INFECTION AND IMMUNITY, Nov. 2011, p. 4472–4479 Vol. 79, No. 110019-9567/11/$12.00 doi:10.1128/IAI.05666-11Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Next-Generation Computational Genetic Analysis: Multiple ComplementAlleles Control Survival after Candida albicans Infection�†

Gary Peltz,1* Aimee K. Zaas,2 Ming Zheng,1 Norma V. Solis,3 Mason X. Zhang,4 Hong-Hsing Liu,1Yajing Hu,1 Gayle M. Boxx,4 Quynh T. Phan,3 David Dill,5 and Scott G. Filler3,6

Department of Anesthesia, Stanford University School of Medicine, Stanford, California1; Department of Medicine, Duke UniversityMedical Center, Durham, North Carolina2; Department of Medicine, Los Angeles Biomedical Research Institute at

Harbor-UCLA Medical Center, Los Angeles, California3; Department of Biological Sciences,California State University-Long Beach, Long Beach, California4; Department of

Computer Science, Stanford University, Stanford, California5; andDavid Geffen School of Medicine at UCLA, Los Angeles, California6

Received 14 July 2011/Returned for modification 10 August 2011/Accepted 16 August 2011

Candida albicans is a fungal pathogen that causes severe disseminated infections that can be lethal inimmunocompromised patients. Genetic factors are known to alter the initial susceptibility to and severity of C.albicans infection. We developed a next-generation computational genetic mapping program with advancedfeatures to identify genetic factors affecting survival in a murine genetic model of hematogenous C. albicansinfection. This computational tool was used to analyze the median survival data after inbred mouse strainswere infected with C. albicans, which provides a useful experimental model for identification of host suscep-tibility factors. The computational analysis indicated that genetic variation within early classical complementpathway components (C1q, C1r, and C1s) could affect survival. Consistent with the computational results,serum C1 binding to this pathogen was strongly affected by C1rs alleles, as was survival of chromosomesubstitution strains. These results led to a combinatorial, conditional genetic model, involving an interactionbetween C5 and C1r/s alleles, which accurately predicted survival after infection. Beyond applicability toinfectious disease, this information could increase our understanding of the genetic factors affecting suscep-tibility to autoimmune and neurodegenerative diseases.

Genetic factors are known to alter susceptibility to and se-verity of Candida albicans infection in mice (1, 3, 22) andhumans (42). Therefore, characterizing genetic factors affect-ing host susceptibility to C. albicans infection is of great im-portance. Since systemic candidiasis in mice closely resemblesthe human disease, inbred mouse strains provide a useful ex-perimental model for identification of host susceptibility fac-tors. Although virtually all organs are infected, the kidney isthe major target, and the histopathology of infected lesions issimilar in mice and humans. Mutations in several immuneresponse genes have been associated with susceptibility tochronic mucocutaneous candidiasis in human families (14, 17,36, 48), and several have been verified in murine models.Differences in survival after hematogenous C. albicans infec-tion among inbred mouse strains have been associated withcomplement factor 5 (Hc or C5) alleles (1, 2, 4, 34). A 2-bpdeletion polymorphism at the 5� end of the C5 transcript shiftsits reading frame and causes �50% of inbred strains to be C5protein deficient (54). Disseminated candidiasis is rapidly fatalin C5-deficient strains because of uncontrolled fungal prolif-eration in most organs (34). Although C5 alleles make animportant contribution, several previous analyses indicated

that there are other genetic factors that affect the severity oftissue damage or survival after C. albicans infection (2, 38).However, no one has yet been able to identify these othergenetic factors.

Since its inception in 2004, haplotype-based computationalgenetic mapping (HBCGM) (30) has been used to identify thegenetic basis for many biomedical trait differences among in-bred mouse strains, including differences in gene expression(30), pharmacogenetic factors (19, 20, 58), susceptibility toinvasive aspergillosis (56) and respiratory syncytial virus infec-tions (47), analgesic medication (43) and inflammatory painresponses (26, 27), incisional wound biology (23, 24), and nar-cotic drug responses (12, 28, 29, 43). In a mapping experiment,a property of interest is measured in �10 inbred mouse strains;genetic factors are then predicted computationally by identi-fying genomic regions where the pattern of genetic variationcorrelates with the distribution of trait values among the in-bred strains (30). Despite multiple successes, this genetic map-ping method has been unable to identify the underlying geneticdifferences in other, more complex biologic systems (59). Thepaucity of genomic regions covered by the genetic map was asignificant contributor to these failures. The previous haplo-type map covered only �15% of the genes in the mouse ge-nome (30), and gene families were selected to enable analysesof specific phenotypes (i.e., drug metabolism). Also, the exist-ing haplotype block construction algorithm (30) rewarded theinclusion of more single-nucleotide polymorphisms (SNPs),penalized the generation of more haplotypes in a block, anddid not allow for overlapping blocks within a region. As aconsequence, a causative block could easily be missed (pro-

* Corresponding author. Mailing address: Department of Anesthe-sia, Stanford University School of Medicine, 300 Pasteur Dr., Stanford,CA 94305. Phone and fax: (650) 721-2487. E-mail: gpeltz@stanford.edu.

† Supplemental material for this article may be found at http://iai.asm.org/.

� Published ahead of print on 29 August 2011.

4472

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

ducing false-negative results) if another block in a region withfewer haplotypes and fewer SNPs was selected. A newHBCGM method with whole-genome coverage and an im-proved method for haplotype block construction were neededto enable a wider range of biomedical phenotypes (includinginfectious disease) to be evaluated. Therefore, we produced anext-generation version of the HBCGM method and used it toanalyze survival after hematogenous C. albicans infection in apanel of inbred mouse strains. The results led us to produce anovel combinatorial, conditional genetic model, involving aninteraction between C5 and C1s alleles, that accurately pre-dicted survival after infection.

