8/3/2019 Acidosis Activates Complement

1/4

Acidosis activates complementsystem in vitro

Michael Emeis,1 Josef Sonntag,1.CA Carsten Willam,2

Evelyn Strauss,1 Matthias M. Walka and

Michael Obladen1

Departments of 1Neonatology and 2Nephrology,Charite, Virchow-Hospital, Humboldt-UniversityBerlin, 13353 Berlin, Germany

CACorre sp onding AuthorTel: (+49) 30 45066463Fax: (+49) 30 45066922

WE investigated the in vitro effect of different formsof acidosis (pH 7.0) on the form ation of an aphy latox -ins C3a and C5a. Metabolic ac idosis due to a ddition of

hydrochloric acid (10mmol/ml blood) or lactic acid(5.5 mmol/ml) to heparin blood (N=12) caused sig-nificant activation of C3a and C5a compared tocontrol (both p=0.002). Res piratory acidos is ac tivatedC3a (p=0.007) and C5a (p =0.003) comp ared to nor mo -capnic controls. Making blood samples with lacticacidosis hyp ocapn ic resulted in a me dian pH of 7.37.In this respiratory compensated metabolic acidosis,C3a and C5a were not increased. These experimentsshow that acidosis itself and not lactate trigger foractivation o f comp lemen t comp onen ts C3 and C5.

Key words:Ac idosis, Alcalosis, Compleme nt syste m, Com-

plement activation, Anaphylatoxins

IntroductionHypoxia and reperfusion cause complement activa-tion in animal experiments and in clinical studies.1 4

It could be shown, that after cytological damage,contact w ith ce llular components such as mitochon-dria, or excess of hydroxyl radicals, was responsiblefor the activation of the complement system.5 ,6 Onthe other hand, in vitro studies showed a comple-ment activation by acidosis or hypoxia only.7 9 In aprevious study, we found complement activation

induced by lactic acid.10

The aim of the present invitro study was to investigate: (1) whether lactateitself or metabolic acidosis is responsible for activa-tion, and (2) whether respiratory acidosis is also ableto activate the c omplement system.

Material and Methods

Blood samples (10 ml) from 12 healthy volunteers (6male and 6 female, age: 28 40 years) we re collecte dand 50 IE Heparin was added. Each sample was

divided into eight portions and placed in poly-propylen tubes.

To investigate the influe nce of resp iratory as w ell asmetabolic change s of pH on anaphylatox in formation,the portions were equilibrated with different gasmixtures and supplemented with acids resulting inmarked changes in pH, pCO2 and bas e de ficit (Table1). Probe s use d as c ontrols are No. 13. Four portions

were equilibrated with different gas mixtures at 37C

to achieve normocapnia and normox emia in controlportions (No. 3; Table 1), respiratory acidosis (No 6;Table 1) or alcalosis (No. 7,8; Table 1). For thispurpose we modified the me thod of Siriwardhana e ta l.11A 6 ml polypropylen tube (6 ml) w as filled 1.5 to2.0 ml blood and the end o f a polypropylen tube w ithan inner diameter of 3 mm w as placed into the blooda few millimetres above the bottom of the tube. Gasmixtures (O2 , N2 , CO2 ) composed from medicalgases by flowmeters were put through the tube intothe blood sample. Gas flow was adjusted to one

bubble pe r sec ond for 20 min. Foam caused bybubbling was continuously removed by a suctioncatheter placed above the blood level. After equilibra-tion, the blood samples were transferred into differ-ent tubes for further processing.

For metabolic acidosis 10 mmol hydrochloric acid(No. 4; Sigma-Aldrich, Deisenhofen, Germany) or5.5 mmol lactate (No: 5; Sigma-Aldrich, Deisenhofen,Germany) w ere added pe r ml of blood. To achievecompensated metabolic acidosis, the sample wasequilibrated with a hypocapnic gas mixture for

20 min and the n 5.5mmol lactate pe r ml of blood w asadded (No. 8).

Blood gas analysis, potassium and lactate conce ntra-tions were measured by a Radiometer Copenhagen

ABL 505 (Willich, Germany) using heparinised syrin-ges after all samples w ere incubated at 37C for 1 h. Tostop complement activation after incubation, weadded 1 mg EDTA disso lved in purified w ater to e achsample. Plasma was separated by centrifugation at

0962-9351/98/060417-04 $9.00 1998 Carfax Publishing Ltd 417

Short Communication

Mediators of Inflammation, 7, 417420 (1998)

8/3/2019 Acidosis Activates Complement

2/4

4C at 3000 g for 10 min and frozen immediate ly at80C.

