of March 26, 2018.This information is current as Association

In Vitro Survival and Human Genetic Infections: Evidence from Bacterialpyogenes

StreptococcusImportant for Pathogenesis of Acquisition of Complement Factor H Is

Seppo Meri, Juha Kere and T. Sakari JokirantaSuvilehto, Satu Massinen, Matti Karppelin, Irma Järvelä, Karita Haapasalo, Jaana Vuopio, Jaana Syrjänen, Jari

http://www.jimmunol.org/content/188/1/426doi: 10.4049/jimmunol.1102545December 2011;

2012; 188:426-435; Prepublished online 2J Immunol 

MaterialSupplementary

5.DC1http://www.jimmunol.org/content/suppl/2011/12/02/jimmunol.110254

Referenceshttp://www.jimmunol.org/content/188/1/426.full#ref-list-1

, 28 of which you can access for free at: cites 46 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2011 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Acquisition of Complement Factor H Is Important forPathogenesis of Streptococcus pyogenes Infections: Evidencefrom Bacterial In Vitro Survival and Human GeneticAssociation

Karita Haapasalo,*,† Jaana Vuopio,‡ Jaana Syrjanen,x,{ Jari Suvilehto,* Satu Massinen,‖

Matti Karppelin,x,{ Irma Jarvela,‖ Seppo Meri,*,† Juha Kere,‖,# and T. Sakari Jokiranta*

Streptococcus pyogenes (or group A streptococcus [GAS]) is a major human pathogen causing infections, such as tonsillitis,

erysipelas, and sepsis. Several GAS strains bind host complement regulator factor H (CFH) via its domain 7 and, thereby, evade

complement attack and C3b-mediated opsonophagocytosis. Importance of CFH binding for survival of GAS has been poorly

studied because removal of CFH from plasma or blood causes vigorous complement activation, and specific inhibitors of the

interaction have not been available. In this study, we found that activation of human complement by different GAS strains (n = 38)

correlated negatively with binding of CFH via its domains 5–7. The importance of acquisition of host CFH for survival of GAS

in vitro was studied next by blocking the binding with recombinant CFH5–7 lacking the regulatory domains 1–4. Using this

fragment in full human blood resulted in death or radically reduced multiplication of all of the studied CFH-binding GAS strains.

To study the importance of CFH binding in vivo (i.e., for pathogenesis of streptococcal infections), we used our recent finding that

GAS binding to CFH is diminished in vitro by polymorphism 402H, which is also associated with age-related macular degener-

ation. We showed that allele 402H is suggested to be associated with protection from erysipelas (n = 278) and streptococcal

tonsillitis (n = 209) compared with controls (n = 455) (p < 0.05). Taken together, the bacterial in vitro survival data and human

genetic association revealed that binding of CFH is important for pathogenesis of GAS infections and suggested that inhibition of

CFH binding can be a novel therapeutic approach in GAS infections. The Journal of Immunology, 2012, 188: 426–435.

Streptococcus pyogenes or group A Streptococcus (GAS) isa Gram+ bacterium colonizing skin and mucous membranesand causing either superficial infections, such as impetigo

and tonsillitis, or invading superficial or deeper tissues in erysipelas,cellulitis, necrotizing fasciitis, or sepsis. Sometimes even superficialinfections can lead to poststreptococcal infection sequelae (1).Tonsillitis and erysipelas are the two best-known infections

caused by GAS. Both have a tendency to recur and need to betreated with antimicrobials to avoid the postinfection sequelae (2).

Erysipelas involves skin and superficial layers of the s.c. tissue,

typically of a leg or a foot, and is usually caused by GAS; some-

times it is caused by group G or C streptococci or rarely by

Staphylococcus aureus (3). Bacterial tonsillitis is usually caused by

GAS and sometimes by group G or C streptococci. Recurrent GAS

infections have been explained mainly by recurrent exposure to the

pathogen, carrier state, and differences in host immune functions (4).GAS strains causing invasive or septic infections must evade the

innate immunity, including the complement (C) system that attacks

the invader by opsonization, formation of cytolytic membrane at-

tack complexes, and induction of inflammation. Activation of any

of the three C pathways, classical, lectin, or the alternative pathway

(AP), leads to covalent surface deposition of C3b, the activation

product of C3. C3b and its fragments iC3b and C3dg act as im-

portant opsonins recognized by specific C receptors on phagocytes

(5). Upon activation of C3 and the terminal C cascade, proinflam-

matory chemotactic and anaphylatoxic protein fragments C3a and

C5a are released, mediating their effects by binding to receptors

on phagocytes (6).AP attacks any surface that is not specifically protected against

it (7); therefore, host cells need downregulation on their surfaces

by plasma protein complement factor H (CFH) (8, 9). CFH is

composed of 20 domains; domains 1–4 mediate inactivation of

deposited C3b or prevent C3b generation by accelerating decay

of the C3-convertase C3bBb. Host cell recognition by CFH is

mediated by domains 19–20 (10–12). The importance of CFH in

controlling AP activation on host structures is exemplified by the

often lethal atypical hemolytic uremic syndrome and dense de-

posit disease caused by malfunction of the domains 19–20 or 1–4

of CFH, respectively (13–16). In addition, polymorphism Y402H

*Department of Bacteriology and Immunology and Infection Biology Program, Haart-man Institute, University of Helsinki, FIN-00290 Helsinki, Finland; †HUSLAB,Helsinki University Central Hospital Laboratory, FIN-00290 Helsinki, Finland;‡Department of Infectious Diseases, National Institute for Health and Welfare, FIN-00271 Helsinki, Finland; xDepartment of Infectious Diseases, Tampere UniversityCentral Hospital, FIN-33521 Tampere, Finland; {School of Medicine, University ofTampere, FIN-33520 Tampere, Finland; ‖Department of Medical Genetics, HaartmanInstitute, University of Helsinki, FIN-00290 Helsinki, Finland; and #Department ofBiosciences and Nutrition, Karolinska Institutet, SE-14183 Stockholm, Sweden

Received for publication September 6, 2011. Accepted for publication November 2,2011.

This work was supported by grants from the Academy of Finland (Project 128646),Helsinki University Central Hospital funds, and the Sigrid Juselius Foundation.

Address correspondence and reprint requests to Mrs. Karita Haapasalo, HaartmanInstitute, University of Helsinki, P.O. Box 21, FIN-00014 Helsinki, Finland. E-mailaddress: karita.haapasalo@helsinki.fi

The online version of this article contains supplemental material.

Abbreviations used in this article: AMD, age-related macular degeneration; AP, alter-native pathway of complement; AU, arbitrary unit; C, complement; CFH, complementfactor H; CI, confidence interval; EIA, enzyme immunoassay; GAS, group A strepto-coccus; MF, multiplication factor; MgEGTA, 10 mMMgCl2 and 10 mM EGTA; NHS,normal human serum; PBS-T, PBS containing 0.05% Tween 20.

Copyright� 2011 by TheAmericanAssociation of Immunologists, Inc. 0022-1767/11/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1102545

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

in domain 7 of CFH is associated with age-related macular de-generation (AMD), the most common cause of visual loss in in-dustrialized countries (17–20). This indicates that domain 7 isimportant for physiological function of CFH, but it is not clearwhether this is due to the role of this domain in binding to neg-atively charged cell surface polyanions, such as heparan sulfate, orin binding C-reactive protein (21, 22).At least 10 pathogenic microbes protect themselves from AP

attack by acquiring host CFH onto their surfaces via domains 6–7,allowing domains 1–4 to regulate C activation on the microbe(e.g., S. pyogenes and Neisseria meningitidis) (23–25). The im-portance of microbial CFH binding is exemplified by the onlytwo successful vaccine candidates against N. meningitidis groupB (under phase III trials), both using the meningococcal CFH-binding protein (26). The importance of CFH binding, particu-larly via the domains 6–7, is supported by the finding that moststrains of S. pyogenes (GAS) bind CFH via these domains, and thepolymorphic variant 402H impairs binding of CFH to the bacteria,reducing GAS multiplication in human blood in vitro (27, 28).

