Scand J Infect Dis 24: 275-282, 1092
Polymerase Chain Reaction with Double Primer Pairs for Detection of Human Parvovirus B19 Induced Aplastic Crises in Family Outbreaks EVA FRIDELL’, ALBERT N. BEKASSY?, BARBRO LARSSON’ and BRITT-MARIE ERIKSSON‘
‘
Departments of Virology. Nationcrl Bnc/wiological Laborutory rind Knrolinsku lnstrtirtc,, Stockholm, Pediatrics, University Hospitril. Litrid, Pediarrics, Central Hospitul. V(jxjo, rirrd ‘infectious D~semes.tinivrrsity Hospirtrl, tipl~scila.Sweden Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Frortr /he
’
’
Parvovirus B19 DNA can be detected by polymerase chain reaction with double primer pairs (nested PCR). Recent infection was documented by a retrospective serological study using Parvoscan-B19 enzyme linked immunosorbent assay (EIA) for detection of B19 human parvovirus IgM and IgG antibodies in serum or plasma specimens. In 3 families B19 outbreaks caused aplastic crises necessitating blood transfusion in 5 children and 1 adult with hereditary sphaerocytosis. Four members from 2 of the families had clinically overt haemolytic anaemia prior to the event. W o members in another family presented with an aplastic crisis disclosing the underlying chronic haemolytic disease. All 7 patients were identified as PCR positive in serum samples taken 3-14 days after the onset of symptoms. Comparison with dot blot hybridization revealed detectable DNA in only 213 PCR positive patients. Thus, nested PCR is more sensitive than the dot blot hybridization method and is therefore a suitable complement to the antibody assay for identifying recent B19 infection. E. Fridell, MD, Dept. of Virology, National Bacteriological Laboratory, S-105 21 Stockholm, Sweden
INTRODUCTION Human parvovirus B19 causes the harmless and common disease erythema infectiosum fifth disease - in otherwise healthy children (1). The virus also causes aplastic crisis in about 90% of children with a chronic haemolytic disease ( 2 4 ) . Patients with an impaired immune system such as newborn and premature babies, the immunosuppressed transplant patients and patients suffering from leukaemia or HIV disease are unable to maintain an adequate immune response to parvovirus and periods of long standing anaemia may occur (5-11). These patients also shed the virus (12) and are considered to be contagious. In order to study different assay techniques available for diagnosis of B19 infection, a group of patients with aplastic crisis and underlying haemolytic anaemia was examined for the presence of B19 D N A . The results obtained from a nested PCR were compared with those from a dot blot hybridization tcchnique. We also measured antibody responses to a synthetic peptide and compared the results with the antibody responses to the whole virus antigen. MATERIALS A N D METHODS Study outline Mciiihers of3 families with hereditary sphaerocytosis cxpcriencing aplastic crisis or other B19 symptoins during oulbreaks of erythcnia infectiosum in Sweden in 1986 and 1989 were investigated. Sera were draun at early onset of thc disease. during the convalescence period and at follow-up about 10 month, eftcr thc onsct (Tables I and IV).
276 E. Fridell et al.
Scand J Infect Dis 24
Table I . Lowest huematologic values in 3 families with hereditary sphaerocytic anaemia and uplastic crisis. NI = not investigated Values during disease (normal values)
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Haemoglobin (g4
Reticulocytes (Yo)
White blood cells ( loy/\)
Platelets (10911)
Family A BOY Mother
54 (110-160) 56 (133-139)
0.6 (11.4) 0.7 (6.6)
6.9 4.3
194 121
Family B Elder boy Younger boy
48 (80-105) 40 (90-100)
0.2 (4.8) 0.0 (2.1)
I .2 2.0
90 R5
Family C Elder girl Younger girl Mother
60 (10&110) 32 (7C- 80) 122
0.2 (4.1) 0.1 (2.7) 1.2
NI NI NI
NI NI NI
Nesred polymerase chain reaction (PCR) Sera were stored at -20°C until use. The DNA from serum was diluted 1:IOO and used for amplification by PCR. A sample known to contain B19 virus (kindly donated from Dr B. Cohen, London) was prepared in the same way as the samples and used as a positive control. Dilution buffer was used as negative control. DNA oligonucleotide primers were chosen from the gene encoding the structural proteins of parvovirus BIY. They were synthesized in a Gene Assembler I (Pharmacia) and purified as described by the manufacturer. Two primer pairs were used. The outer pair consisted of 5’ GGACTGTAGCAGATGAAGAG (390; nucleotides 2955-2974) and 5’ TATGGGACTGATGGTG (490; 3364-3349), the inner pair consisted of 5’ GGGTTTCAAGCACAAGTAG (190; 3002-3220) and 5‘ CCTTATAATGGTGCTCTGGG (290; 3291-3272). The numbers of the nucleotides are given according to Shade et al. (13). The outer primer set, 39W90, amplifies 410 bp and the inner set, 190-290, a 290 bp amplimer. A search in databases revealed no strong nucleotide homologies in this region. PCR was carried out in 2 steps, initiated with the outer primer pair and followed by the inner (i.e. nested) primer pair, and performed in 0.5 ml microfuge tubes in a total volume of 25 pl. The test system was optimised through titration of the Mg2+-and Na+-concentrations together with the proper primer concentration and annealing temperature. The final composition of the reaction mixture was 10 mM Tris-HCI pH 9.6, 2 mM MgCI,, 50 mM NaCI, 0.2 mM of each deoxynucleoside triphosphate (dNTP), 0.1 pM of each primer. 1U of thermus aquaticus (Taq) DNA polymerase (Perkin Elmer-Cetus, CA), 2 PI prediluted sample, and water added to a final volume of 25 pl. The tubes were covered with 25 PI paraffin oil to prevent evaporation. Two pI from the first amplified DNA was transferred to a new microfuge tube containing the same reaction mixture as the outer PCR system. The cycles, consisting of 30 s at 95”C, 60 s at 55°C for the outer and 63°C for the nested PCR, and 60 s at 7 2 T , were performed in an automated thermal cycler (DNA Thermal Cycler, Perkin Elmer-Cetus). The product from the second PCR (10 pl) was analysed by electrophoresis on a 1.8% agarose gel and stained with ethidium bromide. The diluted serum samples from the patients and prepared controls were transferred to the reaction tubes as the last step to avoid contamination of the buffers according to standard sterile procedures. No contamination occurred in more than 300 nested PCRs. The results were compared to findings obtained by dot blot hybridisation using a probe located partly in the same region as the PCR product (University College London, UCL, 14; Central Public Health Laboratory, London. CPHL, 15) and using digoxygenin with a coloured reaction product or 3’P-labelled probes with X-ray measurement as detection systems. For comparison 3 samples were evaluated with dot blot hybridisation, and 1 also in immune electron microscopy (IEM; 6).
