Journal of Hepatok?gy,1992; 15: 387-390 0 1992 ElsevierScience Publishers B.V. All rights reserved. 0168~8278/92/$05.00
387
tIEPAT 01184
Rapid
-rip treat-ed
during
active
polymeras verify hepatitis C virus infection Christoph
JMax-Planck-lnstiturfiir
Publication
Biochemie.
Depurtment
P. Geiger” and Wolfgang of Virus Research, linicersity
of
Martinsried
H. Caselmanna*b and “Department
of Medicine
II.
Klinikutn
Grossltadern,
Munick. Municjz. Germany
All routinely available assays to detect hepatitis C virus (HCV) infection are based on the specific reaction of serum antibodies to corresponding viral antigens. The only method to detect viral nucleic acids in serum or liver is the enzymatic amplification of reversely transcribed HCV-RNA by polymerase chain reaction. Here, we describe a fast and simple method to confirm the HCV specificity of amplified fragments by hybridization with a non-radioactive cDNA probe created during polymerase chain reaction. A sequenced 521 bps HCV fragment derived from serum of an anti-HCV/EIAII-positive patient A with post-transfusion hepatitis C was used to verify HCV infection diagnosed by nested polymerase chain reaction in an anti-HCV/ETAII-seronegative patient B who did not produce anti-HCVspecific antibodies during the observation period of 1 year. The method is applicable to RNAs derived from serum or liver tissue and combines the sensitivity of nested polymerase chain reaction with the specificity of Southern blot hybridization. The procedure is independent of cloned viral sequences. In clinical use it permits fast and accurate confirmation of HCV infection, especially when antibodies detectable with conventional methods are absent. Key words: HCV diagnostics; Azocolors; Chemoluminescence; antibodies
Digoxigenin labeling: AMPPD; Anti-HCV/EIAII
Immunological tests which detect human serum amibodies directed against recombinant hepatitis C virus (HCV) antigens are generally used today to diagnose HCV infection. However, these radioimmuno-or enzyme-linked immunosorbent assays (ElA) often fail to yield positive results in the early phase of infection (I), when HCV-specific antibodies are lacking. Some patients will never produce anti-C-100 antibodies because they are infected by mutants or different subtypes of HCV (2). Although second-generation tests like the 4-antigenrecombinant immunoblot assay or EIAII, comprise one structural and 3 additional non-structural viral antigens that are detectable as early as 8 weeks after infection (3), these assays are also not appropriate for detecting all cases of HCV infection. Until now, reverse transcription of HCV-RNA and enzymatic amplification of HCV-cDNA by polymerase chain reaction (PCR) was the only method to directly
detect viral constituents (4). However, if stringent conditions are not used to ensure the detection of various subtypes with the same primers, false-positive PCR results arise (5). False negatives may also be seen, when only small quantities of low-titer serum are used (6). For these reasons, it is essential to verify both positive and negative PCR results with an independent method.
co: Wolfgang H. Caselmann, M.D., Department 15, D-8000 Munich 70, Germany.
