Repeatability and individual correlates of microbicidal capacity of bird blood
B. Irene Tieleman
a,b,*,Elsemiek Croese
a,Barbara Helm
b,Maaike A. Versteegh
a,*, Animal Ecology Group, Centre for Ecological and Evolutionary Studies, University of Groningen, P.O. Box 14, 9750 AA Haren, The Netherlands
b Max Planck Institute for Ornithology, Von-der-Tann-Strasse 7, 82346 Andechs, Germany
ABSTRACT
Keywords:
Aging
Ecological immunology Immune function Microbicidal ability of blood Repeatability
With the rapid development of the field of ecological and evolutionary immunology, a series of new techniques to measure different components of immune function is becoming commonplace. An important step for the interpretation of these new measures is to understand the kind of information about the animal that they convey. We showed that the microbicidal capacity of Stonechat (Saxicola torquata) blood, an integrative measure of constitutive immune function, is highly repeatable when tested against Escherichia coli and not significantly repeatable when tested against Candida albicans. The low repeatability against C. albicans results from relatively low variation among individuals, providing only low resolution to identify if this interindividual variation is consistent. In addition, we explored the effect of sex and age on microbicidal capacity, and found that over a range of ages from 1 to 7 years the blood of older birds had a better capacity to kill microbes. We concluded that, over a time period of weeks, microbicidal capacity of avian blood is an individual-bound trait, that shows consistent interindividual variation partly related to age, and unaffected by sex. This knowledge is important when interpreting the possible evolutionary mechanism underlying immunological differences, for example among individuals, environments and seasons,
1. Introduction
With the rapid development of the field of ecological and evolutionary immunology, a series of new techniques to measure different components of immune function is becoming commonplace (Merchant et aI., 2003; Matson et aI., 2005; Tieleman et aI., 2005;
Matson et aI., 2006; McGraw et aI., 2006; Lee, 2006; Millet et aI., 2007;
Martin et aI., 2007). An important step for the interpretation of these new measures is to understand the kind of information about the animal that they convey. Some physiological traits are labile and change rapidly within an individual depending on behavior, diet or environmental conditions (Hau, 2001; Wingfield, 2003); others are robust and characteristic for an individual, a population or a species (Masman et aI., 1988; Tinbergen and Dietz, 1994; Piersma et aI., 1996;
Tieleman et aI., 2003). Physiological traits that characterize indivi- duals can be linked to their performance or fitness. Consistent interindividual variation (including interindividual variation of temporal patterns) forms a prerequisite for natural selection; it allows ecological and evolutionary interpretation of temporal and spatial physiological variation, within or between individuals and populations,
One measure recently introduced in ecological immunology, and rapidly spreading, is the microbicidal capacity of whole blood, which
• Corresponding authors. TeI.: +31 503638096; fax: +31 503633408.
E-mail addresses:8.I.Tieleman@rug.nl(8.I.Tieleman).m.a.versteegh@rug.nl (MA Versteegh).
provides an integrative measure of constitutive immunity (Merchant et aI., 2003; Tieleman et aI., 2005; Matson et aI., 2006; McGraw et aI., 2006; Millet et aI., 2007; Martin et aI., 2007). Microbicidal capacity has been used in comparisons among individuals (Tieleman et aI., 2005) and among species (Tieleman et aI., 2005; Matson et aI., 2006; Martin et aI., 2007). The microbicidal capacity of blood can be altered by stress of capture and handling, and by challenging the animal with lipopolysaccharide, suggesting that it may be affected byendocrino- logical and infection status of an individual (Matson et aI., 2006; Millet et aI., 2007). However, whether differences among individuals are consistent, related to other individual-bound traits such as age, clutch size or energy expenditure, and/or have a genetic basis, is currently unknown.
How consistent individuals differ from each other can be expressed by the repeatability of a trait. The repeatability is a measure that expresses how much of the overall phenotypic variation, the sum of inter- and intra-individual variation, is caused by inter-individual variation (Lessells and Boag, 1987). Inter-individual phenotypic variation can be partitioned into environmental variation, arising from external circumstances that affect the animal permanently, and genetic variation, caused by differences in genetic background. lnter- and intra-individual variation may vary between populations, because of environmental and/or genetic differences. Hence repeatability is a population measure.