PCR parvovirus in family outbreaks 277
Scand J Infect Dis 24
Table 11. Signs of infection and duration of symptoms in 7 members of 3 families with
hereditary sphaerocytic anaemia and aplastic crisis
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
No. of pats. Pallor Rash Pyrexia duration Vomiting Mild diarrhoea Headache Weakness Joint and muscle complaintsd
Duration (days)
2- 6
2- 9 1- 3 1- 2 3-10 1-10 6-14
“Females Anlibody detection assays Determination of antibodies to parvovirus was performed in 2 different ways: an immunosorbent assay using a synthetic peptide as antigen (amino acid sequence: FSPAASSCHNASGKEAKVCTISPI; SP assay. 16, 17), and also using native virus as antigen (NV assay, 18, 19).
Patients Family A Family A comprised a 12-year-old boy and his parents who had no known underlying chronic haemolytic disease. Following a week of pyrexia. headache. nausea and mild diarrhoea, the boy was admitted to hospital, He was pale and had generalised lymphadenopathy. Haematological findings are shown in Table 1. With a haemoglobin (Hb) level of 54 g/l and reticulocytopenia of 0.6%, transfusion of packed red cells was given and he recovered (Fdbk 111). On day 17. he developed a rash on his limbs during treatment with cotrimoxazole for pulmonary infiltrates. His mother, 43-year-old, wab admitted to hospital 4 days after the boy with the same clinical symptoms. Her H b level fell to 56 g/l 11 days after the onset of symptoms (Table I), necessitating blood transfusion which lead to an uneventful recovery. The boy and his mother were both suffering from previously undiagnosed hereditary sphaerocytosis which was diagnosed by a further haematologic investigation. The father remained healthy. This family has been described in a previous report (20). Funii1.v B Two bays aged 10 and 8 years, were known to have clinically overt haemolytic anaemia due to hereditary sphaerocytosis which did not require blood transfusion. Their 36-year-old father became splenectomised in his early teens and their 13-year-old sister was splenectomised when she was Y-year-old. Both the father and the sister remained asymptomatic, as did the mother. The elder boy became tired, with pyrexia, headache and nausea, and started vomiting 3 days after the onset of symptoms. On admission, his spleen was enlarged and generalised lymphadenopathy was noted. H e had depressed haematological values (Table I) necessitating transfusion, with a haemoglobin level of 48 g/l and reticulocytes of 0.2%. Half a year later splenectomy was carried out when cholestatic jaundice caused by gallstones developed. The younger boy presented 8 days after his brother’s first symptoms. H e had jaundice and generalised lymphadenopathy. Following blood transfusion with a Hb level of40 g/l and reticulocytes 0 (Table I), he rapidly recovered.
Farnily C The 31-year-old mother was splenectomised in early childhood due to hereditary sphaerocytosis causing severe haemolytic anaemia. Two of her children, 5- and 3-year-old girls, had clinically overt haemolytic disease. The younger girl needed blood transfusion 5 times during her first year of life. Her sister experienced a mild haemolytic crisis 1 year prior to the present event which did not require blood transfusion. The younger girl presented with fever and vomiting and complained of joint and muscle pains. She became tired and pale and was still pyrexic after a week. On admission, haematologic values (Table I)
278 E. Fridell et al.
Scand J Infect Dis 24
Table 111. Intrufumilial spread, individuul treatment and duration of symptoms in the members
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
of 3 ,families with hereditary sphaerocytosis and aplustic crisis Onset of symptoms (Days after the first symptom in the "index" child)
Admission to hospital
Family A Boy Mother
0 4
5 10
Family B Elder boy Younger boy
0 8
3, 4 , 9 3.6
7
3
Family C Younger girl Mother Elder girl
0 4 8
7
78
5
6
9 14 10
Blood transfusion
Duration of main symptoms (Days)
(Days after onset of symptoms)
3
I 7t
8 12
6
Table IV.Antibody responses and D N A detection in 3 families with parvovirus infection. NI = not investigated Days after onset
IgG (A405 values)
Family A
Boy Mother
Family B Elder boy
Younger boy
7 47 4 14 31 71
0.29 0.88 2.00
>TO0 >2.00 1.70 0.50
1.10 1.60 0.60 0.60
IgM
DNA detection (nested PCR)
Scnnd I Infect DI\ 21
PCR parvovirus
itz
family outbreiiks
were very lowg. With Hb 32 g/l and rcticulocytes 0.I"h she was given a transfusion of packed red blood ccllh (Table 111) and recovered rapidly. Four days later, the mother became ill with fever. and vomiting lasting for 2 days. She suffered from joint pains and stiffnes for 2 weeks. Transfusion was not required as her Hb level was stable. Thc onset of pyrexia occurred another 4 days later in the clder child. She also received transfusion of packed red blood cells, with a Hb level of 60 g/l and 0.2% reticulocytes. and recovered uneventfully.
