- Email: [email protected]
Indirect immunofluorescence for serological diagnosis of dengue virus infections in Swedish patients
ELSEVIER
Clinical and Diagnostic Virology 4 (1995) 43-50
Clinical and Diagnostic Virology
Indirect immunofluorescence for serological diagnosis of dengue virus infections in Swedish patients Sirkka Vene a.u.,, Joseph Mangiafico c, Bo Niklasson b.d "Department of Virology, Swedish Institute for Infectious Disease Control, S-105 21 Stockholm, Sweden; b Microbiology and Tumorbiology Centre, Karolinska Institute, Stockholm, Sweden," c Applied Research Division, USAMRIID, Frederick, MD, USA; National Defence Research Establishment, FOA-4, Umedt, Sweden
Received 6 September 1994; accepted 15 November 1994
Abstract Background: An increasing number of travellers to dengue endemic areas has emphasized the need for an easy and reliable serological test for diagnosis of dengue virus infections. Objectives: Indirect immunofluorescence (IF) and hemagglutination inhibition (HI) tests were compared for serological diagnosis of dengue fever (DF). Study design: Sera from patients with clinical symptoms compatible with DF and a travel history comprising dengue endemic areas were included in the study. Paired serum samples from 24 patients and single convalescent sera from five patients were investigated by HI and IF. Paired sera from ten patients were investigated by plaque reduction neutralization test (PRNT) in order to confirm the HI/IF results. Results: In twelve of the patients with paired sera a sero-conversion or a significant (~>4-fold) antibody titer rise was seen with HI as well as with IF. Four patients with stationary HI-titers had significant titer rises when investigated by IF. Stationary or declining HI- and IF-titers were found in the remaining eight patients. The five single sera were antibody-positive by both methods. Investigation by PRNT of paired sera from ten patients confirmed the IF results. A total of 298 serum samples (paired sera or early convalescent samples) were investigated for dengue antibodies by IF. A sero-conversion, significant titer-rise or an IF-titer of />320 was generally seen 7-13 days post onset of disease. During the years 1991-1993 altogether 161 cases of dengue were diagnosed by IF in Swedish patients, a majority of which had travelled to Southeast Asia. Conclusion: Antibody detection by indirect IF proved to be at least as reliable as HI for diagnosis of dengue infections in Swedish patients. The specificity of the IF was ensured by testing the sera of ten patients for serotype specific neutralizing dengue antibodies. Key words." Dengue antibody; IF; HI; PRNT
*Corresponding author at address a. Fax: (+46) 8-272231. 0928-0197/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0928-0197(94)00060-3
44
& Veneet aL/Clinical and Diagn+~stic Virology 4 (1995) 43 50
1. Introduction
Dengue fever (DF) has been recognized for more than 200 years (Rush, 1789) and occurs on all continents. Symptoms range from inapparent to severe hemorrhagic fever, although most patients experience classical DF with a sudden onset of fever, headache, myalgias and rash (Sabin, 1959). The dengue viruses (serotypes 1-4), which belong to the family Flaviviridae and the genus Flavivirus (Westaway et al., 1985) are endemic in tropical areas and are transmitted to humans by mosquitoes of the Aedes species, principally A. aegypti (Gubler, 1988). Serological diagnosis of DF is complicated by the cross-reactive antigenic determinants shared by the four dengue virus serotypes and other flaviviruses. In endemic countries, where distinction between primary and secondary infection is essential, sero-diagnosis of DF is generally based on detection of dengue-specific IgM by/x-capture ELISA (Bundo and Igarashi, 1985; Kuno et al., 1991 ), while most laboratories with only an occasional demand for dengue-serology use a hemagglutination inhibition (HI) test for antibody detection. An increasing number of travellers to dengue endemic areas risk exposure to the virus and subsequent illness (Cunningham and Mutton, 1991; Jacobs et al., 1991; Wittesj6 et al., 1993), a development which emphasized the need for a more rapid and convenient test than the HI for diagnosis of dengue fever in febrile travellers returning home. Although indirect immunofluorescence (IF) has been reported for serological diagnosis of DF (Boonpucknavig et al., 1975), the diagnostic use of IF has been limited in dengue serology. We used an approach suggested by Monath and co-workers for diagnosis of yellow fever (Monath et al., 1981 ) to evaluate the use of IF in the diagnosis of DF in Swedish patients.
