Classical swine fever virus detection

Abstract

The current study reports on a real-time reverse transcription polymerase chain reaction (real-time RT-PCR) ring trial for the detection of Classical swine fever virus (CSFV) genomic RNA undertaken by 10 European laboratories. All laboratories were asked to use their routine in-house real-time RT-PCR protocols and a standardized protocol commonly used by the Friedrich-Loeffler-Institute (FLI) on a panel of well-characterized samples. In general, all participants produced results within the acceptable range. The FLI assay, several in-house assays, and the commercial kits had high analytical sensitivity and specificity values. Nevertheless, some in-house systems had unspecific reactions or suboptimal sensitivity with only a single CSFV genotype. Follow-up actions involved either improvement of suboptimal assays or replacement of specific laboratory assays with the FLI protocol, with or without modifications. In conclusion, the ring trial showed reliability of classical swine fever diagnosis on an international level and helped to optimize CSFV-specific RT-PCR diagnostics.

Keywords

polymerase chain reaction ring trial

Due to its economic impact, classical swine fever (CSF) ranks among the most important diseases of domestic pigs. Clinical and pathological signs are highly variable, and diagnosis must be confirmed by laboratory tests.^5,16 During 2000–2010, the polymerase chain reaction (PCR) technique has become more and more important for routine CSF diagnosis,^1,3,4,13,19 and reverse transcription (RT)-PCR has been accepted by the European Union as an official method for CSF confirmation (Commission Decision 2002/106/EC). Meanwhile, various real-time RT-PCR methods for diagnosing CSF have been described,^* and the first commercial real-time RT-PCR kits are already available.^3,11 Beside the published protocols, a variety of in-house protocols are also used. To guarantee consistent quality and diagnostic reliability, assay protocol optimization, validation, and standardization is needed. In this context, ring trials can help improve protocol standardization and harmonization. For CSF, such ring trials have been performed in the past.^18,19

References 2–4, 6–9, 12, 17, 20, 21, 24

The current study reports on the results of a real-time RT-PCR ring trial for the detection of Classical swine fever virus (CSFV; family Flaviviridae, genus Pestivirus) genomic RNA among 10 European laboratories that are routinely involved in CSF diagnosis. Organized by the Friedrich-Loeffler-Institut (FLI; Riems Island, Germany), the trial was based on well-characterized samples evaluated with 1 standard method by all participating laboratories⁷ in comparison to different routine in-house protocols (published and unpublished). The aim of the ring trial was to evaluate the reliability of CSF real-time RT-PCR diagnosis on an international level, and to optimize CSFV-specific RT-PCR diagnostics.

To this means, a test panel was sent out comprising 48 manually extracted^a viral RNA samples in RNA-safe buffer (50 ng/µl of carrier polyA-RNA,^b 0.05% Tween-20,^c 0.05% sodium azide^d in RNase-free water) with defined copy numbers including 2 negative controls. Viral RNA from CSFV strains of all available genotypes as well as from related pestiviruses were integrated into the test panel to assess analytical specificity in terms of inclusivity and exclusivity. Furthermore, 3 dilution series of CSFV RNA were used to define differences in the analytical sensitivity. Dilution series were prepared from CSFV strains Alfort (genotype 1.1), Spante (genotype 2.3), and Kanagawa (genotype 3.4), respectively. For the standardization of the ring trial, the protocol commonly used by the FLI⁷ was included (hereafter, FLI protocol). The reagents for this assay were transferred together with the viral RNA.

Participating laboratories were asked to use their own routine real-time RT-PCR protocols for CSFV detection (hereafter, in-house protocols, not considering whether such protocols were published, unpublished, or commercially available). In total, 13 additional assays were used, of which only 3 were unpublished. Most protocols were CSFV specific,^{7,11,12,14,15,20} but 3 protocols (1 unpublished) were Pestivirus specific.^22,23 In detail, 3 laboratories used a previously published assay,^14,15 while 1 laboratory used a reference method⁷ with in-house modifications, and another laboratory a different assay.²⁰ Two laboratories used a commercial real-time RT-PCR kit.^e One of the laboratories combined the commercial assay with an additional commercial kit.^f Each of the Pestivirus-specific protocols was used by 1 laboratory.