RESULTS
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Signs of infection and duration of symptoms are summarised in Table 11.
D N A detection The PCR method was used with sera from all family members with symptoms. The first sample from all patients was positive using the PCR technique (Table IV). B19 DNA was found using PCR in the 2 dot blot positive samples and in 1 dot blot negative sample. Sera from the 3 boys in families A and B were examined with the dot blot digoxygenin hybridisation method. In the first serum sample from the boy in family A , B19 DNA could be detected as well as in the first sample of 1 boy in family B. In the latter sample the findings were supported by a positive result using IEM. The first serum sample from the younger boy in family B was tested, first with the digoxygenin method, then with a more sensitive technique using j2P-probes as detecting system, but the sample remained negative for B19
DNA. Syrrtlietic peptirle assay IgM and IgG responses were followed for up to 10 months in the families and are summarised in Table I V . The first sample from all patients except one was positive in the IgM SP assay.Seven samples from h patients were analyzed in the IgM native virus assay. IgM could not he detected in 2 samples from 2 patients on days 3 and 4 after onset of symptoms but they were present in all samples drawn on day 6 or later. In family A, the child had dctectable IgM using both SP and NV assays in the first sample and a late IgG rise. His mother had no detectable IgM using the SP assay. A slow IgG response was provoked, giving a low absorbance value 1 month later. A subsequent IgG absorbance rise was seen 2 months later. In family B, the first samples were taken very early and the second samples after more than 3 months. T h e boys in this family had IgM and a low IgG corresponding to a B19 virus infection using the SP assay, while IgM was not detectable in their first samples using the native virus assay. All members in family C had an IgM response in the first two samples using the SP assay and in samples tested with the NV assay. They rapidly produced IgG and had high absorbance values after 1@14 days which decreased after 2 months, but were still detectable after 10 months. DlSCUSSION We describe here 3 families, in which all members (unless previously splenectomised) with sphaerocytosis experienced aplastic crises due to transient arrest of erythropoiesis induced by human parvovirus B19 infection (2). 617 patients required blood transfusion. Four of them had clinically overt hacmolytic anaemia prior to this event. Two members in one family presented with an aplastic crisis disclosing the underlying chronic haemolytic disease. A
279
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
280 E. Fridell et al.
Scand J lnfect Dis 24
clinically mild human parvovirus B19 infection occurred in the remaining adult who remained well without haemolysis. These patients did not have erythema (1). Pallor was the common symptom in all the patients with parvovirus B19 infection. Pyrexia, headache, vomiting, and diarrhoea also occurred, currently accompanied in 3 females with joint and muscle pains of short duration, The intrafamilial spread and epidemiological data are summarised in Table 111. The index case in each family was a child, followed by a parent and/or another child. The period of incubation between the index case and the parent was short (4 days). The second child seemed to have an incubation period of 8 days. The average incubation period is generally 7-10 days from the first viraemic phase causing aplastic crisis. As expected, the second child infected in both families with siblings became sick after the time of ascribed transmission of parvovirus from the first infected child (Table 111). Contrariwise, 2 adults presented with symptoms 4 days after the index case. They were presumably infected simultaneously with the index case, but had a longer incubation period for reasons unknown. Alternatively, if infection from the index child had occurred, the incubation period in these adults must have been shorter. One could speculate, that the parents were reinfected, and their immune system, therefore, presented a more prompt antibody response, leading to a shorter incubation period. All patients had detectable D N A in the first sample collected at days 3-4 from the onset of symptoms using the nested PCR. The duration of the D N A persistence varied: in family A up to at least 47 days, while in family C for only 15 days. The duration of the virus infection in these patients shows that B19 virus is neutralised more slowly in some of them. It is obvious, that IgM coexists with the viral DNA. The mother and the elder girl in family C both had high IgG absorbance values and detectable B19 DNA. This might indicate that neither IgG nor IgM directed against the synthetic peptide are neutralising antibodies. The patients recovered and had no further symptoms even though B19 D N A was still present. The boy in family A , for example, had recovered after 11 days but still had detectable B19 D N A after 47 days. This shows that the presence of D N A in an asymptomatic patient not only correlates to infection in progress but might also indicate an infection 1-2 months ago. In order to establish diagnosis of B19 infection, antibody measurement and viral detection are useful. We have earlier described an EIA using a synthetic peptide as the antigen for B19 IgM (16), where IgM and IgG responses could be monitored for up to 10 months. All patients who developed IgM responses had IgM in their first samples collected 3-14 days after onset. The IgM reaction disappeared in samples collected on day 27 and later. One patient did not maintain the IgM response to the peptide used as antigen. This might indicate, that the IgM responses to the peptide are directed to different epitopes than in the NV assay, which did not detect IgM in the first sample from the younger boy in family B. The SP assay seems to be able to detect an early response. However, IgM directed to the synthetic peptide cannot be detected for as long a time as IgM directed to the whole virus, which persists for up to 4 months (17). IgG detected by the SP assay showed rises in all 5 patients where the first sample was collected during the first 2 weeks after onset of symptoms and the second sample within 2-3 months. One patient did not produce IgM and developed a very low amount of specific IgG during the first month. The reason for her late antibody response is uncertain, but in samples from this patient the B19 DNA could b e spotted for an unusually long period. This patient might have produced non-neutralising antibodies early in the disease. Low IgG levels to the peptide could be detected in the first samples from 3 boys. IgG values in the SP assay rose for 11-71 days and then decreased in all patients in samples collected later than 6 months after the onset of symptoms. In 2 patients the absorbance was
Scand J
Infect Dis 24
PCR parvovirus in family outbreaks 28 I
very low after 8-9 months showing that thc SP assay might be less sensitive for detection of a previous infection. There are differences in the antibodies found by the S P and NV assays. In the former, a 24 amino acid long peptide derived from a region shown to provoke early IgG is used as antigen (20). In the later the native virus from human sera with a less defined antigenic region i s used.
