473
The energetic consequences of parasitism: effects of a
developing infection of Trichostrongylus tenuis (Nematoda)
on red grouse (Lagopus lagopus scoticus) energy balance,
body weight and condition
R. J. DELAHAy12, J. R. SPEAKMAN l and R. MOSS 2
1
2
Zoology Department. University of Aberdeen, Aberdeen AB9 2TN, Scotland
Institute of Terrestrial Ecology, Hill of Brathens, Banchory, Kincardineshire AB31 4B Y, Scotland
(Received 5 August 1994; revised 22 September 1994 .. accepted 22 September 1994)
SUMMARY
The timing of the energetic consequences of a developing, single-dose infection of Trichostrongylus tenuis larvae was
investigated in captive red grouse Lagopus lagopus scoticus. At 12 days post-infection (p.i.), infected birds had a resting
metabolic rate 16 % greater than controls and thenceforth lost weight at a faster rate than controls. At 16 days p.i. infected
birds consumed 38 % less energy and excreted 33 % less energy than controls. The estimated total daily energy expenditure
and energy expended on activity for infected birds at 16 days p.i. were 36 % and 83 % lower, respectively, than for controls.
Infected birds lost condition from 16 days p.i. onwards. The period of energy imbalance at 12-16 days p.i. coincided with
development of late 4th-stage larvae into adult worms and the onset of patency. After this, the effects on energy balance
diminished. Synchronous development of previously arrested T. tenuis larvae in wild birds in spring probably has similar
effects to those reported here and places grouse under conditions of energy imbalance. The observed effects on energy
balance provide a possible mechanism by which the parasite can reduce fecundity and survival of infected grouse.
Key words: Trichostrongylus tenuis, red grouse, energy balance, pathophysiology, arrested larvae.
INTRODUCTION
Red grouse populations show a high prevalence of
infection with the caecal threadworm T. tenuis
(Cobbett & Graham-Smith, 1910; Wilson, 1983;
Hudson, Dobson & Newborn, 1985; Shaw & Moss,
1989), causal agent of 'grouse disease' (Cobbett &
Graham-Smith, 1910; Committee oflnquiry, 1911).
The parasite has a direct life-cycle. Caecal faeces of
infected birds contain eggs from which free-living
larvae emerge and develop into highly motile
infective larvae, which may then be ingested by birds
feeding on heather tips. The ingested larvae normally
take up to 16 days to develop to adult worms within
the caeca (Shaw, 1988b) and may survive for over 2
years (Shaw & Moss, 1989). However, infective
larvae ingested in late summer-early autumn may
become arrested until the following spring (Shaw,
1988 a). A sharp rise in worm egg output from wild
grouse in spring (Moss et al. 1993) suggests that the
delayed maturation of arrested larvae is synchronous.
Resumed development of arrested larvae may there­
fore be linked with disease, as the most severe
outbreaks usually occur in spring and early summer
(Committee of Inquiry, 1911; Jenkins, Watson &
Miller, 1963). High worm burdens have been
associated with dramatic population crashes
(Macintyre, 1918; Hudson, Newborn & Dobson,
1992).
Parasitology (1995),110,473-482
Experimental work on captive (Shaw & Moss,
1990) and wild birds (Hudson, 1986 a) has shown
that T. tenuis infections can reduce host breeding
success. Infected captive birds also ate less and lost
more weight than controls (Shaw & Moss, 1990): the
greatest effects were observed while larvae were
developing into adults. The authors concluded that
developing burdens of T. tenuis are more pathogenic
than patent infections and the simultaneous resumed
development of many arrested larvae in the spring
might be the main cause of disease. This would also
explain why outbreaks of disease are not observed in
autumn even when autumn burdens are much
greater than spring burdens of adult worms associ­
ated with disease (Moss et al. 1993).
Two aspects of timing are crucial to the Shaw &
Moss (1990) hypothesis. First, the greatest patho­
logical effect to the host is related to the devel­
opmental stage of the parasite and its habits. Second,
the importance of this pathology to the host depends
on the timing and duration of parasite development
relative to the host's annual cycle. This study
describes the timing of the greatest pathological
effects of a developing infection of T. tenuis in red
grouse.
Reduced growth rates and poor energy metabolism
have often been associated with gastrointestinal
worm infections in domestic animals (see MacRae,
1993). In contrast, the energetic consequences of
Copyright © 1995 Cambridge T'niversi ty Press
R. J. Delahay,
J.
474
R. Speakman and R. Moss
parasitic infection in wild animals have been little
studied, despite the possibility that even relatively
small effects upon energy acquisition could signifi­
cantly reduce survival and fecundity. Little is known
about the mechanisms which may link parasite
infection to ecological consequences for the host. One
possibility is that the parasite may affect components
of the host's energy budget, leading it into energy
imbalance. This would reduce the energy available
for reproduction, intraspecific competition and/or
survival.
Any physiological dysfunction is likely to have
energetic consequences. Hence, the pathological
effects of T. tenuis infection in red grouse have been
assessed in this study by their subsequent energetic
consequences. This approach allowed a measure of
the impact of pathology at the whole animal level as
the infection developed, and provided information of
relevance to the ecology of infected individuals. A
single dose of infective larvae was used to model the
effects on the host of the resumed synchronous
development of arrested larvae.
MATERIALS AND METHODS
The host
One-year-old captive-bred red grouse were divided
into 4 pairs of cocks and 3 pairs of hens according to
physical condition (see below) and weight. Within
pairs, 1 bird was then randomly assigned to the
infected or the control group. Throughout the
experiment, and for a pre-experimental acclimation
period of 1 week, birds were individually housed in
wire-floored cages in the open air, as described by
Moss (1969). Fresh water, pelleted food (Moss &
Hanssen, 1980) and grit were provided ad libitum.