Correspondence
Materials and Methods RNA preparation Biopsy tissue was homogenized in buffer I (140 mM NaCl, 1.5 mM MgCl,, 10 mM Tris-HCl (pH X.0) 5 mM dithiothreitol (DTT) and 25 U RNasin (Promega, Germany). After centrifugation, buffer II (200 mM Tris-HCl (pH 7.5), 25 mM EDTA, 300 mM NaCl, 2% SDS, 200 pg/ ml Proteinase K (Merck, Germany) was added to the
of Medicine II, Klinikum Grosshadern,
University of Munich, Marchioninistrasse
C.P. GEIGER and W.H. CASELMANN
388 supernatant. After 1 h of incubation at 37°C RNA was extracted with phenol/chloroform and precipitated. To prepare RNA for serum, 200~1 serum was diluted with an equal volume of buffer II and processed as described above. Oligonucleotide
synthesis
Primers derived from the structural region of a Japanese HCV strain (7) and adapted to other subtypes were synthesized according to the phosphoamidite procedure. Oligonucleotide primers hybridizing to exons 5 and 7 of the human albumin gene served as control. The trityl groups were removed by incubation at 56 “C for at least 4 h. Afterwards, the oligonucleotides were dried and dissolved in 10 mM Tris/l mM EDTA buffer. Nested reverse PCR
A reverse transcriptase mixture containing 10% (v/v) of PCR buffer, IO%,(v/v) of 0. I M DTT, 200 PM deoxynucleotide triphosphates, 80 pmol of the downstream primer, 5% (v/v) hexanucleotides, 20 U RNasin and I U of MMLV reverse transcriptasc (Gibco BRL, Germany) was added to the extracted RNA and incubated at 37 “C for I h. PCR was carried out using 20 ~1 of cDNA mixture, 10% (v/v) of PCR buffer, 200 PM deoxynucleotide triphosphates, 80 pmol of the upstream primer and 80 pmol of the downstream primer and 2 U of AmpliTaq polymerase ( Perkin-Elmer Cetus, Germany). A two-step PCR cycle lvas performed 30 times, with I min denaturation at 95 -C. and 2 min annealing and synthesizing at 50 C slowly rising to 72 ‘C. A second round of amplification using a IO-/d aliquot of the first-round PCR product and internal oligonucleotide primers was performed as described previously (8). Sour/tern blot hyhridizution After electrophoresis the gel was blotted onto a nylon membrane. The membrane was prehybridized for 1 h and hybridized for 3-6 h with 100 ng of a defined hybridization probe created by PCR as described below. El A cmdfilter
stainirtg
After immunoreaction with digoxigenin-specific antibodies (Boehringer, Germany) two different staining reactions were performed alternately. Chernohrminescence using AMPPD Following equilibration with 100 mM Tris-HCl (pH 9.5), 100 mM sodium chloride and 50 mM magnesium chloride, I% (v/v) of AMPPD and IO% (v/v) of EMERALD Enhancer (Dianova, Germany) were added to the filter and removed
after 5 min incubation at room temperature. Bands were detected by X-ray films (Amersham, Germany). Azodye
staining
For multistaining procedure: with different a.zodyes (9) the filter was equilibrated with 50 mM Tris-HCI (Serva, Germany) (pH 8.2); 0.02% (w/v) of Naphthol AS phosphate and 0.1% (w/v) of Fast Blue BB salt (Serva, Germany) were mixed in IO ml 50 mM Tris-HCI (pH 8.2), cleared by filtration and poured on the membrane.
Results Patients’ data
RNA was extracted from the serum and liver biopsy tissue of 2 patients with suspected non-A, non-B hepatitis who differed with respect to their anti-HCV/EIAII positivity. Patient A was a 48-year-old man with acute histologically confirmed hepatitis C. During an observation period of 1 year, his serum had been repeatedly nonreactive in the anti-HCV/EIAII assay. Patient B was an immunosuppressed 47-year-old man, who had undergone orthotopic liver transplantation because of end-stage anti-HCV/EIAII-negative alcoholic cirrhosis. After transplantation the patient had developed chronic, anti-HCV/EIAII-positive post-transfusional hepatitis C. Amplification
of HCV sequences
Reverse nested PCR was performed in both patients using oligonucleotide primers derived from conserved areas of the HCV core (C/M) and envelope (El) regions (Fig.la). The outer upstream primer was CG I .I (5’AAA GAC CCC CGG CGT AGG TC-3’), the corresponding downstream primer was CG 1.2 (S-GCG TCA ACG CCG GCA AAT AG-3’). The inner primers CG 2.1 (5’-GTT GCT CTT TCT CTA TCT TCC-3’) and CG 2.2 (5’-GAC GGC TTG TGG GAT CCG GAG-3’) covered HCV nucleotides 844-864 and 1344- 1364, respectively. After the fragments amplified from the serum or liver of patient A were stained with ethidium bromide, both a weak signal of the expected size of 820 bps and a single band of 521 bps were visible (Fig. 2a). In order to verify the specificity of these signals, Southern blotting was performed. Production
qf HCV hybridization probe A 52 I-bp HCV-specific hybridization probe CG-HYB derived from HCV El region (Fig. la) was synthesized by nested PCR of RNA isolated from patient B’s antiHCV/EIAII-positive serum. The identical sets of primers
VERIFICATION
OF HCV-PCR RESULTS
3x9
(a>
C/M 1
1
333
El
906
CG 1.1
660
CG 2.1
.
