Genetic control of susceptibility to Candida albicans in susceptible A/J and resistant C57BL/6J mice

Publisher’s version / Version de l'éditeur:

Genes and Immunity, 6, 8, pp. 672-682, 2005-12-01

READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE. https://nrc-publications.canada.ca/eng/copyright

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.

Questions? Contact the NRC Publications Archive team at

PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.

Archives des publications du CNRC

This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.

For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.

https://doi.org/10.1038/sj.gene.6364254

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at

Genetic control of susceptibility to Candida albicans in susceptible A/J

and resistant C57BL/6J mice

Tuite, A.; Elias, M.; Picard, S.; Mullick, A.; Gros, P.

https://publications-cnrc.canada.ca/fra/droits

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

NRC Publications Record / Notice d'Archives des publications de CNRC:

https://nrc-publications.canada.ca/eng/view/object/?id=40e47e92-997c-4b4e-81b2-f6fc00cb0e0e https://publications-cnrc.canada.ca/fra/voir/objet/?id=40e47e92-997c-4b4e-81b2-f6fc00cb0e0e

FULL PAPER

Genetic control of suceptibility to Candida albicans

in susceptible A/J and resistant C57BL/6J mice

A Tuite

_{, M Elias}

_{, S Picard}

_{, A Mullick}

_{and P Gros}

1_{Department of Biochemistry, Center for the Study of Host Resistance, McGill University, Montreal, Quebec, Canada;}2_{Biotechnology}

Research Institute, Montreal, Quebec, Canada

The importance of host factors in determining susceptibility to systemic Candida albicans infections is evident in both humans and mice. We have used a mouse model to study the genetic basis of susceptibility, using the inbred strains A/J and C57BL/6J, which are susceptible and resistant, respectively, based on different parameters of the response to infection. To identify genes responsible for this differential host response, brain and kidney fungal load were measured in 128 [A/J
C57BL/6J] F2mice

48 h after infection with 5
104_{C. albicans blastospores. Segregation analysis in this informative population identified}

complement component 5 (C5/Hc) as the major gene responsible for this differential susceptibility (LOD of 22.7 for kidney, 19.0 for brain), with a naturally occurring mutation that causes C5 deficiency leading to enhanced susceptibility. C5 was also found to control heart fungal load, survival time, and serum TNF-a levels during infection. Investigation of the response to C. albicans challenge in a series of AcB/BcA recombinant congenic strains validated the importance of C5 in determining the host response. However, the strains BcA67 and BcA72 showed discordant phenotypes with respect to their C5 status, suggesting additional complexity in the genetic control of the inter-strain difference in susceptibility observed in A/J and C57BL/6J following systemic infection with C. albicans.

Genes and Immunity (2005) 6, 672–682. doi:10.1038/sj.gene.6364254; published online 4 August 2005 Keywords: C. albicans; genetics; infection; C5

Introduction

Candida albicans normally exists as a relatively benign

commensal in the gastrointestinal and genitourinary tracts of healthy humans,1_{although in certain}

immuno-compromised hosts it has the ability to enter into the bloodstream and colonize various organs, including the kidney, digestive tract, lung, and brain.2 _This

opportu-nistic fungal pathogen is a major causative agent of nosocomial infections, accounting for up to 15% of hospital-acquired infections in intensive care units.3_With

an overall prognosis comparable to septic shock with multiple organ failure, systemic candidiasis is a life-threatening infection.3 _{Host genetic factors have been}

proposed to influence infection establishment, progres-sion, and ultimate outcome. For chronic disseminated candidiasis (CDC), a form of invasive infection that occurs primarily in acute leukemia patients undergoing intensive therapy,4_{a genetic contribution to susceptibility}

has been observed.5_{Additionally, there is some evidence}

that there is a genetic component affecting susceptibility to systemic candidiasis in humans, as illustrated by the elevated incidence of candidiasis in patients with

congenital defects affecting phagocyte function, such as myeloperoxidase deficiency.6

Mouse models of disseminated candidiasis represent a valuable tool to study the genetic basis of susceptibility to fungal infections. Intravenous (i.v.) and intraperitoneal (i.p.) infection models have been established and the ensuing pathology in susceptible strains mimics the human infection, with the major targets for colonization being the kidney, brain, and heart. Inbred mouse strains display a spectrum of responses to systemic C. albicans infection, ranging from highly susceptible to resistant, as assessed by fungal burden in affected organs, type and extent of cytokine response elicited and overall survi-val.7,8

Studies in animal models have helped to clarify the aspects of innate and adaptive immunity that are required to protect the host against invasive infections with C. albicans. Neutrophils are rapidly recruited to sites of infection, where they can kill both the yeast and hyphal forms of C. albicans.9,10 _{Components of the}

neutrophil microbicidal system, including NADPH oxidase and myeloperoxidase protect against infec-tion,11,12 _{while members of the Toll-like receptor (TLR)}

family appear to have disparate effects on infection outcome.13–16 _{A T helper (Th)1-type response has been}

shown to be protective against reinfection with C.

albicans, while a Th2 response is more frequently

observed in susceptible strains.17_{Additionally, C. albicans}

has been shown to activate the complement system

Received 9 May 2005; revised 20 June 2005; accepted 6 July 2005; published online 4 August 2005

Correspondence: Dr P Gros, Department of Biochemistry, 3655 Promenade Sir William Osler, Room 907, Montreal, Quebec, Canada H3G 1Y6. E-mail: philippe.gros@mcgill.ca

through the alternative pathway,18_{with the chemotactic}

and inflammatory activities of complement apparently the crucial components for mounting an effective host defence. In particular, a relationship between a naturally occurring deficiency in complement component 5 (C5) and enhanced susceptibility to systemic candidiasis is evident.7 _{However, to date, a genetic analysis of the}

overall contribution of this locus in determining the differential response to infection in a C5-deficient and a C5-sufficient strain has not been undertaken, making it difficult to draw any conclusions about the overall importance of C5 in the host defence against systemic candidiasis.

