Journal of Medical Virology 4681-90 (1995)

Molecular and Serologic Analysis in the Transmission of the GB Hepatitis Agents George G. Schlauder, George J. Dawson, John N. Simons, Tami J. Pilot-Matias, Robin A. Gutierrez, Cynthia A. Heynen, Mark F. Knigge, Gretchen S. Kurpiewski, Sheri L. Buijk, Thomas P. Leary, A. Scott Muerhoff, Suresh M. Desai, and Isa K. Mushahwar Viral Discovery Group, Experimental Biology Research, Abbott Laboratories, North Chicago, Illinois

-

Two flavivirus-like genomes have recently been cloned from infectious tamarin (Saguinus labia- tus) serum, derived from the human viral hepati- tis GB strain, which is known to induce hepatitis in tamarins. In order to study the natural history of GB infections, further transmission studies were carried out in tamarins. Reverse-transcrip- tion-polymerase chain reaction and enzyme- linked immunosorbant assays were developed for the detection of RNA and antibodies associ- ated with the two agents, GB virus-A and GB virus-B. The infectivity of both of these agents was demonstrated in tamarins to be filterable through a 0.1 pn filter. Two distinct genomes were identified in the serum of eight of the in- fected tamarins, while in four tamarins, the ge- nomes were detected independently of each other. Although specific antibodies to the GB vi- rus-B epitopes were detected in the serum of animals inoculated with both agents or GB vi- rus-B alone, antibodies to putative epitopes spe- cific to GB virus-A were not detected in any of the animals. All tamarins inoculated with serum con- taining GB virus-B exhibited an elevation in liver enzyme levels after inoculation. Elevations of se- rum liver enzyme levels did not occur when GB virus-A was the only agent detected in the se- rum. Infection with the original infectious tama- rin inoculum conferred protection from reinfec- tion with GB virus-B but not with GB virus-A. 0 1995 Wiley-Liss, Inc.

KEY WORDS non-A, non-B, non-C, non-D, non-E hepatitis, GB agent, GBV-A, GBV-B, ELISA, PCR

INTRODUCTION Viral hepatitis is known to be caused by a number of

viruses, each with its own distinct mode of transmission and replication. The diseases caused by these viruses can differ significantly with regards to severity of he- patic damage. These viruses are: hepatitis A virus (HAV), hepatitis B virus (HBV), hepatitis C virus 0 1995 WILEY-LISS, INC.

(HCV), hepatitis delta virus (HDV), and hepatitis E virus (HEV). Several lines of epidemiological and labo- ratory evidence have suggested the existence of addi- tional hepatitis agents which may be transmitted parenterally [Peters et al., 1993; Thiers et al., 19931. One potential agent is associated with serum from a surgeon who had developed acute hepatitis. Inoculation of tamarins (Saguinus species) with serum from this patient induced hepatitis in these animals and has been referred to as the “GB” agent [Deinhardt et al., 19673. Additional passage experiments led to the character- ization of GB as a viral agent [Parks et al., 1969; Dein- hardt et al., 1975; Karayiannis et al., 19891.

Recently, the molecular cloning of two flavivirus-like genomes from tamarins experimentally infected with a pool of infectious tamarin GB serum was described [Si- mons et al., 19951. These genomes putatively represent two independent viruses: GB virus-A (GBV-A) and GB virus-B (GBV-B). In this paper, a PCR assay is de- scribed for the detection of GBV-A and GBV-B RNA, and the use of immunoassays to detect antibodies spe- cific for antigens encoded by the GB agent genomes. Also described is the relationship between viremia and serologic profiles obtained from a series of transmission studies in tamarins infected experimentally with the GB agents. Data from these studies provide additional evidence that two independent viral agents are present in the passaged viral stock containing the GB agent.

MATERIALS AND METHODS Animals

Animals for the transmissibility and infectivity stud- ies were secured through the Laboratory for Experi- mental Medicine and Surgery in Primates, Tuxedo, New York (LEMSIP). All animals were maintained and monitored at LEMSIP according to protocols that met all relevant requirements for the humane care and the ethical use of primates in an approved facility. Baseline serum levels in international units per liter (IU/L) were

Accepted for publication December 15, 1994. Address reprint requests to George G. Schlauder, Ph.D., Exper-

imental Biology Research, Abbott Laboratories, North Chicago, IL 60064.

