JOURNALOF CLINICAL MICROBIOLOGY, Apr.1979,p.476-478 Vol.9,No.4 0095-1137/79/04-0476/03$02.00/0
Immunoglobulin M Anti-Hepatitis A Virus Determination by Reorienting Gradient Centrifugation for Diagnosis of Acute
Hepatitis A
G.G. FROSNER, R. SCHEID, H. WOLF,* AND F. DEINHARDT
MaxvonPettenkofer-Institut, University of Munich, 8000 Munich 2, Federal Republic of Germany Received forpublication15December 1978
Thepersistence ofantibody to hepatitis A antigen (anti-HAV) of the immu- noglobulin M(IgM)class wasevaluated in 88 sera of 51 acute hepatitis A patients.
IgMwasseparated from IgGby a 2-hreorientingsucrose gradientultracentrifu- gation, and the titer of anti-HAV wasdeterminedintheIgG-andIgM-containing fractions by solid-phase radioimmunoassay. IgM anti-HAV was thepredominat- ing antibodyat onsetofjaundice and persisted in these patients for at least 60 days, but not longer than 115 days. The demonstration of IgM anti-HAV is therefore a valuable tool for the diagnosis of recent hepatitis Ainfection.
Particleswith an averagediameter of27to 29 nmfound in human stoolspecimens havebeen showntobe theetiological agent of hepatitis A infection (2). Recently, accordingtonucleic acid andphysical data, this virus could beclassified as apicornavirus (10, 11), asalready suggested byProvost andco-workers (8). Differentmeth- ods such as radioimmunoassay (4, 9), immune adherence(6),andcomplementfixation(7) have been used to detect antibody to hepatitis A antigen (anti-HAV). These antibodies can be foundbyradioimmunoassay in all casesofhep- atitisAafteronsetofjaundice,and absence of anti-HAV at this stage of the disease excludes hepatitisA(3).Because of therapidincreasein antibodytiterduringthe first 2 weeks ofillness, frequentlynosignificantfurther increase intiter canbe foundif the firstserumis drawn late in thecourseofthedisease.
The presence of anti-HAV of the immuno- globulin M (IgM) class (IgM anti-HAV) has been shown in acute- and convalescent-phase sera from human hepatitis A cases and from experimentallyinfectedchimpanzees by testing ofIgG- andIgM-containingserumfractions ob- tainedbymolecular-sieve columnchromatogra- phy by radioimmunoassay (1). In serial serum specimens from10patients, IgManti-HAVwas separatedfrom anti-HAV of theIgG class (IgG anti-HAV) bysucrose gradient ultracentrifuga- tion anddeterminedby solid-phaseradioimmu- noassay (5). It was concluded that IgM anti- HAV ispresentinacutehepatitisAin lowtiter, but it maypersistfor almost100daysafter the appearance ofdark urine. Therefore, the dem- onstration of IgM anti-HAV may serve as a diagnosticmarker foracutehepatitisAinfection.
In thispaper the persistence of IgM anti-HAV isevaluated by usingalesstime-consumingre- orienting sucrose gradient ultracentrifugation method fortheseparation of IgG and IgM.
MATERIALS AND METHODS Sera. Fifty-six sera were collected 0 to 200 days after onset ofjaundice from 19 patients during an epidemicofhepatitis AinGomaringen (4) and during anotheroutbreakinNeckartailfingen, Germany. In all patients thediagnosis of hepatitis Awasconfirmedby amore than fourfold increaseinanti-HAV titer in a solid-phase radioimmunoassay inhibition test (4).
Thirty-two additional sera were collected within 40 days afteronsetofjaundicefrompatients suspected ofhavinghepatitisAaccordingtoepidemiologicaland clinical data.
Determination of IgM anti-HAV. IgMwassep- arated fromIgG by reorienting gradient centrifugation (12),usingaTV865rotor(DuPont Instruments, New- ton,Conn.) andaKontroncentrifuge model TGA 50 (Zurich, Switzerland). A50-,ul amount of serum was layered on top of a 4-ml linear 5 to 35% (wt/wt) preformedsucrosegradientinSABAG buffer(Teste- rub,RohmPharmaGmbH, Darmstadt, Germany) and coveredwith0.3mlof mineral oil(Merck, Darmstadt, Germany,no.7174).After2hofcentrifugation(50,000 rpmat4°C),12 to13fractions werecollectedfrom the bottomof the tube. Ineach fraction thetiterofanti- HAV was determined by a solid-phase radioimmu- noassay inhibition test,which hasbeendescribed else- where(4).