Most laboratories carried out 2 runs for each method to obtain a final result. For the purpose of the current ring trial, evaluation of results was done on the basis of mean threshold cycle (Ct) values. In cases where only 1 run was positive or Ct values were higher than the internal cut-off of the respective laboratory, results were reported as inconclusive. The same applied for samples with increasing fluorescent values that did not cross the threshold. Results of the organizing laboratory were only used for comparison and were not included in the comparative evaluation.

Results indicated that CSFV samples representing different genotypes were reliably detected in all participating laboratories that used the FLI protocol (Table 1, samples 1–13). Using the in-house protocols, only minor problems were noticed in this section. One laboratory (D) reported 4 inconclusive results, and 1 test system did not detect 1 of the group 3 CSFV strains in 2 laboratories (I1 and J1). While laboratory I1 failed to detect CSFV strain Kanagawa, laboratory J1 did not detect CSFV strain Congenital tremor. Both commercial kits (laboratories B and G) detected all strains included in the ring trial. Detailed results are presented in Table 2 (samples 1–13).

Table 1.

Classical swine fever virus detection: results of mean threshold cycle values obtained using the protocol routinely used by the Friedrich-Loeffler-Institut as reference method.*

				Participant laboratories‡
Sample	Species	Strain	Quantity†	FLI	A	B	C	D	E	F	G	H	I	J
Analytical specificity—inclusivity
1	CSFV, gt 1.1	C-Stamm	1 × 10⁴	26	25	27	27	27	25	26	26	24	26	25
2	CSFV, gt 1.1	Eystrup91	1 × 10⁴	29	29	29	30	29	28	29	29	26	29	27
3	CSFV, gt 1.1	Alfort187	1 × 10⁴	28	28	29	29	29	27	27	28	26	29	26
4	CSFV, gt 1.1	Koslov1128	1 × 10⁴	27	27	28	29	28	27	27	27	25	28	26
5	CSFV, gt 1.2	Brescia	1 × 10⁴	27	27	28	28	28	26	27	27	25	28	26
6	CSFV, gt 2.1	Schweiz II	1 × 10⁴	27	27	28	29	28	26	29	28	27	26	27
7	CSFV, gt 2.1	Pader	1 × 10⁴	27	27	27	28	27	26	28	27	27	26	26
8	CSFV, gt 2.2	Bergen	1 × 10⁴	27	28	27	29	28	26	31	27	27	26	25
9	CSFV, gt 2.2	D4886/82/Ro	1 × 10⁴	27	27	27	28	28	26	28	27	27	26	24
10	CSFV, gt 2.3	Uelzen	1 × 10⁴	25	26	26	27	27	25	26	26	25	24	23
11	CSFV, gt 2.3	Spante	1 × 10⁴	25	25	26	26	26	24	25	25	23	25	23
12	CSFV, gt 3.1	Congenital tremor	1 × 10⁴	26	25	26	26	25	24	25	25	23	25	24
13	CSFV, gt 3.4	Kanagawa	1 × 10⁴	29	29	30	30	30	29	29	29	27	30	28
Analytical specificity—exclusivity
14	BDV, gt 3	Gifhorn	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
15	BDV, gt 2	137/4	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
16	BDV, gt 1	Rudolph	1 × 10⁴	No Ct	39	No Ct	No Ct	No Ct	No Ct	IC§	No Ct	No Ct	No Ct	No Ct
17	BDV, gt 4	Isard	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
18	BVDV-1	cp7	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
19	BVDV-1	NADL	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
20	BVDV-1	Grub	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	38	No Ct	No Ct	No Ct	No Ct	No Ct
21	BVDV-2	München2	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
22	BVDV-2	CS8644	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
23	BVDV-2	Bure	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	IC¦	No Ct	No Ct	No Ct	No Ct
24	Atypical pestivirus	Hobi	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
25	Atypical pestivirus	Giraffe H138	1 × 10³	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	IC¦	No Ct	No Ct	No Ct	No Ct
Analytical sensitivity
26	CSFV, gt 1.1	Alfort187	4 × 10³	29	30	29	30	30	28	30	30	29	28	28
27	CSFV, gt 1.1	Alfort187	8 × 10²	32	35	31	33	33	31	33	34	32	30	29
28	CSFV, gt 1.1	Alfort187	1.6 × 10²	34	36	34	35	34	33	35	34	34	33	31
29	CSFV, gt 1.1	Alfort187	3.2 × 10¹	36	38	37	37	37	35	37	37	37	34	33
30	CSFV, gt 1.1	Alfort187	6.4 × 10⁰	39	40	39	39	39	37	IC	42	No Ct	38	34
31	CSFV, gt 1.1	Alfort187	1.28 × 10⁰	No Ct	No Ct	No Ct	No Ct	IC	No Ct	IC¦	No Ct	No Ct	No Ct	No Ct
32	CSFV, gt 1.1	Alfort187	2.56 × 10⁻¹	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	IC¦	No Ct	No Ct	No Ct	No Ct
33	CSFV, gt 2.3	Spante	4 × 10³	27	26	27	28	27	26	26	26	24	27	25
34	CSFV, gt 2.3	Spante	8 × 10²	29	29	29	30	30	28	ND	29	26	30	28
35	CSFV, gt 2.3	Spante	1.6 × 10²	31	32	32	33	32	31	31	31	29	32	30
36	CSFV, gt 2.3	Spante	3.2 × 10¹	33	35	33	35	34	32	35	34	33	32	32
37	CSFV, gt 2.3	Spante	6.4 × 10⁰	36	37	37	37	39	35	37	35	36	34	33
38	CSFV, gt 2.3	Spante	1.28 × 10⁰	No Ct	40	No Ct	39	IC	36	39	38	38	36	35
39	CSFV, gt 2.3	Spante	2.56 × 10⁻¹	No Ct	No Ct	No Ct	No Ct	41	No Ct	IC¦	No Ct	39	37	No Ct
40	CSFV, gt 3.4	Kanagawa	4 × 10³	30	32	30	32	32	29	31	31	30	29	28
41	CSFV, gt 3.4	Kanagawa	8 × 10²	33	33	33	34	34	32	32	33	31	32	31
42	CSFV, gt 3.4	Kanagawa	1.6 × 10²	35	36	36	37	36	34	35	35	35	36	34
43	CSFV, gt 3.4	Kanagawa	3.2 × 10¹	37	No Ct	38	38	39	37	37	39	37	38	35
44	CSFV, gt 3.4	Kanagawa	6.4 × 10⁰	39	No Ct	No Ct	40	No Ct	37	38	No Ct	No Ct	42	No Ct
45	CSFV, gt 3.4	Kanagawa	1.28 × 10⁰	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	41	No Ct	No Ct	No Ct
46	CSFV, gt 3.4	Kanagawa	2.56 × 10⁻¹	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
47	Water			No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
48	RSB50			No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct

