Modulation of Complement Activity in Vitro and in Vivo

Folia Microbiol. 51 (1), 27–32 (2006)

http://www.biomed.cas.cz/mbu/folia/

Modulation of Complement Activity in Vitro and in Vivo by Yersinia Wild and Mutant Strains M. YORDANOVa, E. GOLKOCHEVAb, H. NAJDENSKIb aDepartment of Immunology and bDepartment of Pathogenic Bacteria, Institute of Microbiology, 1113 Sofia, Bulgaria

fax +359 2870 0109 e-mail [email protected] Received 22 February 2005 Revised version 3 August 2005

ABSTRACT. The ability of released proteins (Yops) and surface lipopolysaccharides (LPS) from the wildtype strain Yersinia enterocolitica 8081-L2, serotype O:8 to influence the complement activity was determined. Yops and LPS from wild-type and mutant strains showed different ability to affect the classical pathway (CP) functional complement activity in vitro. The serum CP activity was inhibited during the infection induced with six Y. enterocolitica and three Y. pseudotuberculosis strains in rabbits. The changed complement activity might be of importance for the course of Yersinia infections.

Abbreviations AP CFU CP

alternative pathway colony forming units classical pathway

LPS Yops

lipopolysaccharide Yersinia outer proteins

Yersinia enterocolitica and Yersinia pseudotuberculosis have been recognized as enteropathogenic microorganisms (Bottone 1977). The most common and well-known disease caused by these pathogens is gastroenteritis and mesenteric lymphadenitis in humans and a number of domestic animals, suggested as possible sources of these infections (Slee and Button 1990). Moreover, systemic infections with abscesses and granuloma-like lesions in the liver and spleen and immunomorphological sequelae such as reactive arthritis, Reiter’s disease and erythema nodosum are associated with Y. enterocolitica and Y. pseudotuberculosis. Invading pathogens are normally attacked by the complement system which causes their lysis and thus leads to limitation of the infection. During the evolution microbes have developed a number of mechanisms to evade complement-mediated damage. One way is by removing the assembling complement complexes. Another possible mechanism is through complement consumption by surface and secreted microbial proteins. Enteropathogenic Yersinia spp. harbor a 70-kb virulence plasmid, called pYV, responsible for the secretion of more than 14 proteins (Michiels et al. 1990). One of the plasmid-encoded proteins, called YadA, mediates adherence to tissues through binding to cellular fibronectin and various types of collagen (SchulzeKoops et al. 1992). Secreted proteins YopH, YopE and YadA are involved in the phagocytosis, killing and oxidative burst inhibition of granulocytes and macrophages (Fallman et al. 1995; Visser et al. 1995; Visser et al. 1999). Both YopE and YopH are transferred directly into the host cell cytoplasm following contact with mammalian cells by a type III secretion system (Finlay and Falkow 1997; Michiels et al. 1990; Sory et al. 1995). Other proteins also mediate different damaging effects on the host cells. For example, YpkA/YopO disrupts host signaling in phagocytic cells and YopK/YopQ disregulates Yop effectors by affecting the size of the translocation pore (Galyov et al. 1993; Holmstrom et al. 1997). It is established that rough and smooth strains of Y. enterocolitica serotype O:3 which express YadA on their surface are serum resistant. YadA is capable of preventing the activation of late complement components (Pilz et al. 1992). A resistance-conferring chromosomally encoded outer membrane 17-kDa protein Ail has been isolated in Y. enterocolitica. This protein is associated with the bacterium’s ability to attach and enter into host cells. Furthermore, Ail makes the microorganism resistant to normal human serum (Bliska and Falkow 1992). Incubation of bacterial LPS with normal human serum activates the complement system, resulting in a decrease of hemolytic complement. Yersinia enterocolitica escaped the alternative-pathway activation by lower C3–C9 consumption when grown at 22 °C due to LPS, but independently of plasmid-encoded outer membrane proteins (Wachter and Brade 1989).

