Antibody titres in FMD type A strains: comparison of ......IAEA/FAO r1 Values Comparison r1-values VNT (black), NIT (dark grey) and LPBE (light grey) Huge difference between neutralising

IAEA/FAOIAEA/FAO

Antibody titres in FMD type A strains:

comparison of methodologies to predict

cross-protection

Tesfaalem T.Sebhatu, Klaas Weerdmeester, Froukje van Hemert-Kluitenberg, Aldo Dekker

17 – 19 April 2012, Hazyview, South Africa

IAEA/FAO

Outline

� VP1 gene sequence of all 10 FMDV type A strains

� Difference between serological tests

� Virus neutralisation test (VNT)

� Neutralisation index test (NIT)

� Liquid Phase Blocking ELISA (LPBE)

� Statistical analysis

� Mean titres

� r1 values

� Scaling responses

� Cluster analysis

� Principal Component analysis (PCA)

IAEA/FAO

Research goals:

� Comparison of test methodologies

�VNT, NIT, and LPBE

�Which serological methodology gives the highest

confidence?

�Does difference between test methods predict significant

variation?

IAEA/FAO

Source of the 10 FMDV Serotype A strains

used in the study

WRLFMD, Pirbright, U.K.

� A/ERI/2/98

� A/SUD/2/84

� A/KEN/12/2005

� A/ETH/13/2005

� A/MAU/1/2006

CVI, Lelystad, Netherlands

� A22/IRQ/24/64

� A/TUR/20/2006

� A/TUR/14/98

� A/IRN/02/97

� A10/Holland/42

IAEA/FAO

Animal experiment

� 50 cattle plus 5 controls carried out in Eritrea

� 10 different FMDV type A strains

� Cultured on BHK-21 cells

� Inactivated with binary ethylenimine (BEI)

� One cycle of polyethylene glycol (PEG) concentration

� Quantitative sucrose density gradient analysis was used to determine the 146S antigen concentration

� Vaccination with 10 µg antigen in 2 ml Aluminium hydroxide saponin adjuvant (formulation on site)

� Sera 21 day post-vaccination

IAEA/FAO

Serological test methods

� Virus neutralisation

� Varying serum dilution- versus fixed virus concentration

� Neutralisation index

� Fixed serum dilution - versus variable virus concentration

� Liquid Phase Blocking ELISA

� Using subtype specific Guinea-pig and Rabbit semi-purified Ig fractions (saturated Ammonium Sulphate)

IAEA/FAO

Statistical analysis

� Mean titres

� r1-values

� Mean scaled titre (titre – mean)/sd

� Cluster analysis using Euclidian distance

� Principal component analysis (PCA)

IAEA/FAO

VP1 gene sequence of the 10 FMDV type A strains

� A dendrogram

constructed from the VP1

gene of the 10 FMDV

serotype A strains using

BEAST displaying the

posterior value (0 – 1)

indicating the confidence

of the split. FMDV

serotype O1 manisa/1969

was used as an outgroup

to construct the rooted

tree.

IAEA/FAO

Results of the 3 serological

methods

� Huge dataset, difficult to interpret

� On the X-axis the different strains used in the test

� On the Y-axis the 10log

mean antibody titre

� For each vaccine barplots of VNT (black), NIT (grey) & LPBE (Light grey) results

� Very similar titres in VNT and NIT

� Mostly homologous titres are the highest

1 2 3 4 5 6 7 8 9 10

1 - A10 Holland

01

23

4

1 2 3 4 5 6 7 8 9 10

2 - A22 Iraq

01

23

4

1 2 3 4 5 6 7 8 9 10

3 - A SUD/2/84

01

23

4

1 2 3 4 5 6 7 8 9 10

4 - A IRN/2/97

01

23

4

1 2 3 4 5 6 7 8 9 10

5 - A TUR/14/98

01

23

4

1 2 3 4 5 6 7 8 9 10

6 - A ERI/2/98

01

23

4

1 2 3 4 5 6 7 8 9 10

7 - A ETH/13/2005

01

23

4

1 2 3 4 5 6 7 8 9 10

8 - A KEN/12/2005

01

23

4

1 2 3 4 5 6 7 8 9 10

9 - A MAU/1/2006

01

23

4

1 2 3 4 5 6 7 8 9 10

10 - A TUR/20/2006

01

23

4

IAEA/FAO

Comparison of vaccine response (African strains)

- Mostly homologous titres are the highest

- LPBE has a lower discriminatory capacity for differences between strains, but

gives a more consistent result when compared with VN & NI

A/ERI/2/98 Vaccine A/MAU/1/2006 Vaccine

IAEA/FAO

Comparison of vaccine response (Middle East strains)

A/TUR/20/2006 Vaccine A22/IRQ/24/64 Vaccine

IAEA/FAO

VNT Results

� Difference between antibody response induced by the vaccine compared to the response of the virus to the sera made by vaccination against the other strains

� Low response of vaccine against heterologous strains

� Relatively good response by sera from heterologous strains

IAEA/FAO

r1 Values

� Comparison r1-values VNT (black), NIT (dark grey) and LPBE (light grey)

� Huge difference between neutralising and ELISA antibodies

� A/SUD/2/84, A/IRN/2/97 and A/TUR/20/2006 showed high VNT & NIT r1 values (>1.0)