The parasite
Third-stage T. tenuis larvae were cultured from
caecal faeces of infected captive birds, incubated for
8-10 days at 20°C (Shaw, 1988b). Larvae were
recovered using a Baermann apparatus and concen­
trated by washing over a 125 11m sieve. Birds in the
treatment group were dosed orally with a suspension
of approximately 6000 larvae in 1 ml of distilled
water. Control birds were subjected to the same
handling and dosed with an equivalent volume of
distilled water.
The presence of infections was confirmed weekly
by faecal sampling.
Experimental protocol
Measurements of energy utilization (below) by each
bird were recorded 2 days before infection, and
thereafter at 4, 8, 12, 16, 20, 26 and 36 days p.i.
Worms reached patency at 12-16 days. It was
practicable to deal with only 2 birds/day and
consequently each pair of birds began the experiment
at different times over 10 weeks, thereby reducing
potential bias due to environmental effects.
Food intake and faecal output
Food intake and excreta output were measured for
24 h before respirometry. A known quantity of
pelleted food was supplied to the birds in feeding
troughs at 1000 h and the remainder was weighed
24 h later. Spilled food and excreta (faeces and
urine) were collected separately from plastic-covered
boards beneath the wire floors. Dry weight (DW)
of food and excreta was determined by oven drying
samples to constant weight at 80°C. Dry weight
digestibility was calculated as (DW food intake­
DW excreta)/DW food intake.
An adiabatic oxygen bomb calorimeter (Gallen­
kamp Ltd) was used to estimate the calorific values
of food pellets and samples of excreta from infected
(n = 3) and control (n = 4) birds.
Immediately before respirometry birds were
weighed and their physical condition assessed by
palpating sternum musculature and assigning a score
on the scale 1-5 (emaciated to plump condition)
according to muscle mass (Moss, Rothery &
Trenholm, 1985). The birds were transported from
the aviary, in separate darkened compartments of a
carrying case, to the laboratory.
Measurement of resting metabolic rate
A single-channel open-flow respirometry system was
used to measure oxygen consumption and carbon
dioxide production for estimates of resting metabolic
rate and respiratory quotient (RQ = CO 2 produced/
O 2 consumed). Air was drawn from outside the
building at 1800-2000 ml/rnin using an electric
pump (Charles Austin Ltd) and dried with silica gel.
The air flow was directed through a precision
cumulative flowmeter (water displacement: Alex­
ander Wrights Ltd) and re-dried. Birds were placed
in a darkened, sealed, Plexiglas chamber (31 x
13 x 11 em) through which this sample flow was
directed. The chamber was kept inside an incubator
(Gallenkamp Ltd) which allowed precise tempera­
ture control. The chamber temperature was mon­
itored by a thermistor at either end, connected to a
data-logger (Grants Squirrel), and maintained at
19-21 °C. This range had been identified in pre­
liminary trials to lie within the birds' thermo-neutral
zone. The exhaust stream from the chamber was
dried with silica gel, passed through a CO 2 analyser
(LIRA 3000, ~ISA Instruments Ltd) and thereafter
a subsample of 150 ml /rnin was directed into the
sample channel of the oxygen analyser (Servornex
1100h, Servomex Ltd). Analog signals representing
the 02 and CO 2 contents of the sample gases and the
flow rate, were digitized using the A to D converter
(uPD7002) of a BBC microcomputer (Acorn Ltd) at
40 ms intervals. and averaged over I min periods.
Effects of trichostrongv!e s
~.
c....
475
:,."2
'·4
Statistical analysis
":§" '·3
g
>
'"
0>
o
rr:
~
rr:
'·2
,.,
'.01//
0·9
.'
+---------------,--.---.­
6·5
6·'
6·2
6·3
6·4
I
Bodvwsidht (loge g-1)
Fig. 1. The relationship between body weight and
resting metabolic rate for all birds before treatment. The
regression equation was loge R\IR = -4,374+0'869 loge
body weight (r" = 0'362. II = 1+, P < 0'05). The
regression coefficient of 0·869 ± 0·333 indicated that the
scaling of RlVIR to body weight did not differ
significantly from isometry ib = 1,0).
The oxygen consumption of the bird over 1 min was
calculated as the difference in 02 content between
the ambient levels and that in the sample flow,
multiplied by the flow rate. Periodic recalibration of
the analysers with standard gas samples was carried
out when necessary.
Birds were weighed and placed in the chamber at
least 3 h after feeding and measurements were made
for between 1 h 46 min and 3 h 54 min, depending
on how quickly the birds became settled and the
duration of stable periods. Stable oxygen consump­
tion readings were usually obtained within 30 min of
a bird entering the chamber. Oxygen consumption
values for each respirometry run were plotted over
time and stable sections of measurement retained for
analysis. It was possible to hear when a bird was
active inside the box and these periods could be
correlated with high O 2 consumption, and excluded
from the analysis.
All estimates of 02 consumption and CO 2 pro­
duction were corrected to standard temperature and
pressure. RQ estimates from each respirometry
session were used to calculate the energy equivalent
of 02 consumption (Lusk, 1976).
Energy budget calculations
Ingested energy (Ei) and excreted energy (Ee) were
defined as the energy equivalents of measured daily
food intake and excreta output respectively. Total
daily energy expenditure (DEE) was estimated as the
energy equivalent of food consumption, less that of
weight increase and excreta output. Where weight
was lost, the equivalent energy was added to DEE.