844.
1
E2
1500 820
.
1479
CG 1.2
521 + 1364 CG 2.2 CG-HYB
1 CCTTC~TA~TTCCTCT~~~TCC~~AC~TCC~CT 50 8 I, #,#,,,,,,,:::I ::I::II::::III :I:::: !IllllI 850 CTTTTTCTAT~TT~~TCTTGG~T~TGCTGT~CTG~CTGA~~A~~C~AG~T 099
900
101 950 151 1000 201 1050 303 251 ~~~AC~~cC~C~~c~ I(; IIIIIIIIIIIIIIII ,I,,,,,,,I.IIIII ::I:I:::II: :::::III::;:; ( 1199 1100 GACGATACGACGCCACGTCGATCTGCTCGTTGGGGCGTTGGGGCGGCTGCTTTCTGTT 301 1150
Non-radioactit:e staining Anti-digoxigenin antibodies linked to alkaline phosphatase were used to visualize the hybridizing bands by chemoluminescence with AMPPD (Fig. 2b) or lXQhtho$ AS phosphate and Fast color salts (Fig. 2~). One to .5n%n after exposure, both methods provided specific staining of HCV bands, while albumin sequences, which were amplified from reversely transcribed liver RNA extracted from biopsy tissue of patient A as a control (Fig. 2a), showed no hybridizing band (Fig. Zb,c). In additional experiments performed with the serum and liver samples of other hepatitis C patients, the AMPPD chemoluminescence method appeared slightly more sensitive than azodye staining. Non-radioactive hybridization of amplified cDNA fragments to a well-characterized HCV-specific probe, which has been produced by PCR and can be stored at 4’C and re-used for more than 1 year, is an appropriate method to confirm the HCV specificity of amplified bands. The use of this procedure to answer clinical questions shows that chronic HCV infection can exist in the absence of anti-HCVjEIAlI serum antibodies. To recognize this pattern, PCR and other confirmatory procedures such as the method described are required.
351 1200 401
Discussion
1250 451 1300 SUl 1350
Fig. 1. (a) Genomic organization of the structural region of HCV. NC= S-non-coding; C/M =core; El/2 envelope regions I and 2; CG I.I/CG 1.2 and CG 2.1/CG 2.2 = two sets of oligonuclcotide primers used for reverse nestted PCR of HCV-RNA. CG-HYB= 521 bps HCVcDNA fragment used for hybridization. (b) Sequence of CG-HYB hybridization probe (upper sequence in bold letters) in comparison to a Japanese isolate of HCV (lower sequence; Ref. 7). The oligonucleotide primer sequences are underlined.
described above were used. Repeated direct sequencing of the amplified cDNA proved the HCV specificity of the fragment which displayed the best nuclentide homology (90.7%) to the Japanese subtype described by Takamizawa (7) (Fig. 1b). For non-radioactive labeling of this hybridization probe, digoxigenin-linked deoxynucleotide triphosphates (molar ratio dTTP/DIC;-dUTP= 51) were incorporated into. the newly synthesized cDNA strands during the second round of amplification with Taq polymerase. The resulting PCR fragment was purified from a 1.2% agarose gel by QIAEX purification kit (DIAGEN) and used for hybridization.