28

M. YORDANOV et al.

Vol. 51

Here evaluated the in vitro influence of Yops and LPS from Y. enterocolitica wild and mutant strains on CP complement activity. The ability of Y. enterocolitica and Y. pseudotuberculosis wild and mutant strains to change the serum complement activity during the course of oral infection in rabbit model was also studied. MATERIALS AND METHODS Bacterial strains are described in the following table: Strain Yersinia enterocolitica

Characteristics 8081-L2

8081-R2 WA-314 WA-314 sodA

parent strain; serotype O:8, pYV+ (fully virulent restriction-deficient derivative of 8081) derivative of 8081-L2; pYV+ (producing O antigen with uncontrolled chain length) derivative of 8081-L2 (producing O antigen with only one O-unit; Bengoechea et al. 2004) spontaneous rough derivative, pYV+ (with completely missing O antigen) wild type; serotype O:8, pYV+ isogenic mutant of WA-314 (Mn-cofactored superoxide dismutase mutant strain)

Yersinia pseudotuberculosis

pIB102 pIB155∆YopK pIB44∆ypkA

serotype O:3 isogenic mutant of pIB102 ditto

E. coli

O55:B5

Difco, USA

8081-∆wzzGB 8081-∆wbcEGB

Experimental animals. Six-month-old New Zealand rabbits weighing 2.5 ± 0.2 kg were purchased from a high-health-status breeding farm. The animals use protocols were approved by the Animal Care Commission at the Institute of Microbiology according to Helsinki declaration. Infection. One-hundred thirty-five rabbits were orally infected by feeding tube with 5.0 × 109 CFU from each strain. Three infected animals and one control (healthy) rabbit were examined at 7, 14, 30, 45 and 60 d after challenge. Extraction of lipopolysaccharide was performed by using the hot phenol–water extraction method (Zhang and Skurnik 1994). Briefly, bacterial cells from overnight cultures were collected by centrifugation and dried. Then the dried bacteria were resuspended into 0.25 % CCl3COOH and incubated for 3 h at 4 °C. The cells were collected by centrifugation and resuspended into a 1 : 1 mixture of water and 45 % phenol (preheated to 68 °C). The mixture was kept for 15 min at 68 °C and then cooled to 4 °C in an ice bath. The phases were separated by centrifugation and aqueous layer was carefully collected. The procedure was repeated by adding an equal volume cold water and after centrifugation the new aqueous layer was carefully collected. The two aqueous layers were combined and dialyzed overnight against distilled H2O at 4 °C. The fraction was then freed of RNA by adding 2 % sodium acetate, incubated for 2 h at 4 °C, Fig. 1. SDS-PAGE profile of LPS isolaand lyophilized; it was kept at −20 °C until use. Fig. 1 shows ted from wild-type Y. enterocolitica strain 8081-L2 and its isogenic mutants 8081-R2, the differences of core region and O-antigen chains of LPS ∆wbcEGB and ∆wzzGB (10 µg LPS per lane). isolated from the wild Y. entrocolitica strain 8081-L2 and its mutants R2, ∆wbzEGB and ∆wzzGB. For comparison LPS from E. coli O55:B5 was used. Isolation of Yersinia outer proteins. Proteins released from Yersinia cultures were prepared according to Zhang and Skurnik (Heesemann et al. 1984). Briefly, overnight bacterial cultures were diluted 1 : 20 in fresh media supplemented with 20 mmol/L MgCl2 and 20 mmol/L disodium oxalate, and grown for an additional 3 h at 37 °C. Bacterial supernatants were passed through a 0.45 µm pore-size filter and secreted proteins were precipitated with 10 % CCl3COOH. The proteins were collected by centrifugation (167 Hz, 30 min, 4 °C). Pellets were washed with cold acetone, lyophilized and kept at −20 °C until use. Complement assay. The CP activity of complement was determined by a microtiter hemolytic assay (Klerx et al. 1983). The assay was performed in a similar manner in fresh rabbit and human serum. Veronal (25 mmol/L)–0.9 % NaCl buffer (pH 7.2), containing 0.15 mmol/L Ca2+ and 0.5 mmol/L Mg2+ was used as

2006

MODULATION OF COMPLEMENT ACTIVITY BY Yersinia WILD AND MUTANT STRAINS

29

a diluent. Sheep red blood cells (Institute of Infectious and Parasitic Diseases, Sofia, Bulgaria) sensitized by an equal volume of appropriately diluted antiserum served as target cells. For the CP assay, sera from control and infected animals were separated 2 h after bleeding and serially diluted in the appropriate buffer. To 0.2 mL of each dilution, 50 µL of 2 % suspension of target cells was added. After incubation for 1 h at 37 °C, hemolysis was measured in the supernatants at 405 nm in an ELISA reader (Organon Teknika 530). The percentage of hemolysis was calculated according to the formula: Y = [(I – II) / (III – IV)] × 100, where I = A405 of test sera and II = A405 of heat inactivated sera (30 min, 56 °C), and III and IV represent A405 of water-lysed (100 % lysis) and buffer (no lysis) controls, respectively. Data are means ± SD from three determinations. Statistics. Statistical calculation was done by Student’s t-test. RESULTS In vitro changes of CP activity in rabbit and human serum. A fraction containing Yops from the culture medium and a fraction containing surface LPS isolated from the wild strain L2 of Y. enterocolitica (LPSyers) were incubated at a concentration of 100 µg/mL with rabbit serum and the complement activity was determined at different points via CP. Yops and LPSyers enhanced hemolysis starting at 15 min of incubation (Fig. 2 left). LPS isolated from E. coli (LPScoli) exhibited an opposite effect, resulting in suppression of hemolysis.