The energy equivalent of wet weight change was
estimated as described by Thomas (1982), who
calculated that 1 g wet weight willow ptarmigan
(Lagopus lagopus lagopus) muscle would yield 5·34 k]
energy. The energy expended on activity (Ea) was
calculated as DEE minus RMR.
The effect of treatment over the course of the
observations was estimated by analysis of covariance
(SAS (1985): GLM procedure), using a linear model
with class variables pair, treatment, day post­
infection and their interactions, and including initial
body weight as a continuous variable. We tested for
the effect of treatment using the random variation
between individual subjects within pairs as the error
term. This provided a highly conservative test.
Paired t-tests were performed at the point in the time
series for each variable where the greatest difference
between the treatment group means occurred. Analy­
sis of variance was performed on selected summary
statistics for mass and condition change. Inter­
correlations in the time series of other response
variables were controlled for in a canonical dis­
criminant analysis.
RESULTS
Establishment of infection
Infection was confirmed for all dosed birds by the
presence of T. tenuis eggs in the caecal faeces by 16
days p.i. Counts of worm eggs in caecal faeces
(Shaw, 1988b) indicated that the mean adult worm
burden was 2860 ± 540 S.E.M. This is consistent with
establishment rates in previous experimental work
with T. tenuis in red grouse (Shaw, 1988b). No
evidence of infection was detected in the control
birds.
During the course of the experiment 1 infected
bird died at day 18 p.i. The immediate cause of death
was not identified but there were no signs of the
acute caecal pathology usually associated with stron­
gylosis. The observations made before death were
not unusual in any respect and no difference in the
level of infection was detected. As the removal of the
relevant pair of birds from the data set did not
significantly alter the results, they were retained in
all analyses.
There was no significant relationship between the
worm burden (estimated from faecal egg counts) and
mass-specific food intake, body weight, condition or
RMR, confirming that there is great individual
variation in host response to infection with T. tenuis
(Shaw & Moss, 1990).
Scaling mass and metabolic rate
All mass-specific measurements are expressed per
gram live body weight. For RMR this assumes that
metabolic rate scales with mass to the 1·0 power
(Packard & Boardman, 1987). This relationship was
confirmed for the animals in this experiment by a
regression of body weight on estimated metabolic
rate prior to treatment (Fig. 1).
R.
J. Delahay, J. R.
80
Speakman and R. Moss
476
>. 40
A
'"
's
.g
0,
"'0
0,
70
.::.!.
60
:9
30
....
2
50
B­
:::>
"8
40
2Q)
.::.!.
:9
:::>
Q)
.::.!.
c
o
LL
30
o 20
u
-+---.------,-~-,----------.-___,___-...,...-__,-
-5
.~
:B
.~
0·60
5
15
25
35
0·55
l
D
]i 6·0
~
0·50
'"
"'0
~ 5·5
~
~ 0·45
a:
'"
~ 5·0
C
>- 040
.
o 0·35
I
10 -il--~-~-~-..,..--..,..--..,..--..,..----.---~
15
5
25
35
-5
6.5
c
Q)
.~
~
--+---.---r------.~---,-_,__-___.__-_.______,~_____.
-5
600 ­
5
15
25
4·5
~----'----,----~---,---.---r---,.----,-~
I
-5
35
5
15
I
I
25
35
E
2:
E! 550
....
o
.~
'"c
o
:-e
"'0
8
o
s:
~ 500
>­
o 450
III
"'0
6
c
o
U
~-,_____,
-____,-~
-__.__~
400 --+-1-~-__,____-.
I
I
j
I
-5
15
25
5
35
Days post-infection
: J----.---r---r--,---,.----,-------.---.-----,
I
i
I
I
-5
5
15
25
Days post-infection
35
Fig. 2. Effects of Trichostrongylus tenuis infection on red grouse dry weight food intake (A), dry weight excreta output
(B), dry weight digestibility (C), resting metabolic rate (D), body weight (E) and condition score (F). Symbols
represent means j- I S.E.M. for infected (.) and control (D) groups.
Table 1. Mass-specific differences in measured response variables at the point of greatest difference
between treatment group means
Food intake (g/kg)
Excreta output (g/kg)
Digestibility
Oxygen consumption (ml/rnin)
RMR (W/kg)
Body weight (g)
Condition score
Days post­
infection
Control
birds
Infected
birds
Difference
Percentage
difference*
16
16
16
12
12
36
36
7+7±5'2
34·7±2·4
0'54±0'02
15'6±0'9
5·2±0·3
531 ±21
8·5±0·6
45·3 ±9-6
23·0 ± 3·7
0'45±0'04
18·2±0·8
6'0±0'2
480±31
4·5 ± 1·2
29'4±7'3
11·7±2·8
0'09±0'04
2·6 ±0'3
0·8 ±O·l
52±43
4·0± 1·5
-39,4
-33-6
-16,7
+ 16·7
+ 16
-9,7
-47
* (Mean difference/control mean) x 100.
Food intake and faecal output
Analysis of variance revealed a significant treatment
x time interaction for mass-specific food intake
(F7.31 = 3'23, P = 0'01) and mass-specific excreta
output (F 7 , 3 1 = 2'52, P = 0'04). The greatest dif­
ference between infected and control groups (here­
after the treatment groups) occurred 3': day 16 p.i.
