Differing complement-mediated opsonic activity of rabbit interstitial fluids from autologous serum

Comp. Immun. Microbiol. infect. Dis. Vol. 11, No. 1, pp. 1-9, 1988

Printed in Great Britain

0147-9571/88 $3.00+0.00 Pergamon Press pie

DIFFERING COMPLEMENT-MEDIATED OPSONIC ACTIVITY OF RABBIT INTERSTITIAL FLUIDS FROM AUTOLOGOUS SERUM CHARLES LAM, APOSTOLOS GEORGOPOULOS* a n d EBERHARD SCHUTZE Sandoz Forschungsinstitut, Brunner StraBe 59, A-1235 Vienna, Austria (Received 27 April 1987)

Abstract--Lack of appropriate methods for withdrawing extravascular or interstitial fluid from an animal host has limited in vitro study on the role of complement in the local defence of the extravascular space. In the present study, we obtained fluids from membrane diffusion chambers (porosity 0.22/~m) implanted into the kidneys, peritoneal cavity and soft tissues in rabbits. The complement-mediated opsonic activity (CMOA) of these fluids for Staphylococcus aureus ATCC 502A and Escherichia colt 01 was then compared to that of autologous sera. Soft tissue and renal interstitial fluids were as opsonic for E. colt as autologous sera but were however, poor opsonins for S. aureus. The peritoneal fluid was marginally effective in opsonization of both bacterial strains. While chelation of the fluids with MgEGTA (to block the classical pathway) did not diminish CMOA for E. colt, it reduced the CMOA for S. aureus by half. Conversely, heat-inactivation of the fluids and serum eliminated the opsonic activity for E. colt but only decreased the opsonic activity for S. aureus by half. Following a 24 h in vivo growth of E. colt in the implanted chambers, the CMOA was drastically reduced. Concomitant to the reduction in functional complement in the fluids, E. colt recovered from the chambers were found coated, though not maximally, with C3b as evidenced by studies with fluorescent antibody. The differences in opsonic content of extravascular fluids observed here might explain why certain sites of the body may be more vulnerable to attack by some bacterial species which are not effectively opsonized and therefore phagocytized. Key words: Opsonins, regional differences, interstitial fluids, rabbit, autologous serum

Rrsumr---L'absence de m+thodes approprires pour rrcolter le liquide extravasculaire ou interstitiel d'un hrte animal a limit~ l'&ude in vitro du role du compl+ment dans la d~fense locale des espaces extravasculaire. Dans la pr~sente ~tude, les fluides ont +t~ r~colt~s dans des chambres ~i membrane de diffusion (porosit~ 0,22/~m) implant~es dans les reins, la cavit6 prritonrale et les tissus mous des lapins. L'activit6 opsonique induite par le complrment (CMOA = Complement-Mediated Opsonic Activity) de ces fluides pour Staphylococcus aureus ATCC 502A et pour Escherichia colt 01 a +t6 comparre 5. celles des s~rums autologues. Pour E. colt, les fluides des tissus mous et les fluides r~naux interstitiels sont aussi opsoniques que les srrums autologues; par contre, pour S. aureus, ils le sont moins. Le liquide prritonral est marginalement efficace dans ropsonisation des deux souches bact~riennes. Tandis que la chrlation des liquides avec MgEGTA ne diminue pas la CMOA pour E. colt, elle la rrduit de moiti+ pour S. aureus. Au contraire, l'inactivation par la chaleur des fluides et des s~rums 61imine l'activit6 opsonique pour E. colt, mats elle ne la rrduit que de moiti6 pour S. aureus. Aprrs 24 h des croissance in vivo de E. colt dans les chambres implant+es, on constate une rrduction drastique de la CMOA. Parall~lement ~ cette rrduction du complrment fonctionnel dans les fluides, on observe que les E. colt extraits des chambres sont "coated", mats pas au maximum, alors que C3b est mis en +vidence par des &udes aux anticorps fluorescents. Les differences dans le contenu opsonique des tissus extravasculaires observres ici pourraient expliquer pourquoi certains sites du corps sont plus vuln+rables aux attaques de certaines esp+ces bact~riennes qui ne sont pas efficacement opsonisres et donc phagocytres. Mots-clefs: Liquide interstitiel, lapin, opsouihe, serum autologue

*Present address: Universit/itsklinik f/Jr Chemotherapie, Lazarettgasse, A-1090 Vienna, Austria. 1 CIMID I I / I - - A

2

CHARLES LAM et al.