In this study, we used 38 GAS strains to study correlation be-tween streptococcal CFH binding and the resulting C activationand analyzed the importance of host CFH acquisition onto GASby inhibiting the interaction with a recombinant CFH fragmentconsisting of domains 5–7 (CFH5–7). Lack of a proper animalmodel for GAS infections led us to analyze whether CFH bind-ing is important for survival of GAS in vivo by studying the as-sociation between polymorphism Y402H of CFH domain 7 andclinical GAS infections. The results showed that CFH acquisitionis necessary for survival and multiplication of GAS in humanblood in vitro, suggested decreased susceptibility to GAS infec-tions in individuals genetically susceptible to AMD (27), and in-troduced a novel idea for therapeutic intervention in GAS infec-tions by inhibition of CFH binding.

Materials and MethodsProteins

Cloning of the recombinant fragment CFH5–7 and its expression in Pichiapastoris were described earlier, including a figure of the fragment in both

Table I. Strains of Streptococcus pyogenes used and known CFH- and C4BP-binding properties of thesame strains or emm types

Binding toThis Straina

Binding to Another Strain of theSame Genotype

Origin emm/st Type CFH5–7 CFH FHL-1b C4BPb

Rfs NCTC 8326 emm4 — +b + +++Rfs NCTC 12058 emm76 (+54 nt) +Rfs ATCC 12349 emm8 — +b + +++Rfs ATCC 12962 emm28 — —b — +++Rfs NCTC 100085 emm12 — —b — —Rfs NCTC 12056 emm78 — —b — ++Rfs ATCC 12384 emm3 —Rfs NCTC 100064 emm2 + —b +Rfs NCTC 8193 emm5-8193B — +c

Rfs ATCC 12348 emm6-2 (+24 nt) — +c + —Rfs NCTC 100068 emm11 +++ +b + +++Spain 1995 emm1 — +b + —Inv emm22 +Inv emm77 ++Inv emm53-1 (-39 nt) ++Inv emm28 +++ +c — +++Inv emm68-3 ++Inv emm87 +++Inv emm58 +Inv st369 +++Inv emm89 +++AGNs stL1376 —AGNs emm18.8 ++ +b + +++AGNs emm89 +++AGNs emm74.0 +AGNs st212.0 +++AGNs emm55 —AGNs st212 +++AGNs emm95.0 +AGNs st6735.0 +++AGNs st3757.0 ++AGNs stG653 —AGNs st221.0 —AGNs st62.0 ++AGNs emm1.2b +++AGNth emm71.0 +++AGNth st206.0 +AGNth emm25.1 na

aData from Haapasalo et al. (27). Categories for CFH5–7 binding (relative percentage of maximal binding): —, ,14%;+, 14–30%; ++, 31–60%; +++, .60%.

bData from Perez-Caballero et al. (36).cData from Horstmann et al. (25).AGNs, acute glomerulonephritis skin isolate, Ethiopia; AGNth, acute glomerulonephritis throat isolate, Ethiopia; ATCC,

American Type Culture Collection; Inv, invasive blood isolate, Finland; na, not assessed; NCTC, National Collection of TypeCultures; Rfs, reference strain.

The Journal of Immunology 427

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

SDS-PAGE gel and Western blotting. The CFH(402Y) protein was purifiedfrom plasma of individuals homozygous for the 402Y variant of CFH,essentially as described earlier (27). C4BP was purchased from Comple-ment Technologies (Tyler, TX)

Binding of [125I]–CFH/CFH5–7 onto GAS

Binding of CFH5–7 to the GAS strains was described earlier using radi-oligand and direct-binding assays (27). Binding of full CFH and C4BPto the strains was analyzed similarly using the radioligand assay. Briefly,coating of the microtiter plates (Polysorp Breakapart; NUNC, ThermoScientific, Roskilde, Denmark) was done by incubating 1.2 3 108 bacteriain 200 ml PBS for 24 h at 37˚C. The wells were washed three times withPBS and blocked with 300 ml 3% BSA in PBS for 1 h at 37˚C. Afterwashing, 50,000 cpm of [125I]-protein in 200 ml 0.1% BSA-PBS was addedto each well, and the plate was incubated for 1 h at 37˚C. The wells werewashed, separated from each other, and subjected to counting with agamma counter.

For the inhibition assay, 1 3 109 log-phase bacteria (per reaction) werepreincubated in veronal buffered saline containing 0.1% gelatin with in-creasing concentrations (0, 1.2, 2.4, and 4.8 mM) of CFH5–7 and washedonce with 200 ml PBS. Thereafter, [125I]-CFH(402Y) was added (50,000cpm in 100 ml VBS containing 0.1% gelatin; sp. act. 43 106 cpm/mg), andthe mixture was incubated for 30 min under continuous horizontal rotation(Thermomixer) at 37˚C. Each mixture was applied on top of 250 ml 20%sucrose in a 400-ml plastic tube; after centrifuging (10,000 3 g for 3 min),radioactivity in the pellet and supernatant was measured with a gammacounter. The ratios of bound (pellet) to total activity (pellet + supernatant)were calculated. The experiment was performed three times using threeparallel samples.

Bacterial strains and growth conditions

The strains of S. pyogenes (Table I) were taken from 270˚C milk/glycerolsuspensions and grown at 37˚C under 5% CO2 on blood agar plates sup-plemented with colistin and oxolinic acid.

Study populations

Study populations consisted of Finnish patients (n = 487) suffering fromerysipelas (n = 278) or tonsillitis (n = 209). The erysipelas patients wereeither treated in Tampere University Hospital or Hatanpaa City Hospital,Tampere, for an acute episode of erysipelas (n = 90) or they were onbenzathine penicillin prophylaxis because of recurrent erysipelas (n =188). Tonsillitis cases were pediatric patients referred to Helsinki Uni-versity Central Hospital for scheduled tonsillectomy due to chronic orrecurrent tonsillitis and/or pharyngitis (n = 209). Inclusion criterion fortonsillectomy patients was recurrent tonsillitis (at least six episodes/y orthree episodes/y for two consecutive years, with at least one positiveculture for GAS), prolonged tonsillar infection refractory to antimicrobialtherapy, or tonsillar hyperplasia with symptoms (29). Control group (n =455) consisted of Finnish blood donors (n = 350) and individuals withouthistory of erysipelas (n = 105).

Ethics statement

The studies on the erysipelas patients were approved by the Ethical ReviewBoards of Pirkanmaa District and National Public Health Institute, and thestudies on the tonsillitis patients were approved by the Ethical ReviewBoard of the Hospital District of Helsinki and Uusimaa. Awritten informedconsent was provided by the study participants and/or their legal guardians.