(Fig. 2 and Table 1) when infected birds consumed
significantly less food (paired t-test t 6 = 3'49, P =
0'01) and produced significantly less excreta (paired
t-test t 6 = 3'47, P = 0'01) than controls. Despite a
significant treatment x time interaction for DW
digestibility (F 7 , 3o = 2'66, P = 0'03) and a reduction
in the digestibility of food by infected birds at 16
days p.i. (Fig. 2 and Table 1), the point of greatest
477
0·3
day p.i. Ip2ii::-eo t-tests P always> 0'1), but some
divergence ::1 group means was evident after 12 days
p.i. (see Fig . .2 i. From this point onwards infected
birds were consistently lighter than controls. Re­
gression of bodvwcight against time from day 12 p.i.
onwards vielded a regression coefficient for each
bird. These coefficients were used as the dependent
variables in a two-way analysis of variance with
treatment and pair as class variables. The model
indicated a significant effect of treatment on the
regression slope (F 6 5 = 8'06, P = 0'04), showing
that infected birds lost weight faster than controls.
Rates of weight loss were calculated as the change
in weight per day expressed as a percentage of the
weight of the bird at the beginning of each interval.
The rate of weight loss was greater amongst infected
birds up to day 26 p.i. (Fig. 3), with the most rapid
loss occurring at 12-16 days p.i. After day 26 p.i.
infected birds lost weight less rapidly than controls.
The difference in percentage weight change was
significant between days 12 and 16 p.i. (paired t­
tests, t 6 = 2'68, P = 0'04) but not significantly dif­
ferent elsewhere (P always> 0'05).
After day 12 p.i. infected birds began to lose
condition relative to controls. This divergence
increased over time and treatment group means were
significantly different by day 36 p.i. (paired t-test,
t 6 = 2'66, P = 0'04; Table 1). A one-way analysis of
variance, using the regression coefficients for in­
dividual birds, indicated a significant effect of
treatment on condition score (F 6 , 5 = 10'55, P =
0'02). The greatest rates of decrease in the condition
of infected birds were observed at 16-20 days p.i.
There was no significant interaction of the initial
weight or sex of the bird and the treatment for any of
the response variables (P always> 0'1).
0·1
Respirometry
~
1­
s
C
<1l
T
U
... <1l
<1l0l
O-~--~~-
o.c
""
.-c J::
<1lU
u_
CJ::
<1l0>
:D'iii
~~
-1 ­
"0
C
""<1l
::2:
-2-
------~
8
4
12
16
20
Days post-infection
"',---'
36
26
Fig. 3. The mean differences in percentage weight
change between treatment groups. Daily weight change
over each interval between measurements was expressed
as a percentage of the bird's weight at the start. Bars
represent the mean difference in weight change between
pairs ± 1 S.E.M.
1·6 -,
A
1·3
1·0
's 0·7
""
"0
~ 0·4
~
~
0·1
""
.~
~I-~-~-~._~-~-~-~-'---~
-5
5
15
25
35
:J
0­
<1l
>­
e'
<1l
C
UJ
0·5
difference between treatment means was not quite
significant (paired t-test t 6 = 2'15, P = 0'07). Both
before and after day 16 p.i. there were no significant
differences between the treatment groups for these
response variables (paired t-tests, P> 0'05).
No significant treatment X time interaction for mass­
specific oxygen consumption (F 7,28 = 1'97, P = 0'09)
or RMR (F 7 • 28 = 1'96, P = 0'10) was detected al­
though both effects approached significance. Despite
this, the control birds had a slightly higher RMR at
day - 2 p.i. than the birds destined for infection, and
this difference was reversed by day 8 p.i. By day 12
p.i., infected birds exhibited a significantly greater
mass-specific oxygen consumption (paired t-test
t 6 = - 3'39, P = 0'02) and RMR (paired t-test t 6 =
- 3'28, P = 0'02) than controls (Fig. 2 and Table 1).
This was the only point during the experiment when
differences between the two groups of birds were
significant. The RQ was not affected by treatment
and remained between 0·7 and 0·8 for both groups
throughout the experiment.
Body weight and condition
Energy budgets
There was no significant difference in the weight of
birds from the two treatment groups at any single
The energy content of pelleted food was 18·42 k] / g
DW. The energy content of excreta did not differ
-0·1
--1--~-~~-~-
-5
5
15
25
Days post-infection
35
Fig. 4. The effect of Trichostrongylus tenuis infection on
four components of the energy budget: (A) ingested
energy (Ei) and excreted energy (Ee); (B) total daily
energy expenditure (DEE) and daily activity rate (Ea),
expressed /g body weight. Symbols represent treatment
group means for infected (.) and control (D)
birds ± 1 S.E.M.
R.]. Delahay,
J. R. Speakman and R.
478
Nloss
Table 2_ Mass-specific differences in estimated parameters of daily
energy use (k] / g/ d) at the point of greatest difference between
treatment group means
(Ei = measured energy intake, Ee = measured energy excreted, DEE = esti­
mated daily energy expenditure, Ea = estimated energy expended on activity.)
Ei
Ee
DEE
Ea
*
Days postinfection
Control
birds
Infected
birds
Difference
Percentage
difference*
16
16
16
16
1350± 110
490±40
834± 57
382±71
830± 180
340± 50
535 ± 109
64± 120
510± 190
160±60
299±97
318±96
-37-8
-32,7
-35,9
-83,2
(Mean difference/control mean) x 100.