INTRODUCTION One of the crucial mechanisms which regulates the recognition of invading pathogenic microorganisms by polymorphonuclear leucocytes (PMNs) is opsonization. In nonimmune individuals, opsonization is mediated by serum complement system activated through either the classical or alternative pathway leading to the coating of the bacteria with opsonically active C3b [6, 14]. The importance of complement has been unquestionably established experimentally [1, 17] and confirmed clinically [1, 7] that bloodstreamborne dissemination of gram-negative enteric bacteria is concomitant to decreased phagocytosis-promoting activity of serum complement in circulation. However, relatively little is known about the contribution of complement or complement-mediated opsonic activity (CMOA) in the local defence of extravascular space in an animal host. And yet this is a common site for most bacterial infections [2]. It is generally assumed that the amount of complement and other heat-labile opsonins in systemic circulation is similar to and directly exchanges with interstitial fluid. Recent studies, however, suggest that this assumption is erroneous. Fluids which were obtained during bacterial infections from the lungs [5], pleura [13, 14] and meninges [18, 20] were found to be inadequate for effective opsonization of various pyogenic bacteria. Lack of appropriate methods for sampling interstitial fluid from an animal host for in vitro investigations has in the main limited studies on the precise role of complement in the defence of extravascular space. Implanted membrane diffuse chambers (DC) have been used successfully for investigations of drug kinetics in laboratory animals [9]. Chambers of porosity 0.2-0.45 #m can be implanted into various sites in an animal host and within 4 days become filled with a fluid which is biochemically similar to interstitial fluid [16]. The present study was carried out to compare the CMOA of fluids obtained from infected and uninfected chambers which were previously implanted subcutaneously, intraperitoneally and intrarenally in rabbits as described elsewhere [16] with autologous pooled sera. The CMOA was studied as the ability of the fluids to promote the killing of Escherichia coli serotype 01 and Staphylococcus aureus 502A by casein-induced peritoneal PMNs and as the ability of the opsonized bacteria to induced chemiluminescence (CL) in the leucocytes. The results indicate that tissue fluids which accumulate in implanted chambers in rabbits differ significantly as opsonins for gram-negative and gram-positive organisms. MATERIALS AND METHODS Collection and storage of serum and diffusion chamber fluid

About 40 ml blood samples from each of 5 rabbits were obtained by an ear artery and allowed to clot at 4°C for 1 h. A pool of serum was separated thereafter by centrifugation at 1500g for 10min at 4°C and stored in liquid nitrogen in 0.5 ml aliquots. Tissue fluids from the blood donors were collected via catheters from DCs which were previously implanted subcutaneously, intraperitoneally or intrarenally [16], allowed to clot at 4°C for 1 h, centrifuged and the supernates stored under the same conditions as described above for serum. Protein concentrations were measured in all samples of interstitital fluids and pooled serum by the method of Lowry et al. [15]. Heat-inactivated serum was prepared by heating thawed serum at 56°C for 30 min. In some experiments any specific antibody was absorbed from the thawed serum or interstitial fluid by incubation with 1 × 1 0 9 cfu/ml of the test bacteria in an ice bath for 1 h.

