through integrated spatial and molecular analyses

hardtofindcurtainUrban and Civil

Nov 16, 2013 (3 years and 6 months ago)

62 views

Unlocking mysteries of
Rift Valley fever

in
southern
Africa

through
integrated spatial and molecular analyses

Peter Thompson

Department of Production Animal Studies,

University of Pretoria, Onderstepoort, South Africa

Peter
Durr

Australian Animal Health
Laboratory, CSIRO,

Geelong, Australia

Rift Valley fever


RNA
arbovirus



Bunyaviridae
,
Phlebovirus


First described in 1930s in Rift Valley of W Kenya


Main vectors are
Aedes

and
Culex

spp. mosquitoes


Main hosts are sheep, goats and cattle


Important zoonosis


RVF outbreaks


Large outbreaks at long, irregular intervals


Abortion storms and mortality in livestock


Sheep most severely affected


Human disease


contact with diseased animal tissues


mosquito bites


severe complications in 1
-
2%


Role of wildlife uncertain


clinical disease / abortion


subclinical infection

RVF HUMAN DISEASE

INFECTED BY CONTACT WITH DISEASED ANIMAL TISSUES
OR

MOSQUITO BITE

LESS COMMON IN SUB
-
SAHARAN AFRICA
WHERE THE VECTORS ARE SYLVATIC (DO NOT ENTER DWELLINGS)

INCUBATION PERIOD ‹1 WEEK

=
80% OF INFECTIONS SUBCLINICAL OR MILD

‹0.5% FATAL HAEMORRHAGIC FEVER/ENCEPHALITIS

~5% OCULAR SEQUELAE
Weaver &
Reisen

(2010)
Antiviral Res
. 85:328
-
345

Likely origin of Rift Valley fever virus

RVF in South Africa


First recognised in South Africa in 1950/51


±
100,000 sheep died and 500,000 aborted


Central plateau of South Africa


Largest outbreak occurred in 1973/74


Extended across most of the central plateau


Next large series of outbreaks in 2008
-
10


First in eastern part of SA, followed by large
outbreak across central plateau


±
15,000 animal cases


Smaller outbreaks in some intervening years


Both in eastern areas and central plateau

Some unanswered questions


Mechanism(s) of RVFV persistence during the IEP?


Transovarial transmission in
Aedes

spp. with long
-
term survival of eggs


Low
-
level endemic circulation in livestock and/or wildlife


Overwintering of adult
Culex

spp. during mild winters


Mechanism(s) of disease emergence after
IEP?


Emergence of infected
Aedes

eggs


Movement of infected hosts


? Long
-
distance vector dispersal


? Introduction of infected
Aedes

eggs by migratory waterfowl


Mechanism(s) of spread during
outbreaks?


Vector dispersal


Movement of viraemic hosts


? Multiple introductions or emergences


Can prediction models for Southern Africa be improved?


NDVI, ENSO, …


RVF in South Africa, 2008
-
2011

Jan
-
May 2008


Mpumalanga and Gauteng provinces


Clinical disease in African buffalo near Mozambique border


Livestock: mainly cattle affected

Feb
-
Jun 2009


Kwa
-
Zulu Natal, high rainfall area


One farm in Mpumalanga (May 2009)


Mainly cattle affected

Oct
-
Nov 2009


Near Namibian border


Localised to a small area along the Orange River


Not associated with above
-
average rainfall


Flood irrigation

Jan
-
Aug 2010


Across entire central plateau


Associated with high rainfall and flooding


484 reported outbreaks


14,342 animal cases, of which >90% sheep


Many wild antelope species affected

Jan
-
Jul 2011


Mainly Eastern and Western Cape Provinces


135 outbreaks reported

RVF in South Africa,
2008
-
2011: questions…


Where did the virus come from?


What was/is the reservoir?


How was the virus introduced?


What were the mechanisms of spread of the epidemic?


Origin and spread of the 2008
-
2011 outbreaks


Data of RVF outbreaks from OIE WAHID website


Space
-
time permutation scan statistic (
Kulldorf

et al
. 2005) using
SatScan


9.1.1 (
www.satscan.org
)


Detection of space
-
time clusters



Partial (490
-
nt)
Gn

glycoprotein gene sequences
(M
-
segment)
from
2008
-
2010 isolates (mainly human) (
Grobbelaar

et al.