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Conclioions Family outbreaks of human parvovirus B19 infection and its complications in patients with haemolytic anaemia can be monitored by a peptide assay for the detection of specific antibodies, and the P C R technique for the detection of the viral DNA. The nested PCR is a very sensitive method for detection of B19 virus DNA and the method is more sensitive than the dot blot hybridisation test. Provided it is handled properly contamination can be avoided. Nested PCR i s therefore the method of choice in reference laboratories for diagnosis of B 19 infection in patients with aplastic crisis. The SP assay is a good complement to the NV assay. However, the latter is limited by a poor supply of B19 virus to use as antigen. Synthetic antigen is available and can form a part of a fast and sensitive assay for detection of antibodies directed to human parvovirus B19.
REFERENCES 1. Anderson MJ, Lewis E , Kidd l M , Hall SM. Cohen BJ. An outbreak of erythema infectiosum
associated with human parvovirus infection. J Hyg 93: 85-93, 1984. 2. Lefrere J-J, Courouce A-M. Girot R , Bertrand Y, Soulier JP. Six cases of hereditary spherocytosis revcalcd by human parvovirus infection. Br J Haematol 62: 653458, 1986. 3. Serjeant GR, Topley JM. Mason K , Serjeant BE, Pattison JR, Jones SE, Mohammed R. Outbreak of aplastic crises in sickle cell anaemia associated with parvovirus-like agent. Lancet 2: 595-597, 1981. 4. Young N. Haematologic and haematopoietic consequences of B19 parvovirus infection. Semin Hematol 25: 159-172, 1988. 5. Bowman CA. Cohen BJ, Norfolk DR. Lacey CJN. Red cell aplasia associated with human parvovirus B 19 and HIV infection: failure to respond clinically to intravenous immunoglobulin. AIDS 4: 1038-1039, 1990. 6. Field AM, Cohen BJ, Brown KE, Mori J . Clewley JP, Nascimento JB, Hallam NF. Detection of B19 parvovirus in human fetal tissues by electron microscopy. J Med Virol 35: 85-95, 1991. 7. Graeve JLA, de Alarcon PA, Naides SJ. Parvovirus B19 infection in patients receiving cancer chemotherapy: The expanding spectrum of disease. Am J Pediatr Hematol Oncol 11: 441-444. 1989. 8. Kurtzman G J , Ozawa K , Cohen BJ, Hanson G, Oseas R, Young NS. Chronic bone marrow failure due to persistent B19 parvovirus infection. N Engl J Med 317: 287-294, 1987. 9. Malarme M, Vandervelde D , Brasseur M. Parvovirus infection. leukaemia, and immunodeficiency. Lancet 1: 1457, 1989. 10. Pillay D, Patou G, Griffiths P. Rees L. Secondary parvovirus infection in an immunocompromised child. Pediatr Infect Dis J 10: 623-624. 1991. 11. Pattison JK. Jones SE. Hodgson J . Davis LR. White JM. Stroud CE, Murtaxa L. Parvovirus infections and hypoplastic crisis in sickle- cell anaemia. Lancet 1: 664-665, 1981. 12. Kurtzman GJ. Cohen BJ. Meyers P, Amunullah A , Young NS. Persistent B19 parvovirus infection as a cause of severe chronic anaemia in children with acute lymphocytic leukaemia. Lancet 2: 1159-1162, 1988. 13. Shade RO. Blundell MC, Cotmore SF, Tattcrsall P, Astell C. Nucleotide scqucncc and genome organization of human parvovii-us B19 isolated from the serum of a child during aplastic crisis. .I Virol 58: 921-936, 1986. 14. Anderson MJ. Jones SE. Minson AC. Diagnosis of human parvovirus infection by dot-blot hybridization using cloned viral DNA. J Med Virol 15: 163-172, 1985.
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
282 E. Fridell et al.
Scand J lnfect Dis 24
15. Clewley JP. Detection of human parvovirus using a molecularly cloned probe. J Med Viral 15: 173-181, 1985. 16. Fridell E, Trojnar J , Wahrcn B. A new peptide for human parvovirus B19 antibody detection. Scand J Infect Dis 21: 597-603. 1989. 17. Fridell E, Cohen BJ, Wahren B. Evaluation of a synthetic-peptide enzyme-linked immunosorbcnt assay for immunoglobulin M to human parvovirus BIY. J Clin Microbiol 29: 13761381, 1991. 18. Anderson MJ. Davis LR, Jones SE, Pattison JR, Scrjeant GR. The development and use of an antibody capture radioimmunoassay for specific IgM to a human parvovirus-like agent. J Hyg 88: 309-324. 1982. 19. Cohen BJ. Mortimer PP. Pereira MS. Diagnostic assays with monoclonal antibodies for the human serum parvovirus-like virus (SPLV). J Hyg 91: 113-130, 1083. 20. Eriksson B-M, Striimberg A, Kreuger A. Human parvovirus B19 infection with severe anemia affecting mother and son. Scand J Infect Dis 20: 335-337, 1988. 21. Kurtzman G, Frickcnhofen N , Kimball J , Jenkins DW, Nienhuis AW, Young NS. Pure red-cell aplasia of 10 years duration due to persistent parvovirus B19 infection and its cure with immunoglobulin therapy. N Engl J Med 321: 519-523, 1989.