82 Schlauder et al.

established for the enzymes, alanine transaminase (ALT), gamma-glutamyltransferase (GGT) and isoci- trate dehydrogenase (ICD), on serum specimens ob- tained weekly or bi-weekly. A minimum of eight values were obtained for each animal prior to inoculation. Cut- off (CO) values were determined for each animal, based on the mean enzyme value plus 3.75 times the standard deviation. Enzyme values above the cutoff value were interpreted as abnormal and suggestive of liver dam- age. Tamarins (Saguinus labiatus) and Macaques (Macaca fascicularis) were inoculated intravenously with 0.25 ml of serum or plasma unless otherwise stated as described below, and monitored for changes in ALT, GGT and ICD serum levels. Although the abso- lute value of these enzyme levels are different, no sig- nificant differences in the temporal profiles of these enzymes were observed and, therefore, only ALT levels are reported.

Transmission Studies Study #1: Primary inoculations. Four different

inocula were used in the first transmission study. Two inocula were of human origin: a negative control (pool of normal human sera from healthy volunteer blood donors) and a serum specimen obtained three weeks after onset of acute hepatitis from a surgeon who devel- oped hepatitis (the original source of the GB agent). This latter specimen had been stored for over 30 years a t -20°C until utilized in this study. Two inocula were prepared from tamarin sera: a negative control (pool of normal tamarin sera obtained prior to inoculation) and a pool of infectious tamarin sera (designated as H205 GB pass 11) containing the GB agent [Almeida et al., 19761. This latter tamarin pool had been prepared from sera collected during the early acute phase (19 to 24 days postinoculation) of hepatitis from nine tamarins inoculated with tamarin GB serum (passage 10). Ali- quots of this pool were stored for over 14 years under liquid nitrogen until utilized in this study as an inocu- lum into tamarins T-1053, T-1048, T-1061, and T-1057.

Study #2: Passage of T-1053 infectivity. Tama- rin T-1053 developed elevated liver enzymes at 7 days after inoculation with H205. The animal was eutha- nized 12 days postinoculation (PI) and plasma was ob- tained in order to passage the virus. Tamarin T-1055 was inoculated with neat T-1053 plasma. Tamarins T-1038 and T-1051 were inoculated with T-1053 plas- ma which had been diluted in pooled normal tamarin plasma to lo-* or lop5, respectively. Tamarin T-1049 was inoculated with T-1053 plasma which had been passed through a series of filters of decreasing pore size (0.8, 0.45, 0.22, and 0.10 pm) and diluted to lop4 in pooled normal tamarin plasma. Two macaques (5-211 and 5-219) were also inoculated with neat T-1053 se- rum.

Study #3: Challenge with T-1053. Four of the tamarins from studies #1 and #2 were challenged with 0.1 ml of neat T-1053 plasma 97 days (T-1048, T-1057

and T-1061) or 69 days (T-1051) after the H205 or T- 1053 inoculation, respectively.

Study #4: Separation of GBV-A and GBV-B in- fectivity. Tamarin T-1044 was inoculated with T-1057 serum obtained 7 days after the H205 inocula- tion and diluted 1500 in normal tamarin serum. Tam- arins T-1047 and T-1056 were subsequently inoculated with T-1044 serum obtained 14 days PI and diluted 1:2 in normal tamarin serum. Tamarin T-1058 was inocu- lated with neat T-1057 serum obtained 22 days after the challenge with T-1053 serum.

Commercial ELISAs Selected serum samples obtained before and after in-

oculation were tested for antibodies detected frequently following exposure to known hepatitis viruses. Speci- mens were tested for antibodies to hepatitis A virus (HAVAB assay, Abbott Laboratories), the core protein of hepatitis B virus (Corzyme test, Abbott Laborato- ries), hepatitis E virus (HEV EIA, Abbott Laboratories) and hepatitis C virus (HCV second generation test, Ab- bott Laboratories). These tests were carried out accord- ing to the manufacturer’s instructions (Abbott Labora- tories, Abbott Park, IL).