A non-radioactive HCV hybridization probe was crezted by a reverse PCR of RNA extracted from the serum of a patient suffering from anti-HCWEIAII-positive post-transfusional hepatitis C. After incorporation of digoxigenin-labeled deoxynucleotide triphosphates during the second round of nested PCR, this probe could be used for EIA-linked AMPPD or azodye staining of HCV-cDNA. This combination of specific hybridization with enzymatic amplification was used to diagnose hepatitis C in an anti-HCV/EIAIl-negative patient. It may be applicable to similar cases, where, until now, HCV etiology could not be proven. Further characterization of HCV sequences derived from patients who fail to produce anti-HCV antibodies may help to identify mutants or unknown HCV subtypes and to investigate their possible role during the natural course of l-10 infection. The method described is applicable to both serum and cellular RNA and may help exclude false-positive as well as false-negative PCR results. If unspecific hybridization has occurred, no hybridizing band is detectable with the HCV probe. Hybridization after amplification improves the sensitivity by at least one order of magni-
C.P. GEIGER
390
w
(a > Kbp
M
m
1
2 3
C>
(
1’ 2’ 3’
and W.H. CASELMANN
1" 2” 3” Kbp
Fig. 2. (a) Agarose gel with amplified HCV (lanes I and 2) and human albumin (lane 3) cDNAs derived from serum and liver tissue of patient A, respectively. After the first round of PCR using the outer HCV primers a weak band cf 820 bps is seen (lane I ). The second amplification step with inner primers CG 2.1 and CG 2.2 resulted in a 521 bps fragment (lane 2). Amplification of reversely transcribed liver RNA with albuminspecific oligonucleotide primers yielded an expected band of 328 bps. (b) Corresponding HCV-specific signals (lanes I’ and 2’) after AMPPD and (c) azodye staining (lanes I” and 2”). Size markers: M = bacteriophage (lambda) x Hind111 x EcoRt, m -= pUCl9 x Mspl.
tude (10) and can therefore also be used to confirm negative results in PCR-negative patients. The procedure is also still applicable when serum HCV titers are low, as is frequently seen in chronic hepatitis C. Since cloning of the HCV probe used for hybridization is not necessary, the method can be performed under less stringent safety conditions. As shown in Table I, once the hybridization probe is prepared the entire assay can be completed within 2 days. Non-radioactive hybridization, which renders the procedure less expensive and independent of isotope laboratories, has been shown to be as sensitive as radioactive procedures and has been successfully used for the detection of hepatitis R virus DNA (10). The labelled hybridization probe is stable for at least I year and can be stored at -20 ‘-C. As long as no antigen or early IgM antibody assay is available to diagnose HCV infection at the time hepatitis C develops, PCR remains the only method to establish the diagnosis in early stages of disease, Hybridization of amplified sequences to HCV-cDNA probes produced during PCR is a safe, simple, highly sensitive method to verify both positive and negative PCR results. TABLE
I
Acknowledgements The authors wish to thank Ms. Renate Renz for excellent technical assistance. This work was supported by a research grant (Ca 113/S-l) allocated to W.H.C. by the Deutsche Forschungsgcmeinschaft.
References
6
7
Time schedule of HCV detection assay Procedure
Time
RNA preparation Reverse transcription PCR Agarose gel electrophoresis Southern blot Hybridization EIA
2 h 1.5 h 3.5 h 0.5 h overnight 5 h 2 h
Day
8
9
IO
Kuo G. Choo QL. Alter HJ et al. An assay for circulating antibodies to a major etiologic virus of human non-A, non-B hepatitis. Science 1989; 244: 362- 4. Farci P. Alter HJ. Wong D et al. A long-term study of hepatitis C virus replication in non-A, non-B hepatitis. N Engl J Med 1991; 325: 98-104. van der Poel CL. Cuypers HTM. Reesink HW et al. Confirmation of hepdutis C virus infection by a new four antigen recombinant immunoblot assay. Lancet 1991; 337: 317-9. Shimizu JK, Weiner AJ. Rosenblatt J et al. Early events in hepatitis C virus infection of chimpanzees. Proc Nat1 Acad Sci USA 1990; 87: 6441-4. Kwok S. Higuchi R. How to avoid false positives with PCR. Nature 1989; 339: 237-8. Bukh J. Purcell RH, Miller RH. The importance of primer selection for the detection of hepatitis C virus (HCV) RNA by PCR. Hepatology 1991; 14: l6A (Abstract). Takamizawa A. Mori C, Fuke I et al. Structure and organization of hepatitis C virus genome from Japanese patients with non-A, non-l hepatitis. Proc Nat1 Acad Sci USA 1990: 87: 9524-8. Garson JA, Tedder RS. Briggs M et al. Detection of hepatitis C viral sequences in blood donations by ‘nested’ polymerase chain reaction. Lancet 1990; 335: 1419-22. West S, Schrijder J, Kunz W. A multiple staining procedure for detection of different DNA fragments on a single blot. Anal Biochem 1990; 190: 254-8. Naoumov NV, Lau JYN, Daniels HM, Alexander GJM, Williams R. Detection of HBV-DNA using a digoxigenin-labeled probe. J Hepatol 1991: 12: 382-5.