Table 3. Raw canonical coefficients for the four principal response
variables
Days postinfection
Dry weight
food intake
Dry weight
excreta
Digestibility
RMR
-2
4
8
12
16
20
26
36
170
-77
-185
202
-2
-146
90
-46
-1888
686
1223
-358
301
291
-277
876
-48
-33
-23
107
7
-41
16
-24
-2896
-453
-1265
2190
1416
-1640
-1408
2399
Table 4. Correlation of canonical variables with the observations on a
given day for the four principal response variables
Days postinfection
Dry weight
food intake
Dry weight
excreta
-2
4
8
12
16
20
26
36
0·45
0-01
0-18
0·41
0-67
-0-08
0-04
0·10
-0,43
-0-02
-0-09
-0,30
-0-65
-0,01
-0,10
-0,18
significantly relative to treatment group (infected
birds, 14'64±0-01 kJ/g DW; control birds, 14·61 ±
0·04 kJ/g DW)_ There was a significant treatment x
time interaction for Ei (F g • 4 = 2-3, P = 0'03) and Ee
(F g , 4o = 2'2, P = 0-04). The greatest difference be­
tween the treatment group means for mass-specific
Ei and Ee occurred at day 16 p.i. (paired z-tests ; Ei,
t 6 = 2'73, P = 0'03; Ee, t« = 2-42, P = 0'05), when
infected birds consumed and excreted less energy
than controls (Fig. 4(A) and Table 2), By day 20 p.i.
there was once more no difference between the group
means.
There was a simultaneous decrease in total daily
energy expenditure (DEE) and daily activity rates
Digestibility
0-40
~0-04
0-32
0-44
0-55
-0'22
-0,02
-0,05
RMR
-0,63
-0,20
0-17
0-61
0·29
0-24
0'26
0·19
(Ea) amongst infected birds after day 8 p.i. (Fig.
4(B)). However, the treatment x time interaction was
only significant for Ea (F g , 3 3 = 2'39, P = 0'03). The
differences between treatment group means for both
DEE and Ea were significant at 12 and 16 days p.i.
(paired t-tests, P < 0'05; Table 2). During this
trough in the energy expenditure of infected birds,
calculated levels of energy expended on activity were
so low for some birds that they fell below zero. This
was likely to have reflected the errors in measurement
of the components of Ea calculation. By day 20 p.i.,
however, there were no significant differences be­
tween levels of energy expenditure in infected and
control birds (paired t-tests, P> 0-05).
479
Effects of trichostrongyles in red grouse
Canonical discriminant analysis
In order to take into account the inter-correlations between
repeated measurements on the same individuals in a time
series, a repeated measures analysis of variance may be
appropriate. This was not possible in the present study
owing to insufficient error degrees of freedom. Conse­
quently canonical discriminant analysis (SAS, 1985; CAN­
DISC procedure) was used to confirm where the greatest
differences between treatment groups occurred, whilst con­
trolling for the effects of intercorrelations between points
in the time-series.
The observed mass-specific food intake, excreta output,
RlVIR and digestibility for each bird at each day p.i. were
used as variables and the analysis performed by treatment
group. The contribution of each day p.i, to the overall
canonical variable for each bird is given as the raw canoni­
cal coefficient (weighting), and the correlation of these
canonical variables with the observed values for a given day
p.i. is the canonical score. Results show that the greatest
weightirigs are not given to those days when the plots
indicate the greatest difference between treatment group
means (Table 3). However, the canonical variables for food
intake, excreta output and digestibility are highly corre­
lated with observations of the raw data for these variables
at day 16 p.i. (Table 4). The canonical variables for RlVIR
are highly correlated with the observations at days - 2 and
12 p.i. (Table 4) and indicate that the differences between
treatment groups at these times occur in opposite direc­
tions to each other. This confirms the differences between
treatment means observed in Fig. 2. Although the weight­
ing given to day 12 p.i. is relatively large, the analysis
suggests that differences at this point alone are also not the
best criterion for discriminating between RlVIR for the two
groups.
These results suggest that for mass-specific food intake,
excreta output, digestibility and RlVIR, the points of great­
est difference between treatment means alone do not pro­
vide the best criterion for separation of the two groups.
Instead, a combination of days that are themselves corre­
lated with events at the points of greatest difference is
better. This analysis does, however, confirm that relatively
large differences between treatment groups do occur at the
points identified by analysis of covariance, despite inter­
correlations in the data set.
DISCUSSION
Few studies on the impact of parasites on wild
animals have dealt with the energetic consequences
of infection (e.g. Meakins & Walkey, 1975; Munger
& Karasov, 1989; Connors & Nickol, 1991), even
though alterations in the energy balance of a
physiologically stressed animal may affect its per­
formance and provide a useful insight into the
ecological consequences. This study has demon­
strated a significant transient energetic effect of
parasitic infection followed by a loss in body
condition.
Previous work has shown that red grouse infected
with adult T. tenuis burdens lose weight and
condition when compared with uninfected controls
(Wilson & Wilson, 1978; Shaw & Moss, 1990). In
100 -".
/
/--" \
I
\
.i
I
I
:1
:1
I
~
50 ­
f..
;:
/
/\
/
/
/
/
I
o
r
10
20
30
Days post-infection
Fig. 5. Schematic representation of the development of a
Trichostrongylus tenuis burden. Approximations of the
mean percentage of the total burden at the L3 stage (....),
L4 stage (--) and the adult stage (-) (after Shaw,
1988b). The shaded area represents the period of
greatest pathological effect in the present study (12-16
days p.i.).
addition a burden of developing worms was more
detrimental to the breeding production of laying
hens than a burden of adult worms (Shaw & Moss,
1990). Shaw (1988b) showed that the infection
comprises only adult worms after day 16 p.i. The
effects at 12-16 days p.i. in our study (see Fig. 5)
therefore appear to be linked with events during
worm development. This is consistent with obser­
vations of pathophysiology in other developing
nematode infections (e.g. Sadun, 1950; Williams et
al. 1983; Ovington, 1985), associated with damage to
tissues and the host inflammatory response.