Ashman and his co-workers19 _{have used a set of}

recombinant inbred strains to identify a putative gene effect on chromosome 14 that controls tissue damage following i.v. infection, designated C. albicans resis-tance gene 1 (Carg1). An additional gene effect, named

Carg2, was also identified, although it was not localized

to any region within the genome.20 _{Characterization of}

the genes underlying these loci awaits further mapping and cloning.

A detailed analysis of the pathophysiology and host response to acute C. albicans infection demonstrated that A/J mice are extremely sensitive to infection, whereas C57BL/6J (B6) are relatively resistant to this infectious challenge.8 _{The dramatically different response elicited}

in A/J and B6 mice following systemic challenge with

C. albicans prompted us to investigate whether these

phenotypic differences have an underlying genetic basis. To do this, we analyzed infection progression and outcome in A/J and B6 mice following i.v. challenge with C. albicans and performed a genome-wide scan and linkage analysis in an informative [A/J
B6] F2

popula-tion. We further attempted to characterize the genetic basis of the differential response to infection in these strains by surveying the response to infection of a series of AcB/BcA recombinant congenic strains (RCS).21

Results

The inbred strains A/J and B6 show discrete differences in their response to i.v. infection with C. albicans, particularly in terms of tissue damage, fungal replica-tion, and overall survival.22 _{C. albicans replication was}

assessed in different organs from three A/J and B6 mice 48 h after i.v. injection with a low dose of C. albicans (Figure 1a). In both strains, C. albicans was cleared rapidly from the bloodstream, with levels below detec-tion by 1 h postinjecdetec-tion (data not shown). Levels of fungal replication in the liver, lung, and spleen of A/J and B6 mice were low and there was no significant difference between the two strains. However, fungal load in the brain (Po0.05), kidney (Po0.01), and heart (Po0.001) of A/J mice was found to be significantly and reproducibly higher than that observed in B6. The highest fungal loads in both strains of mice were detected in the kidney, with the kidneys of A/J mice con-taining an approximately 100-fold greater number of fungi at this point in the infection process. The levels of several cytokines that had previously been reported to be elevated during systemic C. albicans infection8,23–26_were

measured in A/J and B6. Cytokine levels were measured by ELISA in serum isolated from A/J and B6 mice 48 h

after low dose injection with C. albicans blastospores. Cytokine levels were comparable between strains prior to infection (Figure 1b). No detectable differences in serum IL-12 or IFN-g levels were observed postinfection between the two strains. However, A/J was found to have elevated levels of IL-6 and TNF-a at the 48 h time point compared to B6. Thus, in this model of infection, susceptible A/J mice display high levels of fungal growth in the kidney, heart, and brain, and a correspond-ing elevation of serum levels of IL-6 and TNF-a.

The clear phenotypic differences between A/J and B6 confirmed that these inbred strains do indeed provide a good model system for studying the genetic basis of differential susceptibility to systemic candidiasis. We used an [A/J
B6] F2intercross population to determine

the mode of inheritance of susceptibility to candidiasis, as measured by fungal load in the kidney and brain. These mice, in addition to A/J and B6 controls, were infected i.v. with 5
104_{C. albicans blastospores and the}

extent of fungal replication in the kidney and brain was assessed 48 h later. [A/J
B6] F1 generation mice were

found to be as resistant to infection as the B6 parental controls (data not shown), while the bimodal distribution pattern of 128 F2 mice was indicative of susceptibility

acting as an autosomal recessive trait under simple genetic control (Figure 2). The segregation pattern of the F2population for these traits approximated that expected

for a Mendelian trait if susceptibility to infection is recessive, with an approximately 3 : 1 ratio of resistant to susceptible mice. The mean log10 CFU counts in the F2

population were at 4.1 (kidney) and 2.5 (brain), values closer to the resistant B6 controls (3.8 for kidney, 2.3 for brain) than the susceptible A/J controls (5.9, 4.2), suggesting that susceptibility segregates as a recessive trait in this cross. We observed that the levels of fungal load in the brain were significantly correlated with kidney fungal load in individual mice (Po0.0001), indicating that these two traits are likely under the same genetic control. Finally, male mice displayed higher kidney CFU counts than female mice and this was observed for all experimental groups tested.

The phenotypic data for this F2 cross behave as a

quantitative trait that is amenable to study by linkage analysis. To identify the genetic locus or loci responsible for control of fungal load, we performed a genome-wide scan on the DNA from these F2mice. A total of 138

polymorphic dinucleotide repeat markers informative for A/J and B6 (Table 1) provided an average coverage of 10 centimorgan (cM) along each chromosome. The largest gap was estimated at approximately 29 cM for the proximal portion of chromosome X. Marker mapping and assignment, as well as genome-wide multipoint linkage analysis, was performed using Mapmaker/EXP version 3.0 and Mapmaker/QTL 1.1. The results from this analysis are shown as a multiple point LOD score trace (Figure 3), with numerical data for the significant interval shown in Table 2.

Using either kidney or brain fungal load CFU data as a quantitative trait, a statistically significant linkage was identified on the proximal portion of chromosome 2, encompassing a region of greater than 25 cM. Maximum linkage was found to the marker D2Mit295 (w2

¼ 79.4 (kidney) and 73.2 (brain); LOD ¼ 17.2 (kidney) and 15.9 (brain)), which maps to 17 cM on chromosome 2 (Table 2). Since the gene for complement component 5 (C5/Hc)

mapped less than 7 cM away from this marker27_{and had}

been previously implicated for its role in controlling response to candidiasis in inbred mouse strains,28–30 _we

looked directly at this gene to determine if it was responsible for the linkage identified on this chromo-some. A/J mice are deficient in C5 production, whereas B6 produce normal levels.31 _{We used a unique BsgI}

restriction site that is introduced by the 2-bp deletion in C5-deficient mice to determine the C5 status of the F2

mice. C5 status did in fact show a highly significant linkage with C. albicans susceptibility (w2

¼ 104.6 (kidney) and 87.3 (brain); LOD ¼ 22.7 (kidney) and 19.0 (brain)), accounting for 60 and 50% of the phenotypic variance for kidney and brain, respectively. No other significant linkages were identified in this cross, likely due to the strong effect of C5 in determining infection outcome. Although the initial linkage analysis was conducted using a free genetic model, dominant, recessive, and additive models were subsequently tested for the chromosome 2 interval. The strongest evidence was for a dominant-acting locus (w2

¼ 103.7 (kidney) and 86.9 (brain); LOD ¼ 22.5 (kidney) and 18.9 (brain)), consistent

with the wild-type C5 conferring protection against candidiasis, even in the heterozygous state.