387
tIEPAT 01184
Rapid
-rip treat-ed
during
active
polymeras verify hepatitis C virus infection Christoph
JMax-Planck-lnstiturfiir
Publication
Biochemie.
Depurtment
P. Geiger” and Wolfgang of Virus Research, linicersity
of
Martinsried
H. Caselmanna*b and “Department
of Medicine
II.
Klinikutn
Grossltadern,
Munick. Municjz. Germany
All routinely available assays to detect hepatitis C virus (HCV) infection are based on the specific reaction of serum antibodies to corresponding viral antigens. The only method to detect viral nucleic acids in serum or liver is the enzymatic amplification of reversely transcribed HCV-RNA by polymerase chain reaction. Here, we describe a fast and simple method to confirm the HCV specificity of amplified fragments by hybridization with a non-radioactive cDNA probe created during polymerase chain reaction. A sequenced 521 bps HCV fragment derived from serum of an anti-HCV/EIAII-positive patient A with post-transfusion hepatitis C was used to verify HCV infection diagnosed by nested polymerase chain reaction in an anti-HCV/ETAII-seronegative patient B who did not produce anti-HCVspecific antibodies during the observation period of 1 year. The method is applicable to RNAs derived from serum or liver tissue and combines the sensitivity of nested polymerase chain reaction with the specificity of Southern blot hybridization. The procedure is independent of cloned viral sequences. In clinical use it permits fast and accurate confirmation of HCV infection, especially when antibodies detectable with conventional methods are absent. Key words: HCV diagnostics; Azocolors; Chemoluminescence; antibodies
Digoxigenin labeling: AMPPD; Anti-HCV/EIAII
Immunological tests which detect human serum amibodies directed against recombinant hepatitis C virus (HCV) antigens are generally used today to diagnose HCV infection. However, these radioimmuno-or enzyme-linked immunosorbent assays (ElA) often fail to yield positive results in the early phase of infection (I), when HCV-specific antibodies are lacking. Some patients will never produce anti-C-100 antibodies because they are infected by mutants or different subtypes of HCV (2). Although second-generation tests like the 4-antigenrecombinant immunoblot assay or EIAII, comprise one structural and 3 additional non-structural viral antigens that are detectable as early as 8 weeks after infection (3), these assays are also not appropriate for detecting all cases of HCV infection. Until now, reverse transcription of HCV-RNA and enzymatic amplification of HCV-cDNA by polymerase chain reaction (PCR) was the only method to directly
detect viral constituents (4). However, if stringent conditions are not used to ensure the detection of various subtypes with the same primers, false-positive PCR results arise (5). False negatives may also be seen, when only small quantities of low-titer serum are used (6). For these reasons, it is essential to verify both positive and negative PCR results with an independent method.
co: Wolfgang H. Caselmann, M.D., Department 15, D-8000 Munich 70, Germany.
Correspondence
Materials and Methods RNA preparation Biopsy tissue was homogenized in buffer I (140 mM NaCl, 1.5 mM MgCl,, 10 mM Tris-HCl (pH X.0) 5 mM dithiothreitol (DTT) and 25 U RNasin (Promega, Germany). After centrifugation, buffer II (200 mM Tris-HCl (pH 7.5), 25 mM EDTA, 300 mM NaCl, 2% SDS, 200 pg/ ml Proteinase K (Merck, Germany) was added to the
of Medicine II, Klinikum Grosshadern,
University of Munich, Marchioninistrasse
C.P. GEIGER and W.H. CASELMANN
388 supernatant. After 1 h of incubation at 37°C RNA was extracted with phenol/chloroform and precipitated. To prepare RNA for serum, 200~1 serum was diluted with an equal volume of buffer II and processed as described above. Oligonucleotide
synthesis