GBV-A and GBV-B ELISAs An indirect assay format was developed from var-

ious recombinant proteins from GBV-A and GBV-B that had been expressed in E. coli as fusions with CTP: CMP - 3 - deoxy - D-rnanno - octulosonatecytidylyltrans- ferase (CKS) [Pilot-Matias et al., 19933 and purified using successive pellet washes followed by gel filtration chromatography [Paul et al., 19941. The antigens con- sisted of portions of the putative non-structural (NS) regions, NS3/4 and NS5 of GBV-A and portions of the core, NS3/4 and NS5 regions of GBV-B (Muerhoff et al., manuscript in preparation; Pilot-Matias et al., manu- script in preparation). The ELISAs were carried out essentially as described previously [Dawson et al., 19921. Briefly, serum or plasma was diluted in speci- men diluent and reacted with the antigen-coated solid phase (polystyrene bead). After a washing step, the beads were reacted with horseradish peroxidase (HRP)- labeled antibodies directed against human immunoglo- bulins to detect tamarin or human antibodies bound to the solid phase. Specimens that produced signals above a cutoff value were considered reactive. For each group of specimens, a preliminary cutoff value was set to sep- arate those specimens that presumably contain anti- bodies to the GB epitopeb) from those that did not.

Detection of Viral RNA An assay utilizing reverse-transcription and the

polymerase chain reaction (RTPCR) was used for mon- itoring viremia associated with a GB infection. RNA was extracted from 25 to 50 pl plasma or serum using an SDS/Proteinase K method [Schlauder et al., 19911 or a guanidiniudurea method employing the Ultraspec

Transmission of GB Hepatitis

RNA Isolation System (Biotecx, Houston, TX) essen- tially according to the manufacturer’s instructions.

Ethanol-precipitated nucleic acids were suspended in DEPC-treated H,O. cDNA synthesis and PCR were performed using the GeneAmp RNA PCR Kit from Per- kin-Elmer-Cetus (Norwalk, CT) essentially according to the manufacturer’s instructions. cDNA synthesis was primed with random hexamers. The resulting cDNA was subjected to PCR using primers (0.5 pM final concentration) based on sequences from GBV-A and GBV-B [Simons et al., 19951. The sequences of the GBV-A specific primers were 5‘-CAATAGCA- CAATCTTCCTTGG-3’ (~160 and 5’-GAAAGC’ITGGT- TGGTTGTGG-3’ (c16r). The sequences of the GBV-B specific primers were 5’-GATCCATAGTGAGCCACT- CACS’ (1240 and 5’-CAAAATGTTCCTGTCAT- TATTTG-3’ (1241-1. PCR was performed for 35 cycles (94”C, 1 min; 55”C, 1 min; 72”C, 1 min) followed by an extension cycle of 72°C for 7 min.

PCR products were separated by agarose gel electro- phoresis and visualized by ethidium bromide staining and by hybridization to a radiolabeled probe after Southern blotting to nitrocellulose [Sambrook et al., 1989). The probe, a gel-purified PCR product generated from the H205 inoculum, was radiolabeled using a ran- dom-primer labeling kit (Boehringer Mannheim, India- napolis, IN) with 32P-dCTP (Amersham, Arlington Heights, IL). Hybridizations were carried out at 42°C in 50% formamide, 5X Denhardt’s, 5 X SSPE, 0.5% SDS, 100 Fglml salmon sperm DNA. Alternatively, Hy- bond-N+ was used for Southern blotting and hybridiza- tions were carried out using Rapid-hyb buffer (Amer- sham). Filters were washed in 2 x SSC, 0.1% SDS (15 min, RT) and 0.1 x SSC, 0.5% SDS (30 min, 37°C; 1 hr, 65°C). Autoradiography was carried out using a Molec- ular Dynamics (Sunnyvale, CA) PhosphorImager. Posi- tive and negative control serum or plasma samples were included in each experiment.

RESULTS A flow diagram describing the transmission studies

is shown in Figure 1. The first three tamarin transmis- sion studies described in this report were initiated prior to the discovery that two flaviviruses were associated with the GB agent. Studies #1 through #3 were de- signed to confirm that the H205 inoculum was still infectious, to confirm that the infectivity was caused by a filterable agent, to determine the titer of the plasma obtained from T-1053, and to establish if protection was established against challenge with the GB agent. Study #4 was designed to determine if these agents were pas- sagable as independent agents.