2. Materials and methods
2.1. Serum specimens Paired serum samples were available from 24 patients with clinical symptoms compatible with dengue fever and a travel history comprising dengue endemic areas. Single convalescent sera were available from five patients chosen by the same criteria. 2.2. Hemagglutination inhibition test (HI) All serum samples were tested by a standard HI for arboviral antibodies employing goose erythrocytes and acetone-extracted sera as described by Shope (1974). Dengue virus mouse brain antigen prepared from dengue type 2 strain New Guinea C was kindly provided by N. Karabatsos, CDC, Fort Collins, Colo., USA. 2.3. hnmunofluorescence (IF) Dengue type 2 strain New Guinea C, kindly provided by R. Tesh, Yale Arbovirus Unit, New Haven, Conn, USA, was used to infect a confluent Vero cell culture in a 25 c m 2 tissue culture flask with Eagle's minimal essential medium supplemented
S. Veneet al./Clinical and Diagnostic Virology 4 (1995) 43-50
45
Fig. 1. Typicalcytoplasmicfluorescenceas seen in denguevirus infectedVero cells stained with positive human serum and fluorescein(DTAF)-labelledconjugate. with 200 U penicillin/ml, 200 mg streptomycin/ml and 5% foetal calf serum (MEMFCS). Cells were incubated at 37°C and IF-positivity of cells was checked daily with anti-dengue mouse serum, also provided by R. Tesh, and FITC-labelled rabbit (Fab')2 fragments to mouse Ig (Dakopatts A/S, Denmark). After five days of incubation, infected cells were mixed with uninfected Vero cells at a ratio of 1:3 in 30 ml of MEM-FCS. "Spot slides" were prepared by allowing the cells to attach to cleaned 12-well slides during a 16 h incubation at 37°C and 5% CO2. Cells were fixed to slides by exposure to cold anhydrous acetone and slides were stored at - 7 0 ° C until used. Sera were tested at 2-fold dilutions starting at 1:10 and fluorescein (DTAF)-conjugated goat anti-human IgG (heavy and light chain specific; Jackson Immunoresearch Laboratories, West Grove, Pa., USA) was used to detect bound antibodies. End point titers were set at the reciprocal of the last dilution giving an unequivocal fluorescence. The typical cytoplasmic fluorescence is shown in Fig. 1. 2.4. Plaque reduction neutralization test ( P R N T )
In order to exclude cross-reactions caused by antibodies to other flaviviruses, paired sera from ten patients were investigated for dengue serotype specific neutralizing antibodies, essentially as described by Burke et al. (1977). The PRNT was modified for use with 12-well plates to obtain neutralizing titers to the four dengue serotypes, strains WP (type 1), 25 (type 2), CH53489 (type 3), and H241 (type 4).
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
46
Table 1 Results by different serological methods in 29 patients with dengue infections Patient No.
HI Den
IF Den
PRNT Den
Sero-conversion or four-fold titer rise H I and I F 1.I < 10 80 1.II 40 > 10240 2.I <10 <10 2.II 80 5120 3.I <10 <10 3.11 20 320 4.1 <10 <10 4.I1 20 1280 5.1 < 10 10 5.11 80 5120 6.1 < 10 10 6.11 40 640 7.1 10 20 7.11 80 5120 8.1 < 10 < 10 8.11 40 2560 9.1 < 10 < 10 9.11 40 > 10240 10.I < 10 < 10 10.II 80 2560 ll.I <10 <10 11.II 40 > 10240 12.I < 10 <10 12.I1 40 2560
< 10 160 nt nt nt nt nt nt nt nt <10 2560 < 10 80
Stationary H I but titer r&e IF 13.I 40 13.II 40 14.I 20 14.1I 20 15.1 40 15.II 40 16.I 40 16.II 40
40 640 1280 5120 80 1280 1280 10240
nt nt < 10 640 < 10 160 nt nt
5120 5120 2560 2560 1280 1280 2560 2560 640 640 1280 640
nt nt nt nt nt nt nt nt nt nt nt nt
No titer rise H I or IF 17.I 17.11 18.I 18.II 19.I 19.1I 20.I 20.II 21.I 21.II 22.1 22.II
40 40 80 80 80 80 40 40 40 40 40 20
< 10 320 <10 640
Den serotype
Dengue 3/4 Dengue 1 Dengue 2 Dengue 1
Dengue 2
Dengue 1 Dengue 1
Dengue 1 Dengue 1
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
47
Table 1 continued) Patient No. 23.I 23.11 24.I 24.11
HI Den
IF Den
40 20 80 nt
2560 640 640 80
20 40 20 40 40
> 10240 5120 1280 1280 640
PRNT Den nt nt 640 1280
Den serotype
Dengue3
Sing& sera
25 26 27 28 29
nt nt nt nt nt
ant = not tested. Briefly, dengue virus, serotypes 1 through 4, at approximately 100 plaque forming units (PFU)/0.1 ml was mixed with an equal volume of serial 4-fold dilutions of test sera. Serum-virus mixtures were held at 4°C overnight and inoculated onto confluent monolayers of L L C - M K 2 cells in duplicate wells at 0.1 ml/well. Plates were incubated for 1 h at 37°C and 5% CO2, and each well was covered with 1.5 ml of nutrient overlay (Eagle's basal medium and Earle's salts with 1.1% agar, 5% FCS, 200 U penicillin/ml and 200 mg streptomycin/ml). After incubation at 37°C and 5% CO2 for 4 days, 1.5 ml of a second overlay, identical to the first, but containing neutral red at 1:7500 dilution was added to each well, and plates were incubated for 24-48 h at 37°C and 5% CO2, before plaques were enumerated. Neutralizing antibody titers were expressed as the reciprocal of the highest dilution reducing the number of plaques by 80% as compared to the virus control.