Primers derived from the structural region of a Japanese HCV strain (7) and adapted to other subtypes were synthesized according to the phosphoamidite procedure. Oligonucleotide primers hybridizing to exons 5 and 7 of the human albumin gene served as control. The trityl groups were removed by incubation at 56 “C for at least 4 h. Afterwards, the oligonucleotides were dried and dissolved in 10 mM Tris/l mM EDTA buffer. Nested reverse PCR
A reverse transcriptase mixture containing 10% (v/v) of PCR buffer, IO%,(v/v) of 0. I M DTT, 200 PM deoxynucleotide triphosphates, 80 pmol of the downstream primer, 5% (v/v) hexanucleotides, 20 U RNasin and I U of MMLV reverse transcriptasc (Gibco BRL, Germany) was added to the extracted RNA and incubated at 37 “C for I h. PCR was carried out using 20 ~1 of cDNA mixture, 10% (v/v) of PCR buffer, 200 PM deoxynucleotide triphosphates, 80 pmol of the upstream primer and 80 pmol of the downstream primer and 2 U of AmpliTaq polymerase ( Perkin-Elmer Cetus, Germany). A two-step PCR cycle lvas performed 30 times, with I min denaturation at 95 -C. and 2 min annealing and synthesizing at 50 C slowly rising to 72 ‘C. A second round of amplification using a IO-/d aliquot of the first-round PCR product and internal oligonucleotide primers was performed as described previously (8). Sour/tern blot hyhridizution After electrophoresis the gel was blotted onto a nylon membrane. The membrane was prehybridized for 1 h and hybridized for 3-6 h with 100 ng of a defined hybridization probe created by PCR as described below. El A cmdfilter
stainirtg
After immunoreaction with digoxigenin-specific antibodies (Boehringer, Germany) two different staining reactions were performed alternately. Chernohrminescence using AMPPD Following equilibration with 100 mM Tris-HCl (pH 9.5), 100 mM sodium chloride and 50 mM magnesium chloride, I% (v/v) of AMPPD and IO% (v/v) of EMERALD Enhancer (Dianova, Germany) were added to the filter and removed
after 5 min incubation at room temperature. Bands were detected by X-ray films (Amersham, Germany). Azodye
staining
For multistaining procedure: with different a.zodyes (9) the filter was equilibrated with 50 mM Tris-HCI (Serva, Germany) (pH 8.2); 0.02% (w/v) of Naphthol AS phosphate and 0.1% (w/v) of Fast Blue BB salt (Serva, Germany) were mixed in IO ml 50 mM Tris-HCI (pH 8.2), cleared by filtration and poured on the membrane.
Results Patients’ data
RNA was extracted from the serum and liver biopsy tissue of 2 patients with suspected non-A, non-B hepatitis who differed with respect to their anti-HCV/EIAII positivity. Patient A was a 48-year-old man with acute histologically confirmed hepatitis C. During an observation period of 1 year, his serum had been repeatedly nonreactive in the anti-HCV/EIAII assay. Patient B was an immunosuppressed 47-year-old man, who had undergone orthotopic liver transplantation because of end-stage anti-HCV/EIAII-negative alcoholic cirrhosis. After transplantation the patient had developed chronic, anti-HCV/EIAII-positive post-transfusional hepatitis C. Amplification
of HCV sequences
Reverse nested PCR was performed in both patients using oligonucleotide primers derived from conserved areas of the HCV core (C/M) and envelope (El) regions (Fig.la). The outer upstream primer was CG I .I (5’AAA GAC CCC CGG CGT AGG TC-3’), the corresponding downstream primer was CG 1.2 (S-GCG TCA ACG CCG GCA AAT AG-3’). The inner primers CG 2.1 (5’-GTT GCT CTT TCT CTA TCT TCC-3’) and CG 2.2 (5’-GAC GGC TTG TGG GAT CCG GAG-3’) covered HCV nucleotides 844-864 and 1344- 1364, respectively. After the fragments amplified from the serum or liver of patient A were stained with ethidium bromide, both a weak signal of the expected size of 820 bps and a single band of 521 bps were visible (Fig. 2a). In order to verify the specificity of these signals, Southern blotting was performed. Production
qf HCV hybridization probe A 52 I-bp HCV-specific hybridization probe CG-HYB derived from HCV El region (Fig. la) was synthesized by nested PCR of RNA isolated from patient B’s antiHCV/EIAII-positive serum. The identical sets of primers
VERIFICATION
OF HCV-PCR RESULTS
3x9
(a>
C/M 1
1
333
El
906
CG 1.1
660
CG 2.1
.
844.