MATERIALS AND METHODS

Survival after Candida albicans infection. All mouse experiments were ap-proved by the Los Angeles Biomedical Research Institute Animal Care and UseCommittee and were performed according to the Guide for the Care and Use ofLaboratory Animals (35a). Male mice were obtained from Jackson Laboratoriesand were used in survival studies at approximately 6 weeks of age. C. albicansstrain SC5314 was grown in yeast extract-peptone-dextrose (YPD) broth at 30°C.The yeast-phase organisms were washed twice in phosphate-buffered saline(PBS) and enumerated with a hemacytometer. To induce disseminated candidi-asis, 10 mice of each strain were inoculated via the lateral tail vein with 104 C.albicans cells per gram of body weight. The inoculum was confirmed by quanti-tative culture. In the survival experiments, the mice were monitored at least 3times daily, and moribund animals were euthanized humanely. The kidney fungalburden and myeloperoxidase activity were determined with a separate group ofinfected mice that were sacrificed 1 day after inoculation. The kidneys obtainedfrom these mice were harvested and homogenized in ice-cold PBS containingprotease inhibitor cocktail (Sigma-Aldrich), and an aliquot was cultured quan-titatively to measure the kidney fungal burden. The myeloperoxidase content ofthe kidney homogenates was measured to assess renal phagocyte accumulationas previously described (13). The kidney homogenates were clarified by centrif-ugation at 16,000 � g, and the myeloperoxidase content was determined using anenzyme immunoassay kit (Cell Sciences, Canton, MA). The log rank test wasused to compare the survival curves for different C57BL/6 and B6.CSS6 miceafter C. albicans infection (21).

Computational genetic mapping. The methods for producing the genetic mapand the new mapping method and the characterization of the method are de-scribed in the supplemental material. Using the new genomewide haplotype map,genetic factors were identified computationally using our previously describedmethods (30, 51). In brief, the pattern of genetic variation within each block wascorrelated with the distribution of trait values among the strains analyzed byusing analysis of variance (ANOVA)-based statistical modeling. P values fromthe ANOVA model and the corresponding genetic effect size were calculated foreach block (30, 51). The blocks were then ranked by their P values, and thosebelow an input threshold were used as candidate predictions. The SNPs withinthe blocks were annotated using Ensembl mouse genome annotation information(http://www.ensembl.org; NCBI mouse genome build 37), and our software au-tomatically identified SNPs causing nonsynonymous coding changes. The haplo-type pattern, chromosomal location, presence of nonsynonymous SNPs, andcalculated P values and genetic effect size for each block meeting the inputcriteria were also outputs of this program. For analysis of C. albicans survival, themedian survival was used as the phenotypic input, since the median is moreresistant to occasional outliers than the average. NZW mice had an extremelylong survival time (median, 13 days) relative to the other strains. This compli-cated the computational analysis, since all blocks with NZW-specific haplotypeswere identified. Therefore, the survival data were censored at 7 days for thisanalysis, which indicated that the strains had prolonged survival. This approachis very similar to that used for nonparametric analysis; it enabled genetic map-ping to produce more stable results that were unaffected by the extreme value forone strain.

Determination of complement alleles for inbred strains. Genomic DNAs fromthe following strains were purchased from Jackson Laboratory (ME): DBA/1J,LG/J, MRL/MpJ, NZW/LacJ, PL/J, SJL/J, and SM/J. Next, 9 genomic regionswere PCR amplified and sequenced using the following sets of primers: for C1rbon chromosome 6 (124,469,409 bp), GGAGGGAGTTGGGGGTCTAGT andAAGTGGAGGACACCTGTGCAA for amplification and AGTGGAAACAGGCAAGGGTCT and GCAAGACTTCAGTCGGGCAAT for sequencing; for

C1s on chromosome 6 (124,490,473 bp), CAGTGGCACAAAGCTGGAGTCand AGCACAGGAGGGAGAGGGATG for amplification and GCACAAAGCTGGAGTCTTGGA and AGGGAGAGGGATGGGAGGAG for se-quencing; for C1qb on chromosome 4 (136,436,623 bp), AGCTTCAAGACTACCCCACCTG and GGAGGCTCTAGGAGGCCCATT for amplificationand CTACCCCACCTGTGGTCACCT and CTGAACCCAGAGAGGCACAAG for sequencing; for C1qc on chromosome 4 (136,446,076 bp), CTCATAAGGTATTGATAAATGGCCACA and AGGTGTGGAGGGCCGATACAA for amplification and AAATGGCCACAGGAATAATACCA and GGCCGATACAAACAGAAGCAC for sequencing; for C1qc on chromosome 4(136,446,490 bp), TGGCCGTATGCGATGTGTAGT and TTCCTGGAAGAGGAAACTGGA for amplification and GTAGTAGAGGCCCGGCACTTC andAAGGGGAAGGAGAAAGATCATCA for sequencing; for C1qc on chromo-some 4 (136,448,449 bp), GAGGCTCCTCTTTATTCCCTTCT and AGTTGTGGTTTCCCCTCAGGT for amplification and GCTGCAAGGTCACCAGAGTCA and CTCAGGTACCCTCCTGAACCA for sequencing; for C1qa onchromosome 4 (136,453,706 bp), GGAGGAGAAAGGGGAAAGAGG and GGGGTGGGTGCTAGGGTTAAG for amplification and GGAAAGAGAAGCTGGGGACAA and AGGGCTAGGGGTTAGGGTCAA for sequencing; for C1son chromosome 6 (124,489,825 bp), CTGCCTCTGCTTCCTGAGTGA and TGCCTTGCCTTTGTGTGACTT for amplification and sequencing; and for C1s onchromosome 6 (124,484,450 bp), AGACAACTCTGTCCCGGCATT and GAAAATGTGAGAATGTCTGAAGATGC for amplification and sequencing.