To further investigate the role of C5 in regulating the host response to systemic candidiasis, we used addi-tional F2mice to assess other phenotypic measures that

differ between resistant and susceptible mice during candidiasis. Specifically, we looked at host response parameters that showed clear differences between A/J and B6. Heart fungal load was measured 48 h after i.v. injection with a low dose of C. albicans in 64 F2 mice.

Survival was followed for up to 28 days after i.v. infection with a high dose of C. albicans in 122 F2mice.

In a separate group of 56 F2mice, serum TNF-a levels

were measured 48 h after low dose inoculation. The C5 status of each of these F2 mice was established by

genotyping and there was a significant correlation between C5-deficiency and increased susceptibility to systemic candidiasis, as measured by each of the different phenotypes (Po0.001 for all phenotypes) (Figure 4). MapManager QT was used to perform tests for single-locus association between the genotype at C5 and each of these phenotypes, using a log10transform for Figure 1 Differential susceptibility of A/J and B6 mice to systemic candidiasis. A/J and B6 mice were infected i.v. with a low dose (5
104₎

of C. albicans blastospores. (a) The number of colony-forming units (CFU) was determined in the brain, heart, kidney, liver, lung, and spleen of three mice of each strain 48 h postinjection. Fungal loads in the brain, heart, and kidney of A/J mice were significantly higher than in B6 at this time point. (b) Serum cytokine levels in 4–6 mice of each strain were also measured prior to infection and at the 48 h time point, as described in Materials and methods. TNF-a and IL-6 levels were significantly higher in A/J mice compared to B6. The standard error of the mean (s.e.m.) is indicated for each group. Statistically significant differences are indicated on the graphs (*Po0.05; **Po0.01; ***Po0.001).

all of the data to reduce skewness. C5 status explained greater than 70% of the phenotypic variance observed for survival (w2

¼ 148, LOD ¼ 32.1) and circulating TNF-a levels (w2

¼ 76.0, LOD ¼ 16.5). The weakest evidence of linkage with C5 was observed using the heart fungal load data (w2

¼ 26.3, LOD ¼ 5.7), although the LOD score was significant and the 32% phenotypic variance described by this trait was well above the 20% cutoff that describes a strong, and likely Mendelian QTL.31

To assess the possible involvement of additional genetic determinants (alone or in combination with C5) in the A/J vs B6 inter-strain difference in susceptibility to systemic candidiasis, we investigated the response of a series of AcB/BcA RCS derived by systematic inbreeding from a double backcross (N3) between A/J and B6 parents.21_{In this breeding scheme, each of these strains}

derives 12.5% of its genome from either A/J or B6, fixed as a set of discrete congenic segments on the background of the other parental strain (87.5%). We measured fungal load in the kidney (Table 3 and Figure 5) and heart (data not shown) for these strains. The C5 status of each of these strains was also determined by genotyping and is shown in Table 3. For analysis, mice were separated according to their C5 status. A/J and B6 controls were included in each experimental group of mice and controls from within each experiment were used to

determine statistical significance, in order to account for variations in dose between experiments. In general, we observed that the C5 status of the strains was a strong predictor of their response to C. albicans infection, with a C5 deficiency causing increased susceptibility to infec-tion, even when present on a resistant B6 background, as seen in the strains BcA70 and BcA83. Likewise, the presence of wild-type alleles at C5 on an otherwise susceptible A/J background was associated with resis-tance to infection in the strains AcB55 and AcB63 (Table 3).

Among the C5-deficient strains, mean CFU counts were compared to the appropriate sex-matched C5-deficient A/J controls to identify strains that displayed enhanced resistance to systemic C. albicans infection in the absence of functional C5. This analysis identified the strain BcA72 as displaying a discordant phenotype. In BcA72, there was a strong gender effect on resistance, with male BcA72 mice showing CFU counts 40- to 50-fold greater than those seen in female BcA72. When compared to gender-matched controls, BcA72 were significantly more resistant to infection than their respective A/J counterparts (Figure 5) and this by a factor of 10- to 100-fold in females and males, respec-tively. A similar analysis of C5-sufficient strains identi-fied the strain BcA67 as having a significantly higher kidney fungal burden than the B6 controls. These results indicate that additional susceptibility and resistance genetic factors, distinct from the major effect of C5, have become fixed in the strains BcA67 and BcA72.

Discussion

The onset, response to, and outcome of systemic infection with C. albicans varies greatly among inbred strains. Infection manifestation in susceptible strains resembles human candidiasis, in terms of the organs targeted for colonization and active replication, as well as the ensuing pathology.2_{We focused on two strains that}

display disparate responses to infectious challenge: A/J and C57BL/6J. A/J is extremely sensitive to infection, with overwhelming fungal replication occurring primar-ily in the kidney, and to a lesser extent, the brain and heart. Elevated levels of the proinflammatory cytokines TNF-a and IL-6 are observed early in infection, suggest-ing that these mice undergo a heightened, and possibly unregulated, inflammatory response after challenge with

C. albicans.8 _{In contrast, B6 is relatively resistant to}

systemic candidiasis, with an approximately 100-fold lower fungal load in the kidney and a muted production of IL-6 and TNF-a compared to A/J.

