Relationship between angiotensinogen, a l-protease inhibitor elastase complex, antithrombin III and C-reactive protein in septic ARDS U.l-lilgenfeldt1, W. KeUenllann2, G . Kienapfel !, and M. Jochum3

I Dcpartmem of Pharmacology, University of Hcidelbcrg, 1 Dcpartment of Anesthesjology, CHnic Großhadcm and 1 Dcpartmcnl of CJinical Chemislry and Biochemistry. City Clinic, University of Munieh, FRG

Rccci\'cd: May 23. 1989/Accepted in revised form: September 12, 1989

Summary. The time-course of plasma angiotensinogen (Ao), elastase-Cll-prolease inhibitor complex (Ew\PI), antithrombin 1II (AT lli) aod C-reactive protein (eR!» have been invest igated ofsix patients suffering from adult respiralory distrcss syndrome (AROS).

Thc total plasma Ao level (activc and inactive Ao) vancd in individuals but was increascd up to five-fold. An incrcasing amount of inactive Ao is found. From the be­ginning oe their stay in the intensive care unil up to five days haU of the patients displayed a positive eorrelation bctween the pla~ma eRP and Ao level. Thc CRP and Ao values were either nol or were negativeJy correlated with the AT 111 val ues.ln contrast plasma Ao and AT Iß levels in all patients were positively correlated during a particu­lar period in the subsequent phase of the disease, where there was no or a negative corrclation with CRP. The two acule phase reaelants CRP and ELaIPI were only corre­laled in two palients at the beginning ofthc disease.

The markcdly increased plasma level allhe beginning of the inflam matory diseasc indicates that Ao is an acutc phase reaclanl, and this is supported by the parallel changes in plasma CRP and Ao levels during the early days of ARDS. The relationship between the plasma le­vels of Ao and AT 1lI for more than fourteen days suggcsts similar regulation of these members oi thc serpin family after termination oitbe acutc-phasc.

Key words: angiolensinogen, antithrombin, acute-phase proteins, septie ARDS; elaslase-al-protease inhibitor eomplex, C-rcacüve-protein, renin-angiotensin system, adult respiratory distrcss syndrome, sepsis

Angiote nsinogen (Ao) is thc only known precursor of the vasoactive pepl ide hormone angiolcnsin. This plasma gly­coprotein is synthesizcd and secrelcd by the liver. Various hormones are known to be involved in ils synthesis and secretion, inc1uding angiotensin 11 (Khayyal et al., 1973), oestrogens and glucocorticoids (Crane el al., 1971), and thyroid hormones (Dzau and Hcrrmann, 1982). Sub­seq uent to publication of the cDNA sequence of Ao by

Ohkubo and coworkers (1983) , a sequenccrelationship to the serpin (serine protease inhibitors) family, antithrom­bin 111 (AT 1lI), ovalbumin and cxpecially 10 arprotease inhibitor (alPt) was detccled by Dool ittle (1983). This was supported by the finding oi Tanaka cl al. (1984), showing a common Slructural organizalion ofthe Ao and a lPI genes.alPl interacts with lysosomal enzymes, inc1ud­ing kallikrein, various cathepsins and especially elaSlase to form the elastase-al-protease inhibitor compIex (ELa1PI). A III is tbc most important inhibitor oflheclol­ting system (Travis and Salvescn, 1983). AT III and al PI are acute-phase reactants in various an imals (Koj and Re­goecz.i, 1978; Kushner, 1982), whereas in humans A fII is not (Martfnez-Brot6ns CI a1., 1987) . C-reactive protein (CRP) is a further acute-phase reactant, whieh does not belang to the serpin family. The physiological role ofCRP is not yet comletely undcrstood (McCarty, 1982). Recenl data have shown that CRP-peptides are poten t imm uno­modulators (Robey ct al., ]987). Wilhin hours fo llowi ng acule inflammation, thc syn thesis of these protcins is in­creased in the Iiver and their plasma concentration i5 ele­vatcd by as much as 2000-fold (Koj and Gordon, 1983; Morley und Kushncr, 1982).11 has recently been shown that administration of lipopolysaccharides (LPS) 10 rats resulls ina marked increase in a lP] and Ao mRNA in the liver (Kageyamaet a1. , 1983) and an increase in plasma Ao levels (Hilgenfeldt, 1988). These find ingsindicate that Ao is a member of the acute-phase reaclant dass in rats. To verify its role in inflammatory diseases in men, the plasma levels of Ao have been compared over a 4 weck period with those of EuIPI, AT III and CRP in patients suffer­ing from septic ARDS (Madig, 1986).