All PCRs were performed using Novagen KOD Hot Start DNA polymeraseunder the following conditions: 95°C for 2 min; 40 cycles of 95°C for 20 s, 60°Cfor 10 s, and 70°C for 15 s; and a final step of 70°C for 2 min. The single amplifiedband for each genomic region was purified using a QIAquick PCR purificationkit (Qiagen). The only exception was chr. 6 region 124484450, for which the PCRproducts were gel fractionated and the band at 594 bp was purified for sequenc-ing. The alleles were determined by aligning the resulting sequence with thereference C57BL/6 sequence (52), using the ClustalW alignment program (46).

Measurement of serum anti-Candida antibodies and serum C1q binding ac-tivity. C. albicans SC5314 yeast cells were grown in YNB broth (Difco) at 30°Covernight, washed in PBS, and then frozen at �80°C in PBS as single-usealiquots. The presence of anti-C. albicans antibodies in mouse sera was deter-mined using a method previously used to detect human anti-C. albicans antibody(8). Serum C1q binding to C. albicans cells was also assessed using a modificationof a method previously used to measure human C1q binding to cultured cells(15). C. albicans cells were washed once with Veronal-buffered saline (VBS) andincubated at 1 � 107/ml for 1 h at 1°C with mouse serum that was diluted in 0.1ml VBS containing 0.1% gelatin, 1.5 mM CaCl2, and 1 mM MgCl2. The cellswere washed twice with PBS and incubated for 1 h at room temperature eitherwith fluorescein isothiocyanate (FITC)-labeled anti-mouse Ig(H�L) (Southern-Biotech, Birmingham, AL) for detection of cell-bound mouse antibody or withpolyclonal rabbit anti-mouse C1q antibody (Santa Cruz Biotechnology, SantaCruz, CA) followed by FITC-conjugated goat anti-rabbit antibody (Southern-Biotech) to detect cell-bound C1q. The detection antibodies were used at a 1:50dilution in 1% bovine serum albumin (BSA)-PBS. After being washed twice withPBS, C. albicans cells were resuspended in 1% BSA-PBS containing LDS 751(Invitrogen, Carlsbad, CA). The amount of cell-bound antibody or C1q wasquantified by flow cytometry using a Quanta SC MPL flow cytometer at a 488-nmexcitation wavelength (Beckman Coulter, Miami, FL) as described previously(9). C. albicans cells were identified with LDS 751, and single cells were thendefined by cell size with Coulter electronic volume technology and side scatter.The amount of mouse antibody or C1q bound to 5,000 single yeast cells wasmeasured quantitatively as the fluorescence intensity of cell-bound FITC-labeledantibody. A positive cell was defined as having a fluorescence intensity of �10fluorescence units. As controls, cells treated with FITC-labeled antibody alonewere used to measure the level of background binding, which was less than 10fluorescence units.

RESULTS

Next-generation computational genetic mapping method. A2.7-million-SNP database was generated from analysis of dataobtained from two different sources and was used to providegenomewide coverage (�95% of the genes in the murine ge-nome) for 16 inbred strains: 129S1/SvImJ, A/J, AKR/J, BALB/cJ, C3H/HeJ, C57BL/6J, CBA/J, DBA/2J, LP/J, NOD/ShiLtJ,NZO/HiLtJ, BALB/cByJ, FVB/NJ, BTBR T�tf/J, KK/HlJ,and NZW/LacJ. We also developed a new haplotype block

VOL. 79, 2011 COMPLEMENT ALLELES AND SURVIVAL DURING CANDIDIASIS 4473

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

construction method that reduced the possibility that a com-putational genetic mapping experiment would miss a true caus-ative haplotype block (i.e., produce false-negative results). Thenew “maximal” haplotype block construction method identifiesall patterns of genetic variation within a region by allowing thehaplotype blocks to overlap (see Fig. S1 in the supplementalmaterial). When the allelic data from all 16 strains were eval-uated, the maximal method generated 6-fold more blocks (n �580,565) than the prior method (n � 92,109) (see Table S1).Our previous method generated a single haplotype map byusing all available allelic data for all strains (30), and this mapwas used for all analyses. However, phenotypic data are usuallyavailable only for a subset of the strains in a typical mappingexperiment, and inclusion of irrelevant alleles can disrupt hap-lotypic patterns that are uniform among the strains of interest.The 30,000-fold improvement in the computational efficiencyof this implementation enabled customized haplotype blocksto be produced dynamically in real time for the strains withavailable phenotypic data. We also found that the use ofwhole-genome sequencing data enabled the haplotype map toprovide a more complete representation of the pattern of ge-netic variation for new strains than could be obtained usinggenotyping arrays that characterize only previously knownSNPs (see Tables S2 and S3). As described in the supplementalmaterial, this new genetic map and mapping method exhibitsuperior performances over those of our prior genetic mapping

method (see Fig. S4) and other available methods for associ-ation mapping (see Tables S4 and S5).

Survival after fungal infection. We examined the survival of15 inbred mouse strains after hematogenous C. albicans infec-tion. There was substantial variation in strain survival; themedian survival for 7 strains was �3 days, while other strainssurvived for 10 to 13 days (Fig. 1). One day after infection, wealso measured the kidney fungal burden and kidney myeloper-oxidase activity (as a measure of phagocyte accumulation).Kidney fungal burden (P value, 0.007) and myeloperoxidaseactivity (P value, 0.03) were both inversely correlated withsurvival, while the normalized myeloperoxidase activity (rela-tive to fungal burden) was directly correlated with survival (Pvalue, 0.01). Since both measurements were made at a singletime after infection, they provide an imperfect assessment of adynamic and evolving response to infection. However, thesecorrelations verify the expected result that the host’s abilitiesto recruit phagocytes and limit fungal growth in the kidney aredeterminants of survival. C5 alleles could explain some, but notall, of the observed interstrain differences. All C5-deficientstrains had a median survival of �3 days after infection, whilethe median survival among the C5-sufficient strains rangedfrom 3 to 13 days (Fig. 2). The large variation in survivalamong the C5-sufficient strains indicates that genetic factorsother than C5 could affect survival.