Table 1 Polymorphic microsatellite markers used for genome-wide scan

D1Mit318, 213, 156, 380, 440, 136, 495, 107, 111, 116, 17 D11Mit135, 313, 242, 245, 285, 124, 301, 103 D2Mit81, 295, 458, 380, 37,132, 395, 309, 411, 343, 266 D12Mit60, 285, 52, 118, 233, 8 D3Mit203, 224, 137, 311, 215, 320, 147, 323 D13Mit16, 266, 13, 193, 148, 77, 78 D4Mit149, 2, 236, 214, 348, 255, 338, 148 D14Mit98, 133, 60, 262, 263, 131 D5Mit352, 201, 10, 158, 188, 95, 97, 101, 287 D15Mit266, 100, 122, 239, 107, 171, 161 D6Mit138, 274, 188, 284, 108, 290, 294, 373 D16Mit131, 211, 188, 158, 70 D7Mit294, 270, 232, 323, 285, 332 D17Mit62, 10, 7, 186 D8Mit281, 190, 205, 45, 211, 200 D18Mit163, 123, 186 D9Mit67, 71, 207, 123, 355, 137 D19Mit68, 85, 46, 53, 34, 137 D10Mit106, 214, 198, 230, 70, 14, 103 DXMit55, 166, 64, 172, 132, 186 Figure 2 Kidney and brain fungal load in [A/J
B6] F2 mice

infected with C. albicans. 128 [A/J
B6] F2mice were infected i.v.

with a low dose of C. albicans. Fungal load in the (a) kidney and (b) brain was measured 48 h after infection. F2mice were separated by

gender, with males displaying an overall enhanced susceptibility to infection as measured by kidney (Po0.01) but not brain CFU. Each data point represents an individual mouse, and horizontal bars represent the mean CFU counts for each group. CFU values for parental A/J and B6 mice are shown for comparison.

A genome-wide scan was conducted in an informative F2population derived from A/J and B6, in order to study

the genetic control of susceptibility to C. albicans. Linkage analysis using kidney or brain fungal load as a phenotypic marker of susceptibility identified an interval on chromosome 2 with a significant linkage, with the highest LOD score associated with the gene for comple-ment component 5. Heterozygotes and mice homozy-gous for wild-type C5 displayed a degree of resistance comparable to B6 controls, whereas C5-deficient mice showed an enhanced susceptibility to candidiasis. No additional gene effects were detected in this cross. Single-marker linkage analysis identified C5 as an important regulator of heart fungal load, survival, and

serum TNF-a levels in additional F2 mice. The

associa-tion of C5 with these independent measures of the host response provides strong evidence that a deficiency in C5 is indeed responsible for enhancing susceptibility to infection.

A previous study identified a putative gene effect on chromosome 14 (Carg1) that controlled overall suscept-ibility to tissue damage.19_{We detected no such effect on}

chromosome 14. Although the infection conditions were similar, a number of variables differed between these studies, including the strains of mice used and the phenotypes assessed, which may account for the lack of correlation. Additionally, it has been proposed that both A/J and B6 possess resistance alleles at Carg1, based on a

Figure 3 Linkage analysis by whole-genome scan of susceptibility to intravenous infection with C. albicans. LOD score traces for chromosome 2, for which a highly significant linkage was detected that controls fungal replication in both the brain and kidney after i.v. infection with a low dose of C. albicans. Separate analyses using brain (dashed line) and kidney (solid line) fungal load as phenotypic traits were conducted.

Table 2 Statistical values for significant linkages obtained from whole-genome scan of an [A/J
B6] F2population infected with C. albicans,

using kidney and brain fungal loads as phenotypic traits (n ¼ 128)

Marker Position (cM)a _{Kidney fungal load Brain fungal load}

LOD w2 Variance explained (%)b LOD w2 Variance explained (%)

D2Mit81 13.5 13.5 62.1 39 12.1 55.8 36 D2Mit295 17.0 17.2 79.4 51 15.9 73.2 45 C5c _{23.5 22.7}c _{104.6 60 19.0}c _{87.3 50} D2Mit458 31.7 15.4 70.8 45 12.4 57.2 38 D2Mit380 40.4 3.4 15.4 12 4.6 21.4 15 D2Mit37 45.0 2.4 10.8 8 4.1 18.7 14

a_{The chromosomal position of each marker is expressed as the distance from the telomere (in cM), as determined by the Mouse Genome} Database Project (http://www.informatics.jax.org).

b_{The proportion of total phenotypic variance explained by each marker is indicated as a percentage.} c_{Highest LOD score determined for this scan.}

Figure 4 Effect of C5 status on different measures of susceptibility to candidiasis. (a) Heart fungal load was measured in 64 F2mice 48 h after

low dose infection with C. albicans. (b) Survival was monitored in 122 F2mice for up to 28 days after infection with a high dose of C. albicans.

(c) Serum TNF-a levels were measured 48 h after infection with a low dose of C. albicans in 56 F2mice. For all phenotypes there was a

significant linkage between C5-deficiency and an increased susceptibility to infection (***Po0.001). The response of A/J and B6 parents are included for comparison. C5 alleles are identified as ‘a’ and ‘b’, for A/J and B6, respectively, and were identified by genotyping (see Materials and methods). Horizontal bars indicate the mean value for each of the traits analyzed for each group.

phenotypic evaluation of the patterns of tissue damage observed in these strains.32 _{Although this hypothesis}

could not been tested in the current study, since we did not use tissue damage as a phenotypic measure of

susceptibility, a lack of polymorphism between A/J and B6 at this locus would explain the absence of this gene effect in the present cross. Interestingly, the strains used to identify Carg1, namely AKR and C57/L, differ in their C5 status in a manner similar to A/J and B6, with AKR being C5 deficient and C57/L being C5 sufficient. A comparison of the C5 status of the AKXL strains used in this study33_{indicates that C5 is in fact a strong predictor}

of the response to systemic candidiasis as measured by tissue damage. For 80% of the strains tested, their response to infection followed their C5 status. The existence of three discordant strains could indicate either the presence of additional gene effects or a difficultly in ascertaining the tissue damage phenotype using a bimodal distribution.