We studied short-term repeatability and correlates of individual variation of microbicidal capacity against the gram-negative bacteria Escherichia coli and the yeast Candida albicans in a population of First publ. in: Comparative biochemistry and physiology - part A: molecular & integrative physiology 156 (2010), 4,
pp. 537-540
DOI:10.1016/j.cbpa.2010.04.011
Konstanzer Online-Publikations-System (KOPS) URN: http://nbn-resolving.de/urn:nbn:de:bsz:352-opus-131878
URL: http://kops.ub.uni-konstanz.de/volltexte/2011/13187/
538
captive European Stonechats (Saxicola torquata rubicola) during the winter season. Variation among different measurements is the sum of the methodological error and the physiological differences. Therefore, in this study repeatability is a measure of both methodological and physiological variation, two aspects that are hard to disentangle in the field. We focused on variation within a single season to determine whether microbicidal capacity is characteristic for a given individual at a given time of the year. This knowledge is crucial when interpreting the possible evolutionary mechanism underlying immu- nological differences, for example among individuals, environments and seasons.
2. Material and methods 2.1. Birds and blood sampling
We measured microbicidal capacity of fresh whole blood of 26 European stonechats (5. t. rubicola) (16 females, 10 males), tested against E. coli (n = 22) and C albicans (n = 17). Stonechats belonged to 15 families, i.e. were offspring of 15 pairs. All individuals were raised in captivity, 18 of them were hatched in captivity, eight were brought into captivity as nestlings. Each individual was kept in a separate cage, and 8 to 12 cages were put together in indoor rooms with controlled temperature (20-22°C) and simulated naturally experienced day length conditions (Andechs, 48°N 11 °E). We measured each individual two or three times during the winter period, between 16 January 2006 and 10 February 2006. Winter is a quiescent period for stonechats: none of the birds were molting, migrating, or in reproductive state. Average time between measure- ments equaled 11.3 days (SO
=
3.19, n=
36, range 9-22 days). Birds were sterilely bled within 3 min of entering their room, and blood was processed in the laboratory within an hour after collection (Millet et aI., 2007).2.2. Assays for microbicidal capacity
We tested the repeatability of the microbicidal capacity of blood against two types of microbes, that represented the classes of gram- negative bacteria (E. coli) and yeasts (C albicans) (Millet et aL, 2007).
A third microbe, Staphylococcus aureus, is commonly used in ecological immunology to represent the class of gram-positive bacteria (Millet et aI., 2007). Because the blood of stonechats did not kill this bacterium, we did not include it in our study. These microbes have been selected because they are ubiquitous and thereby minimize the potentially confounding effect of different antigen- exposure histories (Millet et aL, 2007). In addition, they represent different response pathways of the immune system: E. coli is thought to be killed especially by plasma components of the blood, such as the complement cascade, whereas C albicans is killed largely by phagocytosis (Millet et aL, 2007; Janeway et aL, 2004).
All bacterial work was done in a laminar flow hood that provided a sterile environment. We followed the assay described in Tieleman et aL (2005) and Millet et aL (2007). In brief, we reconstituted lyophilized pellets of E. coli (EPower Microorganisms, ATCC #8739, MicroBiol.ogics, St. Cloud, MN, USA) and C albicans (EPower Micro- organisms, ATCC #10231, MicroBiol.ogics, st. Cloud, MN, USA) following manufacturer's instructions. Every day we took a subsample from these stock solutions, that we diluted to obtain a working solution that yielded about 200 microorganisms per 70
J.ll
of medium- blood-microorganisms mixture on our plates (see below). Stock and working solutions were kept on ice at all times. We diluted 30 pl.whole blood in 270
J.ll
COr independent medium (Gibco-Invitrogen, Carlsbad, CA, USA) with 4 mM L-glutamate. The assay was started when 30 l. of E. coli working solution was added to the diluted blood.The blood-microorganism mixture was incubated at 41°C and under continuous shaking. After 7.5 and 15 min (based on previous studies)
70
J.ll
was plated on tryptic soy agar plates (F1uka, St l.ouis, MO, USA), in duplicate. For C albicans we mixed 50J.ll
blood with 150 pI. medium and added 20J.ll
of C albicans working solution to start the assay. The blood-C albicans mixture was incubated for 4 h at 41°C and continuous shaking before plating 70J.ll
per plate, in duplicate, We chose the 4 h incubation time period based on an earlier study on stonechats from the same population, in which we found average values for microbicidal capacity of 27.7 ± 10.1% (SE, n = 17) after 3 h of incubation and 47.9 ± 8.8% (SE, n = 16) after 4 h of incubation (Tieleman et aL unpublished data). To determine the number of bacteria added at time zero and to assess bacterial growth, we made controls with 30J.ll
of E. coli working solution in 300J.ll