FLI = Friedrich-Loeffler-Institut; CSFV = Classical swine fever virus; BVDV = Bovine viral diarrhea virus; BDV = Border disease virus; IC = inconclusive; gt = genotype; ND = not done; Ct = threshold cycle.

†

Quantity refers to the approximate copy number per µl working solution.

‡

The values obtained by the organizer (FLI) are displayed in bold.

Inconclusive based on curve shape.

Inconclusive result with one of the duplicates showing high Ct values, the other none.

Table 2.

Classical swine fever virus detection: results of mean threshold cycle values obtained using the in-house methods.*

				Participant laboratories‡
Sample	Species	Strain	Quantity†	FLI	A¹⁴	B¹²	C⁷	D^14,15	E²⁰	F	G1	G2	H^§	I1¹⁵	I2²³§	J1	J2²²§
Analytical specificity—inclusivity
1	CSFV, gt 1.1	C-Stamm	1 × 10⁴	26	33	26	26	19	28	29	25	25	27	32	24	26	26
2	CSFV, gt 1.1	Eystrup91	1 × 10⁴	29	34	29	30	19	30	29	27	27	29	36	27	29	25
3	CSFV, gt 1.1	Alfort187	1 × 10⁴	28	34	29	29	20	29	29	27	28	29	34	27	29	24
4	CSFV, gt 1.1	Koslov1128	1 × 10⁴	27	33	29	28	21	30	28	27	28	28	34	27	27	24
5	CSFV, gt 1.2	Brescia	1 × 10⁴	27	33	28	28	IC	29	29	27	27	27	33	26	29	26
6	CSFV, gt 2.1	Schweiz II	1 × 10⁴	27	32	29	28	19	29	29	28	28	28	32	25	29	27
7	CSFV, gt 2.1	Pader	1 × 10⁴	27	32	28	27	19	30	29	28	28	28	32	25	29	26
8	CSFV, gt 2.2	Bergen	1 × 10⁴	27	30	28	28	18	29	28	28	28	26	32	25	29	26
9	CSFV, gt 2.2	D4886/82/Ro	1 × 10⁴	27	31	29	28	IC	29	29	28	28	27	31	25	28	26
10	CSFV, gt 2.3	Uelzen	1 × 10⁴	25	29	26	26	IC	28	23	26	27	24	29	24	29	24
11	CSFV, gt 2.3	Spante	1 × 10⁴	25	30	26	26	16	27	23	26	27	26	31	25	28	24
12	CSFV, gt 3.1	Congenital tremor	1 × 10⁴	26	29	25	27	18	33	21	25	25	26	30	34	No Ct	33
13	CSFV, gt 3.4	Kanagawa	1 × 10⁴	29	46	35	30	IC	30	29	32	30	31	No Ct	28	31	25
Analytical specificity—exclusivity
14	BDV, gt 3	Gifhorn	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	IC	No Ct	No Ct	No Ct	30	No Ct	34	32	No Ct
15	BDV, gt 2	137/4	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	IC	No Ct	No Ct	No Ct	28	No Ct	34	31	No Ct
16	BDV, gt 1	Rudolph	1 × 10⁴	No Ct	No Ct	IC	No Ct	No Ct	IC	No Ct	No Ct	No Ct	26	No Ct	29	31	27
17	BDV, gt 4	Isard	1 × 10⁴	No Ct	No Ct	42	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	24	No Ct	25	IC	33
18	BVDV-1	cp7	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	25	No Ct	23	No Ct	26
19	BVDV-1	NADL	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	25	No Ct	25	No Ct	24
20	BVDV-1	Grub	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	25	No Ct	24	No Ct	26
21	BVDV-2	München2	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	24	No Ct	No Ct