Anaphylatoxins C3a and C5a were determined aspreviously described.12 C3a enzyme immunoassay(EIA, Fa. Progen Biotechnik GmbH, Heidelberg, Ger-many) selectively detects C3a-desArg using mono-clonal antibodies.13 C5a was determined with aspecific sandwich EIA (Fa. Behring, Marburg,Germany).14

Statistical analysis

As most of the data were not normally distributed,results w ere show n as me dians w ith quartiles. Differ-ences between controls and study samples wereassessed using the Wilcoxon test. Statistical sig-nificance w as assumed at p

8/3/2019 Acidosis Activates Complement

3/4

those from lactic acidosis (No. 4 versus 5). Theanaphylatoxins were also higher in the hypercapnicsamples c ompared to the normoc apnic c ontrols (No.6 versus 3; Table 1).

Influence of respiratory alkalosisRespiratory alkalosis had no influence on anaphyla-tox in conc entration (No. 7 ve rsus 3; Table 1).

Influence of respiratory compensatedmetabolic acidosis

Although measured median lactate (13.1 versus11.6 mmol/l) w as not different in the lactic sample s(No. 5 vers us 8;p=0.75), anaphylatoxins w ere higherin acidic samples compared to respiratory compen-sated normac idic sample s (No. 5 ve rsus 8; Table 1).

Discussion

The p rese nt in vitro study confirms the results of ourprevious study that acidosis activates complementsystems in heparin blood. The study shows that theacidosis itself is the trigger for activation, bec ause allthree forms of acidosis (hydrochloric or lactic acid orcarbon dioxide) lead to a significant increase of

anaphylatox in concentrations. Lactate did not lead toan activation, when acidosis was prevented byprev ious re sp iratory alcalosis (No. 8; Table 1).

Heparin w as chos en as anticoagulant, bec ause thecomplement may be markedly spontaneously acti-vated in serum and both EDTA and sodium citratepossibly influence pH values and impair complementactivation by calcium binding. Heparin however didnot completely prevent spontaneous complementactivation e ven after incubation for 1 h at 37C andusing a bubble oxygenator for gas equilibration.

Therefore we could only compare the extent ofcomplement activation by acidosis to the sponta-neous activation that occurred during the samplepreparation procedure. As pH levels were similar inlactic and hydrochloric acidos is, the activation of thecomplement system was stronger for hydrochloricacid. The effects of the two acids on the blood gasanalys is were different. Whereas lac tate led to a large rincrease in base deficit, hydrochloric acid caused astronger release of carbon dioxide. This differencemay explain the stronger complement activation by

hydrochloric acid. Respiratory acidosis followingequilibration of blood samples with elevated carbondiox ide conc entrations also led to an activation of thecomplement system compared to equilibration con-trol. In contrast to acidosis, respiratory alcalosis didnot have an influence on the complement system.

With the use d in v itro setting, we could not clarifythe mechanism of anaphylatoxin formation by acid-osis. Complement activation in acidosis mediated by

membrane fragments of damaged erythrocytes6 isimprobable, as we could show that activation alsooccurs in plasma without cell components.1 0 Thereare other arguments against membrane fragment-induced complement activation in ac idosis, inc rease dpotassium values and a slight visible hemolysis

occured only for metabolic but not for respiratoryacidosis, but both forms of acidosis activate thecomplement system. A probable mechanism is thatacidification of plasma inactivate s compleme nt prote-ase inhibitors, this inhibition then results in comple-ment activation in the abs ence or decrease d presenc eof functional inhibitors .

The activation of the complement system byacidosis may explain the elevation of anaphylatoxinsin different diseases, such as perinatal asphyxia,myocardial infarction and shock.3,4,1517 These pro-cesses probably participate in the pathogenesis ofrepe rfusion injury.4,1719

Conclusion

We have shown evidence that acidosis activatescomplement factors C3 and C5. This is independentfrom the type of acidosis which occurs.

References1. Hill J, Lindsay TF, Ortiz F, Ye h CG, Hec htman HB, Moore FD. Solub le

complement receptor type 1 ameliorates the local and remote organinjury after intestinal ischemia reperfusion in the rat. J Immunol 1992;149 : 17231728.

2. Brus F, van Oevere n W, Okken A, Oetomo SB. Activation of circulatingpolymorphonuclear leukocytes in preterm infants w ith seve re idiopathicrespiratory distress syndrome. Pediatr Res1996; 39 : 456463.