Materials ::IOd methods

Patients

Plasma sampies from six patients (3 male, 3 female) with a mean age of 33 (18.5, SO) Y and a mean weight of67.7 kg, admitted 10 the In­tensive Care Uni\, Klinikum Großhadern, University of Munich

Ao! (omol Alfml)

10

0.1 ..•.•

0.1

.. ' .... ....

.... .. '

-/ ... ~" . . '

r=0.869

Aod (nmol!ml)

,,' ..... /"

r=O.935

10

~·ig.l. Plasma angiotensmogen in eontrol group obtained by dircet (Aod) and indireet (Aoi) radioimmunoassay (centre group), n = 24. Mean (SO) 0.88 (0.24) nmol Aolml. Plasma angiotensinogen in preg­mIßt women (right group), 11:: 12,6.2 (1.3) nmol Aofml. The solid liDe is the calculated corrclation betwccn the assays and the dotted line shows the theoretical l: I corrclation

were invcstigated. Qinical data are given in Table 1. Only patients kept in the Intensive Gare Unit for more than 10 days were eligible for thc study. These critical1y iU patients all suffered from bacterial sepsis and the adult respiratory distress syndrome (ARDS) (Modig, 1986).

All patients rcceived ant ibiotics, low dose heparin , diazeparn, morphine and pancuronium bromide. In addition, furosemide was administered ir urine flow was reduced. Epincphrine and dopamine were administered to five of the six patients (exocpt PI0) to maintain blood press ure and kidney function.

Crileria for exclusioll. Patients werc excluded in cases of negative blood cu[ture or no focus of severc infection. Patients wilh liver failurc cvideneed by history and laboratory tests, with ncoplasticdis­ease, with any chron ie illncss, or who rcceived glucocorticoids or oral contraceptives were also exc1uded.

Preparation of plasma. Up to four blood sam pies per day were ob­tained [rom indwelling venous catheters. Blood wa~ eollected in 5 ml pla,>tic tubes containing O.S ml trisodium citrate 0.13 M. Plasma was prcpared by eentrifugation ofcitrated blood at 1000 g for 10 min at room temperature and fr07,.en at - 70'C in plustic tubcs.

Assay of angiolensinogen. Ao was completc1y purified from plasma obtained from the Department of Gynaecology, University of Hei­delberg by a five step procedure. Purity was verified by amino aeid analysis, end group sequence analysis and oomplete liberation of N-teTminal angiotensin I by renin. The protein had a specificactivity of 2J.Jl1g ANG I·mg - t protein. Asuming a moleeular wcight of 6O,0IXl Da for human angiotensinogen, the preparation has a punty of 98% (Kienapfel, 1987). Antibodies against human Aowere raised in rabbits by thc method of Vaitukaitis et al. (1971). Plasma Ao oon­centrations were determincd bya direct radioimmunoassay (RIA) perfonned according to lhe method described for rat Ao (Hilgen­feldt and Schott, 1987). Antiserum was uscd at 1 :2501Xl. The sensi­tivity of the assay was 1 fmol Ao. In addition, an indirect assay fol­lowing enzymatic cleavage o[ Ang l with an exces.s of renin (Hilgenfc1dt and Haekcnthal, 1979), was also used. Angl was measurcd by R1A (Menard and Calt, 1972). Ang l antiserum was raiscd in rabbiL~ using glutaraldehydc-linked angioteusin I. The serum was analY-Led for cross-reaetivity with Ang ]\. The antibody titre for AngI and AngIl differed by S.Hr.

Assay 0/ elastase-a)-proleinase inhibitor complex (EICl.[ Pl). ELatPI complcx was measured using a commercial, solid phase, enzyme­linked, sandwich immunoassay (E. Merck AG, Darmstadt, FRG), as described by Joehum ct al. (1983). The range was from 20 to 150 Ilg (l)P I-complex per Htre.

Assay of plasma anlithrombin (ATlU). Plasma AT UI concentration was determined wüh a commereial thrombin inhibition assay (S-2238, Deutsche Kabi-Vitrum, Munich) using p-nitroaniline as the cromogenic product.