� Because responses in different tests are not on the same scale

1 2 3 4 5 6 7 8 9 10

1 - A10 Holland

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

2 - A22 Iraq

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

3 - A SUD/2/84

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

4 - A IRN/2/97

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

5 - A TUR/14/98

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

6 - A ERI/2/98

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

7 - A ETH/13/2005

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

8 - A KEN/12/2005

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

9 - A MAU/1/2006

0.0

0.4

0.8

1 2 3 4 5 6 7 8 9 10

10 - A TUR/20/2006

0.0

0.4

0.8

IAEA/FAO

Scaled titres

� A clear difference in the scale of the response for each test strain

� For each test strain (titre-mean titre)/sd

� Comparison scaled titres VNT (black), NIT (dark grey) and LPBE (light grey)

� In all tests all homologous scaled titres are the highest

Scaled titre

10log

1 2 3 4 5 6 7 8 9 10

1 - A10 Holland

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

2 - A22 Iraq

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

3 - A SUD/2/84

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

4 - A IRN/2/97

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

5 - A TUR/14/98

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

6 - A ERI/2/98

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

7 - A ETH/13/2005

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

8 - A KEN/12/2005

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

9 - A MAU/1/2006

-2-1

01

2

1 2 3 4 5 6 7 8 9 10

10 - A TUR/20/2006

-2-1

01

2

IAEA/F

AO

A MAU/1/2006

A TUR/20/2006

A22 Iraq

A TUR/14/98

A ETH/13/2005

A SUD/2/84

A ERI/2/98

A KEN/12/2005

A10 Holland

A IRN/2/97

5 6 7 8 9 10

Height

A10 Holland

A IRN/2/97

A SUD/2/84

A ETH/13/2005

A KEN/12/2005

A MAU/1/2006

A ERI/2/98

A TUR/20/2006

A22 Iraq

A TUR/14/98

5 6 7 8 9 10

Height

Cluste

rs b

ased o

n E

uclid

ean d

ista

nce

Based o

n scaled

VNT titre

s

Based o

n scaled

NIT

titres

IAEA/FAO

A T

UR

/20/2

006

A S

UD

/2/8

4

A M

AU

/1/2

006

A22 I

raq

A10 H

olla

nd

A E

TH

/13/2

005

A K

EN

/12/2

005

A E

RI/

2/9

8

A I

RN

/2/9

7

A T

UR

/14/9

8

23

45

67

He

igh

t

Clusters based on Euclidean distance

� Scaled LPBE titre

� African strains (red

circle) cluster more

closely than the Asian

strains (blue circle)

� Different serological techniques produce different clusters

� Different statistical techniques produce different clusters (not shown)

IAEA/FAO

A22 I

raq

A10 H

olla

nd

A K

EN

/12/2

005

A E

RI/

2/9

8

A E

TH

/13/2

005

A I

RN

/2/9

7

A T

UR

/14/9

8

A T

UR

/20/2

006

A S

UD

/2/8

4

A M

AU

/1/2

006

34

56

78

He

igh

t

Clusters based on Euclidean distance

� LPBE scaled r1 cluster

� Clustering of the African and Middle East strains in multi groups indicating no antigenic similarity between the strains

� African strains clustered slightly better than the Asian strains

IAEA/FAO

VNT principal component analysis on strains

-0.35 -0.30 -0.25

-0.2

0.0

0.2

0.4

PC1

PC

2

IAEA/FAO

VNT principal component analysis on sera

-0.2 -0.1 0.0 0.1 0.2

-0.2

-0.1

0.0

0.1

0.2

PC1

PC

2

IAEA/FAO

NIT principal component analysis on strains

NIT

-0.38 -0.36 -0.34 -0.32 -0.30 -0.28 -0.26

-0.4

-0.2

0.0

0.2

0.4

PC1

PC

2

IAEA/FAO

-0.2 -0.1 0.0 0.1 0.2

-0.2

-0.1

0.0

0.1

0.2

PC1

PC

2NIT principal component analysis on sera

IAEA/FAO

LPBE principal component analysis on strains

LPBE

-0.36 -0.34 -0.32 -0.30 -0.28 -0.26 -0.24

-0.4

-0.2

0.0

0.2

0.4

PC1

PC

2

IAEA/FAO

-0.3 -0.2 -0.1 0.0 0.1 0.2

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

PC1

PC

2LPBE principal component analysis on sera

IAEA/FAO

Conclusion

� Neutralisation tests, VNT and NIT, have a higher specificity than LPBE

� Therefore neutralisation tests have a better discrimination capacity of strains

� r1 values are difficult to interpret as values greater than 1 are observed

� Scaled titres can reduce variation between tests

� Cluster analysis does not produce consistent results and not comparable with genetic information

� PCA showed high variations on sera and strains

� Antigenic evolution was found to be more punctuated than genetic evolution, as the change of virion proteins in response to antibody selection is more frequent.

IAEA/FAO

Acknowledgement

� This study was sponsored by the International Atomic Energy Agency (IAEA), Vienna, Austria

� Ministry of Agriculture, Eritrea

� National Veterinary Laboratory staff, Asmara, Eritrea

� FAO Representative Office, Eritrea

IAEA/FAO

THANK YOU