medium, and 20 pI. of C albicans working solution in 200 ~Il. medium. These controls were plated at time zero, i.e. the start of incubation, and after incubation (7.5 and 15 min for E. coli, and 4 h for C albicans). All agar plates were incubated for one (E. coli) or two days (C albicans) at 41°C before colony forming units were counted using a colony counter (BIO, Kobe,Japan), We used average count of duplicate plates for further analyses. Components within blood can prevent microbial growth or kill microbial cells, in which case killing values will be zero or positive. However, microbial cells may divide in blood as indicated by negative killing values (Tieleman et aL, 2005, Buehler et aL, 2008).We realize that different versions of the assay can be and are being used, for example making control plates that include blood both at a zero time point and after incubation. We have chosen to test the version that uses the smallest amount of blood, which is in fact close to the maximum amount of blood that one can take from small animals. When working with small animals one cannot afford to make controls from the blood-microbe mixture at time zero, and has to use control plates without blood to establish the initial density of microbes. The potential for pipetting error may be larger in this version because controls are made separate from the blood-microbe mixture. Yet, this version is frequently applied (e.g. Tieleman et aL, 2005, Matson et aL, 2006, Millet et aL, 2007, Buehler et al.. 2008) and if results are repeatable this version of the assay is meaningfuL
2.3. Statistics
We performed multilevel random effects analyses in MLwiN 2.02, with measurement as level 1 and individual as level 2. Because most individuals came from different families, statistical power to estimate family effects was low, and a model with three levels (family, individual and measurement) did not explain the variation better (C albicans: r =1.25, df=1. P=0.26; E. coli: X2<0.12, df=l, P>0.73), we did not include the level family in the rest of the analyses. We were not able to use a binomial distribution because we had negative values, indicating growth of the microbes in blood, and used a normal distribution after verifying that the residuals of our data were normally distributed (Rasbash et aL, 2005). We included the potential methodologically important factor control count (inocu- lum), and the biologically interesting fixed effects of sex and age. We first calculated the repeatability based on a model with only control count as fixed effect. We then calculated the repeatability based on a model that included also age and sex as fixed effects. Backward elimination of non-significant fixed effects (P> = 0.05) was used as model selection criterion. We used a l.oglikelihood Ratio test to evaluate statistical significance. Repeatability was calculated with the equation (between individual variance)f(within individual variance
+
between individual variance) (Lessells and Boag, 1987).We obtained the between and within individual variances from the multilevel models to calculate repeatability. Standard errors were calculated using Becker (1984). To compare variation in microbicidal capacity between E. coli and C albicans, we calculated interindividual and intraindividual coefficients of variation (C.V.) with the equation C.V. = standard deviationfmean.lnterindividual C.V. was obtained by first calculating average killing values per individual, and
539 Table 1
Repeatability (r, ± SE), interindividual and intraindividual coefficients of variation (CV.), and average (in % colony-forming units killed, ± SD) of microbicidal capacity of whole blood of Stonechats.
Microorganism (incubation time) SE ;f(df=1) p Interindividual CV. Intraindividual CV. Average (%) SD N, n
E. coli (7.5 min) 0.78 0.077 27.0 <0.001 1.04 0.59 31.2 33.66 22,52
E. coli (15 min) 0.63 0.177 14.7 <0.001 0.58 0.42 49.9 31.12 22, 52
C. albicans (4 h) 0.08 0.206 0.18 0.86 0.Q7 0.08 87.4 8.51 17,39
Microbicidal capacity was tested against E. coli (incubation times 7.5 and 15 min) and C. albicans (incubation time 4 h). Calculations are based on repeated measures of individual birds where N is number of individuals and n is number of measurements.
then calculating mean and standard deviation of these individual averages. Intraindividual C.V. was calculated as the average of individual C.V.'s, calculated from means and standard deviations per individual. We repeated all statistics and calculations using the incubated controls for C. albicans instead of the inoculum at time zero, but this did not alter the outcome and we do not report these results.
3. Results
The microbicidal capacity of Stonechat blood against E. coli was significantly repeatable with values of 0.78 and 0.63, when measured after 7.5 and 15 min of incubation, respectively, whereas the microbicidal capacity against C. albicans had an insignificant repeat- ability of 0.08 (Table 1). Variation in control count did not significantly affect microbicidal capacity or repeatability (Table 2). Although insignificant (P= 0.08, Table 2), the direction of the relationship between control count and microbicidal capacity for E. coli after 7.5 min of incubation showed a positive trend.
Over a range of ages from 1 to 7 years, the blood of older birds killed more E. coli and, for birds aged two years and older, tended to kill more C. albicans (Fig. 1). To test if age and sex affected microbicidal capacity, we included them as fixed effects in the model. Age significantly affected microbicidal capacity against E. coli but not against C. albicans, and sex did not have a significant effect on killing of either microbe (Table 2). Taking age into account in the repeatability calculations decreased the repeatability of microbicidal capacity against E. coli to 0.68 and 0.43, after 7.5 and 15 min of incubation, respectively; these values were still significant (both P<O.OOl).