Assay ofC-reaelive pro/ein (CRP) . The conceulration ofC-rcactive protein was mcasured by radial immunodiffusion using a commer­eial test system (Behring Werke AG, Marburg, FRG). Thc coeffi­eient of correlation was ca1culatcd by comparing different parame­ters of evcry plasma sampie of eaeh patient using thc computer program Microsoft Chan .

Rcsults

Contro! plasma Ao levels

Thc values obtained by the direct and indirect assays for Ao were compared by measuring tbe angiotensinogen conte nt in plasma irom 25 heall hy volunteers. The plasma Ao levels were 0.876 (0.278) nmol Ao·ml - J and 0.874 (0.237) nmol Ang I .m1 - 1 (mean (SO). This indicates that in hcalthy man the plasma Ao f1 uctuates within a narrow range. In addition plasma fTom 10 healthy pregnant women, with the pregnancy dependent elevation of Ao was also tested. The mean plasma Ao was6.19(1.25) nmol Ao·ml - 1 and 6.07 (1.6) nmol AngI .ml - 1• As shown in Fig. 1, the correlation coefficicnt betwecn thc direcl and indirect assays for Ao in both groups was dose to].

Patients

Ao, ELcxIPI , AT u r and CRP were analyzed in 209 differ­ent plasma sampies from 6 patients. As shown in Fig.2, plasma Ao followed an unique time course in each pa-

Tahle L a) Traumatic event (days prior admission; Day 0 = bcgin­rung of the sludy, Day of admission to the intensive care uni!; m ", male.f = female) b) Clinicaldiagnosis(n =6)

P9 a) 10 (f), Dear drowning b) Severe ARDS two days prior admission, Legionella

pneumophilia , sepsis syndrome on Days 8-IOof thc study with aeutc renal insuf[iciency, rhabdomyolysis and death on Day 27.

P 10 a) S (m) polytrauma. b) Pneumonia, sepsis (Slaph. al/reus, Ps. aerugil1ol"U),

ARDS. On Day 6 aeute renal insu(fieiency. Second infcction on Day 16.

P I7 a) 8h(m)Polytraumaandsevere posttraumatic ARDS.

b) At Day 10 pneumonia, at Day 18-20sepsis and acute renal insufficicncy.

P45 a) 3 (f), sepsis syndrome 6 h prior admission. b) Shock followed by severe ARDS, reeurrent

infeetions (pneumonia, urinary traet infections). On Day 3 aeute renal insufficiency.

P48 a) OCr), hysterectomy. b) postoperative sepsis (Cl. perjringens), severc

consumptive coagulopathy. Day 3 re-[aparotomy

PS3 a) 7 (m), polytrauma. b) Sepsis, scverc ARDS required artificial respi ration

for 24 days. Day 2 acute renal insufficieucy.

P9 '"

30

1

0, 10 15 20 15

j~ 0510152025

"0 .. '" n.,1/11 1

1.5

• 0.'

00 10 15 20 ~S 30 0

0 1 , , 10 12

" 0 .. 1/.1 Pt7 , .. n.al/I'

PS,

: A 1i-~\~ ..... t~u~r l~ o 5 10 15 20 ~ * ~

time (days)

• , 1

PO ElI I?1 U/.l

"'0

'00 '00

100

o !0-"1,--;--",-C';--;;10,OI1;--;I' time (aays)

.. nlHlllll

• , '"

ti g.2. Plasma Aolevels (nmoVml) oblaincd by the dircci (solid symbols) and indircci (open symbols) radioimmunoassay (nmol A IIml) in six patienlswith septic A RDS

ElalPI U/I\

600

500 <00 300 100 100

Os 15 15 0 O~----~----~~-"O o 10 20

PIO

1

OL---------~--__e o 10 20

.. IWIOl/.l

, 1

PI7

10 20 time (days)

30

ELalPI 11/.1

'" '00 <00

0.'