The finding that C5 is the major gene responsible for the differential susceptibility to systemic candidiasis observed between A/J and B6 is not unexpected, given that A/J has a known C5 deficiency. This C5 deficiency is caused by a 2-bp deletion in exon 6 of the C5 gene, which leads to the introduction of a premature stop codon and the production of a nonfunctional, truncated polypeptide that is not secreted.31_{This deletion occurs in up to 40% of}

the commonly used inbred strains.34 _{A correlation}

between the absence of functional C5 and increased susceptibility to candidiasis among inbred strains has been noted,7,22 _{although this study represents the first}

formal demonstration of this relationship, using linkage mapping in informative crosses.

Complement component 5 is a member of the complement system, a series of plasma proteins and soluble and membrane-bound receptors that react with one another to opsonize invading pathogens and induce various inflammatory responses to prevent infection.35

Following activation of the alternative pathway by C.

albicans, a complement cascade is initiated, leading to C5

cleavage and the generation of C5a and C5b. C5b is a component of the membrane attack complex (MAC),

Table 3 Phenotypic response of AcB/BcA recombinant congenic strains to C. albicans infection

Strain Hc allelea _{Log kidney CFU (mean7s.d.) Phenotype}b

BcA81 1 2.270.2 R BcA78 1 2.270.2 R BcA80 1 2.570.6 R AcB63 1 3.070.6 R BcA68 1 3.071.0 R AcB55 1 3.170.6 R BcA75 1 3.170.7 R BcA87 1 3.370.6 R BcA82 1 3.370.7 R BcA77 1 3.370.8 R BcA69 1 3.370.9 R BcA74 1 3.470.4 R B6 1 3.570.7 R BcA66 1 3.770.1 R BcA84 1 3.770.4 R BcA79 1 3.870.2 R BcA85 1 3.970.3 R AcB65 0 5.670.4 S AcB53 0 5.670.6 S BcA70 0 5.770.4 S BcA83 0 5.870.1 S AcB64 0 6.070.3 S A/J 0 6.070.5 S AcB57 0 6.170.2 S AcB51 0 6.170.3 S AcB52 0 6.170.4 S AcB54 0 6.270.2 S AcB60 0 6.370.1 S a_{0—C5 deficient; 1—C5 sufficient.} b R—resistant; S—susceptible.

Figure 5 Phenotypic response of discordant AcB/BcA recombinant congenic strains. For each strain, a minimum of five mice were tested for

C. albicans replication in the kidney 48 h after infection with a low dose of C. albicans. Horizontal bars represent the mean CFU values for each

strain. Males and females were analyzed separately. Mean CFU counts for each strain were compared to the contemporaneous A/J (C5-deficient) or B6 (C5-sufficient) control of the corresponding gender. The s.e.m. is indicated for each group and strains that differed significantly from the controls are indicated on the graph (*Po0.05; **Po0.01, ***Po0.001).

which joins with additional terminal components of complement to form a pore in the cell membrane of the pathogen, leading ultimately to cell lysis.35_{The relevance}

of the MAC in the clearance of C. albicans has not been appreciably demonstrated and direct lysis of C. albicans in the blood has not been observed.36

It is much more likely that a C5 deficiency has its effect at the level of C5a, a chemotactic agent and anaphyla-toxin.37 _{The most well-defined role for C5a during}

invasive candidiasis is as a chemotactic agent for neutrophils. In the absence of C5a, mice have a reduced recruitment of effector cells to the site of infection.8_The

G-protein coupled C5a receptor (C5aR/CD88) is ex-pressed on neutrophils, monocytes, macrophages, den-dritic cells, basophils, eosinophils, and mast cells, among other cell types.38_{Neutrophils express high levels of this}

receptor,39 _{explaining their rapid infiltration to sites of}

infection and the ensuing phagocytosis of C. albicans in response to C5. C5a signalling pathways in neutrophils lead to the assembly of catalytically active NADPH oxidase at the cell membrane, which has an established role in the killing of phagocytosed C. albicans. Neutro-penia is one of the major risk factors for developing systemic candidiasis in humans,1 _{and the lack of}

functional C5 may mirror this process in mice, with the absence of C5 effectively causing a localized neutrophil-poor response during infection. At 2 h after i.v. infection,

C. albicans deposition per milligram of organ weight is

highest in the lungs, followed by intermediate levels in the liver, spleen, and kidneys, and low levels in the brain and heart.40,41_{It is therefore notable that the major sites of} C. albicans replication in A/J at the 48 h time point are the

kidney, brain, and heart. The ability of the liver, spleen, and lungs to control fungal replication despite relatively high levels of initial infection may reflect the presence of resident macrophages in these organs that are able to contain and control infection in the absence of a C5-induced recruitment of neutrophils.

While the results of the whole-genome scan demon-strate the major effect that C5 has in controlling the response to C. albicans infection, it is expected that additional genes are also integral to the host response to systemic challenge. As an alternate approach to both validate the importance of C5 in infection and identify any additional gene effects, we have used a set of RCS.21

These mice have the advantage of being genetically well-characterized, as they have been genotyped at a high density across the entire genome. The strain distribution patterns can be used to localize any detected gene effects through a comparison of the regions of the genome that are in common among strains that display the pheno-type of interest. The mosaic nature of the genomes of these lines of mice makes them useful tools for mapping, since they contain a mixture of the genomes of the two progenitor strains of interest.