4. Discussion
The microbicidal capacity of stonechat blood is very repeatable when tested against E. coli but not when challenged with C. albicans.
The difference in repeatability between E. coli and C. albicans is due to the small variation in microbicidal capacity among individuals for the latter (Table 1). Small interindividual variation can lead to a low repeatability even if a trait is individual-bound. One possible explanation for the difference in repeatability between the micro- bicidal capacities against E. coli and C. albicans lies in the different immune response pathways involved in the killing process for each microbe. Our data may indicate that the interindividual variation in
Table 2
Results of statistical analyses for effects of control count. sex and age on microbicidal capacity of Stonechat blood.
Variable E. coli (7.5 min) E. coli (15 min) C. albicans (4 h)
;f (dJ= 1) P ;f (dJ=l) P X2 (df= 1) p
Control count 3.08 0.08 0.11 0.74 0.98 0.32
Sex 0.32 0.58 0.33 0.57 0.66 0.42
Age 12.43 0.0004 12.55 0.0004 2.55 0.11
Microbicidal capacity was tested against E. coli (incubation times 7.5 and 15 min) and C. albicans (incubation time 4 h). Values of X2 and significance are based on baci<ward elimination of non-significant fixed effects.
plasma components, especially the complement cascade, involved in E. coli killing is larger than the interindividual variation in phagocy- tosis that is thought to be the main mechanism underlying the killing of C. albicans.
Correct incubation time of the blood-microorganisms mixture seems to affect repeatability of the assay. Over time blood from more individuals may kill close to 100% of the bacterial cells, resulting in a decreasing inter-individual variation, and thereby repeatability. The long incubation time may be an alternative explanation for why we find such a low repeatability in killing of C. albicans. This indicates that correct timing of the assay is important when interested in microbicidal capacity as an individual-bound trait.
The high repeatability of the microbicidal capacity against E. coli suggests that this measure can be interpreted as an individual characteristic at a given time of the year. Therefore, differences among individuals can be meaningfully related to other individual- bound characteristics, such as clutch size (Gwinner et aI., 1995) and
~ .~ '6'
<'>:S2 ~~
m::>
:!2LL
.g ()
~~ CIl
.c
·0_ ~ CIl'O 0.Q)
m::> fl~
:!2LL
.~ ()
e~
<.>'E
100
A
80
-20 100
B
80 60 40 20 0 100
• •
· l~/
~~~~• • •
• •
• I
80
•
60
•
40 20
oC
2
- -
I
•
UQ!L. 7.5 mln incubation
i
•
,_ ...•
~~ ~~ -~- · ~~ ~ ~ •
....-
....-
• •
E~ 15 min incubation
• • •
• •
C. albicans, 4 h incubation
3 4 5 6 7
age (years)
Fig. 1. Microbicidal capacity of whole blood of Stonechats as a function of age of the birds in years. Microbicidal capacity is expressed as percentage of colony forming units (CFU) killed after 7.5 min (A) and 15 min (8) of incubation for E. coli and after 4 h of incubation for C. albicans. Each data point in the graph reflects the average of 2 or 3 measurements on one individual. Negative values indicate microbial growth in stonechat blood.
540
energy expenditure (Versteegh et aI., 2008). In addition, changes over time are likely to reflect physiological adjustments to season or age.
This finding opens possibilities for evolutionary interpretation of microbicidal capacity, although a heritable component, another requisite for evolution to act, still remains to be shown. We caution that repeatability estimates are population measures and that additional studies from different populations will be important to evaluate the general applicability to other species and populations.
Our study shows that microbicidal capacity against E. coli is higher in older Stonechats, but not significantly different between males and females. These findings raise the question of what changes with age, if anything, to make older birds better at fighting of these microbes than younger birds. The individuals in this study reflect a cross-section of a captive population of wild birds that have been kept under constant captive conditions during their entire life. Without a longitudinal study, we cannot distinguish if birds acquire a better microbicidal capacity with age, or if birds with poor microbicidal capacity are more likely to die young, leaving only old birds with high microbicidal capacity in the population. In any case, body condition is not the most likely factor to explain differences with age, because all birds have been fed ad libitum during their entire life. This assay to determine the microbicidal capacity of blood was originally developed for work on wild birds in the field. The current study suggests that part of the interindividual variation found in wild populations may be due to variation in age, a factor that is often unknown in field studies.