3. R ~ In twelve of the patients with paired sera, a sero-conversion or titer rise was seen with HI as well as with IF. Four patients with stationary HI-titers had significant (~>4-fold) titer rises when investigated by IF. Stationary or declining HI- and IF-titers were found in eight patients. Five single sera were antibody-positive by both methods. Paired sera from ten patients were examined by P R N T for dengue type specific antibodies. Sero-conversions by P R N T were seen in seven patients, who also had sero-conversions by HI and/or IF, and in two patients with IF-titer rises but stationary HI-titers. One patient with a reduction in IF-titer (640-80) had a twofold increase in PRNT-titers (640-1280). The results are summarized in Table 1. In order to evaluate at which time point during the course of disease antibodies were detectable by IF, patient sera submitted by physicians to the Department of Virology were investigated for the presence of dengue antibodies. Each patient
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
48
provided paired sera or a single early ( < 2 0 days) convalescent serum. A seroconversion or a significant (~>4-fold) titer rise for paired sera or an IF-titer of ~>320 in single sera were regarded as diagnostic. The time of sampling in relation to onset of disease for 298 sera (paired or single), chosen by the diagnostic criteria above, was provided by the submitting physician. In general, diagnostic results were seen 7-13 days post onset of symptoms. The results are summarized in Fig. 2. During the years 1991-1993 a total of 161 cases of dengue were diagnosed by IF among Swedish patients according to the above criteria. A majority of the patients (138/161 ) had travelled to Southeast Asia, with Thailand the most frequently visited country. All cases as well as geographical origin of infection are shown in Table 2.
Time post onset sampled • 1 month
~1-30days L4-20 days
'-13 days P-6days
10
10
20
40
80
160
320
640
1280
2560
5120
10240
IF-titer
Fig. 2. Indirect immunoftuorescence (IF) titers in relation to post-symptom sampling dates for 298 sera from patients with serologically diagnosed dengue virus infections.
Table 2 Dengue infection in Swedish travellers with geographical origin of infection Year
No. of cases
1991 1992 1993
49 62 50
Total
161
Southeast Asia 39 (29) ~ 57 (34) 42 (22) 138 (85)
Africa
South America
Unknown
2 1 7
2 1
6 4
10
3
10
The number of patients returning from Thailand is given in brackets.
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
49
4. Discussion
The use of HI as the standard test for serological diagnosis of dengue infections, as recommended by WHO, has been questioned, and alternative methods have been suggested. The ELISA procedure reported by Kuno et al. (1991), where detection of specific IgM is combined with IgG detection for classification of serologic responses, exemplifies an approach essential for diagnosis of dengue infections in endemic areas. However, the diagnostic requisites in non-endemic countries are different. Dengue fever is usually a differential diagnosis from e.g. malaria and typhoid in returning travellers. We have previously employed HI for diagnosis of dengue infections, but the difficulties in antigen and erythrocyte supply combined with a time-consuming test procedure prompted us to look for a more rapid and simple test suitable for a relatively small number of specimens. Our data showed that antibody detection by indirect IF proved to be at least as reliable as HI for diagnosis of dengue infections in Swedish patients. In four of the 24 patients from whom paired sera were available, a >~4-fold titer rise was seen with IF, while HI-titers remained stationary. For the remaining patients HI and IF results were in agreement, although HI-titers were occasionally difficult to interpret due to low titer values. The lack of a titer rise by HI in certain patients may be due to the presence of IgM-antibodies, which are detected by HI but not by IF. The fact that the two assays detect antibodies to different viral components provides an additional explanation to the discrepant results, since the HI detects antibodies directed to the envelope glycoprotein, while the IF detects antibodies directed to other antigens as well. The relatively low HI-titers seen in all patients investigated (Table 1) reflect the lack of fine tuning of the HI, mainly due to the difficulty in obtaining reagents and the limited number of tests performed. To ensure specificity of the IF test, paired sera from ten of the patients investigated by HI and IF were also investigated by PRNT. Sero-conversions by PRNT were seen in all patients with