Fig. 2. CP activity (%) in rabbit (left) and human (right) serum. Fresh serum was incubated with 100 µg/mL of Yops, LPSyers and LPScoli and target erythrocytes; at the indicated time intervals (min) 100 µL volumes of the supernatants were taken and the hemolysis was measured; data are expressed as means ± SD from three determinations (* – p < 0.05; *** – p < 0.001); closed circles – control, open circles – Yops, closed squares – LPSyers, open squares – LPScoli.

Under the same conditions Yops, LPSyers and LPScoli were tested in human serum (Fig. 2 right). In contrast to rabbit serum, Yops caused a suppression of hemolysis for the whole period. LPSyers showed an ability to augment the lysis of target cells, while LPScoli after initial inhibition of CP activity reached 100 % hemolysis. Yops and LPS obtained from two wild strains (8081-L2 and WA-314) and from four mutant strains (8081-R2, ∆wzzGB, ∆wbcEGB, and WA-314 sodA) of Y. enterocolitica were tested in vitro for their effect on the complement activity of rabbit serum via CP in a concentration range of 0.015–1 mg/mL (Table I). Dose-response curves were drawn and a concentration giving 50 % activation or inhibition was determined as effective concentration (EC50). Yops from both wild-type strains (8081-L2 and WA-314) and 8081-R2 mutant strain inhibited the hemolytic activity, while the Yops from the remaining three mutant strains (∆wzzGB, ∆wbcEGB and WA-314 sodA) enhanced hemolysis. The LPS fraction from 8081-L2, ∆wzzGB and WA-314 strains caused activation of complement activity at a concentration of 40–50 µg/mL. The Yops from 8081-L2, ∆wzzGB and WA-314 strains showed inhibitory EC50 at a concentration of 20 µg/mL for WA-314, 30 µg/mL for 8081-R2 and 60 µg/mL for 8081-L2. In contrast, LPS from 8081-R2, Yops from ∆wzzGB, both fractions from ∆wbcEGB and LPS from WA-314 sodA failed to change the hemolysis at a concentration ≤1 mg/mL.

30

M. YORDANOV et al.

Vol. 51

Changes in serum CP activity during oral infection induced with different strains of Y. enterocolitica and Y. pseudotuberculosis. During the 60-d period of observation, the mutant strains ∆wzzGB, 8081-R2, 8081-L2, and WA-314 sodA of Y. enterocolitica caused a lowered complement activity as compared with the comTable I. Influence of Yops and LPS from wild-type plement level of noninfected rabbits (Fig. 3 top). Strains and mutant strains of Y. enterocolitica on the CP acti∆wbcEGB and WA-314 expressed elevated CP activity vity of rabbit seruma on day 15 post infection, in comparison with the other groups of infected rabbits. The variations between the Strain Yops LPS inoculated groups were not substantially different 15–60 d L2 –62.5 ± 10.6 50.4 ± 6.4 post infection. R2 –33.2 ± 7.3 0b Changes in serum CP activity were detected duGB 0b 45.2 ± 5.0 ring the course of infection with three Y. pseudotuberEGB 0b 0b culosis strains (Fig. 3 bottom). The wild-type strain WA-314 –24.0 ± 2.8 42.8 ± 4.6 pIB102 caused significant inhibition until 60 d of infecWA-314sod 55.3 ± 10.3 0b tion. At the early phase of infection (7 d) and the late peaFresh serum (0.1 mL) was added to 0.1 mL of Yops riod (45–60 d) all three strains possessed equal supor LPS in a concentration range 0.015–1 mg/mL pressive effect. The mutant strain pIB44∆ypkA strongly and mixed with 50 µL of sheep red blood cells; the increased CP activity after 15 and 30 d and a highly concentration giving 50 % activation or inhibition elevated activity was recorded after 30 d for the mutant was determined from dose-response curves; means pIB155∆YopK strain. ± SD from three determinations. bNot effective at a concentration up to 1 mg/mL.