The median survival data for the 14 strains with available

FIG. 1. Measured survival times, kidney myeloperoxidase activities, and kidney fungal burdens for a panel of inbred mouse strains. (Top)Survival time for each of 10 mice after hematogenous infection with C. albicans for each of the 15 inbred strains tested. The red bars representmedian survival times. (Bottom) Scatter plots for kidney myeloperoxidase activity and kidney fungal burden versus survival time. The x axisrepresents the survival time (days), and the y axis represents the log10-transformed measured kidney myeloperoxidase activity or kidney fungalburden measured 1 day after infection. The red lines represent the fitted regression lines, and both have Pearson’s correlation coefficients of �0.67.Myeloperoxidase activity and kidney fungal burden both show statistically significant inverse correlations with survival (P values of 0.03 and 0.007,respectively). On the right is a scatter plot for normalized myeloperoxidase activity, which is the myeloperoxidase activity divided by the kidneyfungal burden (pg/CFU), relative to the survival time (days). The normalized myeloperoxidase activity was directly correlated with survival time(P value, 0.01).

4474 PELTZ ET AL. INFECT. IMMUN.

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

allelic information were analyzed using the next-generationcomputational genetic mapping method. Two haplotype blocksencoding C1q components (C1qa and C1qc) had the highestcorrelation with median survival (P � 1 � 10�5). C1rb and C1s(P � 0.0002) were also among the 13 most highly correlatedgenes (Fig. 2). All three C1q component genes (C1qa, -b, and-c) are adjacent to each other on chromosome 4 (136.4 Mb),while C1rb and C1s are carried by opposite strands of the samegenomic region on chromosome 6 (124.5 Mb) (see Fig. S2 inthe supplemental material). There was high linkage disequilib-rium among SNP alleles within each of these two genomicregions. SNPs within the four C1 genes gave the followingalterations: Glu293Gln, Asn122Arg, Arg68Lys, and Asp82Glysubstitutions within C1s; an Asn359His substitution within C1r;a Thr16Ile substitution in C1qa; and Val208Ala, Arg70Gln,

and Pro10Gln substitutions in C1qc (see Fig. S2B). C1q bind-ing to an antigen-antibody complex induces the Ca2�-depen-dent assembly of a C1s-C1r-C1s-C1r tetramer (5), which acti-vates the classical complement pathway. C1s is a serineprotease with a modular structure; all 4 nonsynonymous C1sSNPs are located within the NH2-terminal bone morphogenicprotein and epidermal growth factor (EGF)-like modules(CUB1-EGF-CUB2) (reviewed in reference 5). This Ca2�

binding region of C1s mediates the interaction with C1r andC1q, which is essential for C1s activation (18). The haplotypicgroupings created by the C1rs and C1q alleles are very similar(see Fig. S2). However, for the reasons discussed below, theC1s haplotype (Fig. 2) was used in the subsequent analyses.Nevertheless, it is possible that C1q haplotypes (which have avery similar pattern to those for C1rs) have an independent

FIG. 2. (Top) Median survival after hematogenous C. albicans infection for the 14 indicated strains. The C5-deficient strains are shownin red, and the C1s haplotypes of the C5-sufficient strains are indicated. The amino acids at the indicated positions for the two C1s haplotypesare also shown. All C5-deficient strains survived for �3 days after infection, while the C1s haplotype influenced the survival of theC5-sufficient strains. BALB/cByJ mice have a different C1s haplotype (due to a C allele at SNP NES16184810) from the other C5-sufficientstrains, which is why the data for this strain are shown with a bar with a unique color. (Bottom) The survival data were analyzed bycomputational haplotype-based genetic mapping, and the 15 genes with the most correlation are shown. The columns show the gene symbol,calculated P value and genetic effect size, chromosome, and locations of the starting and ending positions for each correlated haplotypeblock. The haplotype pattern within each haplotypic block is also shown; each rectangle represents a single strain that appears in the sameorder as in the bar graph above. Each haplotype within a block is represented by a rectangle of a different color; strains with rectangles ofthe same color have the same haplotype.

VOL. 79, 2011 COMPLEMENT ALLELES AND SURVIVAL DURING CANDIDIASIS 4475

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

effect on survival, but analyzing this would require a muchlarger strain panel.

C1 allelic differences have functional consequences. Thecomplement system can be activated by three distinct mecha-nisms: C1 binding to the Fc region of an antigen-antibodycomplex activates the classical pathway, while the alternativeand mannose-binding lectin pathways are activated by the di-rect binding of other complement proteins to the pathogensurface (49, 50). Of relevance here, the classical pathway playsan important role in host defense against many types of patho-gens (55); it is the dominant pathway for protecting miceagainst Streptococcus pneumoniae infection (10). Bacteria canactivate the classical pathway through direct binding of C1q tothe bacterial surface (11) or via naturally occurring IgM anti-bodies bound to the bacterial surface (10). Similarly, naturallyoccurring mannan-specific human IgG antibodies can activatethe classical complement pathway and induce C3 deposition onC. albicans (57). Since the allelic differences induced significantamino acid changes in C1 components, we compared C1 func-tional activities in sera obtained from 5 strains by measuringC1q binding to Candida albicans. In our initial experiment, theamount of C1q bound was �10-fold higher in SJL/J serum thanin CBA/J serum (Fig. 3A and B). Furthermore, the sera of all3 strains with C1s haplotype 1 (C57BL/6, 129S1/SvImJ, andCBA/J) had low levels of C1q binding activity, while bothstrains (SJL/J and LG/J) with C1s haplotype 2 had high levelsof C1q binding (Fig. 3B and C), and this strain-specific differ-

ence was reproducible in multiple independent experiments.Although LG/J and SJL/J sera had higher C1q binding activi-ties, they had lower levels of anti-Candida antibody thanCBA/J sera (Fig. 3D). Thus, the interstrain differences in thelevel of C1q binding to C. albicans were independent of theamount of naturally occurring anti-Candida antibody but werecorrelated with the C1 haplotype and with survival after C.albicans infection.