Polymerase Chain Reaction with Double Primer Pairs for Detection of Human Parvovirus B19 Induced Aplastic Crises in Family Outbreaks EVA FRIDELL’, ALBERT N. BEKASSY?, BARBRO LARSSON’ and BRITT-MARIE ERIKSSON‘
‘
Departments of Virology. Nationcrl Bnc/wiological Laborutory rind Knrolinsku lnstrtirtc,, Stockholm, Pediatrics, University Hospitril. Litrid, Pediarrics, Central Hospitul. V(jxjo, rirrd ‘infectious D~semes.tinivrrsity Hospirtrl, tipl~scila.Sweden Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Frortr /he
’
’
Parvovirus B19 DNA can be detected by polymerase chain reaction with double primer pairs (nested PCR). Recent infection was documented by a retrospective serological study using Parvoscan-B19 enzyme linked immunosorbent assay (EIA) for detection of B19 human parvovirus IgM and IgG antibodies in serum or plasma specimens. In 3 families B19 outbreaks caused aplastic crises necessitating blood transfusion in 5 children and 1 adult with hereditary sphaerocytosis. Four members from 2 of the families had clinically overt haemolytic anaemia prior to the event. W o members in another family presented with an aplastic crisis disclosing the underlying chronic haemolytic disease. All 7 patients were identified as PCR positive in serum samples taken 3-14 days after the onset of symptoms. Comparison with dot blot hybridization revealed detectable DNA in only 213 PCR positive patients. Thus, nested PCR is more sensitive than the dot blot hybridization method and is therefore a suitable complement to the antibody assay for identifying recent B19 infection. E. Fridell, MD, Dept. of Virology, National Bacteriological Laboratory, S-105 21 Stockholm, Sweden
INTRODUCTION Human parvovirus B19 causes the harmless and common disease erythema infectiosum fifth disease - in otherwise healthy children (1). The virus also causes aplastic crisis in about 90% of children with a chronic haemolytic disease ( 2 4 ) . Patients with an impaired immune system such as newborn and premature babies, the immunosuppressed transplant patients and patients suffering from leukaemia or HIV disease are unable to maintain an adequate immune response to parvovirus and periods of long standing anaemia may occur (5-11). These patients also shed the virus (12) and are considered to be contagious. In order to study different assay techniques available for diagnosis of B19 infection, a group of patients with aplastic crisis and underlying haemolytic anaemia was examined for the presence of B19 D N A . The results obtained from a nested PCR were compared with those from a dot blot hybridization tcchnique. We also measured antibody responses to a synthetic peptide and compared the results with the antibody responses to the whole virus antigen. MATERIALS A N D METHODS Study outline Mciiihers of3 families with hereditary sphaerocytosis cxpcriencing aplastic crisis or other B19 symptoins during oulbreaks of erythcnia infectiosum in Sweden in 1986 and 1989 were investigated. Sera were draun at early onset of thc disease. during the convalescence period and at follow-up about 10 month, eftcr thc onsct (Tables I and IV).
276 E. Fridell et al.
Scand J Infect Dis 24
Table I . Lowest huematologic values in 3 families with hereditary sphaerocytic anaemia and uplastic crisis. NI = not investigated Values during disease (normal values)
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Haemoglobin (g4
Reticulocytes (Yo)
White blood cells ( loy/\)
Platelets (10911)
Family A BOY Mother
54 (110-160) 56 (133-139)
0.6 (11.4) 0.7 (6.6)
6.9 4.3
194 121
Family B Elder boy Younger boy
48 (80-105) 40 (90-100)
0.2 (4.8) 0.0 (2.1)
I .2 2.0
90 R5
Family C Elder girl Younger girl Mother
60 (10&110) 32 (7C- 80) 122
0.2 (4.1) 0.1 (2.7) 1.2
NI NI NI
NI NI NI
Nesred polymerase chain reaction (PCR) Sera were stored at -20°C until use. The DNA from serum was diluted 1:IOO and used for amplification by PCR. A sample known to contain B19 virus (kindly donated from Dr B. Cohen, London) was prepared in the same way as the samples and used as a positive control. Dilution buffer was used as negative control. DNA oligonucleotide primers were chosen from the gene encoding the structural proteins of parvovirus BIY. They were synthesized in a Gene Assembler I (Pharmacia) and purified as described by the manufacturer. Two primer pairs were used. The outer pair consisted of 5’ GGACTGTAGCAGATGAAGAG (390; nucleotides 2955-2974) and 5’ TATGGGACTGATGGTG (490; 3364-3349), the inner pair consisted of 5’ GGGTTTCAAGCACAAGTAG (190; 3002-3220) and 5‘ CCTTATAATGGTGCTCTGGG (290; 3291-3272). The numbers of the nucleotides are given according to Shade et al. (13). The outer primer set, 39W90, amplifies 410 bp and the inner set, 190-290, a 290 bp amplimer. A search in databases revealed no strong nucleotide homologies in this region. PCR was carried out in 2 steps, initiated with the outer primer pair and followed by the inner (i.e. nested) primer pair, and performed in 0.5 ml microfuge tubes in a total volume of 25 pl. The test system was optimised through titration of the Mg2+-and Na+-concentrations together with the proper primer concentration and annealing temperature. The final composition of the reaction mixture was 10 mM Tris-HCI pH 9.6, 2 mM MgCI,, 50 mM NaCI, 0.2 mM of each deoxynucleoside triphosphate (dNTP), 0.1 pM of each primer. 1U of thermus aquaticus (Taq) DNA polymerase (Perkin Elmer-Cetus, CA), 2 PI prediluted sample, and water added to a final volume of 25 pl. The tubes were covered with 25 PI paraffin oil to prevent evaporation. Two pI from the first amplified DNA was transferred to a new microfuge tube containing the same reaction mixture as the outer PCR system. The cycles, consisting of 30 s at 95”C, 60 s at 55°C for the outer and 63°C for the nested PCR, and 60 s at 7 2 T , were performed in an automated thermal cycler (DNA Thermal Cycler, Perkin Elmer-Cetus). The product from the second PCR (10 pl) was analysed by electrophoresis on a 1.8% agarose gel and stained with ethidium bromide. The diluted serum samples from the patients and prepared controls were transferred to the reaction tubes as the last step to avoid contamination of the buffers according to standard sterile procedures. No contamination occurred in more than 300 nested PCRs. The results were compared to findings obtained by dot blot hybridisation using a probe located partly in the same region as the PCR product (University College London, UCL, 14; Central Public Health Laboratory, London. CPHL, 15) and using digoxygenin with a coloured reaction product or 3’P-labelled probes with X-ray measurement as detection systems. For comparison 3 samples were evaluated with dot blot hybridisation, and 1 also in immune electron microscopy (IEM; 6).