Fig. 3. Changes in serum CP activity (U/mL) during infection (d p.i. – days post infection) induced with wild-type and mutant Y. enterocolitica (top; closed circles – EGB, open circles – ∆wzzGB, closed squares – 8081-R2, open squares – 8081-L2, closed triangles – WA-314 sodA, open triangles – WA-314) and Y. pseudotuberculosis (bottom; closed circles – pIB102, open circles – pIB155∆yopK, squares – pIB44∆ypkA) strains; one healthy noninfected and 3 infected animals were used at each point. Differences between individuals in the groups did not exceeded 5 % (error bars are omitted for the sake of clarity).

DISCUSSION Pathogenic Yersinia produce a set of proteins (Yops) encoded by a virulence plasmid (pYV) that enables the bacteria to evade the host’s innate defense mechanisms such as complement-mediated lysis and phagocytosis ensured by macrophages or polymorphonuclear leukocytes (Cornelis et al. 1998). Additionally, factors such as LPS side chains or the ail locus may also contribute to this phenomenon (Falkow 1991; Miller et al. 1989; Pilz et al. 1992). Yops probably do not influence complement activation, as measured by analysis of supernatants, but uptake of C3 complement component may be inhibited by plasmid-encoded proteins. Analysis of the complement activation via the alternative pathway revealed that YadA membrane

2006

MODULATION OF COMPLEMENT ACTIVITY BY Yersinia WILD AND MUTANT STRAINS

31

protein of Y. enterocolitica O:3 inhibits complement activation at the C3 and C9 level (Pilz et al. 1992). As a result, reduced amounts of C5b-9(m) are generated on the surface of YadA-bearing bacteria. In addition, YadA seems to protect against the lytic action of those C5b-9(m) complexes whose deposition could not be prevented. Yops as well as LPS from Yersiniae can affect complement activity of rabbit serum in vitro and in vivo. The Yops and LPS from the parent strain 8081-L2 of Y. entrocolitica rapidly activated CP in rabbit serum, while Yops exerted inhibitory action on the complement in human serum. This result might help in evaluating rabbit model of Yersinia infection and its relevance to human infection. Available literature data on particular molecules and especially on LPS point to structure–activity correlations. The tertiary assembly of LPS subunits determines the AP activation. Lipid A activates CP of complement activity, as this effect is suppressed by the presence of L-glycero-D-manno-heptose in its structure. In our experiments the LPS from the wild L2 strain activated CP, while the 8081-R2 rough strain lacking O-antigen had no effect. Direct structure–activity correlation can not be obtained from the present results, because individual molecules have not been used. The data reflect the overall effect of LPS and Yops fractions. LPS from the wild Y. enterocolitica WA-314 strain with O-antigen and LPS from the mutant ∆wbcEGB strain with only one O-antigen chain did not exhibit any effect. As a whole, 8081-L2 and WA-314 strains might be defined as good activators of complement CP activity through their Yops and LPS. Neither Yops nor LPS from ∆wbcEGB cells could affect complement activity in vitro. This correlated with the previously established serum resistance of ∆wbcEGB strain (Bengoechea et al. 2004). An in vivo effect of Y. enterocolitica and Y. pseudotuberculosis during the course of infection has also been shown. Different Y. enterocolitica strains exhibited, in general, similar inhibitory action on serum CP activity. Y. pseudotuberculosis wild-type strain maintained significantly lower complement level for the whole 60-d period of observation. The effect of the mutant strains passed from inhibition to activation and again to inhibition in different phases of infection. Data show that the functional complement activity might not be changed in vitro, while in vivo under the concerted action with other factors all the strains tested affected the CP activity, although the various strains express different virulent properties. Complement system participates in the early nonspecific responses as well in the late antibody-mediated immune responses. We thank Prof. J. Heessemann, Prof. H. Rüssmann (Max von Pettenkoffer-Institut, University of Munich, Germany), Prof. H. Wolf-Watz (University of Umeå, Sweden) and Dr. M. Skurnik (Turku University, Finland) for kindly providing Y. enterocolitica and Y. pseudotuberculosis strains. REFERENCES BENGOECHEA J.A., NAJDENSKI H., SKURNIK M.: Lipopolysaccharide O antigen status of Yersinia enterocolitica O:8 is essential for virulence and absence of O antigen affects the expression of other Yersinia virulence factors. Mol.Microbiol. 52, 451–469 (2004). BLISKA J.B., FALKOW S.: Bacterial resistance to complement killing mediated by the Ail protein of Yersinia enterocolitica. Proc.Nat. Acad.Sci.USA 89, 3561–3565 (1992). BOTTONE E.J.: Yersinia enterocolitica: a panoramic view of a charismatic microorganism. CRC Crit.Rev.Microbiol. 5, 211–241 (1977). CORNELIS G.R., BOLAND A., BOYD A.P., GEUIJEN C., IRIARTE M., NEYT C., SORY M.P., STAINIER I.: The virulence plasmid of Yersinia, an antihost genome. Microbiol.Mol.Biol.Rev. 62, 1315–1352 (1998). FALKOW S.: Bacterial entry into eukaryotic cells. Cell 65, 1099–1102 (1991). FALLMAN M., ANDERSSON K., H AKANSSON S., MAGNUSSON K.E., STENDAHL O., WOLF -WATZ H.: Yersinia pseudotuberculosis inhibits Fc receptor-mediated phagocytosis in J774 cells. Infect.Immun. 63, 3117–3124 (1995). FINLAY B.B., FALKOW S.: Common themes in microbial pathogenicity revisited. Microbiol.Mol.Biol.Rev. 61, 136–169 (1997). GALYOV E.E., H AKANSSON S., FORSBERG A., WOLF-WATZ H.: A secreted protein kinase of Yersinia pseudotuberculosis is an indispensable virulence determinant. Nature 361, 730–732 (1993). HEESEMANN J., ALGERMISSEN B., LAUFS R.: Genetically manipulated virulence of Yersinia enterocolitica. Infect.Immun. 46, 105– 110 (1984). HOLMSTROM A., PETTERSON J., ROSQVIST R., H AKANSSON S., TAFAZOLI F., FALLMAN M., MAGNUSSON K.E., WOLF-WATZ H., FORS BERG A.: YopK of Yersinia pseudotuberculosis controls translocation of Yop effectors across the eukaryotic cell membrane. Mol.Microbiol. 24, 73–91 (1997). KLERX J.P., BEUKELMAN C.J., VAN D IJK H., WILLERS J.M.: Microassay for colorimetric estimation of complement activity in guinea pig, human and mouse serum. J.Immunol.Methods 63, 215–220 (1983). MICHIELS T., WATTIAU P., BRASSEUR R., RUYSSCHAERT J.M., CORNELIS G.: Secretion of Yop proteins by Yersiniae. Infect.Immun. 58, 2840–2849 (1990). MILLER V.L., FARMER J.J. 3rd, H ILL W.E., FALKOW S.: The ail locus is found uniquely in Yersinia enterocolitica serotypes commonly associated with disease. Infect.Immun. 57, 121–131 (1989). PILZ D., VOCKE T., HEESEMANN J., BRADE V.: Mechanism of YadA-mediated serum resistance of Yersinia enterocolitica serotype O3. Infect.Immun. 60, 189–195 (1992). SCHULZE -KOOPS H., BURKHARDT H., HEESEMANN J., VON DER MARK K., EMMRICH F.: Plasmid-encoded outer membrane protein YadA mediates specific binding of enteropathogenic yersiniae to various types of collagen. Infect.Immun. 60, 2153–2159 (1992).