C57BL/6 and A/J mice have different C1q and C1r/s alleles.Therefore, C1q binding to C. albicans was measured using seraobtained from two chromosome substitution strains (CSS) thatwere derived from these 2 strains and had selectively alteredC1q (chromosome 4) or C1rs (chromosome 6) alleles. EachCSS strain is homosomic for a single specified A/J chromo-some on an otherwise C57BL/6 (C5-sufficient) genetic back-ground (35). At all serum dilutions tested, A/J and B6.CSS4(both C1q haplotype 2) sera had higher levels of C1q bindingthan C57BL/6 (C1q haplotype 1) sera (Fig. 4A). The level ofC1q binding in B6.CSS6 (C1rs haplotype 2 and C1q haplotype1) sera was substantially lower than that in C57BL/6 sera. TheC1qa-C1qb-C1qc and C1r-C1s proteins have different haplo-types in B6.CSS6 mice, and all 4 nonsynonymous C1s SNPs arelocated within the coding sequence for the Ca2� binding re-gion that mediates the interaction of C1s with C1r and C1q (5,18). Thus, the interaction between C1 components with differ-ent haplotypes likely reduces the formation or stability of the

FIG. 3. Candida albicans cells (1 � 107/ml) were incubated for 1 h at 1°C in 0.05 ml calcium-sufficient complement activation buffer containingthe indicated amounts of mouse serum and then washed. Cell-bound C1q was detected using polyclonal rabbit anti-mouse C1q antibody andFITC–goat anti-rabbit antibody. C1q binding was either visualized by immunofluorescence microscopy (A) or quantified by flow cytometry (B andC), where �5,000 cells were analyzed and a positive cell was defined as having a fluorescence intensity of �10. Data for two independentexperiments using dilutions of SJL/J and CBA/J sera are shown in panel B, while the averages standard deviations (SD) for 3 to 5 independentmeasurements for the indicated strain and serum concentrations are shown in panel C. (D) Amounts of naturally occurring anti-Candida antibodyin sera obtained from five inbred strains. After the indicated dilution of serum was incubated with Candida albicans cells, the amount of boundantibody was measured by flow cytometry. Candida cells with �10 fluorescence units were scored as positive cells. Each data point indicates theaverage % positive cells for three independent measurements SD, and data from two different experiments are shown.

4476 PELTZ ET AL. INFECT. IMMUN.

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

C1 complex that is bound to the fungus, thereby decreasingC1q binding.

We also examined the survival of CSS mice after dissemi-nated C. albicans infection. Consistent with their reduced levelof serum C1q binding, B6.CSS6 mice exhibited remarkablypoor survival after disseminated C. albicans infection, whichwas substantially shorter than that of C57BL/6 mice (Fig. 4B);the log rank test P value was 4 � 10�6 when their survivalcurves were compared. In contrast, the survival of B6.CSS4mice was comparable to that of C57BL/6 mice. The CSS resultsand the other data indicate that C1rs alleles have a major effecton C1q binding to the fungal pathogen and on survival after C.albicans infection, which appears to also depend upon the C1qand C5 alleles.

A combinatorial genetic model predicts survival. These re-sults led us to propose a conditional, multiallelic model (in-volving 2 complement genes) for predicting inbred strain sur-vival after hematogenous C. albicans infection. Fourconsiderations formed the basis for this genetic model: (i)because complement pathway activation is essential for con-trolling fungal infection, the effect of a C5 deficiency will pre-dominate over allelic differences within C1 components; (ii)among the C5-sufficient strains, allelic differences within C1components that affect complement binding to the pathogenand the efficiency of classical complement pathway activationcould alter host survival after infection; (iii) C1s alleles causedthe largest number of (and the most significant) amino acidchanges; and (iv) Although C1q alleles had the highest level ofcorrelation with survival among all analyzed strains, C1s allelesexhibited the highest correlation with survival among the sevenstrains that were C5 sufficient. According to this model, allC5-deficient strains will have poor survival (�3 days) afterinfection, while C5-sufficient strains will have a longer survival(�3 days) that is dependent upon the C1s haplotype, as indi-cated by the nonsynonymous amino acid changes shown in Fig.2. According to this model, C5-sufficient strains with C1s hap-lotype 1 will have a median survival time of 4 1 days afterinfection, while those with C1s haplotype 2 will have a longermedian survival time (9 3 days) (Fig. 2). In brief, this modelpredicts that genetic variation within the C1s and C5 genes will

determine whether a strain will have short, medium, or longsurvival after C. albicans infection. This model was tested bymeasuring the survival after hematogenous C. albicans infec-tion of four C5-sufficient strains whose C1s haplotypes wereunknown. The C1s alleles for these 4 strains and for LG/J mice(for which we had survival but not allelic data) were thendetermined. The C1s and C5 allele-based conditional geneticmodel accurately predicted the measured survival for all 5strains (Table 1). Note that the C1q haplotype was the same asthe C1s haplotype in these strains (i.e., all C1q haplotype 2strains were also C1s haplotype 2). Although the C1q haplo-type could have an independent effect on survival after infec-tion, its impact will have to be analyzed using recombinantstrains with targeted substitutions in both C1q and C1rs alleleson a C5-sufficient genetic background.

DISCUSSION

A next-generation computational mapping program ana-lyzed a murine genetic model of survival after hematogenousCandida albicans infection and indicated that genetic variationwithin early classical complement pathway components (C1q,C1r, and C1s) could affect survival. Abundant experimentaldata supported the genetic hypothesis that C1 alleles affectsurvival: C1 binding measurements provided a direct demon-stration that C1 alleles affect C1 binding to the fungal patho-gen, C1 (and C5) alleles could prospectively predict strainsurvival after infection, and C1 allelic substitutions in consomicstrains had a major effect on the level of C1 binding to thefungal pathogen and on survival after infection. These dataindicate that C1 alleles are at least some of the genetic factorsthat were postulated to affect survival but could not be iden-tified using the available genetic discovery methods (2).