Regional differencesin opsonic activityin rabbits

3

Preparation of bacterial strains Staphylococcus aureus strain 502A and Escherichia coli 01 are from our stock-cultured collection. The strains were cultivated in trypticase soy broth (Difco) to mid-logarithmic growth phase, washed 3 times in Hanks balanced salt solution (HBSS) and resuspended to 1 x l 0 9 colony-forming units (cfu)/ml in the same buffer but containing 3% dimethylsulfoxide as a cryoprotector. The cultures were then stored in liquid nitrogen. For opsonic activity assay the bacteria were quickly thawed at 37°C and suspended to 5 × 1 0 7 cfu/ml in HBSS containing DC fluids corresponding to protein content of or to 10% pooled rabbit serum. The bacteria were then allowed to opsonize shaking at 37°C for 30 min. The opsonized bacteria were washed 3 times in cold HBSS by centrifugation at 1600g for 10 min and were suspended in HBSS without phenol-red. Opsonization in the absence of the classical complement pathway was studied using serum in the presence of 10 mM MgC12 and 10 mM ethylene glycol tetraacetate (EGTA, Sigma) at pH 7.2 [8]. Preparation of rabbit peritoneal polymorphonuclear leucocytes (PMNs) Exudative PMNs were induced in chinchilla rabbits weighing about 2.5 kg by intraperitoneal injection of 50 ml of warm 5% casein in 0.1 M phosphate buffered saline (PBS) pH 7.2 as was described by Casiato et al. [4]. After 4 h the exudate cells were obtained by peritoneal lavage using 200 ml of Ca 2÷- and Mg2+-free PBS containing 5 U of heparin/ml. The lavage cells were washed 3 times in HBSS without phenol-red and suspended in the same buffer to 5 x 1 0 6 PMNs/ml. Such preparations were predominantly (> 95%) viable intact PMNs as was revealed by the trypan blue exclusion and differential counts performed on May~Gr/inwald-Giemsa stained smears.

Measurement of the ability of the serum or DC fluids to promote the killing of bacteria by the PMNs 5 X 1 0 6 cFu of opsonized bacteria were shaken in a water bath at 37°C with 5 × 106 PMNs in HBSS as was previously described [12]. At 0, 30, 60 and 120min duplicate incubation mixtures were removed and the leucocytes were lysed with 1.6 ml of cold 0.1% Triton X-100. The surviving bacteria were estimated by rapidly diluting the lysates in PBS and plating out four replicate 25/~1 portions of decimal dilutions on the surface of dried trypticase soy agar plates. The number of viable bacteria were counted after an overnight incubation at 37°C. Chemiluminescence (CL ) assay Complement-mediated opsonic activity of the pooled serum and DC fluids was also studied as the ability of the opsonized bacteria to induce chemiluminescence (CL) in peritoneal rabbit PMNs. Luminol-enhanced phagocytic CL was measured at 37°C in a photometer (Luminometer 1251, LKB Wallac). The reaction mixture contained 5 x 106pMNs, 5 x 10Scfu of opsonized bacteria and 20/~g 5-amino-2,3-dihydro1,4-phthalazinedione (Luminol, Sigma) in 0.5 ml of HBSS without phenol-red. The carousel of the Luminometer, which can take up 25 samples, was automatically operated by a computer program. Each sample was brought in line with photomultiplier tube, where the sample-tube was continuously agitated and the resulting light output in millivolts (mV), the position of the sample and the incubation time were automatically printed out by a

4

CHARLESLAM et al.

teleprinter (Model T 43). At the end of the assay, the reaction mixtures were deposited onto microscope slides by cytocentrifuge (Cytospin, Shandon Elliot). The prepared slides were fixed in methanol, stained in M a y - G r i i n w a l d - G i e m s a stains and were examined microscopically for ingestion of bacteria. RESULTS Luminol-enhanced phagocytic chemiluminescence ( C L ) as a valid assay f o r determining C M O A o f serum