2011)


Construction of Bayesian phylogeny of southern African RVFV
isolates using
MrBayes

(
mrbayes.sourceforge.net
)


Investigation of genetic relationships between clusters


Time aggregation: 1 month

Spatial window: 20% of cases

Temporal window: 20% of study period

1

2

3

4

5

7

8

9

10

6

Cluster

Start

End

Radius

Outbreaks

1

2008/01/14

2008/05/29

380 km

28

2

2009/01/30

2009/06/29

106 km

18

3

2009/09/30

2009/12/29

61 km

26

4

2009/12/30

2010/02/28

93 km

92

5

2009/12/30

2010/02/28

138 km

74

6

2010/01/30

2010/03/29

90 km

144

7

2010/03/30

2010/04/29

160 km

107

8

2010/03/30

2010/04/29

124 km

83

9

2010/11/30

2011/07/29

360 km

119

10

2010/12/30

2011/05/29

229 km

83

Time aggregation: 1 month

Spatial window: 30% of cases

Temporal window: 30% of study period

1

2

3

4

5

6

7

8

Time aggregation: 1 month

Spatial window: 40% of cases

Temporal window: 40% of study period

1

2

3

4

5

6

7

8

Time aggregation: 1 month

Spatial window: 50% of cases

Temporal window: 50% of study period

1

2

3

4

5

6

7

8

Indications from cluster analysis


2008, 2009 (KZN) and 2009 (N Cape) outbreaks form three distinct clusters


2010 epidemic consisted of several clusters, possibly resulting from a
combination of:


short distance spread (vector dispersal)


long
-
distance spread (animal movement)


multiple introductions?


2011 epidemic consisted of two simultaneous clusters:


at
least two separate instances of overwintering

RVF in South Africa, 2008
-
2010

partial M
-
segment sequence available

1

2

3

4, 5, 6, …

Bayesian phylogeny of southern African RVFV isolates

2008

2009

2009

2010

Indications from phylogenetic analysis


2008 and 2009 (KZN) outbreaks were caused by very similar viruses


likely originated from livestock or wildlife reservoir in north
-
east


Mozambique/Zimbabwe?


mechanism of introduction and spread between clusters still unknown


2009 (N Cape) and 2010 outbreaks were caused by very similar viruses


likely originated from N Namibia or N Botswana, mechanism of spread
unknown


N Cape outbreak was a likely source of virus for 2010 outbreak in Free State,
possibly via movement of infected animals

H

C

RVF lineages in South Africa, 2008
-
2010

(
Grobbelaar

et al.

2011)

Grobbelaar

et al.
(2011)
Emerg.Infect.Dis
.

17:2270
-
2276

1999

2008
-
9

2009
-
10

1974
-
5

1951

South African

RVF isolates

B

D

F

I

J

O

A

A

A

C

C

C

C

C

C

C

C

C

E

E

E

G

G

G

G

H

H

K

K

K

K

L

L

L

L

M

M

N

N

N

Data from
Grobbelaar

et al.
(2011)
Emerg.Infect.Dis
.

17:2270
-
2276

RVF virus lineages

L

L

L

L

Data from
Grobbelaar

et al.
(2011)
Emerg.Infect.Dis
.

17:2270
-
2276

RVF virus L
-
lineage

1971
-
4

1969
-
70

1963
-
4

1995

C

C

C

C

C

C

C

C

C

Data from
Grobbelaar

et al.
(2011)
Emerg.Infect.Dis
.

17:2270
-
2276

RVF virus C
-
lineage

1999, 2008
-
9

1985

1977
-
2007

1998

1976
-
98

1991, 2008

1998

2000

2003

H

H

Data from
Grobbelaar

et al.
(2011)
Emerg.Infect.Dis
.

17:2270
-
2276

RVF virus H
-
lineage

2009
-
11

2004

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

Mauritania

×

×

Senegal

×

×

Burkina Faso

×

Guinea

×

CAR

×

×

×

Egypt

×

×

Saudi Arabia

×

Somalia

×

Uganda

×

×

Kenya

×

×

×

×

Tanzania

×

Madagascar

×

×

Angola

×

Zambia

×

Zimbabwe

×

×

×

×

×

×

×

Namibia

×

South Africa

×

×

×

×

×

×

×

×

Data from
Grobbelaar

et al.
(2011)
Emerg.Infect.Dis
.

17:2270
-
2276

RVF lineages in Africa, 1951
-
2010

2008
-
10

2009
-
11

1973
-
4

Directions for further work


Further development of “conventional” spatial analyses


Cluster detection, multivariable models, spread models, etc.


Development of predictive models suited to southern Africa


Longitudinal serological monitoring to detect viral circulation during IEP


Vector population dynamics, host preference and dispersion


Phylogeography / phylodynamics


Reconstruction of timed viral dispersal patterns using
time
-
scaled phylogenies
to accommodate
phylogenetic uncertainty
(
Lemey

et
al.
(
2009)
PLoS

Comp. Biol.
5:
e1000520
)


Reconstruction of
most likely transmission patterns and
infection dates by
combining spatio
-
temporal and genetic data
(
Morelli

et al.
(2012)
PLoS

Comp. Biol.
8:
e1002768)



Acknowledgements


Bob
Swanepoel
, University of Pretoria


Janusz

Paweska
, National Institute for Communicable Diseases


Raphaelle

Métras
, Royal Veterinary College


Kerryne

Graham, Australian Animal Health Laboratory