Although we have shown the combinatorial effect that C1and C5 alleles have on survival in this model, their effect on theresponse to other infectious agents remains to be determined.It is also possible that other genetic factors, especially in dif-ferent genetic backgrounds or with different phenotypes, couldalso impact the outcome after C. albicans infection. For exam-ple, a genetic analysis was performed using kidney fungal bur-den as an outcome after C. albicans infection, utilizing C5-deficient mouse strains, and this revealed that other geneticloci could affect the outcome (38). However, given the impor-tant role of the complement pathway in the response to mul-tiple infectious agents, these findings should stimulate othersto investigate whether this genetic mechanism impacts theresponse to other infectious agents. We do not yet know if a

TABLE 1. Comparison of measured median survival rates after C.albicans infection among 5 strains (10 mice per strain) relative tothose predicted using the C5 and C1s composite genetic modela

Strain C1s haplotypeMedian (SD) survival (days)

Predicted Measured

SM/J 1 4 � 1 3MRL 2 9 3 8DBA/1 2 9 3 6SJL/J 2 9 3 8LG/J 2 9 3 12

a All strains were positive for the C5 allele.

FIG. 4. (A) C1q binding to C. albicans was quantified by flow cy-tometry as described in the legend to Fig. 3, using dilutions of seraobtained from C57BL/6, A/J, or the indicated CSS mice. The C5, C1q,and C1rs alleles for each strain are indicated. (B) Survival after hema-togenous C. albicans infection for A/J, C57BL/6, or the indicated CSSstrain mice (10 mice per group).

VOL. 79, 2011 COMPLEMENT ALLELES AND SURVIVAL DURING CANDIDIASIS 4477

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

similar genetic mechanism involving the complement pathwaywill be found in humans. However, if an allelic effect is iden-tified, genotyping at-risk individuals could identify those thatwould best benefit from increased monitoring or preventativetherapy.

Beyond its potential applicability to other infectious dis-eases, this combinatorial genetic model could provide insightinto the genetic architecture of susceptibility to autoimmuneand neurodegenerative diseases. The relationship betweencomplement alleles and autoimmune disease susceptibility inhumans and mice has been puzzling. While kidney inflamma-tion in human systemic lupus erythematosus and related mu-rine models is driven by immune complex deposition and com-plement activation, paradoxically, C1q-deficient humans havethe highest risk for lupus susceptibility (7, 31). Similarly, a C1qknockout can accelerate the development of renal disease inmice with certain genetic backgrounds, but its effect on auto-immune disease expression is highly strain dependent (6, 33).Investigators have partially explained these paradoxical obser-vations by proposing that early complement proteins play mul-tiple roles in autoimmune disease pathogenesis (reviewed inreference 31). Besides driving complement-dependent tissueinflammation, C1q is also a pattern recognition protein thatfacilitates the clearance of apoptotic cells (6, 15, 16, 33), and adeficiency in this activity could facilitate the development ofautoimmunity.

The combinatorial allelic model described here provides anew mechanism for modulating complement pathway activity,which could explain the significant effect that genetic back-ground has on the autoimmune phenotype in a C1q knockoutmouse (6, 33). Also, the MRL and NZW strains provide pro-totypic models for human lupus, and both have C5 and C1alleles that favor efficient classical complement pathway acti-vation, which may partially explain why they spontaneouslydevelop autoimmune disease (39, 39a). Despite intensive in-vestigation, we do not fully understand the genetic basis for therenal disease that spontaneously develops in F1(NZB �NZW) mice but not in either parent (60). We previously dem-onstrated that the NZB Ifi202 allele promotes autoantibodyproduction. However, congenic C57BL/6 mice expressing theNZB Ifi202 allele (B6.NZBIfi202) produce multiple autoanti-bodies but do not develop renal disease, while NZW miceexpressing the NZB Ifi202 allele develop renal pathology at thesame rate as F1(NZB � NZW) mice (41). Similarly, the lprmutation promotes autoimmunity in MRL/lpr mice, andC57BL/6 mice expressing the lpr mutation develop high auto-antibody titers but do not develop renal disease (53). Thissuggests that the complement alleles in NZW or MRL micecould be required for autoimmune disease expression. It wasalso demonstrated recently that early classical complementpathway components (C1q and C3) regulate synapse formationwithin the central nervous system and retina, that C1q binding“tags” selected synapses for elimination during development,and that C1q is an essential mediator of neurodegeneration ina murine glaucoma model mediated by retina-specific geneticfactors (44). Polymorphisms in human C1q, C1r, and C1s areknown to exist, and allelic associations with systemic lupuserythematosis and early Alzheimer’s disease have been inves-tigated (25, 32, 37, 40). Whether these polymorphisms influ-ence the susceptibility of hospitalized patients to disseminated

candidiasis or the outcome of this disease remains to be de-termined. Nevertheless, characterizing the combinatorial inter-action between allelic variants in different complement pro-teins and with other genetic factors required for autoimmunedisease expression (major histocompatibility complex [MHC],Ifi202, and Fas/lpr) in mice could increase our understanding ofsusceptibility to autoimmune and neurodegenerative diseasesas well as disseminated candidiasis.

ACKNOWLEDGMENTS

G.P. was supported by funding from a transformative RO1 award(1R01DK090992-01) from the NIDDK, and S.G.F. was supported inpart by NIH grants 5R21DE019414 and 5R01AI054928.

REFERENCES

1. Ashman, R. B., E. M. Bolitho, and J. M. Papadimitriou. 1993. Patterns ofresistance to Candida albicans in inbred mouse strains. Immunol. Cell Biol.71:221–225.