0 0

ElalPI Ujll

.. nllOl/.l

1000 , '" 600

1 <00

100

0

pO,

1 , , 10

PS,

1 6 B 10 12 tlle (days)

ElalPI 11/.1

6000

<000

2000

0

" fLalPJ

11/.1

'" 600

.00 100

l·ig. 3. Plasma Aoand ELa, PI levels in six patients suffering (rom sepsis. The Ao "alues in the dircct radioimmunoassay are consistent with those ofFig.2. (Au solid linc; ELa,PI open symbols)

" " UIII " '" Alill 11I01/al • ßIIOII.I • • '00

'" , ". " 80 " , " ..

• .. • " , 15 " • " " '" ATllI " '" Uill

l\JIo l/., • n.ollll • , 15. 1.' "" ,. "" ,

.. , \...., 90

• • 70 • " " • , • , , " "

" '" " '" Il101/111 UIII MO/Ci I ATlI I • • • 3

~r BO fig.4. Plasma Aoand AT II! levels in , "" six paticntssuffcring from sepsis. Thc ,

lOG Ao valucs in the dircci radio-

" immunoassay are consistcnt with those

" ofFig.2 (Ao solid liDe; AT IIl open

• .. 0 - 60 symbols) • " " 30 • , • , , " " " time l~aysJ Ulle Id~Y$1

" " CRP x, '" CRP n.,l/., Ig/dl fYlo/_ ' ~/dl

" • '" • " , " "

0 0 0 0 , 15 " 0 " '" " ". CRP .. '" CO'

MOI/II '9/dl oaol/il ~/dl , L5 "

'" ,

" .. , • 0 0 • • • " '" 30 0 ,

" " '" CRP .. '" CO'

nllDl/l' IlIJ/dl I\l10/11 IIg/dl

• '" " 3

B . ' ig. S. Plasma Ao and CRP levels in six , palienlssuffcring trom septic A RDS. ,

" TIle Ao valucs in lhe di rect rad io-immunoassay are consistent with those of Fig. 2 (Ao solid line; e RP open

0 0 0 0 symbols) 0 " " 30 0 ,

" t lle (days! U .. (days)

T:.ble 2. Corrclation c()(!fficicnt (r) of plasma Ao levels obtained by direct (d) and indirect (i) RI A and corrcsponding plasma ELa1 PI , ATII I and CRP in six patients suffcring rrom septic ARDS. Thc time·period after admission to thc hospital (days) and as thc isolated valucs (rl) dur· ing this pcriod are given. Values in parenthescs arc the correlatiOD cocfficicnts fo r limitcd time periods

Patient No 9 10

Aod·Aoi 0.51 0.96 (0.814) (0.758) 10.4891 {0.955)

Aod·ELo.1I'1 0.500 0.478 ( - 0.479) ( -0.737) 1- 0.6481 10.4381

Aod·ATIII 0.69 0.038 (0.468) ( -0.476) 10.8841 10.7281

Aod·CRJ> - 0.202 0.494 ( - 0.461) (0.778) 1- 0.2461 10.0451

AT III ·CRP - 0.303 0.488 ( - 0.483) ( -0.402) ! - 0.434) 10.323}

ELa,PJ.CRP - 0. 171 0.351 (0.685) (-0.531) {0.235] 10.405)

time a rte r adm ission (days) O-V 1.3-29 (0-186) (1.3- 5.3) {IO-VI {9 -291

" 37 46 (20) (9) 1171 1311

rient. Thc amount of Ao in all palients was incrcascd at SQmc time during Ihe period of observat ion. Jn addition, the values oblai ned by Ihe dircct and indirect measure· ment show marked di(fere nces. In contrast to the control data given in rig. 1, th e two assay systems did oot show a 1: 1 ratio in plasma from Ihese patients, although the values were st ill correlated (Table 2).

Within a cc rtain pc riod, thc values obtained with tbe direct assay were significant ly higherin P9, P 17, P45 and P 53. This indicales increased formation of Angl, leadiog to an iocreased plasma conccntration of des-AngI-Ao. Stimulation of the components of the renin-angiotensin system, e.g. Ao, ren in, Ang J and Ang n , in some of these patients has reeently been reported (HiJgenfc1dt ct al., 1987). The Ao values obtained by the direct assay werc compared with the Icvc lsofELa1PI, AT IJI and CRP. Plas­ma Ao and ELa1PI in a ll six patients are shown in Fig.3. A positive correlalion of Ao and ELa1PI was not observed in most patients. Only in P 10 was there nuctuation of the ELa1PI values around the Ao va lues, although they were not positively correlaled; in P 53 r = 0.734 was found wiLh­in fi ve days after admission.