Before evaluating the C M O A in the rabbit interstitial fluids, it was first necessary to establish the optimal conditions for assessment of serum requirements of E. coli 01 and S. aureus 502A with a luminol-enhanced phagocytic CL assay. An investigation of independent effects of increasing either serum for pre-opsonization of the bacterial targets, luminol, or bacterial:PMN ratio in the reaction mixtures on CL emission revealed a considerable variability in phagocytic CL response. Based on these observations, reaction mixtures containing 5 × 108 cfu bacteria pre-opsonized in 40% pooled serum for 30 min, 5 × 1 0 6 P M N s and 2 0 # g luminol in a total volume of 0.5 ml HBSS (see Materials and Methods) were found to produce suitable and reproducible phagocytic CL responses. Of relevance to the exploitation of the CL emission as a valid assay for evaluating C M O A in serum, was the dependency of phagocytic CL responses on the concentration of serum used for pre-opsonization of the organisms (Fig. 1). There was no phagocytic CL when unopsonized bacteria were used as the test particles. When the bacteria were opsonized with increasing serum concentration, however, there was also a significant generation of CL by the phagocytosing leucocytes. Heat-inactivation of the normal serum eliminated the opsonic CL activity for E. coli ( < 20% CL response) but only reduced that 9000 aooo

•E

i

ilooo 5

tNCO

I0

15

21)

Fig. 1. Luminol-enhanced chemiluminescence of polymorphonuclear leucocytes ingesting Escherichia coli A 1060 opsonized for 30 min in various concentrations of normal serum. (I-q),(m), (~), (A) and (C)) show chemiluminescenceresponse of leucocytesingesting bacteria pre-opsonized with 40, 20, 5, 1 and 0% serum respectively. No chemiluminescence was observed when heat-inactivated serum was used as an opsonin.

Regional differences in opsonic activity in rabbits

5

for staphylococci by about 50% of the normal serum. Adsorption of the pooled serum with homologous organisms at 4°C resulted in a reverse effect. While E. coli opsonized with serum previously adsorbed with the homologous bacteria induced CL to the same extent as that opsonized in normal serum, staphylococci opsonized in staphylococci-adsorbed serum induced only half of CL produced by bacteria opsonized in normal serum. To determine whether this opsonic CL activity was mediated either by C, IgG or both, bacteria opsonized in normal and heat-inactivated serum were examined by immunofluorescence microscopy for the presence of surface C3. C3b were optimally deposited onto the surface of both gram-negative and gram-positive organisms incubated in undiluted normal serum but no deposition occurred in heat-inactivated serum. IgG were undetectable on bacterial surfaces. To further delineate the involvement of complement in the opsonization of the test organisms used in the present study, the normal serum was chelated with 10mM MgEGTA (to block the classical complement pathway) and its opsonic CL activity compared to normal and heat-inactivated serum. Chelation of the serum did not alter the opsonic CL activity for E. coli significantly, but the opsonic activity for S. aureus 502A was decreased to only half of the activity in normal serum. These results suggested that the CL assay was a valid assay for evaluating opsonic activity of serum and that the opsonization of E. coli was dependent on the heat-labile components and of S. aureus 502A was dependent on both heat-labile (complement) and heat-stable serum factors that are different from complement. C M O A o f rabbit interstitial fluids and serum demonstrated by CL assay

To critically compare the CMOA of rabbit interstitial fluids with homologous normal serum, E. coli 01 and S. aureus 502A were opsonized for 30min at 37°C in HBSS containing 40% serum or chamber fluid equal to the serum on the basis of protein content. The degree of opsonization was then evaluated by CL assay (Fig. 2). At similar protein

(A)

9o(111

9000

(a)

~8ooo E

E Z

~ 5ooo hi

0

z i 0

o

-

o

5

10

INCUBATIONTrUE(urn)

15

0

5 10 INCUBATION TIME ODIN)

Fig. 2. Ability of fluids from various anatomical sites in the rabbit in opsonization of Escherichia coli A 1060 (A) and Staphylococcus aureus 502A (B) measured as luminol-enhanced phagocytic chemiluminescence. The fluids were obtained from chambers which were surgically implanted in the kidneys (I-1), subcutaneous tissues (ll) or in the peritoneal cavity (O). /k shows the opsonic activity of homologous serum.