2. Ashman, R. B., A. Fulurija, and J. M. Papadimitriou. 1997. Evidence thattwo independent host genes influence the severity of tissue damage andsusceptibility to acute pyelonephritis in murine systemic candidiasis. Microb.Pathog. 22:187–192.

3. Ashman, R. B., A. Fulurija, and J. M. Papadimitriou. 1996. Strain-depen-dent differences in host response to Candida albicans infection in mice arerelated to organ susceptibility and infectious load. Infect. Immun. 64:1866–1869.

4. Ashman, R. B., et al. 2003. Role of complement C5 and T lymphocytes inpathogenesis of disseminated and mucosal candidiasis in susceptible DBA/2mice. Microb. Pathog. 34:103–113.

5. Bally, I., et al. 2009. Identification of the C1q-binding sites of human C1r andC1s: a refined three-dimensional model of the C1 complex of complement.J. Biol. Chem. 284:19340–19348.

6. Botto, M., et al. 1998. Homozygous C1q deficiency causes glomerulonephri-tis associated with multiple apoptotic bodies. Nat. Genet. 19:56–59.

7. Botto, M., et al. 2009. Complement in human diseases: lessons from com-plement deficiencies. Mol. Immunol. 46:2774–2783.

8. Boxx, G. M., T. R. Kozel, C. T. Nishiya, and M. X. Zhang. 2010. Influence ofmannan and glucan on complement activation and C3 binding by Candidaalbicans. Infect. Immun. 78:1250–1259.

9. Boxx, G. M., C. T. Nishiya, T. R. Kozel, and M. X. Zhang. 2009. Character-istics of Fc-independent human antimannan antibody-mediated alternativepathway initiation of C3 deposition to Candida albicans. Mol. Immunol.46:473–480.

10. Brown, J. S., et al. 2002. The classical pathway is the dominant complementpathway required for innate immunity to Streptococcus pneumoniae infec-tion in mice. Proc. Natl. Acad. Sci. U. S. A. 99:16969–16974.

11. Butko, P., A. Nicholson-Weller, and M. R. Wessels. 1999. Role of comple-ment component C1q in the IgG-independent opsonophagocytosis of groupB streptococcus. J. Immunol. 163:2761–2768.

12. Chu, L. F., et al. 2009. From mouse to man: the 5-HT3 receptor modulatesphysical dependence on opioid narcotics. Pharmacogenet. Genomics 19:193–205.

13. Ejzykowicz, D. E., et al. 2010. Role of Aspergillus fumigatus DvrA in host cellinteractions and virulence. Eukaryot. Cell 9:1432–1440.

14. Ferwerda, B., et al. 2009. Human dectin-1 deficiency and mucocutaneousfungal infections. N. Engl. J. Med. 361:1760–1767.

15. Fraser, D. A., A. K. Laust, E. L. Nelson, and A. J. Tenner. 2009. C1qdifferentially modulates phagocytosis and cytokine responses during inges-tion of apoptotic cells by human monocytes, macrophages, and dendriticcells. J. Immunol. 183:6175–6185.

16. Fraser, D. A., K. Pisalyaput, and A. J. Tenner. 2010. C1q enhances microglialclearance of apoptotic neurons and neuronal blebs, and modulates subse-quent inflammatory cytokine production. J. Neurochem. 112:733–743.

17. Glocker, E. O., et al. 2009. A homozygous CARD9 mutation in a family withsusceptibility to fungal infections. N. Engl. J. Med. 361:1727–1735.

18. Gregory, L. A., N. M. Thielens, G. J. Arlaud, J. C. Fontecilla-Camps, and C.Gaboriaud. 2003. X-ray structure of the Ca2�-binding interaction domain ofC1s. Insights into the assembly of the C1 complex of complement. J. Biol.Chem. 278:32157–32164.

19. Guo, Y. Y., et al. 2007. In vitro and in silico pharmacogenetic analysis in mice.Proc. Natl. Acad. Sci. U. S. A. 104:17735–17740.

20. Guo, Y. Y., et al. 2006. In silico pharmacogenetics: warfarin metabolism. Nat.Biotechnol. 24:531–536.

21. Harrington, D. P., and T. R. Flemming. 1982. A class of rank test proceduresfor censored survival data. Biometrika 69:553–566.

22. Hector, R. F., J. E. Domer, and E. W. Carrow. 1982. Immune responses toCandida albicans in genetically distinct mice. Infect. Immun. 38:1020–1028.

4478 PELTZ ET AL. INFECT. IMMUN.

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

23. Hu, Y., et al. 2010. The role of IL-1 in wound biology. I. Murine in silico andin vitro experimental analysis. Anesth. Analg. 111:1525–1533.

24. Hu, Y., et al. 2010. The role of IL-1 in wound biology. II. In vivo and humantranslational studies. Anesth. Analg. 111:1534–1542.

25. Kamboh, M. I., and R. E. Ferrell. 1987. Genetic studies of low-abundancehuman plasma proteins. VII. Heterogeneity of the C1S subcomponent of thefirst complement component. J. Immunogenet. 14:231–238.

26. LaCroix-Fralish, M. L., et al. 2009. The 3 subunit of the Na�,K�-ATPaseaffects pain sensitivity. Pain 144:294–302.

27. Li, X., et al. 2010. Expression genetics identifies spinal mechanisms support-ing formalin late phase behaviors. Mol. Pain 6:11.

28. Liang, D., et al. 2006. A genetic analysis of opioid-induced hyperalgesia inmice. Anesthesiology 104:1054–1062.

29. Liang, D. Y., G. Liao, G. Lighthall, G. Peltz, and J. D. Clark. 2006. Geneticvariants of the P-glycoprotein gene Abcb1b modulate opioid-induced hyper-algesia, tolerance and dependence. Pharmacogenet. Genomics 16:825–835.

30. Liao, G., et al. 2004. In silico genetics: identification of a functional elementregulating H2-Ea gene expression. Science 306:690–695.