Figures 2-6 summarize the biochemical and sero- logic results obtained from serial serum samples drawn from the tamarins used in these studies and are de- scribed in detail below. An example of an ethidium bromide-stained agarose gel and Southern hybridiza- tion of a RT/PCR analysis using GBV-B primers is shown in Figure 7. RT/PCR results fell into three cate-

83

gories: negative, as exemplified in lanes 1 , 8 , 9 and 10; strong positive (strong ethidium bromide staining and hybridization positive), lanes 4-6; and weak positive (weak or no ethidium bromide staining and hybridiza- tion positive), lanes 3 and 7.

Study #1: Primary Inoculations The four tamarins (T-1053, T-1048, T-1057, and

T-1061) inoculated with H205, the eleventh passage of GB, all showed elevations in liver enzyme levels after inoculation. One of the tamarins (T-1053) had elevated liver enzymes at day 7 PI, ALT = 67 (CO = 65.1) and a t day 11 PI, ALT = 172. The animal was sacrificed on the following day. The plasma obtained from T-1053 at the time of sacrifice served as the biological source leading to the discovery of the GBV-A and GBV-B genomes [Simons et al., 19951. In addition, this plasma was used as the inocula for studies #2 and #3.

Plasma obtained from T-1053 a t the time the animal was exsanguinated was positive for both GBV-B and GBV-A sequences as determined by RT/PCR. While liver enzyme values were elevated, there were no de- tectable antibody responses to recombinant GBV-B or GBV-A proteins (results not shown). The course of in- fection was followed in the three remaining tamarins (Figs. 2 4 ) and is summarized in Table I. Tamarins T-1048, T-1061 and T-1057 developed elevated liver enzyme values beginning 14 days PI, which remained above the cutoff to day 56, 84, and 42 PI, respectively. GBV-B sequences were detected by RTlPCR in all three animals beginning on day 7 PI and persisted for 28 to 137 days PI. Antibodies were detected against one or more of the three GBV-B recombinant proteins in the period following inoculation. However, the antibody re- sponses were transient, beginning as early as 21 days PI and extending as long as 112 days PI. Although GBV-A sequences were present in the H205 inoculum [Simons et al., 19951, GBV-A sequences were not de- tected by RT/PCR in the serum of these three tamarins during the three months following the H205 inocula- tion. In addition, there were no detectable antibody responses to the GBV-A recombinant proteins. None of the tamarins inoculated with either the pool of normal human or normal tamarin serum developed biochemi- cal evidence of hepatitis following inoculation (results not shown) indicating that the elevations in enzyme levels were associated with the H205 inoculum. These data suggested that GBV-B alone may produce hepati- tis in inoculated tamarins.

None of the tamarins had detectable antibodies to HCV or to HEV either before or after inoculation (Table 11). Therefore, infection with the GB passage inoculum, H205, does not elicit detectable antibody against HCV or HEV. One of the tamarins (T-1061) was positive for antibodies to HAV before and after inoculation with H205, suggesting a previous exposure to HAV. Tama- rins T-1048 and T-1057 had no HAV-specific antibodies after inoculation. Therefore, infection with the GB agent does not elicit an anti-HAV response. Tamarin

84 Schlauder et al.

passage 11 Q 150 diln

t t t m I

0.1 urn filtered unfiltered unfiltered unfiltered Q 1:10,000 diln. @ no diln. I Q 1:10,000 diln. I Q 1:100,000 diln

IT1049 I (TI055 I (TI038 I IT1051 I T-1044 14 dpi I

I T I 047 I IT10561

Fig. 1. Flow diagram of GBV tamarin transmission experiments. The top box indicates the inoculum or recipient of the inoculum. The lower attached boxes indicate the presence of sequences as detected by RTiPCR for GBV-A, A or GBV-B, B. The absence of the sequence is indicated by a black box. dpi, days postinoculation; dpc, days postchallenge.

T-1048 was negative for antibodies to HBV core anti- gen (anti-HBc) both before and after inoculation with H205. Two of the tamarins (T-1057 and T-1061) were positive for anti-HBc, both prior to and after inocula- tion with H205. Based on these data, there is no evi- dence that infection with the H205 inoculum induced an anti-HBc response. These data are consistent with a unique viral agent that is immunologically distinct from the viruses commonly associated with hepatitis in humans.