RESULTS
Figure1shows thetypicalresults ofatitration of anti-HAV from the fractions of four sera obtained from patient GBM which were col- lected 0, 9, 57, and 193daysafter onset ofjaun- dice. Atday0 (serumI)peakanti-HAVactivity wasfound in fractions2and 3. Thiscorresponds
476
IgM ANTI-HAV IN ACUTE HEPATITIS A 477
IgMImg%) 97 116 129 94 58 38 29
BG lmg%I - - - - 33 78 128 --160 ,160 160 84 39
FIG. 1. Reorienting sucrosegradient ultracentrifugation of four sera from patient GBM obtained at differenttimesafteronsetofhepatitisA.Anti-HA V titers determinedby radioimmunoassayaregiven forall serumfractions in additiontothe concentrations of IgMandIgG from fractions ofserumI. The relative amountsof IgM anti-HAVwere approximated by dividingthe sum ofthe anti-HAV titersof fractions 1 through4bythesumofthetitersofallfractionsmultiplied by100toequalthe percentIgManti-HA Voftotal anti-HA V.
with thepeak concentrations of IgM determined by radial immunodiffusion (Partigen, Behring- werke AG, Marburg,Germany). Fractions 1,4, 5,and6alsoshowed low anti-HAVactivityand lowIgM levels.
IgG was detected in fractions 5 through 12, withpeak concentrations in fractions 8, 9, and 10. In serum IV, collected 193days after onset ofillness, theanti-HAV titerswerecloselypar- alleltotheIgGconcentrations.Alow anti-HAV titer was also found in fraction 4, which was
usually devoid of antibody activity whenever lower- and medium-titered sera from patients withpasthepatitis A infectionsweretested.
Addition ofanequalvolume of0.2Mmercap- toethanol for45 minat370Creduced the anti- bodyactivity infraction2more than eightfold in serum Ibut didnotchangetheantibodytiter infraction8ofserum IV.The datasuggestthat antibodyfound in fractions 1, 2, and 3, and to some extent in fraction 4, belongs to the IgM class,whereas anti-HAV in fractions 5through 12ismainlyoftheIgGclass. To obtain arough estimate of the percentage of IgM anti-HAV present at different stages of the disease, the antibody titers of fractions 1 through 4 were added andexpressedas apercentage of the sum of the antibody titers of all fractions (Fig. 1, Table 1).
At 0 to 5daysafter onset ofjaundice,between 29and100%(meanvalue, 54%)of the anti-HAV
TABLE 1. Persistenceof anti-HA V of the IgMclass
No.pOsitive %IgM Variationof Days after on- forIgManti- anoftoHtalV IgM anti-HAV set ofjaundice HAV/no. anti-HAV as % oftotal
testeant-HAV anti-HAV
VBtI (mean)
0-5 20/20 54 29-100
6-10 24/24 42 21-67
11-20 9/9 29 23-40
21-40 9/9 24 14-41
41-60 9/9 15 5-21
61-80 3/6 5 0-17
81-120 2/8 3 0-14
>120 0/3 0 0
present was of the IgM class (Table 1). The percentageof IgMantibody decreased withtime afteronsetofjaundice.Betweendays41and 60, 5 to 21% anti-HAV was of the IgM class. Be- tween days 61 and 120, only someof the sera, andafter120daysnosera, wereIgManti-HAV positive. Theserumwith the longestpersistence ofIgM anti-HAV (115days) was drawn from a patient (SH) with elevation of serumtransami- nases for morethan 3 months.
DISCUSSION
Frequently no significant increasein titer of anti-HAV can be detected if the first serumis drawnlate inthe course ofhepatitisA. There- fore,the demonstration of anti-HAV of the IgM VOL. 9,1979
478 FROSNERET AL.
class maybe a valuable tool for thediagnosis of afresh hepatitis A infection.
IgM was separated from IgG by reorienting gradient ultracentrifugation, and the titer of anti-HAV wasdetermined in the different frac- tions. Thismethod has considerable advantage over conventional sucrose gradient centrifuga- tioninswinging-bucketrotors.Theprocedure is far less time consuming, due to the fact that first, thecentrifugation time is shorter (2 h in- steadof16 to20h) and, second, the rotorused (TV 865) hasa capacity of eight tubesas com- paredto threeto, maximally, six tubes incon- ventional swinging-bucket rotors. As a conse- quence,the running time isnolongeralimiting factor, and 16to 24samplescanbesetup, run, and harvested per day using one centrifuge.
Moreover, the separation in this special fixed- anglerotorisatleastasgoodasinalong-column swinging-bucketrotor.Thedetermination of ti- tersafterseparation ofimmunoglobulins by dif- ferentsedimentation behavior is alsoavery sen- sitivemethod, becausenocompetitionbetween IgG and IgM antibody for the antigen takes place,as maybefoundinIgM-specific radioim- munoassay orinimmunofluorescence methods.
One objection to this method is the possible occurrence offalse-positive results for IgM an- tibody byIgGaggregates.However, thiswas not observed inourstudyandcanbecontrolledby evaluating antibody titers with and without treatmentof the fractions with mercaptoetha- nol.