In this study, the C5 status of all of the strains was known and each RCS was tested for its response to systemic challenge in the context of the absence or presence of C5. C5 status is a very strong predictor of susceptibility to candidiasis, as demonstrated by the resistance of the C5-sufficient strain AcB55, despite being on a susceptible background. However, the identification of discordant strains indicates that additional mechan-isms may be involved in the host response to systemic candidiasis. In particular, the strain BcA72 displays an

enhanced resistance to infection, despite its C5-defi-ciency. This resistance could be caused by a number of different events, such as the transfer of a novel resistance locus, de-repression of expression of a resistance allele from one strain when it is introduced onto the reciprocal genetic background, or the suppression of A/J- or B6-derived susceptibility alleles upon transfer to a different genetic background. This strain also displays a gender-specific effect, with females tending to be more resistant than their male counterparts. The effect of gender on the ability of a host to resist infectious challenge with C.

albicans has been noted previously, with male mice of

different inbred strains tending to be more susceptible to infection, even among C5-sufficient strains, where serum C5 concentrations are higher in males than in females.42,43

This gender specificity appears to be much stronger in BcA72 than in the A/J and B6 controls. Among the C5-sufficient strains, BcA67 displayed a marginally enhanced susceptibility, compared to the B6 controls. The pursuit of further gene mapping in BcA67 would necessitate the generation of crosses with a highly resistant inbred strain, such as BALB/c. Based on the failure to identify any highly susceptible C5-sufficient mice, it appears that wild-type C5 alone is sufficient to confer resistance to infection with C. albicans in the experimental model used herein, and that the presence of any additional putative susceptibility or resistance alleles may only become relevant in the absence of functional C5.

The results of the whole-genome scan and the survey of the set of AcB/BcA RCS clearly illustrate the importance of C5 as the major gene controlling the differential susceptibility to systemic C. albicans infec-tions seen in A/J and B6 mice. The physiological relevance of a C5 deficiency in humans with systemic candidiasis is unclear, since although this deficiency renders patients increasingly susceptible to a range of microbial infections, an increased incidence of candidia-sis has not been reported.44_{As resistance to antifungal}

agents becomes a growing concern, a better under-standing of the host–pathogen interaction is invaluable for the development of any new therapies. Animal models provide a useful tool to study the response to systemic candidiasis and studies in mice have indicated the importance of the innate immune system, particu-larly the complement pathway, in controlling infection. The involvement of neutrophils in the initial stages of infection parallels the importance of neutrophils in human infections and validates the mouse as a physio-logically relevant model of systemic candidiasis.

Materials and methods

Mice

Inbred, pathogen-free 8- to 12-week-old A/J and C57BL/ 6J (B6) mice were purchased from The Jackson Labora-tory (Bar Harbor, ME, USA). [A/J
B6] F2progeny were

bred by systematic brother–sister mating of [A/J
B6] F1

mice. The AcB/BcA RCS were purchased from Emerillon Therapeutics (Montreal, Quebec, Canada). All housing and experimental procedures were approved by the Biotechnology Research Institute (BRI) Animal Care Committee, operating under the guidelines of the Canadian Council of Animal Care. Mice were age and sex matched for all experiments.

C. albicans_infections

We performed infections with C. albicans as described previously.8 _{C. albicans strain SC5314 was grown}

over-night in YPD medium (1% yeast extract, 2% Bacto Peptone and 2% dextrose) at 301C and harvested by centrifugation. Blastospores were washed twice and resuspended in phosphate-buffered saline (PBS) at the required density. For experimental infections, mice were injected via the tail vein with 200 ml of either 5
104_(low

dose) or 3
105_{(high dose) of C. albicans blastospores in}

PBS. For survival experiments, we monitored infected animals daily for up to 28 days after injection of a high dose of C. albicans. Moribund animals were killed and survival times were recorded for use in statistical analysis. For the determination of organ fungal load, we killed mice 48 h after injection with a low dose of C.

albicans. At the time of killing, mice were anesthetized

and exsanguinated by cardiac puncture. Serum was isolated by collection of blood in serum-gel microtubes (Sarstedt, Montreal, Quebec, Canada), followed by centri-fugation and storage at 201C until used to measure cytokine levels. To measure TNF-a levels during infec-tion, mice were injected with a low dose of C. albicans and killed at 48 h, with serum collected as described above. Tail biopsies were taken from all mice at the time of killing for all experiments and stored at 801C. Determination of organ fungal load

Target organs were removed aseptically and homoge-nized in 4 ml of PBS. We determined fungal load in the resulting suspensions by making serial 10-fold dilutions in PBS and plating 40 ml of each dilution on YPD-agar plates containing 34 mg/l chloramphenicol. Plates were incubated at 301C for 24–48 h. Total colony-forming units (CFU) were determined and counts were expressed as the log10CFU per organ. For the initial determination of

differences in fungal burden between the organs of A/J and B6, where we tested brain, heart, kidney, lung, liver, and spleen, the minimum detectable number of CFU was 17 per organ. For all other experiments the maximum sensitivity of this assay was 100 CFU per organ. For subsequent analyses, mice below the detection limit of the assay were assigned an arbitrary value of 100 CFU, the minimum number of detectable colonies. This assumption overestimates the mean for groups having titres below detectable levels.

Cytokine detection

We determined levels of IFN-g, IL-6, IL-12, and TNF-a in the serum of healthy and infected mice using a sandwich enzyme-linked immunosorbant assay (ELISA) (BD Bio-sciences, Mississauga, Ontario, Canada). Measurements were performed according to the manufacturer’s instruc-tions.

Genotyping

Genomic DNA was isolated from the tail tips of individual F2 mice, which we obtained at the time of

killing, as described previously.45 _{A total of 138}

poly-morphic microsatellite markers informative for A/J and B6, and distributed over all chromosomes except chromosome Y, were selected to give an approximately 10 cM coverage of the genome (Invitrogen, Burlington, Ontario, Canada) (Table 1). The genetic locations of these microsatellite markers were obtained from the Mouse

Genome Database Project (http://www.informatics.jax. org).27 _{We used the nonpolymorphic Y chromosome}

marker, SRY, to verify the gender of all the F2 mice.