Analysis of AP activation in plasma

GAS strains were grown in 10 ml Todd–Hewitt broth at 37˚C under 5% CO2

atmosphere until late log phase OD600 = 0.6, and the pellet was frozen at270˚C in the broth containing 30% glycerol. After thawing, the aliquotswere washed three times with PBS, OD600 was adjusted to 0.6 (2 3 108

bacteria/ml), 1 ml the suspension was pelleted, and each pellet wasresuspended into 200 ml 50% normal human serum (NHS) diluted usingice-cold PBS with 10 mMMgCl2 and 10 mM EGTA (MgEGTA) on ice. Asa negative control for the C3a assay, the pellet was resuspended into 200 ml50% NHS diluted in ice-cold PBS containing 10 mM EDTA. The NHSused was obtained by pooling serum from six individuals. The mixtureswere incubated at 37˚C for 30 min under continuous shaking, and thereactions were stopped on ice with 15 ml 0.2 M EDTA. The bacteria werepelleted, and the supernatants were frozen at 270˚C for C3a, Bb, andSC5b-9 analyses. C3a and Bb concentrations were measured usingMicroVue C3a and Bb Plus EIA kits (Quidel, San Diego, CA), accordingto the manufacturer’s instructions, using dilutions of 1:2000 (C3a EIA) or1:200 (Bb EIA).

For the SC5b-9 ELISA, anti-C9 neoepitope Ab WU13-15 (2 mg/ml;Hycult Biotech, Uden, The Netherlands) in 100 ml 0.2 M Na2CO3 (pH10.6) was added to wells of a microtiter plate (Maxisorp NUNC, ThermoScientific) and incubated for 17 h at 4˚C. For blocking, 0.5% BSA (Sigma-Aldrich, Steinheim, Germany) in PBS was incubated for 2 h at 22˚C,followed by three washes with PBS containing 0.05% Tween 20 (PBS-T).The plasma samples (100 ml 1:400 in PBS-T) were added, incubated for2 h at 4˚C, washed five times, and incubated for 1 h at 22˚C with 1:1000diluted goat anti-C7 (Organon Teknika, West Chester, PA). After fivewashes, HRP-conjugated donkey anti-goat IgG (1:3000; Jackson Immu-noResearch, West Grove, PA) in PBS-T was incubated for 1 h at 22˚C,followed by five washes and the addition of HRP substrate (OPD; DAKO,Glostrup, Denmark). The reaction was stopped with 100 ml 0.5 M H2SO4,and absorbance was detected at 490 nm. Arbitrary units (AU)/ml werecalculated using a standard curve obtained by activating NHS with Zy-mosan A (Sigma-Aldrich; 1 mg/ml, 60 min at 37˚C) (normalized to 1000AU/ml).

Analysis of AP activation in full blood

Bacteria grown overnight in Todd–Hewitt broth were diluted in the samebroth (1:50) and grown to OD600 of 0.15. The suspension was then diluted1:10,000 in 100 ml PBS (to obtain 200 CFU) and incubated under con-tinuous horizontal rotation (Thermomixer) for 15 min at 37˚C in thepresence or absence of 20 mg CFH5–7 (final concentration 4.8 mM). Thesamples were mixed with 1.2 ml human blood from healthy volunteers andincubated at 37˚C for 1 h under continuous rotation (resulting in CFH5–7/CFH molar ratio ∼1:3). We chose the CFH5–7 concentration to inhibit, but

FIGURE 1. Relationship between CFH binding by GAS and generation

of C-activation marker C3a upon exposure of 38 GAS strains to serum.

Each GAS strain was incubated in 50% NHS in the presence of Mg2+ and

EGTA (each 10 mM), and C3a release was quantified using an EIA. The

results were compared with binding of each strain to 125I-labeled CFH

(using a radioligand assay) (A) or previously published data on binding

of each strain to [125I]-CFH5–7 (using direct binding assay) (B) (27). The

values indicate relative binding of CFH compared with a negative and

positive control in the CFH radioligand assay or the percentage of radio-

activity in pellet versus total radioactivity in pellet and supernatant in

the direct CFH5–7 binding assay. Pearson correlations were calculated,

showing statistically significant negative correlations between binding of

each strain to either CFH or CFH5–7 and generation of complement AP-

activation marker C3a (p , 0.05). The C3a results are presented as mean

of triplicate samples in an EIA in which an NHS pool from six individuals

was used.

428 FACTOR H BINDING IN S. PYOGENES INFECTIONS

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

not to block, binding of CFH to the GAS surface to avoid possible com-plement attack against red cells if CFH binding to blood cells was blockedwith high CFH5–7 concentration. The blood samples (n = 2) were anti-coagulated with lepirudin (Refludan, Schering, Berlin, Germany) anddrawn from individuals who were genotyped CFH402Y homozygous orCFH402Y/CFH402H heterozygous. At 0, 30, and 60 min, 400-, 50-, and50-ml samples were transferred into tubes containing 50, 15, or 15 ml 0.2M EDTA, respectively. The cells were separated by centrifugation, and thesupernatant was stored at 270˚C before measuring the amounts of Bb andSC5b-9, as described above. Blood from different individuals was used inthe two experiments performed for calculating the statistics.

Multiplication assay

Bacteria were prepared and incubated in lepirudin-anticoagulated blood, asdescribed above. The assay was performed from blood obtained from threeindividuals (n = 3). The multiplication factor (MF) of the bacteria in bloodwas measured, as described previously (30, 31). Tubes were incubated at37˚C for 3 h under continuous orbital rotation. At 0, 60, 120, and 180 min,400-ml samples were transferred into tubes containing 50 ml 0.2 M EDTA.At 60 and 120 min, 400 ml the same blood was added after taking thesamples. For each 400-ml sample, 100 ml the mixture (or its 1:10–1:1000dilution) was inoculated on blood agar plates and incubated for 17 h (37˚C,5% CO2). The CFU were counted using the pour-plate method, and MFwas calculated for each time point. In each single assay, two parallelsamples with two parallel inoculations were tested; in each of the threeexperiments used for calculating the statistics, blood from a different in-dividual was used.

To monitor whether addition of CFH5–7 disturbs C activation in blood,the assay was performed by mixing 1.2 ml blood with 200 ml PBS or 20 mgCFH5–7 in PBS (final concentration of CFH5–7 was 0.7 mM, and theCFH5–7/CFH ratio was ∼1:3). At three time points, 200-ml blood sampleswere taken and centrifuged for 3 min at 10,0003 g, and the C activation inthe separated supernatant was analyzed using the C3a EIA.

Factor H Y402H genotyping

The sample DNA was isolated and genotyped for Y402H polymorphism(T1277C, rs1061170) by sequencing, as previously described (32).

Statistical methods

Pearson correlation was used to measure the dependence between C ac-tivation and binding of CFH5–7, CFH, or C4BP, as well as among SC5b-9,Bb, and C3a concentrations (version 15.0 of SPSS for Windows; Analyt-ical Software, IBM, Chicago, IL). For the CFH-inhibition assay, multi-plication assay, and the assay measuring Bb or SC5b-9 formation in blood,

the difference between means of each experiment within one, two, or threeindividual experiments was compared using the Student t test. Geneticassociations between different patient and control groups were analyzed bythe Pearson x2 test, and the relative risk and the 95% confidence interval(CI) for each comparison were calculated using SPSS software. Deviationsfrom Hardy–Weinberg equilibrium were analyzed using the x2 test ofMicrosoft Excel software.