1
E2
1500 820
.
1479
CG 1.2
521 + 1364 CG 2.2 CG-HYB
1 CCTTC~TA~TTCCTCT~~~TCC~~AC~TCC~CT 50 8 I, #,#,,,,,,,:::I ::I::II::::III :I:::: !IllllI 850 CTTTTTCTAT~TT~~TCTTGG~T~TGCTGT~CTG~CTGA~~A~~C~AG~T 099
900
101 950 151 1000 201 1050 303 251 ~~~AC~~cC~C~~c~ I(; IIIIIIIIIIIIIIII ,I,,,,,,,I.IIIII ::I:I:::II: :::::III::;:; ( 1199 1100 GACGATACGACGCCACGTCGATCTGCTCGTTGGGGCGTTGGGGCGGCTGCTTTCTGTT 301 1150
Non-radioactit:e staining Anti-digoxigenin antibodies linked to alkaline phosphatase were used to visualize the hybridizing bands by chemoluminescence with AMPPD (Fig. 2b) or lXQhtho$ AS phosphate and Fast color salts (Fig. 2~). One to .5n%n after exposure, both methods provided specific staining of HCV bands, while albumin sequences, which were amplified from reversely transcribed liver RNA extracted from biopsy tissue of patient A as a control (Fig. 2a), showed no hybridizing band (Fig. Zb,c). In additional experiments performed with the serum and liver samples of other hepatitis C patients, the AMPPD chemoluminescence method appeared slightly more sensitive than azodye staining. Non-radioactive hybridization of amplified cDNA fragments to a well-characterized HCV-specific probe, which has been produced by PCR and can be stored at 4’C and re-used for more than 1 year, is an appropriate method to confirm the HCV specificity of amplified bands. The use of this procedure to answer clinical questions shows that chronic HCV infection can exist in the absence of anti-HCVjEIAlI serum antibodies. To recognize this pattern, PCR and other confirmatory procedures such as the method described are required.
351 1200 401
Discussion
1250 451 1300 SUl 1350
Fig. 1. (a) Genomic organization of the structural region of HCV. NC= S-non-coding; C/M =core; El/2 envelope regions I and 2; CG I.I/CG 1.2 and CG 2.1/CG 2.2 = two sets of oligonuclcotide primers used for reverse nestted PCR of HCV-RNA. CG-HYB= 521 bps HCVcDNA fragment used for hybridization. (b) Sequence of CG-HYB hybridization probe (upper sequence in bold letters) in comparison to a Japanese isolate of HCV (lower sequence; Ref. 7). The oligonucleotide primer sequences are underlined.
described above were used. Repeated direct sequencing of the amplified cDNA proved the HCV specificity of the fragment which displayed the best nuclentide homology (90.7%) to the Japanese subtype described by Takamizawa (7) (Fig. 1b). For non-radioactive labeling of this hybridization probe, digoxigenin-linked deoxynucleotide triphosphates (molar ratio dTTP/DIC;-dUTP= 51) were incorporated into. the newly synthesized cDNA strands during the second round of amplification with Taq polymerase. The resulting PCR fragment was purified from a 1.2% agarose gel by QIAEX purification kit (DIAGEN) and used for hybridization.
A non-radioactive HCV hybridization probe was crezted by a reverse PCR of RNA extracted from the serum of a patient suffering from anti-HCWEIAII-positive post-transfusional hepatitis C. After incorporation of digoxigenin-labeled deoxynucleotide triphosphates during the second round of nested PCR, this probe could be used for EIA-linked AMPPD or azodye staining of HCV-cDNA. This combination of specific hybridization with enzymatic amplification was used to diagnose hepatitis C in an anti-HCV/EIAIl-negative patient. It may be applicable to similar cases, where, until now, HCV etiology could not be proven. Further characterization of HCV sequences derived from patients who fail to produce anti-HCV antibodies may help to identify mutants or unknown HCV subtypes and to investigate their possible role during the natural course of l-10 infection. The method described is applicable to both serum and cellular RNA and may help exclude false-positive as well as false-negative PCR results. If unspecific hybridization has occurred, no hybridizing band is detectable with the HCV probe. Hybridization after amplification improves the sensitivity by at least one order of magni-
C.P. GEIGER
390
w
(a > Kbp
M
m
1
2 3
C>
(
1’ 2’ 3’
and W.H. CASELMANN
1" 2” 3” Kbp
Fig. 2. (a) Agarose gel with amplified HCV (lanes I and 2) and human albumin (lane 3) cDNAs derived from serum and liver tissue of patient A, respectively. After the first round of PCR using the outer HCV primers a weak band cf 820 bps is seen (lane I ). The second amplification step with inner primers CG 2.1 and CG 2.2 resulted in a 521 bps fragment (lane 2). Amplification of reversely transcribed liver RNA with albuminspecific oligonucleotide primers yielded an expected band of 328 bps. (b) Corresponding HCV-specific signals (lanes I’ and 2’) after AMPPD and (c) azodye staining (lanes I” and 2”). Size markers: M = bacteriophage (lambda) x Hind111 x EcoRt, m -= pUCl9 x Mspl.