Tbc plasma levclsof Aoand AT llJ are shown in Fig.4. A positive corre lat ion of Ao and AT III was evident in plasma from three pa ti ents during Ihe enlire observation period . with r bctwecn 0.69 and 0.71 . The other three pa ­tients (P 10, P48 and P53) show a good correlation within a linliled period (Tab le 2, dala in parentheses) . alPT and AT III are mcmbers orl he same gelle family and so are rc­lated to angiotcnsioogcn. However, the acute phase pro­te in CRP does not belang to this group. The plasma lcvels

17 45 48 53

0.649 0.932 0.709 0.74 (0.967) (0.827) 0.876) 10.5781 ( - 0.5831 10.474)

- 0.580 - 0587 - 0.357 0.489 ( - 0.699) ( - 0.509) (0.734) ( - 0.1351 1- 0.791) 1-0.1231

0.709 0.711 - 0.39 - 0.351 (0.851) ( - 0.679) ( -0.636) 1- 0.0961 {0.8691 10.7151

- 0.294 0.213 0.531 0.579 (0.567) (0.789) (0.784) 10.2371 [0.1071 {0.1441

- 0.417 - 0.190 - 0.793 - 0.825 (0.354) ( - 0.617) ( - 0.706) [ - 0.7111 {- 0.794] {0.0701

0.083 - 0.300 0.233 0.736 ( - 0.233) (- 0.350) (0.634) {- 0.6671 {0.3071 {- 0.4991

0-34.2 0-23.7 0-11 0-13 (0-14) (0- 5) (0- 5) {14-241 15-111 15-131

35 49 16 26 (30) (10) (12) 1191 161 1141

of Ao and CRP arc shown in Fig.5. lt is interesting to note that in P 10, P48 and P S3 the Ao and CRP values were positively corrc1aled att he bcginning of Ihe illness. In the time inlerval in which plasma Ao and CRP were positively correlated, however, Ao aod ATllI showed a negative or no corre lation (see Table 1). A eorrclation between CRP and ELalPI was Icss common. In addition , th ere was 00 or almost a negative correlation between CRP and AT JIJ in plasma [rom all the paticnts.

Discussioll

The mean plasma Ao Icvel in healthy volunteers was 0.9 (0.3) llM. An increase in plasma Ao is observed in preg­nancy. in wornen taki ng o ral contraceptives, and in hyper­tension (Gordon, 1983). The possibilit y that Ao may be­have as an aeute-phase protein was first raised by Billg (1972), although he did not rind an increase in plasma Ao Icvels following turpentine injection in the rat. Thc idca that Ao may be an aeute-phase reactant was revivcd after the finding that Ao was gcnetically dosei)' related to (X1PI (Dooliltle, 1983 and Taoaka et al., 1984). The idea was supported by the find ing of as much as a five-fold increase in Ao mRNA in liver (Kageyama et al. 1983) and a three­fold increase in plasma Ao levels (Hilgenfeldt, 1988) fol­lowing lipopolysaccharide administration to rats. A1-thougb these datH are consistent with the hypothesis thai Ao is an acute-phase reaclant in hurnans, caulion is needed in inferring thaI da la Crom one spccics can be ap­plied to another; for example, al Pl is not an acute·phase

reaetant in mice (Baumann el al., 1986), and CRP dem on­strates striking species variations in response 10 tissue in­jury (Pepys ct al., 1979).

Nielscn and Knudsen (1987) reeently reported a 70% inerease in plasma Ao levels in patients with aeute inflam­matory disease, indieating that Ao may be an aeute-phase reaetanl in man. Sioec the regulation of plasma Ao levels is com plex, an increascd level on its own cannot be ac­eepted as proof of this hypothesis. Thc acute-phase re­sponse oceurs in the period immediately following injury and infeetion. Thercfore, thc prcsenl study was designed tofollow the time course of change in Ao levels in patients with bacterial sepsis. Tbe changes were compared with those in known aeute-phase reaetants.

Inereased plasma levels of Ao were found during the first 3 to 5 days of observation in five o[ the six patients. The correlation between Ao and the other aeUle-phase proteins in our study'was based on the direet RlA for Ao. This assay recognizes the total Ao eoneentration and ig­nores Ao eatabolism by plasma reoin. In addition, both proteins, intaet Ao and des-AngI-Ao, have an almOsl identieal plasma half life in vivo (Hilgenfeldt, 1988) and should be eliminated to a similar exlen t.