ResultsBinding of CFH via CFH5–7 protects GAS from complementactivation

To determine whether acquisition of CFH, via its domains 5–7, byGAS is associated with decreased AP activation, we subjected 38GAS strains (Table I) to AP attack in serum and quantified theresulting activation products C3a and C5b-9 (SC5b-9). Our pre-vious data on binding of CFH5–7 to those strains (27) and pre-viously unpublished data on binding of full-length CFH to thosestrains correlated negatively with the new data for which theproduction of both C3a (Pearson correlations = 20.38 and 20.44for CFH and CFH5–7, p = 0.017 and 0.007, respectively) (Fig. 1)and SC5b-9 (Pearson correlation = 20.30 and 20.31, p = 0.034and 0.034, respectively) was quantified (Fig. 2A, 2B). Binding ofa regulator of the classical pathway, C4b-binding protein (C4BP),to GAS showed no correlation to C3a or SC5b-9 production(p = 0.163 and 0.596, respectively; Fig. 3). Because all of theactivation pathways may act simultaneously in vivo, the C3a assaywas also performed in NHS where all of the pathways were active(in the absence of MgEGTA). The correlation analyses showedsimilarly that C3a formation is negatively correlated with bindingof CFH via domains 5–7 (Pearson correlation = 20.42; p = 0.009)but not with C4BP binding to the bacteria (p = 0.541) (Supple-mental Fig. 1).To verify that SC5b-9 resulted from AP activation, we compared

SC5b-9 with C3a and Bb concentrations and found a significantpositive correlation between SC5b-9 and C3a (Pearson correla-tion = +0.4, p = 0.04) (Fig. 2C) or Bb (Pearson correlation = +0.7,p = 0.002) concentration (Fig. 2D). Unlike the others, four strains(emm55, emm6, st221, and stG653) showed practically no C3a

FIGURE 2. Correlation between CFH binding by

GAS and C-activation markers SC5b-9 and Bb. Each

GAS strain (n = 38) was incubated in 50% NHS in the

presence of Mg2+ and EGTA (each 10 mM), and

SC5b-9 formation was measured from the supernatant

and compared with previously published data on the

binding of 125I-labeled CFH (A) or [125I]-CFH5–7 (B)

to each strain (27). The values indicate relative

binding of CFH compared with a negative and posi-

tive control in the CFH radioligand assay or the

percentage of radioactivity in pellet versus total ra-

dioactivity in pellet and supernatant in the direct

CFH5–7-binding assay, described earlier (27). Pro-

duction of C3a (C) and Bb (D) fragment during the

serum exposure was measured; the results showed

statistically significant positive correlation with the

production of SC5b-9 (p , 0.05), as expected. For

measuring the SC5b-9 concentration, an Ab against a

C9-neoepitope was used; therefore, full C-activation–

related AU are used (AU/ml, full Zymosan activation

equals 1000 AU/ml). The results are presented as

mean of triplicate samples in a single EIA in which

the NHS pool from six individuals was used.

The Journal of Immunology 429

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

formation, although they failed to bind CFH5–7; however, thesestrains bind full-length CFH (Table II).

Innate immune attack against GAS is enhanced by CFH5–7

Next, the fragment CFH5–7 was used to inhibit binding of wholeCFH on the bacteria in an assay in which a representative CFH-binding strain (st369) was preincubated with increasing concen-trations of CFH5–7. Preincubation with CFH5–7 resulted in sig-nificant dose-dependent inhibition of CFH binding on GASsurface (Fig. 4A). To further investigate whether the inhibition ofCFH binding could have an effect on GAS survival, the bacteriawere exposed to CFH5–7 in a full-blood survival assay (i.e., an exvivo sepsis model). Preincubation of a representative CFH-bindingGAS strain (st369) with recombinant fragment CFH5–7 (lackingthe regulatory domains 1–4) resulted in increased Bb and SC5b-9release upon exposure of the bacteria to blood (Fig. 4B, 4C). Asexpected, when the same amount of CFH5–7 was incubated inwhole blood in the absence of bacteria, no increase in the C ac-tivation products was seen.Because incubation of the bacteria in blood with the CFH5–7

fragment resulted in increased C-activation products, we next de-termined whether five of the GAS strains binding both CFH andCFH5–7 (Fig. 5A) could be more easily opsonophagocytosed in

the presence of the CFH5–7 fragment. Using fresh anticoagulatedhuman blood, we could evaluate C-mediated inhibition of GASmultiplication because the Gram+ bacteria are resistant to directkilling by complement. By counting CFU using the pour-platemethod, we observed a significant decrease in MF of all of thestudied strains when the bacteria were incubated in the presence,versus the absence, of CFH5–7 (p , 0.001, for all strains by the180-min time point) (Fig. 5B–F).

Genetic association between CFH polymorphism and GASinfections

Next, we wanted to analyze whether CFH binding is important forcomplement evasion of GAS in vivo. Unfortunately murine CFHdoes not bind to any of the tested GAS strains (Supplemental Fig.2A), compromising the ability to use a mouse model. Therefore,we designed an analysis of the importance of streptococcal CFHbinding during human infections based on the data that thepolymorphism Y402H within domain 7 of CFH leads to impairedbinding of the variant 402H to GAS (27). To test possible asso-ciation of the polymorphism with human GAS infections, weanalyzed the Y402H genotype from patients with a history oferysipelas (n = 278) or recurrent tonsillitis (n = 209), as well asfrom 455 control subjects (350 blood donors and 105 non-erysipelas controls). The allele frequencies in both patient groupsand both control groups were consistent with the Hardy–Weinbergequilibrium, as expected (Table III). The frequency of the variantallele responsible for the expression of 402H (1277C) in thecombined patient group (all cases) was significantly lower than inthe combined group of control subjects (p , 0.05); the odds ratio(OR) for the risk for erysipelas or tonsillitis in carriers of the allelewas 0.831 (95% CI, 0.693–0.998) (Table IV). A significant dif-ference (p , 0.05) between the corresponding allele frequencieswas also found when all of the cases were compared with thenonerysipelas controls (OR, 0.735; 95% CI: 0.545–0.991).Next, genotype frequencies were compared; the frequency of the

CC genotype (homozygosity for the allele C coding for the 402Hallotype) was significantly lower in the patients than in the controlsubjects (p = 0.029; OR, 0.653; 95% CI: 0.446–0.958) (Table IV).Also, when all of the cases were compared with the nonerysipelascontrols, the frequency of the CC genotype was significantly lower(p , 0.05; OR, 0.513; 95% CI: 0.274–0.958).

DiscussionIn this study, we showed that acquisition of host CFH is importantfor immune evasion of GAS in blood ex vivo, and it could beblocked by using CFH5–7, resulting in impaired survival of thebacteria. We also showed that the impaired GAS–CFH interactionmediating allele 402H (1277C) is suggested to be associated withprotection from human erysipelas and recurrent tonsillitis. Thisallele was shown to be associated with AMD (17–20), and di-minished binding of 402H variant to heparin and/or C-reactiveprotein was suggested to explain the biological basis for the as-sociation (33–35).Binding of CFH by different clinical GAS isolates has been

studied, but at least two studies showed a significant variation inCFH binding between different strains and isolates (29, 36).Therefore, the effect of CFH binding on complement evasion ofGAS was analyzed by comparing CFH binding and markers ofcomplement activation upon exposure of the bacteria to serum.When 38 GAS strains that were previously analyzed for CFH5–7binding (27) were subjected to serum, a significant negative cor-relation was seen between CFH5–7 binding and an increase in C-activation markers C3a (Fig. 1B) and SC5b-9 (Fig. 2B). Therewere four strains that did not bind CFH5–7 but failed to cause any

FIGURE 3. Correlation between C4BP binding by GAS and comple-

ment activation markers C3a and SC5b-9. Each GAS strain (n = 38) was

incubated in 50% NHS in the presence of Mg2+ and EGTA (each 10 mM).