sero-conversions or significant titer rises in IF; however, one patient with a significant reduction in IF-titer had a rise in PRNT-titers. This patient was sampled 13 days and 5 months, respectively, post onset of symptoms. The late date of sampling for the second specimen could provide an explanation for the increase in neutralizing antibody and decrease in IF-titer, as antibodies to viral envelope proteins involved in neutralization often develop later than antibodies to nucleoproteins (Linde et al., 1987; Lundkvist et al., 1993) The investigation of dengue IF-titers in relation to sampling dates post onset of symptoms showed great individual variation. Titers /> 1280 were observed in several sera sampled from patients during their first week of illness, while lower titers were often seen in sera from other patients sampled > 2 weeks post onset (Fig. 2). When paired sera were not available, an IF-titer of >/320 in a single early convalescent sample (< 20 days) was regarded as diagnostic. This is only possible in a previously unexposed population, and the practice may have to be re-evaluated in the future when re-infections may occur in travellers who frequently visit dengue endemic areas. A possible role of a previous vaccination against other flaviviruses, e.g. Japanese encephalitis or yellow fever, cannot be ruled out as a cause of rapid antibody
50
s. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
d e v e l o p m e n t in certain patients. Vaccination alone, however, only seems to elicit low a n t i b o d y dengue IF-titers ( < 8 0 ) , if any, albeit a small n u m b e r of vaccinees (eight) was studied (S. Vene, u n p u b l i s h e d observations). It is also i m p o r t a n t to p o i n t out that, in all probability, serological cross-reactions with other flaviviruses, as seen with HI, also occur when I F is employed, a n d a positive serology alone should n o t form the basis for a diagnosis. Thus, the patient's clinical s y m p t o m s a n d travel history should be considered when evaluating the serological response.
References Boonpucknavig, S., Vuttivirojana, O., Siripont, J., Futrakul, P. and Nimmannitya, S. (1975) Indirect fluorescent antibody technic for demonstration of serum antibody in dengue hemorrhagic fever. Am. J. Clin. Pathol., 64, 365-371. Bundo, K. and Igarashi, A. (1985) Antibody-capture ELISA for detection of immunoglobulin M antibodies in sera from Japanese encephalitis and dengue hemorrhagic fever patients. J. Virol. Methods, 11, 15 22. Burke, D.S., Ramsburg, H.H. and Edelman, R. (1977) Persistence in humans of antibody to subtypes of Venezuelan equine encephalomyelitis (VEE) virus after immunization with attenuated (TC-83) VEE virus vaccine. J. Inf. Dis., 136(3), 354-359. Cunningham, R. and Mutton, K. ( 1991 ) Dengue haemorrhagic fever. BMJ, 302, 1083 1084. Gubler, D.J. (1988) Dengue. The Arboviruses: Epidemiology and Ecology. CRC Press, Boca Raton, FL, pp. 223-260. Jacobs, M.G., Brook, M.G., Weir, W.R.C. and Bannister, B.A. (1991) Dengue haemorrhagic fever: a risk of returning home. BMJ, 302, 828-829. Kuno, G., G6mez, I. and Gubler, D.J. (1991) An ELISA procedure for the diagnosis of dengue infections. J. Virol. Methods, 33, 101-113. Linde, G.A., Granstr6m, M. and Orvell, C. (1987) lmmunoglobulin class and immunoglobulin G subclassenzyme-linkedimmunosorbent assays compared with microneutralization assay for serodiagnosis of mumps infection and determination of immunity. J. Clin. Microbiol., 25(9), 1653-1658. Lundkvist, ,~., H6rling, J. and Niklasson, B. (1993) The humoral response to Puumala virus infection (nephropatia epidemica)investigatedby viral protein specificimmunoassays.Arch. Virol., 130, 121-130. Monath, T.P., Cropp, C.P., Muth, D.J. and Calisher, C.H. (1981) Indirect fluorescent antibody test for the diagnosis of yellow fever. Trans. R. Soc. Trop. Med. Hyg., 75(2), 282-286. Rush, B. (1789) An account of the bilious remitting fever, as it appeared in Philadelphia in the summer and autumn of the year 1780. In: Medical Inquiries and Observations, 2nd edn. Pritchard and Hall, Philadelphia, PA, pp. 104-117. Sabin, A.B. (1959) Dengue. Viral and Rickettsial Infections of Man. Lippincott, Philadelphia, PA, pp. 361 373. Shope, R.E. (1974) Arboviruses. Manual of Clinical Microbiology. American Society for Microbiology, Washington, DC, pp. 740-745. Westaway, E.G., Brinton, M.A., Gaidamovich, S.Y., Horzinek, M.C., Igarashi, A., K~i~iri~iinen,L., Lvov, D.K., Porterfield, J.S., Russell, P.K. and Trent, D.W. (1985) Flaviviridae. Intervirology, 24, 183-192. WittesjO, B., Eitrem, R. and Niklasson, B. (1993) Dengue fever among Swedish tourists. Scand. J. Inf. Dis., 25, 699-704.