22	BVDV-2	CS8644	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	24	29	25
23	BVDV-2	Bure	1 × 10⁴	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	26	No Ct	24
24	Atypical pestivirus	Hobi	1 × 10⁴	No Ct	No Ct	IC	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	33	No Ct	No Ct	34	No Ct
25	Atypical pestivirus	Giraffe H138	1 × 10³	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	39	39	30	No Ct	23	No Ct	25
Analytical sensitivity
26	CSFV, gt 1.1	Alfort187	4 × 10³	29	34	30	30	21	31	30	28	29	28	35	26	30	27
27	CSFV, gt 1.1	Alfort187	8 × 10²	32	37	32	32	25	33	31	31	31	30	36	28	33	30
28	CSFV, gt 1.1	Alfort187	1.6 × 10²	34	40	34	35	IC	36	32	33	34	32	39	31	32	31
29	CSFV, gt 1.1	Alfort187	3.2 × 10¹	36	42	37	38	No Ct	38	32	36	36	36	No Ct	34	36	33
30	CSFV, gt 1.1	Alfort187	6.4 × 10⁰	39	41	38	38	IC	40	33	38	40	No Ct	No Ct	37	35	34
31	CSFV, gt 1.1	Alfort187	1.28 × 10⁰	No Ct	No Ct	No Ct	No Ct	IC	No Ct	IC^¦	39	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
32	CSFV, gt 1.1	Alfort187	2.56 × 10⁻¹	No Ct	No Ct	IC	No Ct	No Ct	No Ct	No Ct	40	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
33	CSFV, gt 2.3	Spante	4 × 10³	27	32	28	27	17	29	28	27	28	28	32	27	29	26
34	CSFV, gt 2.3	Spante	8 × 10²	29	34	30	30	20	31	29	30	30	30	35	29	31	29
35	CSFV, gt 2.3	Spante	1.6 × 10²	31	37	32	32	22	34	30	32	33	34	37	32	33	32
36	CSFV, gt 2.3	Spante	3.2 × 10¹	33	41	35	35	25	36	31	34	35	33	No Ct	33	IC	34
37	CSFV, gt 2.3	Spante	6.4 × 10⁰	36	41	36	37	No Ct	38	32	36	37	IC	40	35	36	33
38	CSFV, gt 2.3	Spante	1.28 × 10⁰	No Ct	No Ct	39	No Ct	No Ct	No Ct	32	39	38	No Ct	No Ct	No Ct	IC	No Ct
39	CSFV, gt 2.3	Spante	2.56 × 10⁻¹	No Ct	No Ct	IC	IC	No Ct	IC	33	40	No Ct	IC	No Ct	No Ct	35	33
40	CSFV, gt 3.4	Kanagawa	4 × 10³	30	No Ct	35	31	IC	31	31	34	31	30	39	26	30	27
41	CSFV, gt 3.4	Kanagawa	8 × 10²	33	No Ct	39	33	IC	34	30	35	33	42	No Ct	30	34	No Ct
42	CSFV, gt 3.4	Kanagawa	1.6 × 10²	35	No Ct	43	36	No Ct	27	31	37	36	No Ct	No Ct	30	39	27
43	CSFV, gt 3.4	Kanagawa	3.2 × 10¹	37	No Ct	IC	38	No Ct	39	32	39	37	No Ct	No Ct	36	38	29
44	CSFV, gt 3.4	Kanagawa	6.4 × 10⁰	39	No Ct	No Ct	38	No Ct	No Ct	IC^¦	44	No Ct	No Ct	No Ct	No Ct	38	No Ct
45	CSFV, gt 3.4	Kanagawa	1.28 × 10⁰	No Ct	No Ct	No Ct	40	No Ct	No Ct	No Ct	43	No Ct	No Ct	No Ct	No Ct	39	No Ct
46	CSFV, gt 3.4	Kanagawa	2.56 × 10⁻¹	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	IC^¦	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct
47	Water			No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	39	No Ct
48	RSB50			No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	No Ct	IC	No Ct