C3a and SC5b-9 formation was measured from the serum and compared

with binding of 125I-labeled C4BP. The values indicate relative binding of

C4BP compared with a negative and positive control in the C4BP radio-

ligand assay. Production of C3a (A) and SC5b-9 (B) during the serum

exposure was measured. No statistically significant correlation with the

C-activation markers was seen (p . 0.05). For measuring the SC5b-9

concentration, an Ab against a C9-neoepitope was used; therefore, full

C-activation–related AU are used (AU/ml, full Zymosan activation equals

1000 AU/ml). The results are presented as mean of triplicate samples in

a single EIA in which the NHS pool from six individuals was used.

430 FACTOR H BINDING IN S. PYOGENES INFECTIONS

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

increase in C3a concentration (Table II). In principle, this couldresult from expression of C5a-peptidase by these strains, becauseit might also cleave C3a, which is very similar to C5a (37). Be-cause three of these strains also did not lead to generation ofSC5b-9, a more probable explanation is that these GAS strainscould bind CFH via a site elsewhere than in the domains 5–7.Existence of an additional binding site on CFH for GAS is sup-ported by the finding that those four strains (emm55, emm6-2,st221.0, and stG653) bound full-length CFH but not CFH5–7(Table II), and previous data showed that emm55 and emm6 useScl1 protein (not the M protein responsible for CFH5–7 binding)to bind CFH via its domains 18–20 (38). One strain, emm5, causedmassive C3a formation but showed clear binding of full-size CFH(Table II). However, this strain failed to bind CFH5–7, suggestingthat the benefit of acquisition of CFH onto the bacteria may de-pend on the site mediating the binding. The site specificity issupported by the fact that $12 microbial species acquire CFH via

either domains 6–7 or 19–20 and only one via domains 8–10 and13–15 (39). Acquisition via other domains might disturb properrecognition and regulation of surface-bound C3b by CFH. Thiscould explain why the observed CFH binding does not protectGAS of emm5 type from opsonophagocytosis, as described pre-viously (30).We analyzed the effect of CFH5–7 on CFH binding of GAS in

a direct binding assay, using purified proteins, as well as on thesurvival of GAS in fresh full blood, not in serum only, because thethick cell wall protects Gram+ bacteria from direct C-mediatedlysis. To avoid interpretation of the results on the basis of onegenotype only, we used blood from three individuals, one ofwhom was homozygous for CFH402Y and another was hetero-zygous (genotype of the third was not assessed). Preincubation ofbacteria with CFH5–7 resulted in significant reduction in CFHbinding (Fig. 4A) and significant increase in SC5b-9 and Bb (Fig.4B, 4C) within 60 min, indicating that CFH5–7 interferes with

Table II. Summary of the results from complement activation and protein-binding assays with different GAS strains

Generation in SerumInhibition ofGrowth inBlood byCFH5–7

Binding of NHS MgEGTA MgEGTA MgEGTA

emm/st-Type CFHa CFH5-7b C4BPc C3ad C3ad Bbe SC5b-9f

emm4 — — + +++ ++ ++ ++ naemm76 (+54 nt) — + ++ — — ++ +++ naemm8 — — ++ +++ +++ + ++ naemm28 [rfs strain]g + — ++ +++ +++ + ++ naemm12 + — ++ — — + + naemm78 + — ++ ++ ++ + ++ naemm3 ++ — +++ + ++ ++ ++ naemm2 ++ + +++ — — + ++ naemm5-8193Bh ++ — ++ +++ +++ + ++ naemm6-2 (+24 nt)h ++ — ++ — — — — naemm11 + +++ ++ — — — — Yesemm1 — — + — — — + naemm22 — + ++ — + — ++ naemm77 — ++ + — + — ++ naemm53-1 (239 nt) — ++ + — — + + naemm28 [inv strain]g — +++ +++ — — + — naemm68-3 — ++ ++ + + + + naemm87 — +++ ++ — — — + naemm58 + + ++ + + — +++ nast369 ++ +++ ++ — — — — Yesemm89 ++ +++ + — — + ++ YesstL1376 + — + +++ +++ + +++ naemm18.8 ++ ++ ++ — — + ++ naemm89 ++ +++ +++ + — na na naemm74.0 ++ + — — — — ++ nast212.0 ++ +++ — — — — + Yesemm55h ++ — + — — — + nast212 ++ +++ — — — — — Yesemm95.0 ++ + + — — + — nast6735.0 ++ +++ + — — — + nast3757.0 +++ ++ + — — + + nastG653h +++ — ++ — — + ++ nast221.0h +++ — ++ — — — — nast62.0 +++ ++ ++ — — + + naemm1.2b +++ +++ ++ — — + + naemm71.0 +++ +++ + — — + +++ nast206.0 +++ + ++ — — — ++ naemm25.1 +++ na ++ — — — ++ na

aCategories for CFH and CFH5–7 binding (relative percentage of maximal binding): —, ,14%; +, 14–30%; ++, 31–60%; +++, .60%.bData from Haapasalo et al. (27).cCategories for C4BP binding (relative percentage of maximal binding): —, ,25%; +, 26–45%; ++, 46–70%; +++, .70%.dCategories for C3a release (mg/ml): —, ,20; +, 20–40; ++, 40–60; +++, .60.eCategories for Bb release (mg/ml): —, ,15; +, 15–25; ++, 25–35; +++, .35.fCategories for SC5b-9 generation (AU/ml): —, ,150; +, 150–220; ++, 220–270; +++, .270 mg/ml.gThe emm28 strains were separated by indicating the origin in brackets (see Table I).hStrain shows exceptional combination of results from the CFH/CFH5–7-binding versus C3a-release assays, as indicated in the Discussion.na, not analyzed.

The Journal of Immunology 431

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

CFH binding and AP evasion of GAS. An increase in Bb con-centration was already detected at the 30-min time point, whereasthe increase in C5b-9 concentration was seen only at 60 min. Thiscould result, for example, from the possible role of erythrocyte orleukocyte CD59 in eliminating small amounts of C5b-8 or C5b-9complexes that have not been assembled on the bacteria and,therefore, released to the fluid phase.All of the strains multiplied well in blood in the absence of

CFH5–7 but failed to do so in the presence of CFH5–7 (Fig. 5). Ofthe five tested GAS strains, multiplication of strain emm89, whichbinds CFH less efficiently than the other four strains, was totally

inhibited. This showed for the first time, to our knowledge, thatCFH binding via domains 5–7 is a very important innate immune-evasion strategy for GAS. This probably was not shown previ-ously, because the impact of CFH binding to the bacteria cannotbe measured by removing CFH from serum or blood because ofthe resulting vigorous C activation, and removal of the M proteinby bacterial mutagenesis could also impair its other functions,such as C4BP binding (40). Our approach of showing the impactof CFH acquisition for the bacteria is novel and free of thesebiases. Furthermore, some GAS strains were shown to expressother CFH-binding ligands, such as FbaA, which might also in-teract with CFH via CFH5–7, because it binds FHL-1 containingthe CFH domains 1–7 (41). One study also questioned the rele-vance of the interaction between CFH and GAS, because bindingof full-length CFH to some GAS strains is affected by physio-logical salt concentration, unlike binding of FHL-1 and C4BP(36). In our study, we did not analyze the effects of CFH versusFHL-1 in protection of GAS, because CFH5–7 could inhibitbinding of both of these proteins. Although C4BP was also shownto enhance complement and phagocytosis resistance of some GASstrains (M22, M60) (40, 42), we found no correlation betweenC4BP binding and generation of C-activation markers. This couldbe explained by the fact that C4BP is a regulator of the classicalpathway in which Abs play a major role in initiation of activation;therefore, the binding of C4BP by GAS is probably more relevantwhen specific Abs are present. However, the contribution of C4BPon AP evasion of some strains cannot be excluded on the basis ofour results, but the lack of correlation between C3a release andC4BP binding indicates that this is not a common phenomenon.This study showed that recombinant CFH5–7 interferes with C