Clinical and Diagnostic Virology 4 (1995) 43-50
Clinical and Diagnostic Virology
Indirect immunofluorescence for serological diagnosis of dengue virus infections in Swedish patients Sirkka Vene a.u.,, Joseph Mangiafico c, Bo Niklasson b.d "Department of Virology, Swedish Institute for Infectious Disease Control, S-105 21 Stockholm, Sweden; b Microbiology and Tumorbiology Centre, Karolinska Institute, Stockholm, Sweden," c Applied Research Division, USAMRIID, Frederick, MD, USA; National Defence Research Establishment, FOA-4, Umedt, Sweden
Received 6 September 1994; accepted 15 November 1994
Abstract Background: An increasing number of travellers to dengue endemic areas has emphasized the need for an easy and reliable serological test for diagnosis of dengue virus infections. Objectives: Indirect immunofluorescence (IF) and hemagglutination inhibition (HI) tests were compared for serological diagnosis of dengue fever (DF). Study design: Sera from patients with clinical symptoms compatible with DF and a travel history comprising dengue endemic areas were included in the study. Paired serum samples from 24 patients and single convalescent sera from five patients were investigated by HI and IF. Paired sera from ten patients were investigated by plaque reduction neutralization test (PRNT) in order to confirm the HI/IF results. Results: In twelve of the patients with paired sera a sero-conversion or a significant (~>4-fold) antibody titer rise was seen with HI as well as with IF. Four patients with stationary HI-titers had significant titer rises when investigated by IF. Stationary or declining HI- and IF-titers were found in the remaining eight patients. The five single sera were antibody-positive by both methods. Investigation by PRNT of paired sera from ten patients confirmed the IF results. A total of 298 serum samples (paired sera or early convalescent samples) were investigated for dengue antibodies by IF. A sero-conversion, significant titer-rise or an IF-titer of />320 was generally seen 7-13 days post onset of disease. During the years 1991-1993 altogether 161 cases of dengue were diagnosed by IF in Swedish patients, a majority of which had travelled to Southeast Asia. Conclusion: Antibody detection by indirect IF proved to be at least as reliable as HI for diagnosis of dengue infections in Swedish patients. The specificity of the IF was ensured by testing the sera of ten patients for serotype specific neutralizing dengue antibodies. Key words." Dengue antibody; IF; HI; PRNT
*Corresponding author at address a. Fax: (+46) 8-272231. 0928-0197/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0928-0197(94)00060-3
44
& Veneet aL/Clinical and Diagn+~stic Virology 4 (1995) 43 50
1. Introduction
Dengue fever (DF) has been recognized for more than 200 years (Rush, 1789) and occurs on all continents. Symptoms range from inapparent to severe hemorrhagic fever, although most patients experience classical DF with a sudden onset of fever, headache, myalgias and rash (Sabin, 1959). The dengue viruses (serotypes 1-4), which belong to the family Flaviviridae and the genus Flavivirus (Westaway et al., 1985) are endemic in tropical areas and are transmitted to humans by mosquitoes of the Aedes species, principally A. aegypti (Gubler, 1988). Serological diagnosis of DF is complicated by the cross-reactive antigenic determinants shared by the four dengue virus serotypes and other flaviviruses. In endemic countries, where distinction between primary and secondary infection is essential, sero-diagnosis of DF is generally based on detection of dengue-specific IgM by/x-capture ELISA (Bundo and Igarashi, 1985; Kuno et al., 1991 ), while most laboratories with only an occasional demand for dengue-serology use a hemagglutination inhibition (HI) test for antibody detection. An increasing number of travellers to dengue endemic areas risk exposure to the virus and subsequent illness (Cunningham and Mutton, 1991; Jacobs et al., 1991; Wittesj6 et al., 1993), a development which emphasized the need for a more rapid and convenient test than the HI for diagnosis of dengue fever in febrile travellers returning home. Although indirect immunofluorescence (IF) has been reported for serological diagnosis of DF (Boonpucknavig et al., 1975), the diagnostic use of IF has been limited in dengue serology. We used an approach suggested by Monath and co-workers for diagnosis of yellow fever (Monath et al., 1981 ) to evaluate the use of IF in the diagnosis of DF in Swedish patients.