The few existing measurements of the effects of
intestinal parasites on host metabolic rates have
provided diverse findings. Studies on domestic
ruminants have generally failed to show an effect of
nematode infection on host metabolic rates (see
Holmes, 1987). Similarly, Munger & Karasov (1989)
recorded no change in field metabolic rate in white­
footed mice (Peromyscus leucopusi infected with the
tapeworm Hymenolepis citelli. However, Connors &
Nickol (1991) observed a reduction in metabolic
rate, despite an increase in the metabolizable energy
of the food, in starlings (Sturnus vulgaris) infected
with the acanthocephalan Plagiorhynchus cylindra­
ceus. In contrast, a significant increase in metabolic
rate was observed in sticklebacks (Gasterosteus
aculeatus) infected with the tapeworm Schisto­
cephalus solidus (Meakina & Walkey, 1975). In­
creased metabolic rates have been associated with an
increase in body temperature, or fever, during
infections of protozoan, bacterial and viral pathogens
(Kluger, 1979). Some parasites produce toxins (see
Chappell, 1980) which could potentially alter host
metabolic rates, but the secretion of such substances
by T. tenuis has not been investigated.
Alternatively, increased RMR could reflect the
cost of mounting an immune response (Keymer &
Read, 1991). Wilson & Wilson (1978) showed that
R. J. Delahay,
J. R.
Speakman and R. Moss
red grouse have only a weak immunological response
to primary infections of T. tenuis, peaking at 20-50
days p.i., too late to explain the effects observed in
the present study. However, T. tenuis larvae enter
tissue phase between days 2 and 8 p.i. and this may
result in a host inflammatory response and significant
damage to the caecal mucosa (Shaw, 1988b). In a
study on the effects of trematode infection in bluegill
sunfish, Lemly & Esch (1984) reported a significant
increase in host oxygen consumption in newly
infected hosts. This was attributed to the trauma
induced by migrating trematode larvae and the
associated host inflammatory response. Elevated
RMR levels in the present study may reflect the high
costs of the inflammatory response and the repair of
damaged tissue, and the reduction in digestibility
may be a direct consequence of the physical damage
to the caecal mucosa. However, in the current study
a lag of 7 days exists between the time at which most
worms were in the caecal mucosa and the effects on
RMR. As there is insufficient information on the
precise timing of parasite-induced pathology and
host response over this period it was not possible to
make any firm links with the observed effects.
In the present study the effects on energy balance
caused by an increase in RMR were exacerbated by
the reduction in Ei and digestibility at days 12-16
p.i. Inappetance is a well-documented phenomenon
of T. tenuis infection in red grouse (Shaw & Moss,
1990) and other gastro-intestinal parasites have been
found to alter host digestive efficiency (see Holmes,
1987), sometimes, as observed here, with no com­
pensatory increase in food intake (Munger & Kara­
sov, 1989).
The sustained depression in condition and higher
rate of weight loss in infected birds began during the
development of 4th-stage larvae into adult worms.
Rates of weight loss amongst infected birds de­
creased after this period, however, suggesting that
the pathological impact of infection had decreased.
The reduction in sternum musculature amongst
infected birds suggested that they may have been
metabolizing protein when in energy imbalance. RQ
values were consistent with this hypothesis since
they remained between 0·7 and 0·8 for most of the
respirometry trials (Knoebel, 1971). Another possi­
bility is that pectoral proteins were being mobilized
to repair worm-induced damage of the caecal
mucosa.
The ecological consequences of the parasite­
induced energy imbalance and loss of condition are
potentially far-reaching for wild birds with sub­
stantial burdens of arrested larvae. The energetic
costs and depressed body condition imposed by the
synchronous development of worms in the spring, in
combination with the depressed food intake, could
reduce the host's ability to defend a territory and
might therefore reduce survival and breeding suc­
cess. At day 16 p.i., when energy expenditure beyond
480
resting metabolic requirements was virtually zero in
infected birds, these birds would probably be at a
competitive disadvantage to their uninfected counter­
parts. Rau (1984) demonstrated a loss of behavioural
dominance in mice infected with Trichinella spiralis,
associated with reduced exploratory activity. For
much of the year red grouse cocks compete to gain
and defend territories, and any reduction in energy
available for such purposes is likely to reduce their
success. Red grouse of lower social status had higher
T. tenuis burdens in a study by Jenkins et al. (1963)
although the cause of this association was unclear.
Experimental work with captive grouse demon­
strated a transient reversal in dominance rank at day
21 p.i. as a consequence of T. tenuis infection
(Wilson, 1979). This is consistent with the ob­
servation in the present study that infected birds
weighed 7 % less and had a significantly reduced
condition, relative to control birds by day 20 p.i.
Increased susceptibility to predators may also
result from reductions in available energy, body
weight and condition if predators select weaker
individuals from a prey population (Temple, 1987).
Hudson, Dobson & Newborn (1992) reported that
grouse with higher worm burdens were more likely
to be predated, whereas Moss et al. (1990) found that
having more worms did not make hens more likely to
be killed. This inconsistency may be a result of
differences in the type of predation at the respective
study sites. The sternum muscle mass used as a
measure of the condition of birds in this study is of
primary importance in flight. The reduction of this
muscle mass and available energy in infected birds
might increase their susceptibility to predators. In
contrast the reduced activity of parasitized birds may
make them less conspicuous and therefore less
susceptible to predation. The relationship between
parasitized hosts and their predators is unlikely to be
simple.