evasion by GAS in vitro by inhibiting binding of CFH to GASsurface (Fig. 4A). CFH binding on GAS emm1 surface was studiedpreviously using a mAb against FbaA protein, supporting ourhypothesis that competitive inhibition of CFH binding to GAScould enhance elimination of GAS (43). Inhibition of CFHbinding to GAS by CFH5–7 in our study was tested using theCFH402Y allotype, which was shown to interact better with GASthan the CFH402H allotype (27). It is very likely that the inhib-itory effect of CFH5–7 on the binding of the CFH402H allotypewould be similar or better than on the binding of the CFH402Yallotype. It is also obvious that 100% inhibition of CFH bindingto GAS surface is not required for elimination of the bacterium;in our assays, the CFH5–7 concentration was approximately onethird of the CFH concentration. The dramatic effect of partialinhibition of CFH binding on reduced survival of GAS probablyresults from changing the balance between the binding of factorB versus CFH to the surface-bound C3b, because relatively in-creased factor B binding can lead to activation of the AP. Thiskind of effect is also caused by several known mutations in thesurface-recognition site on CFH domains 19–20, where only aminor change in affinity of CFH to C3b leads to severe tissuedamage in atypical hemolytic uremic syndrome (44).Effect of the CFH5–7 fragment in vivo was not easy to assess,

because we could not find any GAS strains that bound murineCFH (Supplemental Fig. 2A), although majority of the tested GASstrains binds human CFH (Supplemental Fig. 2B). This is notunexpected, because GAS is not a natural pathogen for mice.Species specificity of GAS could be at least partially dependent onCFH binding, similar to Neisseria gonorrheae, which does notinfect or bind CFH of primates other than humans (45). It might bepossible to use recombinant CFH5–7 as a biological antimicrobialagent against GAS or, preferably, against CFH-binding microbesthat are more resistant to antibiotics (e.g., Pseudomonas aerugi-nosa). In principle, the strategy is very good for an antimicrobial

FIGURE 4. Effect of CFH5–7 on CFH binding and complement activation

by GAS in human blood. A, A representative GAS strain binding CFH5–7,

st369, was preincubated with increasing concentrations of CFH5–7, followed

by incubation with 125I-labeled CFH. The values indicate relative binding

(percentage of the maximal binding observed in each assay). The effect of

CFH5–7 on C activation by GAS was tested by preincubating the same strain

with PBS (GAS+PBS) or 0.7 mM of CFH5–7 in PBS (GAS+CFH5–7), fol-

lowed by incubation with fresh lepirudin-anticoagulated human blood. For-

mation of Bb (B) and SC5b-9 (C) was measured from samples taken at three

time points (0, 30, and 60 min). As a control, the effect of CFH5–7 in blood in

the absence of GAS (PBS+CFH5–7) did not show an increase in either Bb or

SC5b-9 concentration. Mean 6 SD calculated from triplicate samples in

three (A), one (B), or two (C) individual experiments is shown. *p , 0.05,

***p , 0.001.

432 FACTOR H BINDING IN S. PYOGENES INFECTIONS

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

agent, because the later-emerging resistance to CFH5–7 (i.e.,failure to bind CFH) would simultaneously make the CFH5–7-resistant strain prone to C attack.Our results suggested that CFH variant 402Y is associated with

a patient history of erysipelas or recurrent tonsillitis. It is knownthat GAS is the main pathogen causing these infections. The mainreasons for selecting these infections in this study were that thepatients are readily available because of the high incidence, andsusceptibility to these infections seems to have a genetic com-ponent, as evidenced by a high recurrence rate and occasionalfamilial occurrence. When all of the patients were pooled andcompared with the controls, statistically significant differences inthe frequency of the 402H (1277C) allele and the CC genotype

(homozygosity for the allele C coding for 402H allotype) wereseen. This was not the case when either of the patient groups(erysipelas or recurrent tonsillitis) was compared separately withthe control groups. In the erysipelas group, the reason for therelatively small differences between the patient and control groupsmight be dilution of the association because of indirect risk factors,such as interdigital skin ulceration, diabetes, or impaired lymphaticdrainage in extremities, or because these infections can also becaused by group G Streptococcus (46) not necessarily bindingCFH. Dilution of the association could also be explained by thelack of clinical history of erysipelas or tonsillitis of the blooddonor control subjects. Because interview-based information onprevious streptococcal tonsillitis is often unreliable, we did not

FIGURE 5. Restriction of GAS multiplication

in blood by CFH5–7. A, After analyzing binding of

38 GAS strains to CFH and CFH5–7 (32 strains

from 38 indicated by gray diamonds), five strains

binding CFH via domains 5–7 (binding both

CFH5–7 and CFH) were chosen for multiplication

assay (d). These five strains, emm11 (B), emm89

(C), two strains of st212 (D, E), and st369 (F) were

preincubated with PBS or 20 mg of CFH5–7 in

PBS, followed by exposure to C activation and

opsonophagocytosis in fresh human lepirudin-

anticoagulated blood. At three time points (60,

120, and 180 min), samples were taken for CFU

counting, and the result was compared with CFU

at the 0 time point to obtain the MF. The MF

values for all five GAS strains were significantly

lower when the bacteria were first incubated with

CFH5–7 in PBS (gray bars) than with only PBS

(white bars). A non-CFH or CFH5–7-binding

strain emm8 (s in A) failed to multiply in blood;

therefore, its MF could not be calculated. Mean 6SD was calculated from duplicate inocula of du-

plicate samples from three individual experiments,

each performed with blood from one individual

(one homozygous for CFH402Y, one heterozy-

gous, and one not assessed). *p , 0.05, **p ,0.01, ***p , 0.001.

Table III. Allele and genotype frequencies of patients and controls

Hardy–Weinberg Test Allele Frequency Genotype Frequency

n x2 p C T CC TC TT

All patients 487 0.409 0.591 0.167 0.483 0.350Erysipelas 278 0.17 0.68 0.410 0.590 0.168 0.484 0.378Tonsillitis 209 1.72 0.19 0.407 0.593 0.165 0.483 0.352

Controls 455 0.455 0.545 0.207 0.496 0.297Nonerysipelas controls 105 0.79 0.489 0.486 0.514 0.236 0.500 0.264Blood donors 350 0.05 0.824 0.446 0.554 0.199 0.494 0.307

The Journal of Immunology 433

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

want to create a possible bias in selecting the controls on the basisof tonsillitis in general. Although we did not exclude history ofstreptococcal tonsillitis in the control group, we could demonstratethat 402Y (1277T) allele and the TT genotype were suggested tobe associated with an increased risk for erysipelas/tonsillitis.In summary, our study shows for the first time, to our knowledge,

that CFH binding by GAS is essential for bacterial survival inhuman blood and that a recombinant fragment of CFH that lacks theregulatory functions can be used to inhibit CFH binding on bac-terial surface and impair bacterial survival in blood ex vivo. Thelatter suggests a novel possible strategy for treating streptococcalinfections. In addition, we suggest that a CFH variant 402H, whichimpairs binding of CFH onto GAS, is associated with decreasedsusceptibility to erysipelas and recurrent tonsillitis. Our findingsuggested that individuals who are at risk for AMD are protectedfrom GAS infections. Therefore, it can be hypothesized that, be-fore the antibiotic era, the variant 402H could have becomeenriched in the population because it has provided protectionagainst streptococcal infections during childhood and early adultlife, despite its harmful effect in predisposing to visual loss in theelderly.