2. Materials and methods
2.1. Serum specimens Paired serum samples were available from 24 patients with clinical symptoms compatible with dengue fever and a travel history comprising dengue endemic areas. Single convalescent sera were available from five patients chosen by the same criteria. 2.2. Hemagglutination inhibition test (HI) All serum samples were tested by a standard HI for arboviral antibodies employing goose erythrocytes and acetone-extracted sera as described by Shope (1974). Dengue virus mouse brain antigen prepared from dengue type 2 strain New Guinea C was kindly provided by N. Karabatsos, CDC, Fort Collins, Colo., USA. 2.3. hnmunofluorescence (IF) Dengue type 2 strain New Guinea C, kindly provided by R. Tesh, Yale Arbovirus Unit, New Haven, Conn, USA, was used to infect a confluent Vero cell culture in a 25 c m 2 tissue culture flask with Eagle's minimal essential medium supplemented
S. Veneet al./Clinical and Diagnostic Virology 4 (1995) 43-50
45
Fig. 1. Typicalcytoplasmicfluorescenceas seen in denguevirus infectedVero cells stained with positive human serum and fluorescein(DTAF)-labelledconjugate. with 200 U penicillin/ml, 200 mg streptomycin/ml and 5% foetal calf serum (MEMFCS). Cells were incubated at 37°C and IF-positivity of cells was checked daily with anti-dengue mouse serum, also provided by R. Tesh, and FITC-labelled rabbit (Fab')2 fragments to mouse Ig (Dakopatts A/S, Denmark). After five days of incubation, infected cells were mixed with uninfected Vero cells at a ratio of 1:3 in 30 ml of MEM-FCS. "Spot slides" were prepared by allowing the cells to attach to cleaned 12-well slides during a 16 h incubation at 37°C and 5% CO2. Cells were fixed to slides by exposure to cold anhydrous acetone and slides were stored at - 7 0 ° C until used. Sera were tested at 2-fold dilutions starting at 1:10 and fluorescein (DTAF)-conjugated goat anti-human IgG (heavy and light chain specific; Jackson Immunoresearch Laboratories, West Grove, Pa., USA) was used to detect bound antibodies. End point titers were set at the reciprocal of the last dilution giving an unequivocal fluorescence. The typical cytoplasmic fluorescence is shown in Fig. 1. 2.4. Plaque reduction neutralization test ( P R N T )
In order to exclude cross-reactions caused by antibodies to other flaviviruses, paired sera from ten patients were investigated for dengue serotype specific neutralizing antibodies, essentially as described by Burke et al. (1977). The PRNT was modified for use with 12-well plates to obtain neutralizing titers to the four dengue serotypes, strains WP (type 1), 25 (type 2), CH53489 (type 3), and H241 (type 4).
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
46
Table 1 Results by different serological methods in 29 patients with dengue infections Patient No.
HI Den
IF Den
PRNT Den
Sero-conversion or four-fold titer rise H I and I F 1.I < 10 80 1.II 40 > 10240 2.I <10 <10 2.II 80 5120 3.I <10 <10 3.11 20 320 4.1 <10 <10 4.I1 20 1280 5.1 < 10 10 5.11 80 5120 6.1 < 10 10 6.11 40 640 7.1 10 20 7.11 80 5120 8.1 < 10 < 10 8.11 40 2560 9.1 < 10 < 10 9.11 40 > 10240 10.I < 10 < 10 10.II 80 2560 ll.I <10 <10 11.II 40 > 10240 12.I < 10 <10 12.I1 40 2560
< 10 160 nt nt nt nt nt nt nt nt <10 2560 < 10 80
Stationary H I but titer r&e IF 13.I 40 13.II 40 14.I 20 14.1I 20 15.1 40 15.II 40 16.I 40 16.II 40
40 640 1280 5120 80 1280 1280 10240
nt nt < 10 640 < 10 160 nt nt
5120 5120 2560 2560 1280 1280 2560 2560 640 640 1280 640
nt nt nt nt nt nt nt nt nt nt nt nt
No titer rise H I or IF 17.I 17.11 18.I 18.II 19.I 19.1I 20.I 20.II 21.I 21.II 22.1 22.II
40 40 80 80 80 80 40 40 40 40 40 20
< 10 320 <10 640
Den serotype
Dengue 3/4 Dengue 1 Dengue 2 Dengue 1
Dengue 2
Dengue 1 Dengue 1
Dengue 1 Dengue 1
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
47
Table 1 continued) Patient No. 23.I 23.11 24.I 24.11
HI Den
IF Den
40 20 80 nt
2560 640 640 80
20 40 20 40 40
> 10240 5120 1280 1280 640
PRNT Den nt nt 640 1280
Den serotype
Dengue3
Sing& sera
25 26 27 28 29
nt nt nt nt nt
ant = not tested. Briefly, dengue virus, serotypes 1 through 4, at approximately 100 plaque forming units (PFU)/0.1 ml was mixed with an equal volume of serial 4-fold dilutions of test sera. Serum-virus mixtures were held at 4°C overnight and inoculated onto confluent monolayers of L L C - M K 2 cells in duplicate wells at 0.1 ml/well. Plates were incubated for 1 h at 37°C and 5% CO2, and each well was covered with 1.5 ml of nutrient overlay (Eagle's basal medium and Earle's salts with 1.1% agar, 5% FCS, 200 U penicillin/ml and 200 mg streptomycin/ml). After incubation at 37°C and 5% CO2 for 4 days, 1.5 ml of a second overlay, identical to the first, but containing neutral red at 1:7500 dilution was added to each well, and plates were incubated for 24-48 h at 37°C and 5% CO2, before plaques were enumerated. Neutralizing antibody titers were expressed as the reciprocal of the highest dilution reducing the number of plaques by 80% as compared to the virus control.