According to this technique, at 0 to 5 days after onset ofjaundice more than half of the total anti-HAV present belonged to the IgM class, and itwas found topersist forup to 60 daysineverycase.Inseracollected between60 and 100 days after onset of jaundice it was
present in theserum ofpart of the patientsin low titers. The patient in whom the longest persistence ofIgManti-HAVwasdemonstrated (115days) also hadelevated transaminases for more than3 months. Itmay be thatIgM anti- HAV persists somewhat longerin more severe orprolongedcasesofhepatitisA,butthis needs further study. Nevertheless, sera of all cases obtained 120 days after onset ofjaundice had anti-HAV of the IgG class only, which makes the demonstration of IgM anti-HAV a good marker fortherapid diagnosisofafresh hepa- titis A infection from a single serum sample drawnduringtheacute orconvalescentstageof the disease.
Reorienting sucrose gradient centrifugation with subsequent testing of fractions for anti-
HAV isaveryreliable and sensitivemethodfor the detection of specific antibody of the IgM class. However, still less cumbersome methods shouldbe developedfor routinetestingof clini- calspecimens.
ACKNOWLEDGMENTS
We thank V. Messelberger forexcellent technical assist- ance.
This work wassupported by the Bundesministerium fur Jugend, Familie und Gesundheit (IA 2-5/176295) and by the DeutscheForschungsgemeinschaft (Fr 400/6).
LITERATURE CITED
1. Bradley, D. W., J. E. Maynard, S. H.Hindman, C. L.
Hornbeck,H. A.Fields, K. A.McCaustland, and E.
H.Cook, Jr. 1977. Serodiagnosis of viral hepatitis A:
detectionof acute-phaseimmunoglobulinManti-hep- atitis Avirusby radioimmunoassay. J.Clin. Microbiol.
5:521-530.
2. Feinstone,S.M.,A. Z.Kapikian, and R. H. Purcell.
1973.Hepatitis A: detection by immune electron mi- croscopy of a viruslike antigen associated with acute illness.Science182:1026-1028.
3.Frosner,G. G.1977. Nachweis vonHepatitis-A-Antigen undAntikorpernzurDiagnose derHepatitis-A-Infek- tion. Muench. Med. Wochenschr.119:825-829.
4. Fr6sner,G.G.,L. R.Overby, B. Flehmig, H.-J. Gerth, H.Haas,R.H.Decker, C.-M. Ling, A. J. Zucker- man, and H.-R.Frosner. 1977.Seroepidemiological investigation of patients and family contacts in an epi- demic ofhepatitisA. J. Med.Virol. 1:163-173.
5. Locarnini,S.A.,A. A.Ferris,N. I.Lehmann, and I.
D.Gust. 1977.Theantibodyresponsefollowinghepa- titis A infection.Intervirology8:309-318.
6.Miller, W. J.,P. J. Provost, W. J. McAleer,0. L.
Ittensohn, V. M.Villarejos, and M. R. Hilleman.
1975.Specific immune adherence assay for human hep- atitisAantibody.Applicationtodiagnostic and epide- miologic investigations. Proc. Soc. Exp. Biol. Med. 149:
254-261.
7. Provost, P.J.,0.L.Ittensohn, V. M. Villarejos, and M. R.Hilleman. 1975. Aspecific complement-fixation testfor human hepatitis A employing CR 326 virus antigen. Diagnosisandepidemiology. Proc. Soc.Exp.
Biol.Med. 148:962-969.
8. Provost,P.J., B. S.Wolanski, W. J. Miller, 0. L.
Ittensohn, W. J. McAleer, and M. R. Hilleman.
1975.Physical, chemicaland morphological dimensions of humanhepatitis A virus strain. Proc. Soc. Exp. Biol.
Med. 148:532-539.
9. Purcell, R. H., D. C. Wong, Y. Moritsugu, J. L. Dien- stag, J. A.Routenberg, and J. D. Boggs. 1976. A microtitersolid-phaseradioimmunoassay for hepatitis Aantigen andantibody. J. Immunol. 116:349-356.
10.Siegl,G., and G. G.Frosner.1978.Characterization and classification of virusparticles associated with hepatitis A.I.Size, density, andsedimentation.J.Virol. 26:40- 47.
11.Siegl,G., and G. G.Frosner.1978.Characterization and classification ofvirusparticles associated with hepatitis A.II.Type and configuration of nucleic acid. J. Virol.
26:48-53.
12.Wolf, H.J., G. G.Frosner,andF. Deinhardt. 1979.
Method forrapid separationofimmunoglobulinM from immunoglobulinG antibodiesby usingreorientinggra- dients inverticalrotors.J.Chin.Microbiol.9:544-546.
J. CLIN. MICROBIOL.