PCR parvovirus in family outbreaks 277
Scand J Infect Dis 24
Table 11. Signs of infection and duration of symptoms in 7 members of 3 families with
hereditary sphaerocytic anaemia and aplastic crisis
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
No. of pats. Pallor Rash Pyrexia duration Vomiting Mild diarrhoea Headache Weakness Joint and muscle complaintsd
Duration (days)
2- 6
2- 9 1- 3 1- 2 3-10 1-10 6-14
“Females Anlibody detection assays Determination of antibodies to parvovirus was performed in 2 different ways: an immunosorbent assay using a synthetic peptide as antigen (amino acid sequence: FSPAASSCHNASGKEAKVCTISPI; SP assay. 16, 17), and also using native virus as antigen (NV assay, 18, 19).
Patients Family A Family A comprised a 12-year-old boy and his parents who had no known underlying chronic haemolytic disease. Following a week of pyrexia. headache. nausea and mild diarrhoea, the boy was admitted to hospital, He was pale and had generalised lymphadenopathy. Haematological findings are shown in Table 1. With a haemoglobin (Hb) level of 54 g/l and reticulocytopenia of 0.6%, transfusion of packed red cells was given and he recovered (Fdbk 111). On day 17. he developed a rash on his limbs during treatment with cotrimoxazole for pulmonary infiltrates. His mother, 43-year-old, wab admitted to hospital 4 days after the boy with the same clinical symptoms. Her H b level fell to 56 g/l 11 days after the onset of symptoms (Table I), necessitating blood transfusion which lead to an uneventful recovery. The boy and his mother were both suffering from previously undiagnosed hereditary sphaerocytosis which was diagnosed by a further haematologic investigation. The father remained healthy. This family has been described in a previous report (20). Funii1.v B Two bays aged 10 and 8 years, were known to have clinically overt haemolytic anaemia due to hereditary sphaerocytosis which did not require blood transfusion. Their 36-year-old father became splenectomised in his early teens and their 13-year-old sister was splenectomised when she was Y-year-old. Both the father and the sister remained asymptomatic, as did the mother. The elder boy became tired, with pyrexia, headache and nausea, and started vomiting 3 days after the onset of symptoms. On admission, his spleen was enlarged and generalised lymphadenopathy was noted. H e had depressed haematological values (Table I) necessitating transfusion, with a haemoglobin level of 48 g/l and reticulocytes of 0.2%. Half a year later splenectomy was carried out when cholestatic jaundice caused by gallstones developed. The younger boy presented 8 days after his brother’s first symptoms. H e had jaundice and generalised lymphadenopathy. Following blood transfusion with a Hb level of40 g/l and reticulocytes 0 (Table I), he rapidly recovered.
Farnily C The 31-year-old mother was splenectomised in early childhood due to hereditary sphaerocytosis causing severe haemolytic anaemia. Two of her children, 5- and 3-year-old girls, had clinically overt haemolytic disease. The younger girl needed blood transfusion 5 times during her first year of life. Her sister experienced a mild haemolytic crisis 1 year prior to the present event which did not require blood transfusion. The younger girl presented with fever and vomiting and complained of joint and muscle pains. She became tired and pale and was still pyrexic after a week. On admission, haematologic values (Table I)
278 E. Fridell et al.
Scand J Infect Dis 24
Table 111. Intrufumilial spread, individuul treatment and duration of symptoms in the members
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
of 3 ,families with hereditary sphaerocytosis and aplustic crisis Onset of symptoms (Days after the first symptom in the "index" child)
Admission to hospital
Family A Boy Mother
0 4
5 10
Family B Elder boy Younger boy
0 8
3, 4 , 9 3.6
7
3
Family C Younger girl Mother Elder girl
0 4 8
7
78
5
6
9 14 10
Blood transfusion
Duration of main symptoms (Days)
(Days after onset of symptoms)
3
I 7t
8 12
6
Table IV.Antibody responses and D N A detection in 3 families with parvovirus infection. NI = not investigated Days after onset
IgG (A405 values)
Family A
Boy Mother
Family B Elder boy
Younger boy
7 47 4 14 31 71
0.29 0.88 2.00
>TO0 >2.00 1.70 0.50
1.10 1.60 0.60 0.60
IgM
DNA detection (nested PCR)
Scnnd I Infect DI\ 21
PCR parvovirus
itz
family outbreiiks
were very lowg. With Hb 32 g/l and rcticulocytes 0.I"h she was given a transfusion of packed red blood ccllh (Table 111) and recovered rapidly. Four days later, the mother became ill with fever. and vomiting lasting for 2 days. She suffered from joint pains and stiffnes for 2 weeks. Transfusion was not required as her Hb level was stable. Thc onset of pyrexia occurred another 4 days later in the clder child. She also received transfusion of packed red blood cells, with a Hb level of 60 g/l and 0.2% reticulocytes. and recovered uneventfully.