3. R ~ In twelve of the patients with paired sera, a sero-conversion or titer rise was seen with HI as well as with IF. Four patients with stationary HI-titers had significant (~>4-fold) titer rises when investigated by IF. Stationary or declining HI- and IF-titers were found in eight patients. Five single sera were antibody-positive by both methods. Paired sera from ten patients were examined by P R N T for dengue type specific antibodies. Sero-conversions by P R N T were seen in seven patients, who also had sero-conversions by HI and/or IF, and in two patients with IF-titer rises but stationary HI-titers. One patient with a reduction in IF-titer (640-80) had a twofold increase in PRNT-titers (640-1280). The results are summarized in Table 1. In order to evaluate at which time point during the course of disease antibodies were detectable by IF, patient sera submitted by physicians to the Department of Virology were investigated for the presence of dengue antibodies. Each patient
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
48
provided paired sera or a single early ( < 2 0 days) convalescent serum. A seroconversion or a significant (~>4-fold) titer rise for paired sera or an IF-titer of ~>320 in single sera were regarded as diagnostic. The time of sampling in relation to onset of disease for 298 sera (paired or single), chosen by the diagnostic criteria above, was provided by the submitting physician. In general, diagnostic results were seen 7-13 days post onset of symptoms. The results are summarized in Fig. 2. During the years 1991-1993 a total of 161 cases of dengue were diagnosed by IF among Swedish patients according to the above criteria. A majority of the patients (138/161 ) had travelled to Southeast Asia, with Thailand the most frequently visited country. All cases as well as geographical origin of infection are shown in Table 2.
Time post onset sampled • 1 month
~1-30days L4-20 days
'-13 days P-6days
10
10
20
40
80
160
320
640
1280
2560
5120
10240
IF-titer
Fig. 2. Indirect immunoftuorescence (IF) titers in relation to post-symptom sampling dates for 298 sera from patients with serologically diagnosed dengue virus infections.
Table 2 Dengue infection in Swedish travellers with geographical origin of infection Year
No. of cases
1991 1992 1993
49 62 50
Total
161
Southeast Asia 39 (29) ~ 57 (34) 42 (22) 138 (85)
Africa
South America
Unknown
2 1 7
2 1
6 4
10
3
10
The number of patients returning from Thailand is given in brackets.
S. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
49
4. Discussion
The use of HI as the standard test for serological diagnosis of dengue infections, as recommended by WHO, has been questioned, and alternative methods have been suggested. The ELISA procedure reported by Kuno et al. (1991), where detection of specific IgM is combined with IgG detection for classification of serologic responses, exemplifies an approach essential for diagnosis of dengue infections in endemic areas. However, the diagnostic requisites in non-endemic countries are different. Dengue fever is usually a differential diagnosis from e.g. malaria and typhoid in returning travellers. We have previously employed HI for diagnosis of dengue infections, but the difficulties in antigen and erythrocyte supply combined with a time-consuming test procedure prompted us to look for a more rapid and simple test suitable for a relatively small number of specimens. Our data showed that antibody detection by indirect IF proved to be at least as reliable as HI for diagnosis of dengue infections in Swedish patients. In four of the 24 patients from whom paired sera were available, a >~4-fold titer rise was seen with IF, while HI-titers remained stationary. For the remaining patients HI and IF results were in agreement, although HI-titers were occasionally difficult to interpret due to low titer values. The lack of a titer rise by HI in certain patients may be due to the presence of IgM-antibodies, which are detected by HI but not by IF. The fact that the two assays detect antibodies to different viral components provides an additional explanation to the discrepant results, since the HI detects antibodies directed to the envelope glycoprotein, while the IF detects antibodies directed to other antigens as well. The relatively low HI-titers seen in all patients investigated (Table 1) reflect the lack of fine tuning of the HI, mainly due to the difficulty in obtaining reagents and the limited number of tests performed. To ensure specificity of the IF test, paired sera from ten of the patients investigated by HI and IF were also investigated by PRNT. Sero-conversions by PRNT were seen in all patients with