The individual course of the disease in eaeh patient makes it difficult 10 find a eommon prineiple. 11 is im­possible at present to model such eomplexeurvesof differ­ent plasma parameters, so it is only possible to eompare different parameters !rom eaeh plasma sampie and to fit them in a linear eorrelation . A relationship between Ao and alPI was seen in only one patient during five days after admission (P 53; r = 0.73). However, thc assay for Ct.IPI only measured the ELCt.IPI-eomplex and not the total amount of aIPI. Thcrefore , a eorrelation between Ao and alPI would only have been dcteeled if the amount o[ the EwlPI-eomplex and of free a lPI were linked. A far stronger relationship was found betwcen Ao and AT 111, with a signifieant eorrelation during lhe entire observa­tion period in tillee patients (P 9, P 17, P45).ln those pa­tients Da eorrelatiOD of Ao and CRPwas found. The other three patienls, PIO, P48, P53, displayed a good corre· lation between Ao and CRP duriog the initial part of the study (r = 0.78, 0.79, 0.78). OuTing this peTiod bolb parameters were negatively eorrelated wilh ATlll (Ao-ATTII: r~~0.48, ~ 0.68 , and ~ O.64 ; CRP-ATIIl: r = -0040, - 0.62, -0.71). In lhe subsequent period Ao and ATIII displayed a good correlation (r = 0.73, 0.87, 0.72) without being related to CR P.

Tbe good correlation of the plasma levels of Ao and CRP during the first sevcral days of hospitalization sug­gests that Ao is an aeute-phase reaelan! in man. That this correlation was not fouod in all six patients was mainly due to the fact that data from Ihe initial aeute-phase could only be obtained from one patient (P48). On the other hand, plasma Ao and ATJJI levels showed a farcloscr re­lationship, although ATIJI is not an aeute-phase protein in man (Martinez-Brot6ns el al., 1987). However this eorrelation is observed during a time period when Ao and eRP are nOI eorrclated. This indicates that in the laner period the aeutc-phase has alrcady been terminated. A eorrelation in the plasma levels of gene tieally related pro­teins like Ao, alP I and Arm, indicates not only a similar

synlhetie response to stimuli, but in addition it points 10

similar rates of catabolism and elimination, despite tbeir molecular and funetional differenees. There isevidence of inereased Ao eatabolism in acute inflammation, since there is a deerease in the in vivo half-life of o ne form oe Ao in rats following LPS Slimulation. Deereased renal elimi­nation, whieh might be eaused by deereased renal blood flow, has also becn diseussed (Hilgenfeldt, 1988).

AI present a common stimulus for aeute-phase pro­teins cannot be defined. Yery reeently it has been shown that the synthesis of various aeute-phase proteins is medi­aled by cytokines such as interleukin-l (iL-l), inter­Icukin-6 (IL-6) and tumor neerosis [aetor-o. (TNF-a), by a direct action on hepatoeytes (Ramadori et al., 1988). However while the seeretion of a r maeroglobulin is mar· kedly aeeeierated by TL-6, the seere li on ofeystcine protei­nase inhibitor and alPI is unaffeeted (Andus Cl al., 1988). On the othcr band, a maerophage derived faetor is able 10 stimulale the synthesis and seeretion both of ATIIT and o.lPI (Hoffmann ct al., 1986). Thc biosynlhcsis ofthe mOSI prominent aeute-phase protein CRP by tbc liver is prc· sumably mediated by inlerleukin-l (Kaplan, 1982; Syin cl al., 1986 and Goldberger et al., 1987). Yery reeently it has been shown that the aeute-phase response of Ao is medi­ated by interleukin-6 in rat hepatoma eeUs (ltoh et al., 1989), bul this requires the presence of glucocortieoids, suggesting that the laller are essential for Ihe stimulatory effeet.

The role of angiotcnsinogen in inflammation cannot yet be defined. It appeared (rom a previousstudy (Hilgen­feldt el al. , 1987) that thc renin-angiolensinsyslem did not playa role in thc maintenanee of blood pressurc in such patients. Howevcr, angiolensin may be importaot in the immune response during intlammation (Fcrnandez-Cas+ tello el al., 1987).