AcknowledgmentsWe thank Marjatta Ahonen, Kirsti Widing, Pirkko Kokkonen, and Karita

Viita-aho for excellent technical assistance, Dr. Ayman Khattab and Anna

Heinzman for providing help with the SC5b-9 assay, Dr. Wezenet Tewodros

for providing several GAS strains, and Pirkka T. Pekkarinen for providing

mouse serum.

DisclosuresThe authors have no financial conflicts of interest.

References1. Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections.

Clin. Microbiol. Rev. 13: 470–511.2. Bisno, A. L. 1996. Acute pharyngitis: etiology and diagnosis. Pediatrics 97:

949–954.3. Bisno, A. L., and D. L. Stevens. 1996. Streptococcal infections of skin and soft

tissues. N. Engl. J. Med. 334: 240–245.4. Pichichero, M. E., and J. R. Casey. 2007. Systematic review of factors con-

tributing to penicillin treatment failure in Streptococcus pyogenes pharyngitis.Otolaryngol. Head Neck Surg. 137: 851–857.

5. van Lookeren Campagne, M., C. Wiesmann, and E. J. Brown. 2007. Macrophagecomplement receptors and pathogen clearance. Cell. Microbiol. 9: 2095–2102.

6. Sacks, S. H. 2010. Complement fragments C3a and C5a: the salt and pepper ofthe immune response. Eur. J. Immunol. 40: 668–670.

7. Schreiber, R. D., M. K. Pangburn, P. H. Lesavre, and H. J. Muller-Eberhard.1978. Initiation of the alternative pathway of complement: recognition of acti-vators by bound C3b and assembly of the entire pathway from six isolatedproteins. Proc. Natl. Acad. Sci. USA 75: 3948–3952.

8. Weiler, J. M., M. R. Daha, K. F. Austen, and D. T. Fearon. 1976. Control of theamplification convertase of complement by the plasma protein beta1H. Proc.Natl. Acad. Sci. USA 73: 3268–3272.

9. Meri, S., and M. K. Pangburn. 1990. Discrimination between activators andnonactivators of the alternative pathway of complement: regulation via a sialicacid/polyanion binding site on factor H. Proc. Natl. Acad. Sci. USA 87: 3982–3986.

10. Ferreira, V. P., A. P. Herbert, H. G. Hocking, P. N. Barlow, and M. K. Pangburn.2006. Critical role of the C-terminal domains of factor H in regulating com-plement activation at cell surfaces. J. Immunol. 177: 6308–6316.

11. Kuhn, S., C. Skerka, and P. F. Zipfel. 1995. Mapping of the complement regu-latory domains in the human factor H-like protein 1 and in factor H1. J.Immunol. 155: 5663–5670.

12. Gordon, D. L., R. M. Kaufman, T. K. Blackmore, J. Kwong, and D. M. Lublin.1995. Identification of complement regulatory domains in human factor H. J.Immunol. 155: 348–356.

13. Warwicker, P., T. H. Goodship, R. L. Donne, Y. Pirson, A. Nicholls, R. M. Ward,P. Turnpenny, and J. A. Goodship. 1998. Genetic studies into inherited andsporadic hemolytic uremic syndrome. Kidney Int. 53: 836–844.

14. Levy, M., L. Halbwachs-Mecarelli, M. C. Gubler, G. Kohout, A. Bensenouci,P. Niaudet, G. Hauptmann, and P. Lesavre. 1986. H deficiency in two brotherswith atypical dense intramembranous deposit disease. Kidney Int. 30: 949–956.

15. Marder, H. K., T. H. Coleman, J. Forristal, L. Beischel, and C. D. West. 1983. Aninherited defect in the C3 convertase, C3b,Bb, associated with glomerulone-phritis. Kidney Int. 23: 749–758.

16. Meri, S., V. Koistinen, A. Miettinen, T. Tornroth, and I. J. Seppala. 1992.Activation of the alternative pathway of complement by monoclonal lambdalight chains in membranoproliferative glomerulonephritis. J. Exp. Med. 175:939–950.

17. Hageman, G. S., D. H. Anderson, L. V. Johnson, L. S. Hancox, A. J. Taiber,L. I. Hardisty, J. L. Hageman, H. A. Stockman, J. D. Borchardt, K. M. Gehrs,et al. 2005. A common haplotype in the complement regulatory gene factor H(HF1/CFH) predisposes individuals to age-related macular degeneration. Proc.Natl. Acad. Sci. USA 102: 7227–7232.

18. Edwards, A. O., R. Ritter, III, K. J. Abel, A. Manning, C. Panhuysen, andL. A. Farrer. 2005. Complement factor H polymorphism and age-related maculardegeneration. Science 308: 421–424.

19. Haines, J. L., M. A. Hauser, S. Schmidt, W. K. Scott, L. M. Olson, P. Gallins,K. L. Spencer, S. Y. Kwan, M. Noureddine, J. R. Gilbert, et al. 2005. Comple-ment factor H variant increases the risk of age-related macular degeneration.Science 308: 419–421.

20. Klein, R. J., C. Zeiss, E. Y. Chew, J. Y. Tsai, R. S. Sackler, C. Haynes,A. K. Henning, J. P. SanGiovanni, S. M. Mane, S. T. Mayne, et al. 2005.Complement factor H polymorphism in age-related macular degeneration. Sci-ence 308: 385–389.

21. Jarva, H., T. S. Jokiranta, J. Hellwage, P. F. Zipfel, and S. Meri. 1999. Regulationof complement activation by C-reactive protein: targeting the complement in-hibitory activity of factor H by an interaction with short consensus repeatdomains 7 and 8-11. J. Immunol. 163: 3957–3962.

22. Blackmore, T. K., T. A. Sadlon, H. M. Ward, D. M. Lublin, and D. L. Gordon.1996. Identification of a heparin binding domain in the seventh short consensusrepeat of complement factor H. J. Immunol. 157: 5422–5427.

23. Blackmore, T. K., V. A. Fischetti, T. A. Sadlon, H. M. Ward, and D. L. Gordon.1998. M protein of the group A Streptococcus binds to the seventh short con-sensus repeat of human complement factor H. Infect. Immun. 66: 1427–1431.

24. Schneider, M. C., B. E. Prosser, J. J. Caesar, E. Kugelberg, S. Li, Q. Zhang,S. Quoraishi, J. E. Lovett, J. E. Deane, R. B. Sim, et al. 2009. Neisseria men-ingitidis recruits factor H using protein mimicry of host carbohydrates. Nature458: 890–893.

25. Horstmann, R. D., H. J. Sievertsen, J. Knobloch, and V. A. Fischetti. 1988.Antiphagocytic activity of streptococcal M protein: selective binding of com-plement control protein factor H. Proc. Natl. Acad. Sci. USA 85: 1657–1661.