15

6

CHARLES LAM et al. Table 1. Comparative binding of C3 component on Escherichia coli 01 Opsonin source a Hanks BSS Normal serum Peritoneal fluid Renal interstitial fluid Subcutaneous interstitial fluid

Fluorescence intensity b 0~ 2+ 1+ 3+ 3-4+

aOrganisms were incubated at 37°C for 30min in body fluids corresponding to protein content of or to 10% normal and heat-inactivated serum and washed before the fluorescent antibody test. bThe bound C3 components were quantitated by phase-contrast immunofluorescence microscopy: 0, no fluorescence; 1 + , weakly positive; 2 + positive; 3 - 4 + strongly positive. CSimilar result was obtained with heat-inactivated body fluids.

concentration all the interstitial fluids were marginally less effective as opsonins for S. aureus 502A (Fig. 2B). Conversely, renal and subcutaneous interstitial fluids were more effective in opsonization of E. coli 01 as normal serum (Fig. 2A). To determine whether C was the main opsonin in these fluids we examined the deposition of C3b on the surface of E. coli 01 which had been incubated in normal and heat-inactivated fluids using immunofluorescence. Bacteria recovered from the implanted diffusion chambers 24 h after their inoculation were also checked for the presence of activated C3 components. No C3 or IgG components were detectable on the surfaces of bacteria which were incubated in heat-inactivated fluids (Table 1) but the bacteria which were incubated in normal serum were maximally coated with C3, indicating that most of the opsonic CL activity was complement (classical and alternative pathway) factors. The organisms recovered from implanted diffusion chambers were also found to be coated with C3 but not to the same extent as those incubated in vitro with normal interstitial fluids. Taken together, these observations indicated that opsonization of E. coli in the interstitial fluids, like their opsonization in serum, was mediated mainly by complement components, activated possibly through the alternative complement pathway. Ability of the interstitial fluids and serum in promoting the killing of bacteria by rabbit peritoneal P M N s To determine the correlation between opsonic CL activity and ability of interstitial fluids to promote the killing of bacteria by leucocytes, the fluids were also tested in a bactericidal assay (Table 2). It can be seen that the ability of the fluids to promote the killing of the test bacteria by PMNs paralleled a higher opsonic CL activity. Thus renal and soft tissue interstitial fluids promoted the killing of E. coli 01 by PMNs better than normal serum, whereas the peritoneal fluid was marginally as effective as serum. No bacterial activity of the leucocytes could be demonstrated against E. coli when the body fluids were heatinactivated, confirming that most of the opsonic activity of rabbit interstitial fluids for E. co/i 01 was complement. Although more C3 could be demonstrated on the surface of staphylococci than on E. coli, relatively fewer interstitial fluid-opsonized staphylococci were killed by peritoneal PMNs in comparison to the killing of the gram-negative organisms. Again blockade of the classical complement pathway with MgEGTA during opsonization in normal and heat-inactivated serum produced the expected effects as predicted from the opsonic CL activity. Thus while the killing of the gram-negative

R e g i o n a l differences in o p s o n i c a c t i v i t y in r a b b i t s

Table 2. The ability of fluids from chambers implanted at various anatomical sites in the rabbits in promoting the killing of Escherichia coli 01 and Staphylococcus aureus 502A by polymorphonuclear leucocytes Reduction in viable bacteria (log10 C F U + SD per ml) in the presence of PMNs at 37°C for 2 h a Opsonin source Hanks balanced salts solution Normal pooled serum Peritoneal chamber fluid Renal chamber fluid Subcutaneous chamber fluid

Escherichia coli 01

Staphylococcus aureus 502A

0b 1.77 + 0.22 0.30 + 0.17* 2.62 _+0.20 1.89 + 0.15

0 4.02 + 0.24 1.26 __+0.30** 2.22 _+0.21"** 1.95 +_ 0.22***

'Bacteria were opsonized at 37°C for 30 min with 40% serum or chamber fluids equal to the serum on the basis of protein content. 5 x 106 cfu/ml of opsonized bacteria were incubated with 5 x 106 PMNs/ml at 37°C for 2 h before determining the viable counts. bSimilar results were obtained when heat-inactivated body fluids were used. *P < 0.025; **P < 0.01; ***P = 0.01; by Student's t-test for differences in killing bacteria opsonized with serum and those opsonized with other body fluids.

organisms by the PMNs in the presence of normal and MgEGTA-chelated serum did not differ significantly, it was completely eliminated in the presence of heat-inactivated fluids. In contrast, chelation or heat-inactivation of the interstitial fluids reduced the leucocytic killing of the staphylococci by about 50% in comparison to those killed in the presence of normal body fluids (results not presented). These findings suggested that both heat-labile factors (possibly complement) and heat-stable opsonins (possibly antibodies) were critically necessary for effective coating of S. aureus 502A for their optimal killing by the leucocytes whereas the coating of E. coli 01 by complement components was apparently sufficient to promote their effective killing by the PMNs.