Genotyping was performed using a standard PCR-based method with trace amounts of [32_{P]a-dATP, followed by}

separation on denaturing polyacrylamide gels.45

The A/J strain has a 2-bp deletion in exon 6 of the

Hc gene31 _{that introduces a BsgI restriction site (5}0

-GTGCAG(N)16-30). To determine the C5 status of the F2

mice, this polymorphism was analyzed by PCR ampli-fication of genomic DNA using the forward primer: 50_{-CCGAAGTTATCATTGGTCCTTT-3}0 _{and the reverse}

primer: 50_{-CCCCACCCTCTTCTGGTACT-3}0_{. The PCR}

product was digested with BsgI and visualized by electrophoresis on a 2% agarose gel followed by staining with ethidium bromide. The expected fragment size for wild-type samples was 446 bp, while the sizes for samples containing the deletion were 318 and 126 bp. Statistical analysis

An unpaired, two-tailed t-test was used to establish the significance of differences in mean CFU per organ bet-ween A/J and B6. One-way ANOVA and Dunnett’s multi-ple comparison test were used to determine the effect of genotype at C5 on the various phenotypes assessed in the F2and RCS mice. A P-value of 0.05 or less was

con-sidered significant. All statistical analyses were performed using GraphPad Prism version 4 (San Diego, CA, USA).

Genetic markers were assigned to and mapped within the chromosomes using multipoint linkage analysis in Mapmaker/EXP version 3.0.46_{A log}

10transformation of

the data for brain and kidney fungal load was performed in order to approximate a normal distribution. Genome-wide simple interval mapping between the transformed kidney and brain fungal load phenotypes and genetic markers was performed using Mapmaker/EXP and Mapmaker/QTL 1.1,46_{to identify quantitative trait loci}

(QTLs). Logarithm of odds (LOD) scores were calculated using the expectation/maximization algorithm in Map-maker/QTL. Initial linkage analyses were conducted using a free genetic model. Map Manager QT47_{was used}

to test for single-locus association between C5 genotype and the heart fungal load, serum TNF-a, and survival phenotypes assessed in additional F2 mice. To reduce

skewness and more closely approximate a normal distribution, we performed a log10 transformation on

the phenotypic data for each of these traits. Single-locus associations were tested by simple regression analysis between trait values and genotypes at the C5 locus and the significance of each potential association was measured using the likelihood ratio statistic (LRS, w2_).

LOD scores were calculated as w2_/2ln(10).

Acknowledgements

This work was supported by research grants from the Canadian Genetic Diseases Network to PG. PG is a James McGill Professor of Biochemistry and is a distinguished Investigator of the Canadian Institutes of Health Re-search. AT is supported by a studentship from the Natural Sciences and Engineering Research Council of Canada. Work in AM’s lab was supported by the Genomics and Health Initiative of the National research Council of Canada.

References

1 Mathews HL, Witek-Janusek L. Host defense against oral, esophageal, and gastrointestinal candidiasis. In: Calderone RA (ed). Candida and candidiasis. ASM Press: Washington, DC, 2002, pp 179–192.

2 Odds FC. Candida and candidosis 2nd edn. Baillia´ere Tindall: London, 1988.

3 Eggimann P, Garbino J, Pittet D. Epidemiology of Candida species infections in critically ill non-immunosuppressed patients. Lancet Infect Dis 2003; 3: 685–702.

4 Kullberg BJ, Filler SG. Candidemia. In: Calderone RA (ed).

Candida and candidiasis. ASM Press: Washington, DC, 2002, pp

327–340.

5 Choi EH, Foster CB, Taylor JG et al. Association between chronic disseminated candidiasis in adult acute leukemia and common IL4 promoter haplotypes. J Infect Dis 2003; 187: 1153–1156. 6 Lanza F. Clinical manifestation of myeloperoxidase deficiency.

J Mol Med 1998; 76: 676–681.

7 Tuite A, Mullick A, Gros P. Genetic analysis of innate immunity in resistance to Candida albicans. Genes Immun 2004; 5: 576–587.

8 Mullick A, Elias M, Picard S et al. Dysregulated inflammatory response to Candida albicans in a C5-deficient mouse strain.

Infect Immun 2004; 72: 5868–5876.

9 Cockayne A, Odds FC. Interactions of Candida albicans yeast cells, germ tubes and hyphae with human polymorphonuclear leucocytes in vitro. J Gen Microbiol 1984; 130 (Part 3): 465–471. 10 Baccarini M, Vecchiarelli A, Cassone A, Bistoni F. Killing of yeast, germ-tube and mycelial forms of Candida albicans by murine effectors as measured by a radiolabel release micro-assay. J Gen Microbiol 1985; 131 (Part 3): 505–513.

11 Aratani Y, Kura F, Watanabe H et al. Critical role of myeloperoxidase and nicotinamide adenine dinucleotide phosphate-oxidase in high-burden systemic infection of mice with Candida albicans. J Infect Dis 2002; 185: 1833–1837. 12 Aratani Y, Koyama H, Nyui S, Suzuki K, Kura F, Maeda N.

Severe impairment in early host defense against Candida

albicans in mice deficient in myeloperoxidase. Infect Immun

1999; 67: 1828–1836.

13 Netea MG, Van Der Graaf CA, Vonk AG, Verschueren I, Van Der Meer JW, Kullberg BJ. The role of toll-like receptor (TLR) 2 and TLR4 in the host defense against disseminated candidia-sis. J Infect Dis 2002; 185: 1483–1489.

14 Villamon E, Gozalbo D, Roig P, O’Connor JE, Fradelizi D, Gil ML. Toll-like receptor-2 is essential in murine defenses against

Candida albicans infections. Microbes Infect 2004; 6: 1–7.

15 Netea MG, Sutmuller R, Hermann C et al. Toll-like receptor 2 suppresses immunity against Candida albicans through induction of IL-10 and regulatory T cells. J Immunol 2004; 172: 3712–3718.

16 Bellocchio S, Montagnoli C, Bozza S et al. The contribution of the toll-like/IL-1 receptor superfamily to innate and adaptive immunity to fungal pathogens in vivo. J Immunol 2004; 172: 3059–3069.

17 Romani L. Immunology of invasive candidiasis. In: Calderone RA (ed) Candida and candidiasis. ASM Press: Washington DC, 2002, pp 223–241.