In eonc1usion, up 10 a five-fold inereasc in plasma Ao levels has been found in paticnts sufferi ng from scptie ARDS. The positive eorrelat ion bctween Ao and CRP during the first days after admittanee to the intensive eare unh indicates that Ao is an acute-phase reactant in man. There was also a good eorrelation between plasma Aoand AT lllieveis during the later stage of disease, suggesting similarities in the synthesis and elimination kinetics of these proteins. Further studies are required 10 define the funetional role of Ao in inflammatory diseascs.

Acknowledgemem. The work was supporlcd by a grant from the Deutsche Forschungsgemeinschaft , Bonn -Bad Godesberg, Fed. Rep. Germany.

References

Andus '1', Geiger T, Hirano T, Kishimoto T. Tran-Thi TA, Deeker K, Heinrich pe (1988) Regulation of synthesis aud seeretion of major acute-phase proteins by reeombinant human inlerleukin·6 (BSF-211L-6) in hepatocyle primary culturcs. Eur J Bioehem 173: 287- 293

Baumann H, Latimer 11, Glibetic MD (1986) Mouse (ll-prolease in­hibitoris not an aeute-phase reactant. Arch Biachem Biophys246: 488-493

Bing 1 (1972) Relation betwecn renin substrate and aeutc phasc pro­teins. Acta Pathol Microbiol l mmunol Scand (Al 80: 646-650

Crane MG, Harris JJ , Winsor \V I 11 (1971) H ypcrtension, oral con tra­ccptive agents,andconjugatcscstrogens. Alln Intern Me<! 74: 13

Doolittle RF (1983) Angiotcnsinogcn is related to the antitrypsin­antithrombin-ovalbumin family. Science 222: 417-419

Dzau VJ , Herrmann HC (1982) Hormonal contral of angiotensi­DOgcn production. Life Sdcnces 30: 577-584

Femandez-Castello S, Arzt ES, Pesce A, Criscuolo ME, OiazA , Rn­kielman S, Nahmod VE (1987) Angiotensin 11 regulates inter­feron produetion, J Interferon Res 7: 261-268

Goldbcrger G, Bing Ol-l, Sipc 10, Rits M, Colten HR (1987) Tran­seriptionfll regulation of genes encoding the acutc-phase proteins CR P, SAA and C3. Ilmmunol138: 3967-3971

Gordon O B (1983) T he role of renin substrate in hypertension . H ypertension 3: 353-362

Hilgenfeldt U, I-Iackenthal E (1979) Purification and charac­terization of rat angiotensinogen. Biochirn ßiophys Acta 579: 375-385

Hilgenfc\dt U, Schott R (1987) D ifferences in plasma angiotensi­nogen pattern of native and nephrectomized rats. Hypertension 9: 339-344

I-l ilgenfeldt U, Kienanpfel G , Kellennann W, Schot! R, Sehmitt M (1987) Renin Angiotensin System in Sepsis. Clin Exp H ypertens A 9 (8&9): 1493- 1504

]-J iI genfc\d t U (1988) Half -l ife of rat angiolensi nogen: in n uenceof ne­phrectomy and lipopolysaccharide stimulation. Mol Cell Endocri­noI56:91 -98

Ho(fmann M, Fuchs HE, Pizw SV (1986) l be macrophagc-medi­aled rcgulation of hepatocyte synthesis 01' amithrombin !!I and al-protease inhibitor. Thromb Res 41: 707-715

Itoh N, Matsnda T, Ohtani R, Okamoto H (1989) AngiotensiDogen production by rat hepatoma cells is stimulated by B-ccU stimula­tory faetor-2 interleukin-6. FE BS Leiters 224: 6-10

Jochum M, Duswald KH , NeumannS, Wille J, Fritz H , SecmüUer U (1983) Proteinasesand thei r inhibitors in inl1ammation: basic con­cepts and eliniea! implications. In: Katunnma N, Umezawa H, Holzer H (cds) Proteinase inhibitors: Medical and biologica! as­peets. Japan Scientifie Soeieties Press, Tokyo/Springer, Berlin , pp 85-95

Kageyama R, Ohknbo H, Nakanishi S (1983) Induction of rat liver angiotensinogen mR NA fo llowing acute inl1ammation. Biochim Biophys Res Commun 129: 826--832