DISCUSSION Lack of appropriate methods for sampling fluids which bathe the extravascular space, a common site for most bacterial infections in an animal host, has partly limited studies on the role of the host defence mechanisms of interstitial fluids. It is generally thought that such fluids, like plasma in systemic circulation, play a crucial role in the defence of the extravascular space through their ability to coat bacteria with opsonically active complement components for optimal phagocytosis by resident or recruited professional phagocytes. However, the high risk of bacterial infection seen in the vicinity of foreign bodies, e.g. sutures [3, 19] and in intraabdominal cavity with gut flora [10] despite the presence of normal phagocytic response suggest that the levels of functional opsonins in the extravascular space might be different from that in serum. The results of the present study support this view and indicate that for any given organisms, interstitial fluids sampled from diffusion chambers implanted in various sites in the rabbits may or may not be opsonic even though autologous serum is opsonically optimal. Thus, while peritoneal fluids were almost devoid of opsonins for S. aureus renal and soft tissue interstitial fluids contained only about a quarter to half of the activity present in autologous serum (Fig. 2). The low opsonic activity in the renal and soft tissue interstitial fluids could be abolished by heating the fluids at 56°C for 30 min, indicating that complement rather than specific antibody was the phagocytosis-promoting activity of the fluids. Further support for this interpretation came from fluorescent-antibody study of the staphylococci which were incubated in the interstitial fluids. While no IgG molecules were detectable on the

8

CHARLES LAM et al.

organisms, a small amount of C3 was fixed on the surface of the bacteria incubated in renal and soft tissue interstitial fluids but not during incubation with peritoneal fluids. In contrast to S. aureus, E. coli 01 was adequately opsonized in the interstitial fluids. There was a close correlation between the opsonic CL activity, the amount of bacteria killed by rabbit PMNs in vitro and the amount of C3 fixed on the surface of the bacteria. Again, there were regional differences in the amounts of phagocytosis-promoting activity of the fluids. Fluids from the kidneys and soft tissues were relatively better opsonin sources than pooled normal serum containing also autologous serum (Fig. 2), whereas peritoneal fluids were less active than normal serum. All the fluids were devoid of any detectable specific antibody to E. coli 01 as judged by the indirect fluorescent antibody technique. The differing CMOA in interstitial fluids for E. coli 01 and S. aureus 502A is not clear. Blockade of the classical complement pathway with MgEGTA did not alter the opsonization of E. coli but abolished the low activity for S. aureus. The low CMOA in the chamber fluids for S. aureus appears to be related to the inability of the staphylococci to activate the available complement through the alternative pathway, with little or no contribution from specific antibodies. Previous work with this staphylococcal strain has established that a small amount of specific antibody is necessary for its optimal opsonization in normal serum [18]. In conclusion, the results of the present study indicate that the local levels of CMOA in the extravascular space vary from site to site and are different from those in systemic circulation. The differing opsonic content of the interstitial fluids and the opsonic requirements of the infecting organisms may be important factors in explaining the tropisms of the extravascular space to most bacterial infections. Acknowledgement--We are grateful to Dr C. Janata and Miss E. Wenzel for excellent secretarial assistance.