sero-conversions or significant titer rises in IF; however, one patient with a significant reduction in IF-titer had a rise in PRNT-titers. This patient was sampled 13 days and 5 months, respectively, post onset of symptoms. The late date of sampling for the second specimen could provide an explanation for the increase in neutralizing antibody and decrease in IF-titer, as antibodies to viral envelope proteins involved in neutralization often develop later than antibodies to nucleoproteins (Linde et al., 1987; Lundkvist et al., 1993) The investigation of dengue IF-titers in relation to sampling dates post onset of symptoms showed great individual variation. Titers /> 1280 were observed in several sera sampled from patients during their first week of illness, while lower titers were often seen in sera from other patients sampled > 2 weeks post onset (Fig. 2). When paired sera were not available, an IF-titer of >/320 in a single early convalescent sample (< 20 days) was regarded as diagnostic. This is only possible in a previously unexposed population, and the practice may have to be re-evaluated in the future when re-infections may occur in travellers who frequently visit dengue endemic areas. A possible role of a previous vaccination against other flaviviruses, e.g. Japanese encephalitis or yellow fever, cannot be ruled out as a cause of rapid antibody
50
s. Vene et al./Clinical and Diagnostic Virology 4 (1995) 43-50
d e v e l o p m e n t in certain patients. Vaccination alone, however, only seems to elicit low a n t i b o d y dengue IF-titers ( < 8 0 ) , if any, albeit a small n u m b e r of vaccinees (eight) was studied (S. Vene, u n p u b l i s h e d observations). It is also i m p o r t a n t to p o i n t out that, in all probability, serological cross-reactions with other flaviviruses, as seen with HI, also occur when I F is employed, a n d a positive serology alone should n o t form the basis for a diagnosis. Thus, the patient's clinical s y m p t o m s a n d travel history should be considered when evaluating the serological response.
References Boonpucknavig, S., Vuttivirojana, O., Siripont, J., Futrakul, P. and Nimmannitya, S. (1975) Indirect fluorescent antibody technic for demonstration of serum antibody in dengue hemorrhagic fever. Am. J. Clin. Pathol., 64, 365-371. Bundo, K. and Igarashi, A. (1985) Antibody-capture ELISA for detection of immunoglobulin M antibodies in sera from Japanese encephalitis and dengue hemorrhagic fever patients. J. Virol. Methods, 11, 15 22. Burke, D.S., Ramsburg, H.H. and Edelman, R. (1977) Persistence in humans of antibody to subtypes of Venezuelan equine encephalomyelitis (VEE) virus after immunization with attenuated (TC-83) VEE virus vaccine. J. Inf. Dis., 136(3), 354-359. Cunningham, R. and Mutton, K. ( 1991 ) Dengue haemorrhagic fever. BMJ, 302, 1083 1084. Gubler, D.J. (1988) Dengue. The Arboviruses: Epidemiology and Ecology. CRC Press, Boca Raton, FL, pp. 223-260. Jacobs, M.G., Brook, M.G., Weir, W.R.C. and Bannister, B.A. (1991) Dengue haemorrhagic fever: a risk of returning home. BMJ, 302, 828-829. Kuno, G., G6mez, I. and Gubler, D.J. (1991) An ELISA procedure for the diagnosis of dengue infections. J. Virol. Methods, 33, 101-113. Linde, G.A., Granstr6m, M. and Orvell, C. (1987) lmmunoglobulin class and immunoglobulin G subclassenzyme-linkedimmunosorbent assays compared with microneutralization assay for serodiagnosis of mumps infection and determination of immunity. J. Clin. Microbiol., 25(9), 1653-1658. Lundkvist, ,~., H6rling, J. and Niklasson, B. (1993) The humoral response to Puumala virus infection (nephropatia epidemica)investigatedby viral protein specificimmunoassays.Arch. Virol., 130, 121-130. Monath, T.P., Cropp, C.P., Muth, D.J. and Calisher, C.H. (1981) Indirect fluorescent antibody test for the diagnosis of yellow fever. Trans. R. Soc. Trop. Med. Hyg., 75(2), 282-286. Rush, B. (1789) An account of the bilious remitting fever, as it appeared in Philadelphia in the summer and autumn of the year 1780. In: Medical Inquiries and Observations, 2nd edn. Pritchard and Hall, Philadelphia, PA, pp. 104-117. Sabin, A.B. (1959) Dengue. Viral and Rickettsial Infections of Man. Lippincott, Philadelphia, PA, pp. 361 373. Shope, R.E. (1974) Arboviruses. Manual of Clinical Microbiology. American Society for Microbiology, Washington, DC, pp. 740-745. Westaway, E.G., Brinton, M.A., Gaidamovich, S.Y., Horzinek, M.C., Igarashi, A., K~i~iri~iinen,L., Lvov, D.K., Porterfield, J.S., Russell, P.K. and Trent, D.W. (1985) Flaviviridae. Intervirology, 24, 183-192. WittesjO, B., Eitrem, R. and Niklasson, B. (1993) Dengue fever among Swedish tourists. Scand. J. Inf. Dis., 25, 699-704.