Acknowledgements
We are grateful for the technical support ofSylvia Kuhn, Lisa Trost, Willi Jensen, Erich Koch, and other staff at the Max Planck Institute for Ornithology. This study was financially supported by a VENI-grant of the Netherlands Organisation for Scientific Research to B.I.T.
References
Becker, W.A., 1984. A Manual of Quantitative Genetics, 4th ed. Academic Enterprises, Pullman, WA.
Buehler, D.M., Piersma, T" Matson, K" Tieleman, 8.1" 2008. Seasonal redistribution of immune function in a migrant shorebird: annual-cycle effects override adjustments to thermal regime. Am. Nat. 172, 784-796.
Gwinner, E., Konig, S., Haley, CS" 1995. Genetic and environmental factors influencing clutch size in equatorial and temperate-zone stonechats. (Saxico/a torquata axillaris and S. t. rubico/a): an experimental study. Auk 112,748-755.
Hau, M., 2001. Timing of breeding in variable environments: tropical birds as model systems. Horm. Behav. 40, 281-290.
janeway, CA" Travers, P" Walport, M., Shlomchik, M., 2004. Immunobiology: the immune system in health and disease, 6th edn. Garland, New York.
Lee, KA., 2006. Linking immune defenses and life history at the levels of the individual and the species.lntegr. Comp. BioI. 46,1000-1015.
Lesselis, CM., Boag, P.T., 1987. Unrepeatable repeatabilities: a common mistake. Auk 104,116-121.
Martin, L.B" Weil, Z.M., Nelson, Rj., 2007. Immune defense and reproductive pace of life in Peromyscus mice. Ecology 88, 2516-2528.
Masman, D., Daan, S., Beldhuis, H.j.A., 1988. Ecological energetics of the Kestrel: daily energy expenditure throughout the year based on time-energy budget, food intake and doubly labeled water methods. Ardea 76, 64-81.
Matson, KD., Ricklefs, RE., Klasing, KC" 2005. A hemolysis-hemagglutination assay for characterizing constitutive innate humoral immunity in wild and domestic birds.
Dev. Comp. Immunol. 29, 275-286.
Matson, KD" Tieleman, 8.1., Klasing, KC, 2006. Capture stress and the bactericidal competence of blood and plasma in five species of tropical birds. Physiol. Biochem.
Zool. 79, 556-564.
McGraw, K.j., Cri no, Q.L., Medina-jerez, W., Nolan, P.M., 2006. Effect of dietary carotenoid supplementation on food intake and immune function in a songbird with no carotenoid coloration. Ethology 112, 1209-1216.
Merchant, M.E., Roche, C, Elsey, RM., Prudhomme, j., 2003. Antibacterial properties of serum from the American alligator (Alligator mississippiensis). Comp. Biochem.
Physiol. B 136, 505-513.
Millet, S" Bennett, j" Lee, KA., Hau, M., Klasing, KC" 2007. Quantifying and comparing constitutive immunity across avian species. Dev. Comp. Immunol. 31. 188-201.
Piersma, T., Bruinzeel, L" Drent, R" Kersten, M., Van der Meer, j., Wiersma, P., 1996.
Variability in basal metabolic rate of a long-distance migrant shorebird (Red knot, Ca/idris canutus) reflects shifts in organ sizes. Physiol. Zool. 69, 191-217.
Rasbash, j" Steele, F., Browne, W., Prosser, 8.,2005. A user's guide to MLwiN. Centre for Multilevel Modelling, University of Bristol, Bristol.
Tieleman, B.I., Wiliiams, j.B., Bloomer, p" 2003. Adaptation of metabolism and evaporative water loss along an aridity gradient. Proc. BioI. B 270, 207-214.
Tieleman, B.I" Wiliiams, j.B" Ricklefs, RE., Klasing, KC, 2005. Constitutive innate immunity is a component of the pace-of-life syndrome in tropical birds. Proc. Natl.
Acad. Sci. B 272, 1715-1720.
Tinbergen, j.M" Dietz, M.W., 1994. Parental energy expenditure during brood rearing in the great tit (Parus major) in relation to body mass, temperature, food availability and clutch size. Funct. Ecol. 8,563-572.
Versteegh, MA, Helm, B" Dingemanse, N.j., Tieleman, B.I., 2008. Repeatability and individual correlates of basal metabolic rate and total evaporative water loss in birds: a case study in European stonechats. Comp. Biochem. Physiol. AlSO, 452-457.
Wingfield, j.C, 2003. Control of behavioural strategies for capricious environments.
Anim. Behav. 66, 807-816.