The present study has identified an effect on the
host of the synchronous development of approxi­
mately 3000 worms. Moss et al . (1990) identified a
mean spring egg rise equivalent to the rapid
maturation of 2000-3000 worms in a year when
grouse hatched unusually late but clutches were no
smaller than usual and there were no obvious signs of
disease. The burdens of developing worms in the
current study therefore provide a model of the
situation in the wild when the parasite is having a
subclinical effect.
Hen red grouse usually begin laying in late April;
the number of eggs laid is closely related to food
intake and weight gain in the preceding 5-6 weeks
(Savory, 1975). Trichostrongyles could reduce avail­
able energy for reproduction through the observed
effects on food intake, digestibility and RMR. This
could help explain the impaired breeding production
of infected birds in the wild (Hudson, 1986 a) and in
captivity (Shaw & Moss, 1990). However, if burdens
Effects of trichostrongyles in red grouse
of previously arrested worms begin to develop
synchronously in mid-April (see Moss et at. 1993)
then the present study suggests the greatest impact
on wild grouse will be whilst hens are finishing
laying and beginning to incubate their eggs. During
incubation hens eat less and lose weight (Savory,
1978; Committee of Inquiry, 1911, respectively), so
the effects of a developing infection at this time may
be particularly harmful. This may help explain
observations of desertion and death of emaciated
hens on the nest during disease outbreaks (Jenkins et
at. 1963; Hudson, 1986b). The food intake and
weight of red grouse cocks decreases throughout the
spring (Savory, 1978; Committee of Inquiry, 1911)
as they become increasingly vigorous in their
territorial behaviour, suggesting that even healthy
cock birds may also be in a temporary energy deficit.
Both sexes are consequently at particular risk from
parasite-induced pathology at this time of year.
The mean level of T. tenuis infection during this
experiment was approximately 3000 adult worms
per host. Hudson (1986 b) showed that T. tenuis
infections began to have a significant effect on host
body condition when burdens we re above 3000-4000
adult worms. The significant deleterious effects to
the host observed at the lower end of this range in the
present study, suggest that when many worrns are
developing synchronously, burdens smaller than
3000 may be harmful.
In conclusion, the pathological effects of a burden
of developing T. tenuis larvae in red grouse resulted
in traumatic, short-term energy imbalance, weight
reduction and subsequent loss of condition. These
effects corresponded to the development of -lth-stage
larvae into adult worms and the onset of patency. It
is not clear whether the observed energy imbalance
was a consequence of these events or the earlier
tissue phase. The observed effects are likely to occur
in wild grouse, when previously overwintered T.
tenuis larvae resume their development in the spring.
We are grateful to Dave Elston and Don French for
statistical advice and to Duncan Sivell who helped with
practical work. We thank two anonymous referees who
provided helpful comments. This work was funded by an
AFRC studentship awarded to R. J. D.
REFERENCES
(1980). Physiology of Parasites. Glasgow
and London: Blackie & Son Ltd.
COBBETT, L. & GRAHAM-SMITH, G. s, (1910). The
pathology of grouse disease. Journal of Hygiene 10.
1-27.
COMMITTEE OF INQUIRY ON GROUSE DISEASE. (1911, The
Grouse in Health and in Disease, 2 Vols. London:
Smith, Elder & CO.
CONNORS, V. A. & NICKOL, B. B. (1991). Effects of
Plagiorhynchus cylindraceus (Acanthocephala) on trie
energy metabolism of adult starlings, Sturnusrulear:s.
Parasitology 103, 395-402.
CHAPPELL, L. H.
481
(1987). Pathophysiology of nematode
infections. International Journal for Parasitology 17,
443-51.
HUDSON, P. J. (1986a). The effects of a parasitic
nematode on the breeding production of red grouse.
Journal of Animal Ecology 55, 85-92.
HUDSON, P. J. (1986 b). Red Grouse: the Biology and
Management of a Wild Gamebird. Fordingbridge: The
Game Conservancy Trust.
HUDSON, P. J., DOBSON, A. P. & NEWBORN, D. (1985).
Cyclic and non-cyclic populations of red grouse: a
role for parasitism? In Ecology and Genetics of
Host-Parasite Interactions (ed. Rollinson, D. &
Anderson, R. M.), pp. 77-89. London: Academic
Press.
HUDSON, P. J., DOBSON, A. P. & NEWBORN, D. (1992). Do
parasites make prey vulnerable to predation? Red
grouse and parasites. Journal of Animal Ecology 61,
681-92.
HUDSON, P. J., NEWBORN, D, & DOBSON, A. P. (1992).
Regulation and stability of a free-living host-parasite
system: Trichostrongylus tenuis in red grouse. 1.
lVIonitoring and parasite reduction experiments.
Journal of Animal Ecology 61, 477-86.
JENKINS, D., WATSON, A. & MILLER, G. R. (1963).
Population studies on red grouse (Lagopus lag opus
scoticus) in north-east Scotland. Journal of Animal
Ecology 32,317-76.
KEYMER, A. E. & READ, A. F. (1991). Behavioural ecology:
the impact of parasitism. In Parasite-Host
Associations: Coexistence or Conflict? (ed. Toft, C. A.,
Aeschlimann, A. & Bolis, L.), pp. 37-61. Oxford:
Oxford University Press.
KLUGER, M. J. (1979). Fever, its Biology, Evolution and
Function. Princeton, New Jersey: Princeton
University Press.
