specimens d e t e r m i n e d to harbor HPV, 6/11 alone or in conjunction with HPV 16/18 by FISH (Table 4), only two (13%) were confirmed by PCR to contain either HPV 6 or 11. Conversely, of the 11 first specimens determ i n e d by FISH to harbor HPV 16/18 alone or in conjunction with HPV 6/11, all were confirmed by PCR to contain either HPV 16 or HPV 18 DNA. The lack of specificity of the HPV 6/11 probe u s e d in FISH was also observed for the second specimens (Table

5). TABLE 4

C o m p a r i s o n of HPV Detection and T y p i n g by Parallel FISH and PCR Analysis of First Cervical Cell Specimen

DISCUSSION

FISH PCR No HPV 16 18 16+18 18+6 Total

No HPV

6 / 1 1 1 6 / 1 8 6/11/16/18

Total

4 8 0 2 0

1 1 0 2 0

0 1 0 0 0

0 6 2 1 1

5 16 2 5 1

14

5

1

10

30

HPV, human papiUomavirus; FISH, filter in situ hybridization; and PCR, polymerase chain reaction.

TABLE 5

C o m p a r i s o n of HPV Detection and T y p i n g b y Parallel FISH and PCR Analysis of Second Cervical Cell Specimen FISH

PCR No HPV 16 18 16+6 6+11 Total

No HPV

6 / 1 1 1 6 / 1 8 6/11/16/18 Total

8 2 0 0 0

1 0 1 0 0

0 3 0 0 0

0 2 1 1 1

9 7 2 1 1

10

2

3

5

20

HPV, human papillomavirus; FISH, filter in situ hybridization; and PCR, polymerase chain reaction.

This investigation supports our previous finding that uterine cervical HPV infection is d y n a m i c (McNicol et al., 1990). The variant pattern of infection makes consistent detection and typing of HPV difficult, and it becomes obvious that n u m e r o u s sequential specimens are required to d e t e r m i n e accurately HPV infection status. With two sequential analyses by PCR, a m e a n of 69.2% of w o m e n were HPV positive. A similar high prevalence of HPV infection has b e e n n o t e d in our geographic area (McNicol and D o d d , 1990) and no d o u b t reflects the sensitivity of PCR analysis and of the "high risk" d e m o g r a p h i c characteristics of the cohort. The m a x i m u m n u m b e r of w o m e n d e m o n strating positivity for a single PCR analysis was 25 (83%) of 30. If these 25 tests are c o n s i d e r e d to be true positives, then, with a single test, one w o u l d expect to observe a prevalence of 57.5% (83% x 0.692). The prevalence o b s e r v e d at the second analysis was 55%. With two sequential analyses b y FISH, a m e a n of 51.6% of w o m e n were HPV positive. The m a x i m u m n u m b e r of w o m e n d e m o n s t r a t i n g positivity by FISH analysis was 16 (53%) of 30. Once again, by considering these to be true positives, the prevalence of HPV observed with a single test should be 27.3%. The prevalence o b s e r v e d at the second analysis by FISH was 50%. The discrepancy b e t w e e n predicted and observed values m a y result from the subjective nature of interpretation of FISH results, as well as a lack of specificity observed with the HPV 6/11 probe. The PCR p r o c e d u r e appears to be the more accurate test w h e n analyzing a single specimen based u p o n the above

Latent HPV

predictions. Similar to our findings, Schiffman et al. (1991) f o u n d that the increase in sensitivity of PCR c o m p a r e d with conventional S o u t h e r n analysis, resulted in additional detection of HPV infection among w o m e n w i t h o u t a p p a r e n t cervical neoplasia. H o w ever, w e o b s e r v e d that definitive assessment of HPV infection status w o u l d require analysis of several sequential specimens. The d y n a m i c nature of HPV infection raises important questions concerning the risk of cervical neoplasia associated with sporadic detection of HPV. Are w o m e n with HPV infections consistently detectable by less sensitive diagnostic m e t h o d s such

683

as FISH at greater risk of cervical disease because the virus continues to elude host defenses? By prospectively tracking b o t h latent and clinically apparent HPV infections, we m a y begin to u n d e r s t a n d the dynamics of HPV infection that influence the natural history of cervical diseases.

This work was supported by the Medical Research Council of Canada. Cloned HPV genomes used in FISH analysis were provided by Dr. H. zur Hausen, Dr. L. Gissmann, and Dr. E.-M. de Villiers.

REFERENCES De Villiers E-M, Schneider A, Miklaw H, Papendick U, Wagner D, Wesch H, Wahrendorf J, Zur Hausen H (1987) Human papillomavirus infections in women with and without abnormal cervical cytology. Lancet 2:703706. Erlich HA, Bugawan TL (1990) HLA DNA typing. In PCR Protocols: A Guide to Methods and Applications. Eds, MA Innis, DH Gelfand, JJ Sninsk, and TJ White. New York, Academic Press, pp 261-271. McNicol PJ, Dodd JG (1990) Detection of human papillomavirus DNA in prostate gland tissue by using the polymerase chain reaction amplification assay. J Clin Microbiol 28:409-412. McNicol PJ, Guijon FB, Paraskevas M, Brunham RC (1989) Comparison of filter in situ deoxyribonucleic acid hybridization with cytologic, colposcopic and histopathologic examination for detection of human papillomavirus infection in women with cervical intraepithelial neoplasia. Am J Obstet Gynecol 160:265-270. McNicol PJ, Guijon FB, Paraskevas M, Heywood E, Gray MJ (1990) Effect of the menstrual cycle on detection and typing of human papillomavirus in uterine cervical cells. Am ] Obstet GynecoI 162:1037-1041. Rando RF, Lindheim S, Hasty L, Sedlacek TV, Woodland M, Eder C (1989) Increased frequency of detection of human papillomavirus deoxyribonucleic acid in exfol-

iated cervical cells during pregnancy. Am J Obstet Gynecol 161:50-55. Reeves WC, Arosemena JR, Garica M, Loo de Lao S, Cuevas M, Quiroz E, Caussy D, Rawls WE (1989a) Genital human papillomavirus infection in Panama City prostitutes. J Infect Dis 160:599-603. Reeves WC, Brinton LA, Garcia M, Brenes MM, Herrero R, Gaitan E, Tenoria F, de Britton RC, Rawls WE (1989b) Human papillomavirus infection and cervical cancer in Latin America. N Engl J Med 320:1437-1441. Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich HA, Arnheim N (1985) Enzymatic amplification of betaglobin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230:13501354. Schiffman MH, Bauer HM, Lorincz AT, Manos MM, Byrne JC, Glass AG, Cadell DM, Howley PM (1991) Comparison of Southern blot hybridization and polymerase chain reaction methods for the detection of human papillomavirus DNA. J Clin Microbiol 29:573-577. Schneider A, Kirchrnayr R, Gissrnan L (1988) Human papillornavirus preceding intraepithelial neoplasia in serial cervical smears. Lancet 1:989. Shibata DK, Arnheim N, Martin WJ (1988) Detection of human papilloma virus in paraffin-embedded tissue using the polymerase chain reaction. J Exp Med 167:225230.