Immuno-epidemiology of coccidiosis

1
Immuno-epidemiology of coccidiosis

• Don Klinkenberg
• Maite Severins
• Hans Heesterbeek

2
Coccidiosis

• Caused by Eimeria spp
• Protozoan
• Intestinal infection
• sometimes lesions
• main problem production loss
• Seven species in chickens
• location in the intestine
• no cross-immunity

3
Parasite classification

• After lecture notes by Kretschmar (micro/macro)

Microparasite Macroparasite Eimeria
Parasite lifespan Short Long Short
Reproduction within host Rapid None Rapid (but dose effect)
Transmission Direct Indirect Indirect
Infection events One Multiple Multiple
Immunity Complete Partial, slowly acquired Accumulative, slowly acquired
Model type SIR type Parasite load ???
4
Essential characteristics

• Transmission through environment
• Dose-dependent infectivity
• Slowly acquired immune response
• stronger upon re-infection
• reduces parasite excretion
• Within-host dynamics!

5
This presentation

• Model of within-host dynamics
• relation between uptake and excretion of
  infectious material (oocysts)
• interaction with immune system
• Model of between-host dynamics (I)
• coupling excretion and uptake of oocysts
• interaction chickens and environment
• Model of between-host dynamics (II)

6
Within-host model

• Eimeria characteristics
• transmission through oocysts
• Eimeria parasitises gut epithelial cells
• limited number of asexual generations

7
Eimeria cycle
8
Eimeria cycle
9
Eimeria cycle
10
Adding immunity

• Primarily T cell immunity
• Immunity evoked by schizonts
• Immunity inhibits schizont development
• Keeping the model simple one immunity variable Y

11
Eimeria cycle with immunity
12
Eimeria cycle with immunity
Oocyst uptake (W)
Schizont I (X(1))


Schizont II (X(2))
Immunity (Y)


Oocyst excretion (Z)
13
Eimeria cycle with immunity
Oocyst uptake (W)
Schizont I (X(1))


Schizont II (X(2))
Immunity (Y)


Oocyst excretion (Z)
14
Eimeria cycle with immunity
15
Model summary

• Discrete time
• Two asexual schizont generations
• T cell immunity against schizont development

16
Model analysis

• Compare model experiments to data
• relation single dose and excretion
• saturation followed by decrease
• excretion during trickle infections
• excretion terminates after some time
• immunising effect of trickle and single
  immunisation
• trickle immunisation gives better protection

17
Single dose and excretion
E. tenella
18
Model analysis

• Model experiments
• single dose and excretion
• relation between W0 and Z4
• trickle infections
• trickle vs single immunisation

19
Analysis single dose
20
Analysis single dose
E. tenella
21
Analysis single dose
E. acervulina
22
Analysis single dose
E. maxima
23
Model analysis

• Model experiments
• single dose and excretion
• relation between W0 and Z4
• b gt 0 (naïve immunity growth)
• m ? 1 (non-linear immune effectiveness)
• trickle infections immunisation
• conclusions on g and a

24
Conclusions within-host model

• Simple model of parasite input-output behaviour
• Single immunity variable can explain experimental
  data
• Solid basis for studying re-infection and
  between-host transmission

25
Between-host model

• Relate excretion to uptake with oocyst level in
  environment V
• Simplifying assumption average chicken

26
Eimeria cycle
27
Eimeria cycle
28
Two new parameters

• Per time step of ca. 2 days
• Uptake rate a0
• estimate from a single experiment 0.01
• Oocyst degradation rate
• estimate from couple of articles 0.5

29
Interesting variables

• Oocyst level in environment
• decrease due to degradation ( uptake)
• increase due to excretion
• Immunity level in average chicken
• increase due to presence of schizonts
• decrease by fixed rate
• Number of infected cells as measure of damage
• numbers of schizonts and gamonts

30
Basic dynamics
31
5
0
5
4
2
1
3
3
2
32
Dynamics in single chicken cohort

• First dose of each infection generation most
  important
• major change compared to previous dose
• fast decay of oocysts in environment
• Dynamics can be described in terms of infection
  generations

33
Damage in single chicken cohort

• Cumulative damage maximum damage

logdmax
logv0
34
Conclusion on damage

• Production damage is reflected by the maximum
  number of infected cells
• Damage may take local minimum with intermediate
  oocyst level V0
• Mechanism
• maximum damage if a single infection generation
  dominates
• minimum when generation dominance switches

35
Damage in single chicken cohort

• Cumulative damage maximum damage

logdmax
1
2
3
4
schizonts II
gamonts
logv0
36
Cleaning after each chicken cohort

• Minimizing damage requires optimal cleaning of
  the shed.
• What happens if a proportion r of all oocysts are
  removed after each cohort?
• Study relation between logv0 and logvend

37
Final oocyst level
logvend
logv0
38
Removal of proportion 1 - r
0
-2
logr
-3
39
Minimum damage
Maximum damage
logr
logv0
40
Conclusion on cleaning

• Removal of a proportion 1 - r of oocysts after
  each chicken cohort cannot minimize damage
• Minimizing damage may be done by maximal removal
  adding oocysts

41
Discussion of the model

• Single average chicken
• Deterministic model
• No spatial effects

42
Different approach

• Individual chickens
• Stochastic model
• Spatial model
• Cost
• No continuous infection/immune level

43
Individual based model

• Patches interact with walking chickens
• Patches
• oocyst level empty, low, medium, high (0 103
  105 107)
• level rises if chicken excretes higher level
• level falls after 14 days without excretion

44
Individual based model

• Chickens
• walk or shuffle each hour
• pick up maximum daily exposure (0, 101 3 5)
• excrete once per day depending on
• uptake -4 days
• level of immunity (no, partial, full)
• regulated by excretion templates
• immunity level may increase depending on
• time since first dose
• number and level of doses

45
Example fit to data (Galmes)
46
(No Transcript)
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
(No Transcript)
62
(No Transcript)
63
damage related to initial level
64
Local minimum

• Mechanism?
• High excretion due to serial medium doses
• medium doses require serial low doses
• If initial level is
• high early excretion of many medium, so serial
  medium doses before immunity
• intermediate early exposure for start-up
  immunity, but less serial medium exposure
• low many chicks are not immune while others
  already shed medium doses

65
More generalized mechanism for local minimum
damage

• Low initial level exposure of naive chickens to
  large oocyst quantities excreted by first
  infection generation
• Intermediate initial level immunity builds up
  before large oocyst quantities are available
• High initial level large oocyst quantities
  available before immunity is reached
• However relation to level of mixing yet unclear

66
Our coccidiosis modellers

• Deterministic continuous model
• Don Klinkenberg, Hans Heesterbeek
• Stochastic discrete model
• Maite Severins, DK, HH
• Stochastic continuous model (not shown)
• Andriy Rychahivskyy, DK, HH