†

Quantity refers to the approximate copy number per µl working solution.

‡

Coded participants with superscript figures refer to the reference list for all published protocols. The reference values obtained by the organizer FLI are displayed in bold.

Pestivirus-specific assays.

Inconclusive result with one of the duplicates showing high Ct values, the other none but a curve below cut-off.

In general, neither Border disease virus (BDV) nor Bovine viral diarrhea virus (BVDV), or other related pestiviruses, caused diagnostic problems using the FLI protocol. Only 2 laboratories (A and E) obtained a high Ct value in 1 sample each. Laboratory A reported high Ct values for BDV strain Rudolph, whereas laboratory E had a weak positive result for BVDV strain Grub (Table 1, samples 14–25). The results were not confirmed by the in-house methods employed by laboratories A and E.

One CSFV-specific in-house protocol had problems with 3 out of 4 BDV isolates as well as with BVDV strain CS8644 and Pestivirus Hobi (laboratory J1). In addition, some inconclusive or false-positive results with rather high Ct values occurred in 1 laboratory with 1 commercial test kit (B) and 1 in-house protocol (E). Not all pestiviruses were detected by the Pestivirus-specific protocols (laboratories H, I2, J2). Details can be found in Table 2.

Using the FLI protocol, all laboratories were able to detect the CSFV strain Alfort dilution with 3.2 × 10¹ copies/µl working solution. Eight out of 10 participating laboratories also detected dilutions containing 6.4 × 10⁰ copies/µl; an additional laboratory scored this dilution as inconclusive. Furthermore, all laboratories were able to detect the dilution 6.4 × 10⁰ copies/µl in the CSFV strain Spante dilution series. Eight out of 10 participating laboratories also scored the 1.28 × 10⁰ copies/µl dilution positive. Three positive results were reported for the highest dilution of 2.56 × 10⁻¹ copies/µl working solution. With the CSFV strain Kanagawa dilution series, all laboratories were able to detect the dilution containing 1.6 × 10² copies/µl using the reference method.⁸ All but 1 laboratory also detected the dilution with 3.2 × 10¹ copies/µl, and 4 laboratories were able to detect the dilution containing 6.4 × 10⁰ copies/µl. One positive result with a very high Ct value occurred in the dilution containing 1.28 × 10⁰ copies/µl after a negative result for the previous dilution. Details are given in Table 1 (samples 26–46).

Most in-house systems did not have problems with the CSFV strain Alfort dilution series. All laboratories detected the second dilution containing 8 × 10² copies/µl with their in-house systems. The third dilution, containing 1.6 × 10² copies/µl, was detected by all but 1 assay. The fourth dilution (3.2 × 10¹ copies/µl) was detected by 11 out of 13 assays, and 10 assays also detected the fifth dilution containing 6.4 × 10⁰ copies/µl. One assay scored all dilutions positive.