3. Sonntag J, Wagne r MH, Strauss E, Obladen M. Complem ent and c ontactactivation in term newborns after fetal acidosis. Arch Dis Child FetalNeonatal Ed1998; 78 : F125F128.

4. Horstick G, He imann A, Gotze O, Hofner G, Berg O, Boehmer P, et al.Intracoronary application of C 1 esterase inhibitor improves cardiacfunction and reduce s myocardial nec rosis in an expe rimental model ofischemia and repe rfusion. Circulation 1997; 95 : 701708.

5. Turnage RH, Magee JC, Guice KS, Myers SI, Oldham KT. Compleme nt

activation by hydroxyl radicals during intestinal reperfusion. Shock1994; 2: 445450.

6. Rossen RD, Michae l LH, Kagiyima A, Savage HE, Hanson G, Reisbe rg MA,et al. Mechanism of complement activation after coronary arteryocclusion: Evidence that myocardial ischemia in dogs c auses release ofconstituents of myocardial subcellular origin that complex with h umanC1q in v ivo . Circulat ion Resea rch 1988; 62 : 72584.

7. Fishe lson Z, Horstman RD, Muller-Eberhard HJ. Regulation of thealtemative pathway of complement by pH. J Immunol 1987; 138 :33923395.

8. Hamm er CH, Hans ch G, Gres ham HD, Shin ML. Activation o f the fifth andsixth components of the human complement system: C6-dependentcleavage of C5 in acid and the formation of a bimolecular lytic complex.

J Imm uno l 1983; 131 : 892898.9. Collard CD, Vake va A, Bukuso glu C, Zund G, Spe rati CJ, Colgan SP, et al.

Reoxygenation of hypoxic human umbilical vein endothelial cells

activates the classic compleme nt pathway. Circulation 1997; 96: 226.10. Sonntag J, Emeis M, Strauss E, Obladen M. In vitro activation of

complement and contact system by lactic acidosis. Mediat Inflamm1998; 7: 4951.

11. Siriwa rdhana SA, Kaw as A, Yates S, Gulden H, Lipton JM, Giese cke AH.Oxygen mediated complement activation: an in-vitro study. Can J

Anaesth 1990; 37 : 116.12. Sonntag J, Stiller B, Walka MM, Maier R. Anaphylatox ins in fresh frozen

plasma. Trans fus ion 1997; 37 : 798803.13. Burger R, Zilow G, Bader A, Friedlein A, Nase r W. The C terminus of the

anaphylatoxin C3a generated upon c omplement activation repre sents aneoan tige nic determinant with diagnos tic potential. J Immunol 1988;14 : 553558.

Acidosi s activates complement sy stem

Mediators of Inflammation Vol 7 1998 419

8/3/2019 Acidosis Activates Complement

4/4

14. Klos A, Ihrig V, Messme r M, Grabbe J, Bitter-Suermann D. Detection ofnative human complement components C3 and C5 and their primaryactivation peptides C3a and C5a (anaphylatoxic peptides) by ELISAs

with monoclonal antibodies. J Immunol Methods1988; 111 :241252.

15. He ideman M, Hugli TE. Anaphylatox in gene ration in multisystem organfailure. J Trau m a 1984; 24 : 10381043.

16. We ise r MR, Williams JP, Moore FD, Kobz ik L, Ma M, Hec htma n HB, et al.Reperfusion injury of ischemic skeletal muscle is mediated by natural

antibody and compleme nt. J Exp Med1996; 183: 23432348.17. Buerke M, Murohora T, Lefer AM. Cardioprotective effec ts of a C1

esterase inhibitor in myocardial ischem ia and rep erfusion. Circulation1995; 91 : 393402.

18. Ikai M, Itoh M, Joh T, Yokoyama Y, Okada N, Okada H. Complemen t playsan esse ntial role in shock follow ing intes tinal ische mia in rats. Clin ExpI m m u n o l 1996; 106 : 156159.

19. Crawford MH, Grove r ML, Kolb WP, McMaha n A, ORourke RA, McManusLM, e t a l . Complement and neutrophil activation of the pathogenesis ofischemic myocardial injury. Circulation 1988; 78 : 144148.

ACKNOWLEDGEMENTS. This w ork w as s upp orte d by th e Ge rma n Rese archSociety (DFG-0b 43/62).

Received 10 September 1998;accepted 5 October 1998

M. Em ei set al.

420 Mediators of Inflammation Vol 7 1998