KNOEBEL, L. K. (1971). Energy metabolism. Physiology
(ed. Selkurt, E. E.), pp. 635-650. Boston: Little
Brown.
LEMLY, A. D. & ESCH, G. w. (1984). Effects of the
trematode Uvulijer ambloplitis on juvenile bluegill
sunfish, Lepomis macrochirus : ecological implications.
Journal of Parasitology 70,475-92.
LUSK, G. (1976). The Elements of the Science of Nutrition,
4th Edn. New York: Academic Press.
MACINTYRE, D. (1918). Heather and grouse disease.
British Birds 12, 53-60.
MACRAE, J. c. (1993). Metabolic consequences of
intestinal parasitism. Proceedings of the Nutrition
Society 52, 121-30.
MEAKINS, R. H. & WALKEY, M. (1975). The effects of
parasitism by the plerocercoid of Schistoceph alus
solidus j uller 1776 (Pseudophyllidea) on the
respiration of the three-spined stickleback Gasterosteus
aculeatus L. Journal of Fish Biology 7, 817-24.
MOSS, R. (1969). Rearing red grouse and ptarmigan.
At.iculture Magazine 75, 256-61.
l\IOSS, R. & HANSSEN, I. (1980). Grouse nutrition.
Nutrition Abstracts and Reviews, Series B 50, 555-67.
)'IOSS, R., ROTHERY, P. & TRENHOLI\I, I. B. (1985). The
inheritance of social dominance rank in red grouse
(Lagopus lagopus scoticus). Aggressii:e Bchauiour 11,
253-9.
)'IOSS, R., TRENHOLM, I. B., W.-\TSO:", ,-\. & PARR, R. (1990).
HOLMES, P. H.
R. J. Delahay,
J. R. Speakman and R. Moss
Parasitism, predation and survival of hen red grouse
Lagopus lagopus scoticus in spring. Journal of Animal
Ecology 59, 631--42.
MOSS, R., WATSON, A., TRENHOLM, I. B. & PARR, R. (1993).
Caecal threadworms Trichostrongylus tenuis in red
grouse Lagopus lagopus scoticus: effects of weather
and host density upon estimated worm burdens.
Parasitology 107, 199-209.
MUNGER, J. C. & KARASOV, W. H. (1989). Sublethal
parasites and host energy budgets: tapeworm infection
in white-footed mice. Ecology 70, 904--2l.
OVINGTON, K. s. (1985). Dose-dependent relationships
between Nippostrongylus brasiliensis populations and
rat food intake. Parasitology 91, 157-67.
PACKARD, G. C. & BOARDMAN, T. J. (1987). The misuse of
ratios to scale physiological data that vary
allometrically with body size. In New Directions in
Ecological Physiology (ed. Feder, M. E., Bennett,
A. F., Burggren, W. W. & Huey, R. B.), pp. 216-236.
Cambridge: Cambridge University Press.
RAU, M. E. (1984). Loss of behavioural dominance in
male mice infected with Trichinella spiralis.
Parasitology 88,371-3.
SADUN, E. H. (1950). Studies on the pathogenicity in
chickens of single infections of variable size with the
nematode, Ascaridia galli. Poultry Science 29, 712-22.
SAS (1985). SAS User's Guide: Statistics, Version 5
Edition. Cary, North Carolina: SAS Institute Inc.
SAVORY, c. J. (1975). Seasonal variations in the food
intake of captive red grouse. British Poultry Science
16,471-9.
SAVORY, c. J. (1978). Food consumption of red grouse in
relation to the age and productivity of heather.
Journal of Animal Ecology 47, 269-82.
SHAW, J. L. (1988 a). Arrested development of
482
Trichostrongylus tenuis as third stage larvae in red
grouse. Research in Veterinary Science 45, 256-8.
SHAW, J. L. (1988b). Epidemiology of the caecal
threadworm Trichostrongylus tenuis in red grouse
(Eagopus lagopus scoticus Lath.). Ph.D. thesis,
University of Aberdeen.
SHAW, J. L. & MOSS, R. (1989). The role of parasite
fecundity and longevity in the success of
Trichostrongylus tenuis in low density red grouse
populations. Parasitology 99, 253-8.
SHAW, J. L. & MOSS, R. (1990). Effects of the caecal
nematode Trichostrongylus tenuis on egg-laying by
captive red grouse. Research in Veterinary Science 48,
253-8.
s. A. (1987). Do predators always capture sub­
standard individuals proportionately from prey
populations. Ecology 68, 669-74.
THOMAS, v. G. (1982). Energetic reserves of Hudson Bay
willow ptarmigan during winter and spring. Canadian
Journal of Zoology 60,1618-23.
TEMPLE,
J. C., SHEEHAN, D. S., FUSELIER, R. H. & KNOX,
J. w. (1983). Experimental and natural infection of
calves with Bunostomum phlebotomum, Veterinary
Parasitology 13, 225-37.
WILSON, G. R. (1979). Effects of the caecal threadworm
Trichostrongylus tenuis on red grouse. Ph.D. thesis,
WILLIAMS,
University of Aberdeen.
(1983). The prevalence of caecal
threadworms (Trichostrongylus tenuis) in red grouse.
Oecologia 58, 265-8.
WILSON, G. R. & WILSON, L. P. (1978). Haematology,
weight and condition of captive red grouse (Lagopus
lagopus scoticus) infected with caecal threadworms
(Trichostrongylus tenuis). Research in Veterinary
Science 25, 331-6.
WILSON, G. R.
1
)