26. Granoff, D. M. 2010. Review of meningococcal group B vaccines. Clin. Infect.Dis. 50(Suppl. 2): S54–S65.

27. Haapasalo, K., H. Jarva, T. Siljander, W. Tewodros, J. Vuopio-Varkila, andT. S. Jokiranta. 2008. Complement factor H allotype 402H is associated withincreased C3b opsonization and phagocytosis of Streptococcus pyogenes. Mol.Microbiol. 70: 583–594.

Table IV. Association of CFH Y402H polymorphism with erysipelas and tonsillitis (x2 test)

Comparison Control Group Statistics Erysipelas Tonsillitis All Cases

C versus T Nonerysipelas controls x2 3.387 n.a. 4.095p value 0.07 n.a. 0.043*

OR (95% CI) 0.742 (0.54–1.02) n.a. 0.735 (0.55–0.99)C/C versus T/T Nonerysipelas controls x2 3.557 n.a. 4.479

p value 0.059 n.a. 0.034*OR (95% CI) 0.528 (0.27–1.03) n.a. 0.513 (0.27–0.96)

C versus T All controls x2 2.601 2.706 3.934p value 0.107 0.1 0.047*

OR (95% CI) 0.839 (0.68–1.04) 0.82 (0.65–1.04) 0.831 (0.69–1.00)C/C versus T/T All controls x2 3.0 3.338 4.776

p value 0.083 0.068 0.029*OR (95% CI) 0.67 (0.43–1.05) 0.63 (0.38–1.04) 0.653 (0.45–0.96)

*p , 0.05.n.a., not applicable.

434 FACTOR H BINDING IN S. PYOGENES INFECTIONS

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

28. Yu, J., P. Wiita, R. Kawaguchi, J. Honda, A. Jorgensen, K. Zhang, V. A. Fischetti,and H. Sun. 2007. Biochemical analysis of a common human polymorphismassociated with age-related macular degeneration. Biochemistry 46: 8451–8461.

29. Suvilehto, J., H. Jarva, M. Seppanen, T. Siljander, J. Vuopio-Varkila, andS. Meri. 2008. Binding of complement regulators factor H and C4b bindingprotein to group A streptococcal strains isolated from tonsillar tissue and blood.Microbes Infect. 10: 757–763.

30. Kotarsky, H., M. Gustafsson, H. G. Svensson, P. F. Zipfel, L. Truedsson, andU. Sjobring. 2001. Group A streptococcal phagocytosis resistance is independentof complement factor H and factor H-like protein 1 binding. Mol. Microbiol. 41:817–826.

31. Thern, A., M. Wastfelt, and G. Lindahl. 1998. Expression of two differentantiphagocytic M proteins by Streptococcus pyogenes of the OF+ lineage. J.Immunol. 160: 860–869.

32. Seitsonen, S., S. Lemmela, J. Holopainen, P. Tommila, P. Ranta, A. Kotamies,J. Moilanen, T. Palosaari, K. Kaarniranta, S. Meri, et al. 2006. Analysis ofvariants in the complement factor H, the elongation of very long chain fattyacids-like 4 and the hemicentin 1 genes of age-related macular degeneration inthe Finnish population. Mol. Vis. 12: 796–801.

33. Clark, S. J., V. A. Higman, B. Mulloy, S. J. Perkins, S. M. Lea, R. B. Sim, andA. J. Day. 2006. His-384 allotypic variant of factor H associated with age-relatedmacular degeneration has different heparin binding properties from the non-disease-associated form. J. Biol. Chem. 281: 24713–24720.

34. Laine, M., H. Jarva, S. Seitsonen, K. Haapasalo, M. J. Lehtinen, N. Lindeman,D. H. Anderson, P. T. Johnson, I. Jarvela, T. S. Jokiranta, et al. 2007. Y402Hpolymorphism of complement factor H affects binding affinity to C-reactiveprotein. J. Immunol. 178: 3831–3836.

35. Hakobyan, S., C. L. Harris, C. W. van den Berg, M. C. Fernandez-Alonso,E. G. de Jorge, S. R. de Cordoba, G. Rivas, P. Mangione, M. B. Pepys, andB. P. Morgan. 2008. Complement factor H binds to denatured rather than tonative pentameric C-reactive protein. J. Biol. Chem. 283: 30451–30460.

36. Perez-Caballero, D., S. Albertı, F. Vivanco, P. Sanchez-Corral, and S. Rodrıguezde Cordoba. 2000. Assessment of the interaction of human complement reg-ulatory proteins with group A Streptococcus. Identification of a high-affinitygroup A Streptococcus binding site in FHL-1. Eur. J. Immunol. 30: 1243–1253.

37. Wexler, D. E., D. E. Chenoweth, and P. P. Cleary. 1985. Mechanism of action ofthe group A streptococcal C5a inactivator. Proc. Natl. Acad. Sci. USA 82: 8144–8148.

38. Reuter, M., C. C. Caswell, S. Lukomski, and P. F. Zipfel. 2010. Binding of thehuman complement regulators CFHR1 and factor H by streptococcal collagen-like protein 1 (Scl1) via their conserved C termini allows control of the com-plement cascade at multiple levels. J. Biol. Chem. 285: 38473–38485.

39. Ferreira, V. P., M. K. Pangburn, and C. Cortes. 2010. Complement controlprotein factor H: the good, the bad, and the inadequate.Mol. Immunol. 47: 2187–2197.

40. Berggard, K., E. Johnsson, E. Morfeldt, J. Persson, M. Stalhammar-Carlemalm,and G. Lindahl. 2001. Binding of human C4BP to the hypervariable region of Mprotein: a molecular mechanism of phagocytosis resistance in Streptococcuspyogenes. Mol. Microbiol. 42: 539–551.

41. Pandiripally, V., E. Gregory, and D. Cue. 2002. Acquisition of regulators ofcomplement activation by Streptococcus pyogenes serotype M1. Infect. Immun.70: 6206–6214.

42. Thern, A., L. Stenberg, B. Dahlback, and G. Lindahl. 1995. Ig-binding surfaceproteins of Streptococcus pyogenes also bind human C4b-binding protein(C4BP), a regulatory component of the complement system. J. Immunol. 154:375–386.

43. Ma, C., Y. Guo, H. Gu, L. Zhang, H. Liu, H. Feng, and L. Wei. 2011. A novelmonoclonal antibody against FbaA can inhibit the binding of the complementregulatory protein factor H to group A streptococcus. Clin. Vaccine Immunol. 18:552–558.

44. Kajander, T., M. J. Lehtinen, S. Hyvarinen, A. Bhattacharjee, E. Leung,D. E. Isenman, S. Meri, A. Goldman, and T. S. Jokiranta. 2011. Dual interactionof factor H with C3d and glycosaminoglycans in host-nonhost discrimination bycomplement. Proc. Natl. Acad. Sci. USA 108: 2897–2902.

45. Shaughnessy, J., S. Ram, A. Bhattacharjee, J. Pedrosa, C. Tran, G. Horvath,B. Monks, A. Visintin, T. S. Jokiranta, and P. A. Rice. 2011. Molecular char-acterization of the interaction between sialylated Neisseria gonorrhoeae andfactor H. J. Biol. Chem. 286: 22235–22242.

46. Siljander, T., M. Karppelin, S. Vahakuopus, J. Syrjanen, M. Toropainen, J. Kere,R. Vuento, T. Jussila, and J. Vuopio-Varkila. 2008. Acute bacterial, non-necrotizing cellulitis in Finland: microbiological findings. Clin. Infect. Dis. 46:855–861.

The Journal of Immunology 435

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from