tude (10) and can therefore also be used to confirm negative results in PCR-negative patients. The procedure is also still applicable when serum HCV titers are low, as is frequently seen in chronic hepatitis C. Since cloning of the HCV probe used for hybridization is not necessary, the method can be performed under less stringent safety conditions. As shown in Table I, once the hybridization probe is prepared the entire assay can be completed within 2 days. Non-radioactive hybridization, which renders the procedure less expensive and independent of isotope laboratories, has been shown to be as sensitive as radioactive procedures and has been successfully used for the detection of hepatitis R virus DNA (10). The labelled hybridization probe is stable for at least I year and can be stored at -20 ‘-C. As long as no antigen or early IgM antibody assay is available to diagnose HCV infection at the time hepatitis C develops, PCR remains the only method to establish the diagnosis in early stages of disease, Hybridization of amplified sequences to HCV-cDNA probes produced during PCR is a safe, simple, highly sensitive method to verify both positive and negative PCR results. TABLE
I
Acknowledgements The authors wish to thank Ms. Renate Renz for excellent technical assistance. This work was supported by a research grant (Ca 113/S-l) allocated to W.H.C. by the Deutsche Forschungsgcmeinschaft.
References
6
7
Time schedule of HCV detection assay Procedure
Time
RNA preparation Reverse transcription PCR Agarose gel electrophoresis Southern blot Hybridization EIA
2 h 1.5 h 3.5 h 0.5 h overnight 5 h 2 h
Day
8
9
IO
Kuo G. Choo QL. Alter HJ et al. An assay for circulating antibodies to a major etiologic virus of human non-A, non-B hepatitis. Science 1989; 244: 362- 4. Farci P. Alter HJ. Wong D et al. A long-term study of hepatitis C virus replication in non-A, non-B hepatitis. N Engl J Med 1991; 325: 98-104. van der Poel CL. Cuypers HTM. Reesink HW et al. Confirmation of hepdutis C virus infection by a new four antigen recombinant immunoblot assay. Lancet 1991; 337: 317-9. Shimizu JK, Weiner AJ. Rosenblatt J et al. Early events in hepatitis C virus infection of chimpanzees. Proc Nat1 Acad Sci USA 1990; 87: 6441-4. Kwok S. Higuchi R. How to avoid false positives with PCR. Nature 1989; 339: 237-8. Bukh J. Purcell RH, Miller RH. The importance of primer selection for the detection of hepatitis C virus (HCV) RNA by PCR. Hepatology 1991; 14: l6A (Abstract). Takamizawa A. Mori C, Fuke I et al. Structure and organization of hepatitis C virus genome from Japanese patients with non-A, non-l hepatitis. Proc Nat1 Acad Sci USA 1990: 87: 9524-8. Garson JA, Tedder RS. Briggs M et al. Detection of hepatitis C viral sequences in blood donations by ‘nested’ polymerase chain reaction. Lancet 1990; 335: 1419-22. West S, Schrijder J, Kunz W. A multiple staining procedure for detection of different DNA fragments on a single blot. Anal Biochem 1990; 190: 254-8. Naoumov NV, Lau JYN, Daniels HM, Alexander GJM, Williams R. Detection of HBV-DNA using a digoxigenin-labeled probe. J Hepatol 1991: 12: 382-5.