The CSFV strain Spante dilution series had similar results. All assays successfully detected the dilution containing 1.6 × 10² copies/µl, and 11 out of 13 had positive results in the next dilution step as well (3.2 × 10¹ copies/µl). The dilution with 6.4 × 10⁰ copies/µl was detected in 11 assays again, but 2 results were inconclusive due to negative results in the previous dilution and unexplainable Ct values. The sixth dilution step (1.28 × 10⁰ copies/µl) was detected with 4 assays. Four laboratories recorded high Ct values for the highest dilution step containing 2.56 × 10⁻¹ copies/µl. A comparative analysis can be found in Table 2.

Several in-house assays had problems with the CSFV strain Kanagawa dilution series (Table 2). Two assays did not detect any of the dilutions, and another assay picked up only the first dilution containing 4 × 10³ copies/µl. Nevertheless, 9 assays detected the third (1.6 × 10² copies/µl), 8 assays the fourth (3.2 × 10¹ copies/µl), 3 assays the fifth (6.4 × 10⁰ copies/µl), and 3 assays the sixth (1.28 × 10⁰ copies/µl). Some positive results in higher dilution steps were not accompanied by positive results in the previous dilution steps, or Ct values did not correspond with further dilution. Only 1 assay had problems with both negative controls.

Summarizing the results of the current ring trial, all participants produced results within the acceptable range, and the results indicate that the real-time RT-PCR assay is a rapid and extremely sensitive tool for CSFV detection. Combined with the possibility of using RNA extraction robots and real-time PCR machines with a 96- or 384-sample platform, high throughput, which is required during a crisis or outbreak situation, seems feasible.

In detail, the FLI protocol and several in-house assays (including the commercial kits) had high analytical sensitivity and specificity; however, some in-house systems also amplified related pestivirus RNA (unspecific reactions) or had suboptimal sensitivity with single CSFV genotypes (low sensitivity). The main problems were encountered with CSFV strain Kanagawa, which belongs to genotype 3.4 and is used as an out-group strain in phylogenetic analyses due to its variation from other CSFV strains. In order to understand the apparent failure or inconsistent results of some protocols, further investigations regarding primer and probe sequences as well as other laboratory variables, such as machines, sample handling, enzymes, and human factors, are needed. In general, the ring trial sample panel as well as the experiences with the real-time RT-PCR assays tested in the current ring trial can be used for further development and validation of molecular diagnostic assays for the identification of CSFV and other pestiviruses.¹⁰

In the aftermath of the current ring trial, laboratories with a suboptimal protocol either improved their methods or replaced them with the highly sensitive and specific FLI method, with or without in-house modifications or use of commercial kits. In addition, the current study found that CSFV RNA is very stable in RNA-safe buffer. In conclusion, the current ring trial helped to optimize CSFV-specific real-time RT-PCR diagnostics within the European Union and confirmed that the real-time RT-PCR assay is a reliable and robust tool for CSF diagnosis. Based on the presented findings, minimum standards for diagnostic real-time RT-PCR assays for CSF diagnosis can be discussed in the future.

Footnotes

Acknowledgements

The authors thank all technical assistants involved in assembly of aliquots, standardization of samples, packaging, and shipping. The authors Bernd Hoffmann and Sandra Blome contributed equally.

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

a.

QIAamp® Viral RNA Kit, Qiagen GmbH, Hilden, Germany.

b.

RNA homopolymer, Amersham Biosciences Europe GmbH, Freiburg, Germany.

c.

Tween 20, Sigma-Aldrich Chemie GmbH, Munich, Germany.

d.

Sodium azide, Sigma-Aldrich Chemie GmbH, Munich, Germany.

e.

TaqVet PPC, Laboratoire Service International, Lissieu, France.

f.

Adiavet, Adiagene, Saint-Brieuc, France.

The Network of Excellence for Epizootic Disease Diagnosis and Control or EPIZONE (contract no. FOOD-CT-2006-016236) supported the current work.

References

Cheng

Zhao

.: 2008, Simultaneous detection of Classical swine fever virus and North American genotype Porcine reproductive and respiratory syndrome virus using a duplex real-time RT-PCR. J Virol Methods 151:194–199.

Depner

Bunzenthal

Heun-Münch

.: 2006, Diagnostic evaluation of a real-time RT-PCR assay for routine diagnosis of classical swine fever in wild boar. J Vet Med B Infect Dis Vet Public Health 53:317–320.

Depner

Hoffmann

Beer

: 2007, Evaluation of real-time RT-PCR assay for the routine intra vitam diagnosis of classical swine fever. Vet Microbiol 121:338–343.