REFERENCES I. Alexander J. W., McClellan M. A., Ogle C. K. and Ogle F. D. Consumptive opsoninopathy: possible pathogenesis in lethal and opportunistic infection. Ann. Surg. 184, 672-678 (1976). 2. Barza M. and Weinstein L. Some determinants of the distribution of penicillins and cephalosporins in the body. Practical and theoretical considerations. Ann. N.Y. Acad. Sci. 235, 613-617 (1974). 3. Bernhard V. M. Management of infected vascular prostheses. Surg. Clin. Nth Am. 55, 1411-1417 (1975). 4. Casciato D. A., Goldberg L. S. and Bluestone R. Collection of peritoneal exudate cells from small laboratory animals. Vox. Sang. 31, 25-31 (1976). 5. Conrad J. D. Pulmonary opsonins in Klebsiella pneumoniae pneumonia in rats. Infect. Immun. 33, 533-539 (1981). 6. Fearon D. T. and Austen K. F. The alternative pathway of complement--a system for host resistance to microbial infection. N. Engl. J. Med. 303, 259 263 (1980). 7. Fearon D. T., Ruddy S., Schur P. H. and McCabe W. R. Activation of the properdin pathway of complement in patients with gram negative bacteria. N. Engl. J. Med. 292, 937-940 (1975). 8. Forsgren A. and Quie P. G. Influence of the alternative complement pathway on the opsonization of several bacterial species. Infect. Immun. 10, 402-404 (1974). 9. Georgopoulos A. and Sch/itze E. Concentrations of various antibiotics in serum and fluids accumulated in diffusion chambers implanted in various sites in rabbits. Antimicrob. Agents Chemother. 17, 779-783 (1980). I0. Hurley R. M., Muogabo D., Wilson G. W. and Mam A. Cellular composition of peritoneal effluent: response to bacterial peritonitis. Can. mecl. Ass. J. 117, 1061-1062 (1977). 1 I. Johnston R. B. Jr, Klemperer M. R., Alper C. A. and Rosen F. S. The enhancement of factorial phagocytosis by serum: the role of complement components and two factors. J. exp. Med. 129, 1275 (1969). 12. Lam C., Georgopoulos A., Laber G. and Sch/itze E. Therapeutic relevance of penicillin-induced hypersensitivity of S. aureus to killing by polymorphonuclear leucocytes. Antimicrob. Agents Chemother. 26, 149-I 54 (1984). 13. Lew D. P., Despont J.-P., Perrin L. H., Aguado M.-T., Lambert P. H. and Waldvogel F. A. Demonstration of a local exhaustion of complement components and an enzymatic degradation of immunoglobulins in

Regional differences in opsonic activity in rabbits

14. 15. 16. 17. 18. 19. 20.

9

pleural empyema: a possible factor favouring the persistance of local bacterial infections. Clin. exp. Immun. 42, 506-514 (1980). Lew P. D., Zubler R., Vaudaux P., Farque J. J., Waldvogel F. A. and Lambert P. H. Decreased heat-labile opsonic activity and complement levels associated with evidence of C3 breakdown products in infected pleural effusions. J. clin. Invest. 63, 326-334 (1979). Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. Protein measurement with Folin phenol reagent. J. biol. Chem. 193, 265-275 (1951). Schmook F. P., Nefzger M., Laber G., Georgopoulos A., Czok R. and Schiitze E. Composition of fluids from diffusion chambers implanted in the soft tissue and kidneys of rabbits. Infection 8, 156-161 (1980). Schreiber A. D. and Frank M. M. Role of antibody and complement in the immune clearance and destruction of erythrocytes. II. Molecular nature of IgG and IgM complement fixing sites and effects on their interaction with serum. J. din. Invest. 51, 583-589 (1972). Toftte R. W., Peterson P. K., Kim Y. and Quie P. G. Opsonic activity of normal human cerebrospinal fluid for selected bacterial strains. Infect. Immun. 26, 1093-1098 (1979). Zimmerli W., Waldvogel F. A., Vaudaux P. and Nydegger U. E. Pathogenesis of foreign body infection: Description and characteristics of an animal model. J. infect. Dis. 146, 487-497 (1982). Zwahlen A., Nydegger U. E., Vaudaux P., Lambert P. H. and Waldvogel F. A. Complement mediated opsonic activity in normal and infected human cerebrospinal fluid: Early response during bacterial meningitis. J. infect. Dis. 145, 635-646 (1982).