31. Manderson, A. P., M. Botto, and M. J. Walport. 2004. The role of comple-ment in the development of systemic lupus erythematosus. Annu. Rev. Im-munol. 22:431–456.

32. Martens, H. A., et al. 2009. Analysis of C1q polymorphisms suggests associ-ation with systemic lupus erythematosus, serum C1q and CH50 levels anddisease severity. Ann. Rheum. Dis. 68:715–720.

33. Mitchell, D. A., et al. 2002. C1q deficiency and autoimmunity: the effects ofgenetic background on disease expression. J. Immunol. 168:2538–2543.

34. Mullick, A., et al. 2004. Dysregulated inflammatory response to Candidaalbicans in a C5-deficient mouse strain. Infect. Immun. 72:5868–5876.

35. Nadeau, J. H., J. B. Singer, A. Matin, and E. S. Lander. 2000. Analysingcomplex genetic traits with chromosome substitution strains. Nat. Genet.24:221–225.

35a.National Research Council. 1996. Guide for the care and use of laboratoryanimals. National Academy Press, Washington, DC.

36. Puel, A., et al. 2011. Chronic mucocutaneous candidiasis in humans withinborn errors of interleukin-17 immunity. Science 332:65–68.

37. Racila, E., et al. 2008. A polymorphism in the complement component C1qAcorrelates with prolonged response following rituximab therapy of follicularlymphoma. Clin. Cancer Res. 14:6697–6703.

38. Radovanovic, I., A. Mullick, and P. Gros. 2011. Genetic control of suscep-tibility to infection with Candida albicans in mice. PLoS One 6:e18957.

39. Reilly, C. M., and G. S. Gilkeson. 2002. Use of genetic knockouts to mod-ulate disease expression in a murine model of lupus, MRL/lpr mice. Immu-nol. Res. 25:143–153.

39a.Rigby, R. J., M. M. Fernando, and T. J. Vyse. 2006. Mice, humans andhaplotypes—the hunt for disease genes in SLE. Rheumatology 45:1062–1067.

40. Rosenmann, H., et al. 2003. A polymorphism in the complement componentC1r is not associated with sporadic Alzheimer’s disease. Neurosci. Lett.336:101–104.

41. Rozzo, S. J., et al. 2001. Evidence for an interferon-inducible gene, Ifi202, inthe susceptibility to systemic lupus. Immunity 15:435–443.

42. Shepherd, M. G., R. T. Poulter, and P. A. Sullivan. 1985. Candida albicans:biology, genetics, and pathogenicity. Annu. Rev. Microbiol. 39:579–614.

43. Smith, S. B., et al. 2008. Quantitative trait locus and computational mappingidentifies Kcnj9 (GIRK3) as a candidate gene affecting analgesia from mul-tiple drug classes. Pharmacogenet. Genomics 18:231–241.

44. Stevens, B., et al. 2007. The classical complement cascade mediates CNSsynapse elimination. Cell 131:1164–1178.

45. Thiel, S. 2007. Complement activating soluble pattern recognition moleculeswith collagen-like regions, mannan-binding lectin, ficolins and associatedproteins. Mol. Immunol. 44:3875–3888.

46. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W:improving the sensitivity of progressive multiple sequence alignment throughsequence weighting, position-specific gap penalties and weight matrix choice.Nucleic Acids Res. 22:4673–4680.

47. Tregoning, J. S., et al. 2010. Genetic susceptibility to the delayed sequelae ofRSV infection is MHC-dependent, but modified by other genetic loci. J. Im-munol. 185:5384–5391.

48. van de Veerdonk, F. L., et al. 2011. STAT1 mutations in autosomal dominantchronic mucocutaneous candidiasis. N. Engl. J. Med. 365:54–61.

49. Walport, M. J. 2001. Complement. First of two parts. N. Engl. J. Med.344:1058–1066.

50. Walport, M. J. 2001. Complement. Second of two parts. N. Engl. J. Med.344:1140–1144.

51. Wang, J., and G. Peltz. . Haplotype-based computational genetic analysis inmice, p. 51–70. Computational genetics and genomics: new tools for under-standing disease. Humana Press Inc., Totowa, NJ.

52. Waterston, R. H., et al. 2002. Initial sequencing and comparative analysis ofthe mouse genome. Nature 420:520–562.

53. Watson, M. L., et al. 1992. Genetic analysis of MRL-lpr mice: relationship ofthe Fas apoptosis gene to disease manifestations and renal disease-modifyingloci. J. Exp. Med. 176:1645–1656.

54. Wetsel, R. A., D. T. Fleischer, and D. L. Haviland. 1990. Deficiency of themurine fifth complement component (C5). A 2-base pair gene deletion in a5�-exon. J. Biol. Chem. 265:2435–2440.

55. Winkelstein, J. A. 1984. Complement and the host’s defense against thepneumococcus. Crit. Rev. Microbiol. 11:187–208.

56. Zaas, A. K., et al. 2008. Plasminogen alleles influence susceptibility to inva-sive aspergillosis. PLoS Genet. 4:e1000101.

57. Zhang, M. X., D. M. Lupan, and T. R. Kozel. 1997. Mannan-specific immu-noglobulin G antibodies in normal human serum mediate classical pathwayinitiation of C3 binding to Candida albicans. Infect. Immun. 65:3822–3827.

58. Zhang, X., et al. 2011. In silico and in vitro pharmacogenetics: aldehydeoxidase rapidly metabolizes a p38 kinase inhibitor. Pharmacogenomics J.11:15–24.

59. Zheng, M., S. S. Shafer, G. Liao, H.-H. Liu, and G. Peltz. 2009. Computa-tional genetic mapping in mice: ‘the ship has sailed.’ Sci. Transl. Med. 1:3ps4.

Editor: G. S. Deepe, Jr.

VOL. 79, 2011 COMPLEMENT ALLELES AND SURVIVAL DURING CANDIDIASIS 4479

on March 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from