Dewulf

Koenen

Mintiens

.: 2004, Analytical performance of several classical swine fever laboratory diagnostic techniques on live animals for detection of infection. J Virol Methods 119:137–143.

Floegel-Niesmann

Bunzenthal

Fischer

Moennig

: 2003, Virulence of recent and former classical swine fever virus isolates evaluated by their clinical and pathological signs. J Vet Med B Infect Dis Vet Public Health 50:214–220.

Gaede

Reiting

Schirrmeier

.: 2005, [Detection and species-specific differentiation of pestiviruses using real-time RT-PCR]. Berl Munch Tierarztl Wochenschr 118:113–120. In German. Abstract in English.

Hoffmann

Beer

Schelp

.: 2005, Validation of a real-time RT-PCR assay for sensitive and specific detection of classical swine fever. J Virol Methods 130:36–44.

Hoffmann

Depner

Schirrmeier

Beer

: 2006, A universal heterologous internal control system for duplex real-time RT-PCR assays used in a detection system for pestiviruses. J Virol Methods 136:200–209.

Jamnikar Ciglenecki

Grom

Toplak

.: 2008, Real-time RT-PCR assay for rapid and specific detection of classical swine fever virus: comparison of SYBR Green and TaqMan MGB detection methods using novel MGB probes. J Virol Methods 147:257–264.

10.

LeBlanc

Leijon

Jobs

.: 2010, A novel combination of TaqMan RT-PCR and a suspension microarray assay for the detection and species identification of pestiviruses. Vet Microbiol 142:81–86.

11.

Le Dimna

Vrancken

Koenen

.: 2008, Validation of two commercial real-time RT-PCR kits for rapid and specific diagnosis of classical swine fever virus. J Virol Methods 147:136–142.

12.

Le Potier

Le Dimna

Kuntz-Simon

.: 2006, Validation of a real-time RT-PCR assay for rapid and specific diagnosis of classical swine fever virus. Dev Biol (Basel) 126:179–186; discusssion 326–327.

13.

Mackay

Arden

Nitsche

: 2002, Real-time PCR in virology. Nucleic Acids Res 30:1292–1305.

14.

McGoldrick

Bensaude

Ibata

.: 1999, Closed one-tube reverse transcription nested polymerase chain reaction for the detection of pestiviral RNA with fluorescent probes. J Virol Methods 79:85–95.

15.

McGoldrick

Lowings

Ibata

.: 1998, A novel approach to the detection of classical swine fever virus by RT-PCR with a fluorogenic probe (TaqMan). J Virol Methods 72:125–135.

16.

Moennig

Floegel-Niesmann

Greiser-Wilke

: 2003, Clinical signs and epidemiology of classical swine fever: a review of new knowledge. Vet J 165:11–20.

17.

Ophuis

Morrissy

Boyle

: 2006, Detection and quantitative pathogenesis study of classical swine fever virus using a real time RT-PCR assay. J Virol Methods 131:78–85.

18.

Paton

McGoldrick

Belak

.: 2000, Classical swine fever virus: a ring test to evaluate RT-PCR detection methods. Vet Microbiol 73:159–174.

19.

Paton

McGoldrick

Bensaude

.: 2000, Classical swine fever virus: a second ring test to evaluate RT-PCR detection methods. Vet Microbiol 77:71–81.

20.

Rasmussen

Uttenthal

Reimann

.: 2007, Virulence, immunogenicity and vaccine properties of a novel chimeric pestivirus. J Gen Virol 88:481–486.

21.

Risatti

Holinka

.: 2005, Diagnostic evaluation of a real-time reverse transcriptase PCR assay for detection of classical swine fever virus. J Clin Microbiol 43:468–471.

22.

Uttenthal

Storgaard

Oleksiewicz

de Stricker

: 2003, Experimental infection with the Paderborn isolate of classical swine fever virus in 10-week-old pigs: determination of viral replication kinetics by quantitative RT-PCR, virus isolation and antigen ELISA. Vet Microbiol 92:197–212.

23.

Vilcek

Herring

.: 1994, Pestiviruses isolated from pigs, cattle and sheep can be allocated into at least three genogroups using polymerase chain reaction and restriction endonuclease analysis. Arch Virol 136:309–323.

24.

Zhao

Cheng

.: 2008, Evaluation of a multiplex real-time RT-PCR for quantitative and differential detection of wild-type viruses and C-strain vaccine of Classical swine fever virus. Vet Microbiol 126:1–10.