RESULTS
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Signs of infection and duration of symptoms are summarised in Table 11.
D N A detection The PCR method was used with sera from all family members with symptoms. The first sample from all patients was positive using the PCR technique (Table IV). B19 DNA was found using PCR in the 2 dot blot positive samples and in 1 dot blot negative sample. Sera from the 3 boys in families A and B were examined with the dot blot digoxygenin hybridisation method. In the first serum sample from the boy in family A , B19 DNA could be detected as well as in the first sample of 1 boy in family B. In the latter sample the findings were supported by a positive result using IEM. The first serum sample from the younger boy in family B was tested, first with the digoxygenin method, then with a more sensitive technique using j2P-probes as detecting system, but the sample remained negative for B19
DNA. Syrrtlietic peptirle assay IgM and IgG responses were followed for up to 10 months in the families and are summarised in Table I V . The first sample from all patients except one was positive in the IgM SP assay.Seven samples from h patients were analyzed in the IgM native virus assay. IgM could not he detected in 2 samples from 2 patients on days 3 and 4 after onset of symptoms but they were present in all samples drawn on day 6 or later. In family A, the child had dctectable IgM using both SP and NV assays in the first sample and a late IgG rise. His mother had no detectable IgM using the SP assay. A slow IgG response was provoked, giving a low absorbance value 1 month later. A subsequent IgG absorbance rise was seen 2 months later. In family B, the first samples were taken very early and the second samples after more than 3 months. T h e boys in this family had IgM and a low IgG corresponding to a B19 virus infection using the SP assay, while IgM was not detectable in their first samples using the native virus assay. All members in family C had an IgM response in the first two samples using the SP assay and in samples tested with the NV assay. They rapidly produced IgG and had high absorbance values after 1@14 days which decreased after 2 months, but were still detectable after 10 months. DlSCUSSION We describe here 3 families, in which all members (unless previously splenectomised) with sphaerocytosis experienced aplastic crises due to transient arrest of erythropoiesis induced by human parvovirus B19 infection (2). 617 patients required blood transfusion. Four of them had clinically overt hacmolytic anaemia prior to this event. Two members in one family presented with an aplastic crisis disclosing the underlying chronic haemolytic disease. A
279
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
280 E. Fridell et al.
Scand J lnfect Dis 24
clinically mild human parvovirus B19 infection occurred in the remaining adult who remained well without haemolysis. These patients did not have erythema (1). Pallor was the common symptom in all the patients with parvovirus B19 infection. Pyrexia, headache, vomiting, and diarrhoea also occurred, currently accompanied in 3 females with joint and muscle pains of short duration, The intrafamilial spread and epidemiological data are summarised in Table 111. The index case in each family was a child, followed by a parent and/or another child. The period of incubation between the index case and the parent was short (4 days). The second child seemed to have an incubation period of 8 days. The average incubation period is generally 7-10 days from the first viraemic phase causing aplastic crisis. As expected, the second child infected in both families with siblings became sick after the time of ascribed transmission of parvovirus from the first infected child (Table 111). Contrariwise, 2 adults presented with symptoms 4 days after the index case. They were presumably infected simultaneously with the index case, but had a longer incubation period for reasons unknown. Alternatively, if infection from the index child had occurred, the incubation period in these adults must have been shorter. One could speculate, that the parents were reinfected, and their immune system, therefore, presented a more prompt antibody response, leading to a shorter incubation period. All patients had detectable D N A in the first sample collected at days 3-4 from the onset of symptoms using the nested PCR. The duration of the D N A persistence varied: in family A up to at least 47 days, while in family C for only 15 days. The duration of the virus infection in these patients shows that B19 virus is neutralised more slowly in some of them. It is obvious, that IgM coexists with the viral DNA. The mother and the elder girl in family C both had high IgG absorbance values and detectable B19 DNA. This might indicate that neither IgG nor IgM directed against the synthetic peptide are neutralising antibodies. The patients recovered and had no further symptoms even though B19 D N A was still present. The boy in family A , for example, had recovered after 11 days but still had detectable B19 D N A after 47 days. This shows that the presence of D N A in an asymptomatic patient not only correlates to infection in progress but might also indicate an infection 1-2 months ago. In order to establish diagnosis of B19 infection, antibody measurement and viral detection are useful. We have earlier described an EIA using a synthetic peptide as the antigen for B19 IgM (16), where IgM and IgG responses could be monitored for up to 10 months. All patients who developed IgM responses had IgM in their first samples collected 3-14 days after onset. The IgM reaction disappeared in samples collected on day 27 and later. One patient did not maintain the IgM response to the peptide used as antigen. This might indicate, that the IgM responses to the peptide are directed to different epitopes than in the NV assay, which did not detect IgM in the first sample from the younger boy in family B. The SP assay seems to be able to detect an early response. However, IgM directed to the synthetic peptide cannot be detected for as long a time as IgM directed to the whole virus, which persists for up to 4 months (17). IgG detected by the SP assay showed rises in all 5 patients where the first sample was collected during the first 2 weeks after onset of symptoms and the second sample within 2-3 months. One patient did not produce IgM and developed a very low amount of specific IgG during the first month. The reason for her late antibody response is uncertain, but in samples from this patient the B19 DNA could b e spotted for an unusually long period. This patient might have produced non-neutralising antibodies early in the disease. Low IgG levels to the peptide could be detected in the first samples from 3 boys. IgG values in the SP assay rose for 11-71 days and then decreased in all patients in samples collected later than 6 months after the onset of symptoms. In 2 patients the absorbance was
Scand J
Infect Dis 24
PCR parvovirus in family outbreaks 28 I
very low after 8-9 months showing that thc SP assay might be less sensitive for detection of a previous infection. There are differences in the antibodies found by the S P and NV assays. In the former, a 24 amino acid long peptide derived from a region shown to provoke early IgG is used as antigen (20). In the later the native virus from human sera with a less defined antigenic region i s used.