Kaplan MH (1982) C-reaclive protein: relation todisease and patho­logicaI significance. Ann NY Acad Sei 389:419-422

Khayyal MI, Mac Gregor J , Brown Jl , Lever Ar. Robcrtson IS (1973) Increasc or plasma renin substrate concentration after in­fusion of angiotensin in tlle ral. Clin Sci 44: 87-<xl

Kienapfel G (1987) Human Angiotensinogen: Rcinigung,Charakte­risierung und direkter Radioimmunoassay. Ooc{orate thesis, me­dica! faeulty, University of Heidelberg

Koj A, Regoeczi E (1978) Effeet or experimental inl1ammation on the synthesis and distribution of antithrombin III and u ,-antilryp­sin in rabbits. Br J Exp Pathol59: 473-481

Koj A, Gordon AH (1983) The acnte phase response to injury and infeetion. Res Monogr Ceil Tiss Physiol 10. Elsevier,Amsterdam

Kushner I (1982) Thc phenornenon of acute phase response. Ann NY Acad Sei 389: 39-48

Mart inez-Brol6ns F, Oneins JR, Mcstrcs J, Amarg6s and Reynaldo C (1987) Plasma Kallikrein-K inin system in patients with uncom­plicated sepsis and seplic shock - comparison with cardiogenic shock. Thormb Haemoslas 58: 709-713

McCarty M (1982) Historiea! pcrspective on C-reactive protein. Ann NY Acad Sei 389: 1-10

Menard J, Catt KJ (1972) Mcasurement of renin activity, coneentra­tion and subSlrate in Tat plasma by radioimmnnoassay of an­giotensin I. Endocrinology 90: 422-430

Modig J (Im) Clinical Review: Adult respirator)' dislIess syn­drome, pathogenesis and treatment. Acta Chir Scand 152: 241-249

Morley 1J, Kushner I (1982) Serum C-reactive protein levels in dis­ease,Ann NY Acad Sei 389:406-418

Mortcnsen F, Sarlo K, Le PT (1984) Monokine-induccd hepatocyte synthesis of aculC-phase rcactants in mice. Lymphokine Res 3: 308-322

Nielsen AH, K nudsen F (1987) Angiotensinogen is an aeute-phase protein in man. Scand 1 Clin Lab lnvest 47: 175-178

Ohkubo H, Kageyama R, Ujihara M, Hirose T, Inayama S, Nakan­ishi S (1983) Cloning and sequencc analysis of cDNA of rat an­giotcnsinogen. Proc Natl Acad Sei USA 80:2196-2200

Pcpys MB, 8ahz M , Gomer K, Davls AJS, Docnhoff M (1979) Serum amyloid-componem is an acute-phase reaClant in the mouse. Nature 278: 259-261

Ramadori G, van Damme J, Rieder H , Meyer zum Buschenfelde KH (1988) Interleukin-6, the third mediator o[ aeute-phase reac­tiOll , rnodulates hepatic protein synthesis in hnman and mouse. Comparison with interleukin-l ß and tumor nenosis faetor-er:. Eur J Immunol18: 1259- 1264

Robey FA, Ohura K, Futaki S, Fnjii N, Yajima, Goldman N, Jones KD, Wahl S (1987) Proteolysis of human C-rcactive protein pro­duces peptides with potent immunomodulating act ivity. J Biol Chcm 262: 7053-7057

Sy!n C, Gotschlich EC, Liu TY (1986) Rabbit C-reactive protein, biosynthesis and characlerization uf cDNA clones, J Biol Chern 261: 5473-5479

Tanaka T. Ohkubo H, Nakanishi S (1984) Common structural orga­nization of the angiotensinogen and the er:,-antitrypsin genes. 1 Biol Chem 259: 8063-8065

Travis J, Salvescn GS (1983) H uman plasma proteinase inhibiton;. Ann Rev l3iochem 52: 655-709

Vaitukaitis J, Robbins JB. Nicschlag E, Ross GT (1987) A method for producing specific antisera with small doses of immunogen . J Clin Endon 33: 988- 991

Prof. Dr. U. Hilgenfeldt Pharmakologisches Institut Universität Heidelberg Im Nenenhcimer Feld 366 0 -6900 Heidelberg, FRG