32

M. YORDANOV et al.

Vol. 51

SLEE K.J., BUTTON C.: Enteritis in sheep and goats due to Yersinia enterocolitica infection. Austral Vet.J. 67, 396–398 (1990). SORY M.P., BOLAND A., LAMBERMONT I., CORNELIS G.R.: Identification of the YopE and YopH domains required for secretion and internalization into the cytosol of macrophages, using the cyaA gene fusion approach. Proc.Nat.Acad.Sci.USA 92, 11998– 12002 (1995). VISSER L.G., ANNEMA A., VAN FURTH R.: Role of Yops in inhibition of phagocytosis and killing of opsonized Yersinia enterocolitica by human granulocytes. Infect.Immun. 63, 2570–2575 (1995). VISSER L.G., SEIJMONSBERGEN E., N IBBERING P.H., VAN DEN BROEK P.J., VAN FURTH R.: Yops of Yersinia enterocolitica inhibit receptor-dependent superoxide anion production by human granulocytes. Infect.Immun. 67, 1245–1250 (1999). WACHTER E., BRADE V.: Influence of surface modulations by enzymes and monoclonal antibodies on alternative complement pathway activation by Yersinia enterocolitica. Infect.Immun. 57, 1984–1989 (1989). ZHANG L., SKURNIK M.: Isolation of an R– M+ mutant of Yersinia enterocolitica serotype O:8 and its application in construction of rough mutants utilizing mini-Tn5 derivatives and lipopolysaccharide-specific phage. J.Bacteriol. 176, 1756–1760 (1994).