18 Gelfand JA, Hurley DL, Fauci AS, Frank MM. Role of complement in host defense against experimental dissemi-nated candidiasis. J Infect Dis 1978; 138: 9–16.

19 Ashman RB. A gene (Carg1) that regulates tissue resistance to

Candida albicans maps to chromosome 14 of the mouse. Microb Pathogenesis 1998; 25: 333–335.

20 Ashman RB, Fulurija A, Papadimitriou JM. A second Candida

albicans resistance gene (Carg2) regulates tissue damage, but

not fungal clearance, in sub-lethal murine systemic infection.

Microb Pathogenesis 1998; 25: 349–352.

21 Fortin A, Diez E, Rochefort D et al. Recombinant congenic strains derived from A/J and C57BL/6J: a tool for genetic dissection of complex traits. Genomics 2001; 74: 21–35.

22 Ashman RB. Candida albicans: pathogenesis, immunity and host defence. Res Immunol 1998; 149: 281–288; discussion 494–496.

23 Balish E, Wagner RD, Vazquez-Torres A, Pierson C, Warner T. Candidiasis in interferon-gamma knockout (IFN-gamma/) mice. J Infect Dis 1998; 178: 478–487.

24 Lavigne LM, Schopf LR, Chung CL, Maylor R, Sypek JP. The role of recombinant murine IL-12 and IFN-gamma in the pathogenesis of a murine systemic Candida albicans infection.

J Immunol 1998; 160: 284–292.

25 Netea MG, van Tits LJ, Curfs JH et al. Increased susceptibility of TNF-alpha lymphotoxin-alpha double knockout mice to systemic candidiasis through impaired recruitment of neu-trophils and phagocytosis of Candida albicans. J Immunol 1999; 163: 1498–1505.

26 Romani L, Mencacci A, Cenci E et al. Impaired neutrophil response and CD4+ T helper cell 1 development in interleukin 6-deficient mice infected with Candida albicans. J Exp Med 1996; 183: 1345–1355.

27 Blake JA, Richardson JE, Bult CJ, Kadin JA, Eppig JT. MGD: the mouse genome database. Nucleic Acids Res 2003; 31: 193–195.

28 Ashman RB, Fulurija A, Papadimitriou JM. Strain-dependent differences in host response to Candida albicans infection in mice are related to organ susceptibility and infectious load.

Infect Immun 1996; 64: 1866–1869.

29 Ashman RB, Bolitho EM, Papadimitriou JM. Patterns of resistance to Candida albicans in inbred mouse strains. Immunol

Cell Biol 1993; 71 (Part 3): 221–225.

30 Ashman RB, Papadimitriou JM, Fulurija A et al. Role of complement C5 and T lymphocytes in pathogenesis of disseminated and mucosal candidiasis in susceptible DBA/2 mice. Microb Pathogenesis 2003; 34: 103–113.

31 Wetsel RA, Fleischer DT, Haviland DL. Deficiency of the murine fifth complement component (C5). A 2-base pair gene deletion in a 50_{-exon. J Biol Chem 1990; 265: 2435–2440.}

32 Ashman RB, Fulurija A, Papadimitriou JM. Evidence that two independent host genes influence the severity of tissue damage and susceptibility to acute pyelonephritis in murine systemic candidiasis. Microb Pathogenesis 1997; 22: 187–192.

33 D’Eustachio P, Kristensen T, Wetsel RA, Riblet R, Taylor BA, Tack BF. Chromosomal location of the genes encoding complement components C5 and factor H in the mouse.

J Immunol 1986; 137: 3990–3995.

34 Cinader B, Dubiski S, Wardlaw AC. Distribution, inheritance, and properties of an antigen, Mub1, and its relation to hemolytic complement. J Exp Med 1964; 120: 897–924. 35 Janeway C, Travers P, Walport M, Shlomchik M.

Immunobiol-ogy. 5: The Immune System in Health and Disease 5th edn.

Garland Pub.: New York, 2001.

36 Kozel TR. Activation of the complement system by pathogenic fungi. Clin Microbiol Rev 1996; 9: 34–46.

37 Gerard C, Gerard NP. C5A anaphylatoxin and its seven transmembrane-segment receptor. Annu Rev Immunol 1994; 12: 775–808.

38 Blatteis CM, Li S, Li Z, Perlik V, Feleder C. Signaling the brain in systemic inflammation: the role of complement. Front Biosci 2004; 9: 915–931.

39 Ward PA. The dark side of C5a in sepsis. Nat Rev Immunol 2004; 4: 133–142.

40 Papadimitriou JM, Ashman RB. The pathogenesis of acute systemic candidiasis in a susceptible inbred mouse strain.

J Pathol 1986; 150: 257–265.

41 Ashman RB, Papadimitriou JM. Murine candidiasis. Patho-genesis and host responses in genetically distinct inbred mice.

Immunol Cell Biol 1987; 65 (Part 2): 163–171.

42 Rifkind D, Frey JA. Sex difference in antibody response of CFW mice to Candida albicans. Infect Immun 1972; 5: 695–698. 43 Ashman RB, Kay PH, Lynch DM, Papadimitriou JM. Murine

candidiasis: sex differences in the severity of tissue lesions are

not associated with levels of serum C3 and C5. Immunol Cell

Biol 1991; 69 (Part 1): 7–10.

44 Figueroa JE, Densen P. Infectious diseases associated with complement deficiencies. Clin Microbiol Rev 1991; 4: 359–395. 45 Mitsos LM, Cardon LR, Fortin A et al. Genetic control of

susceptibility to infection with Mycobacterium tuberculosis in mice. Genes Immun 2000; 1: 467–477.

46 Lander ES, Green P, Abrahamson J et al. MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations.

Genomics 1987; 1: 174–181.

47 Manly KF, Olson JM. Overview of QTL mapping software and introduction to map manager QT. Mamm Genome 1999; 10: 327–334.