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
Conclioions Family outbreaks of human parvovirus B19 infection and its complications in patients with haemolytic anaemia can be monitored by a peptide assay for the detection of specific antibodies, and the P C R technique for the detection of the viral DNA. The nested PCR is a very sensitive method for detection of B19 virus DNA and the method is more sensitive than the dot blot hybridisation test. Provided it is handled properly contamination can be avoided. Nested PCR i s therefore the method of choice in reference laboratories for diagnosis of B 19 infection in patients with aplastic crisis. The SP assay is a good complement to the NV assay. However, the latter is limited by a poor supply of B19 virus to use as antigen. Synthetic antigen is available and can form a part of a fast and sensitive assay for detection of antibodies directed to human parvovirus B19.
REFERENCES 1. Anderson MJ, Lewis E , Kidd l M , Hall SM. Cohen BJ. An outbreak of erythema infectiosum
associated with human parvovirus infection. J Hyg 93: 85-93, 1984. 2. Lefrere J-J, Courouce A-M. Girot R , Bertrand Y, Soulier JP. Six cases of hereditary spherocytosis revcalcd by human parvovirus infection. Br J Haematol 62: 653458, 1986. 3. Serjeant GR, Topley JM. Mason K , Serjeant BE, Pattison JR, Jones SE, Mohammed R. Outbreak of aplastic crises in sickle cell anaemia associated with parvovirus-like agent. Lancet 2: 595-597, 1981. 4. Young N. Haematologic and haematopoietic consequences of B19 parvovirus infection. Semin Hematol 25: 159-172, 1988. 5. Bowman CA. Cohen BJ, Norfolk DR. Lacey CJN. Red cell aplasia associated with human parvovirus B 19 and HIV infection: failure to respond clinically to intravenous immunoglobulin. AIDS 4: 1038-1039, 1990. 6. Field AM, Cohen BJ, Brown KE, Mori J . Clewley JP, Nascimento JB, Hallam NF. Detection of B19 parvovirus in human fetal tissues by electron microscopy. J Med Virol 35: 85-95, 1991. 7. Graeve JLA, de Alarcon PA, Naides SJ. Parvovirus B19 infection in patients receiving cancer chemotherapy: The expanding spectrum of disease. Am J Pediatr Hematol Oncol 11: 441-444. 1989. 8. Kurtzman G J , Ozawa K , Cohen BJ, Hanson G, Oseas R, Young NS. Chronic bone marrow failure due to persistent B19 parvovirus infection. N Engl J Med 317: 287-294, 1987. 9. Malarme M, Vandervelde D , Brasseur M. Parvovirus infection. leukaemia, and immunodeficiency. Lancet 1: 1457, 1989. 10. Pillay D, Patou G, Griffiths P. Rees L. Secondary parvovirus infection in an immunocompromised child. Pediatr Infect Dis J 10: 623-624. 1991. 11. Pattison JK. Jones SE. Hodgson J . Davis LR. White JM. Stroud CE, Murtaxa L. Parvovirus infections and hypoplastic crisis in sickle- cell anaemia. Lancet 1: 664-665, 1981. 12. Kurtzman GJ. Cohen BJ. Meyers P, Amunullah A , Young NS. Persistent B19 parvovirus infection as a cause of severe chronic anaemia in children with acute lymphocytic leukaemia. Lancet 2: 1159-1162, 1988. 13. Shade RO. Blundell MC, Cotmore SF, Tattcrsall P, Astell C. Nucleotide scqucncc and genome organization of human parvovii-us B19 isolated from the serum of a child during aplastic crisis. .I Virol 58: 921-936, 1986. 14. Anderson MJ. Jones SE. Minson AC. Diagnosis of human parvovirus infection by dot-blot hybridization using cloned viral DNA. J Med Virol 15: 163-172, 1985.
Scand J Infect Dis Downloaded from informahealthcare.com by Mcgill University on 01/06/15 For personal use only.
282 E. Fridell et al.
Scand J lnfect Dis 24
15. Clewley JP. Detection of human parvovirus using a molecularly cloned probe. J Med Viral 15: 173-181, 1985. 16. Fridell E, Trojnar J , Wahrcn B. A new peptide for human parvovirus B19 antibody detection. Scand J Infect Dis 21: 597-603. 1989. 17. Fridell E, Cohen BJ, Wahren B. Evaluation of a synthetic-peptide enzyme-linked immunosorbcnt assay for immunoglobulin M to human parvovirus BIY. J Clin Microbiol 29: 13761381, 1991. 18. Anderson MJ. Davis LR, Jones SE, Pattison JR, Scrjeant GR. The development and use of an antibody capture radioimmunoassay for specific IgM to a human parvovirus-like agent. J Hyg 88: 309-324. 1982. 19. Cohen BJ. Mortimer PP. Pereira MS. Diagnostic assays with monoclonal antibodies for the human serum parvovirus-like virus (SPLV). J Hyg 91: 113-130, 1083. 20. Eriksson B-M, Striimberg A, Kreuger A. Human parvovirus B19 infection with severe anemia affecting mother and son. Scand J Infect Dis 20: 335-337, 1988. 21. Kurtzman G, Frickcnhofen N , Kimball J , Jenkins DW, Nienhuis AW, Young NS. Pure red-cell aplasia of 10 years duration due to persistent parvovirus B19 infection and its cure with immunoglobulin therapy. N Engl J Med 321: 519-523, 1989.