A tamarin inoculated with the serum obtained three weeks after onset of acute hepatitis from a surgeon who developed hepatitis (the original source of the GB agent) did not exhibit any evidence of disease indicat- ing that the sample used as inoculum may have lost its infectivity due to storage at -20°C for over 30 years (results not shown). Alternatively, this sample may have been obtained after the viremic stage of the dis- ease. RT/PCR on this inoculum was negative, suggest- ing that the GBV-A and GBV-B sequences are absent or a t least present a t a level below the limit of detection

of the GBV-A and GBV-B primers. This again could be a result of the duration and conditions of storage of this sample.

Study #2: Passage of T-1053 Infectivity All four tamarins inoculated with T-1053 plasma (T-

1055, T-1049, T-1038, and T-1051) developed elevated liver enzyme levels (Table I). Tamarin T-1055, which was inoculated with neat T-1053 plasma, had elevated liver enzyme values on days 7 and 11 PI. The animal was sacrificed on day 12 PI. T-1038 and T-1051, which were inoculated with and dilutions of T-1053 plasma, respectively, and T-1049, which was inoculated with filtered T-1053 plasma, had elevated liver enzyme values on days 11 on 14 PI. T-1038 and T-1049 were euthanised on day 14 PI. There were no detectable anti- body responses to the GBV-A or GBV-B recombinant proteins in these animals. These data indicate that the infectivity of T-1053 serum is at least 4 X lo5 tamarin infectious doses (TID) per ml. Both GBV-B and GBV-A sequences were detected by RTiPCR in the four tama-

Transmission of GB Hepatitis

~

85

h

E 1.5% z K

140

120

100

80

60

40

h

2

a

T- 1 048

GBV-B 0 mmMBB8. 13 U ~ I E I E i ~ O ~ B l Id

G BV-A 0 oooooooooo 0

TI 053 1

Sac 1

20

0

200

h

d 150 2 k v

2 100

50

0

" -50 0 50 100

Days Post Inoculation Figure 2.

T-I 057 ~

GBV-B m o o 0 0 o m 0 G B V - A O O O W O O O O O

TI 053 w'k '

I ut,.- f F I , \ b-*

-1 00 0 100 200 300 Days Post Inoculation

Figure 3.

Figs. 2 4 . Tamarin serial profiles. ALT values (solid lines). Anti-GBV-B ELISA NS3/4 (closed circles); NS5 (open circles); Core (closed squares). GBV-B PCR Negative (open squares); weak positive (shaded squares); strong positive (closed squares). GBV-A PCR Negative (open diamond); weak positive (shaded diamond); strong positive (closed diamond). Figure 2, T-1048; Figure 3, T-1057; Figure 4, T-1061; Figure 5, T-1051; Figure 6, T-1044.

rins (T-1055, T-1038, T-1049, and T-1051) inoculated with T-1053 (Table I). Since GBV-A and GBV-B were detected in T-1051, which received the 10~-5 dilution of the T-1053 inoculum, the titers of the GBV-B and the GBV-A genomes are equal to or greater than the infec- tivity titer determined for the T-1053 plasma (i.e., 4 x lo5 TID/ml). In addition, the viral nature of both GBV-A and GBV-B is demonstrated by the presence of infectivity after passage through a 0.10 pm filter indi- cating that they are agents of less than 100 nm in diameter.

The fourth tamarin, T-1051, developed elevated liver enzyme values between days 14 and 63 PI (Fig. 5). Antibodies to the GBV-B recombinant antigens were detected between days 28 and 113 PI (the day of sacri- fice). There were no detectable antibody responses to the GBV-A recombinant proteins. Serum from tamarin T-1051 was also negative for antibodies for HAV, HCV and HEV both before and after inoculation (Table 11). A weak anti-HBc reactivity was noted prior to inocula- tion but was absent after inoculation. These results indicate that the infection associated with T-1053 is

86

T-1061

Schlauder et al.

100

80 h

i v 60

Q

3

5 40

20

0

300

n . 1

2 - 200 5 a

100

0

~

GBV-B m 0 0 0 0 0 0 GBV-A O@w G840 Q 0 9 0

H205 TI053 Sac v A 1 1

A i\ - - I u \

3

E 2.5 c n

c\I 0-J

2 9 8

1.5 6 e 2 1 9 a

0.5

0 -1 00 0 100 200 300 400

Days Post Inoculation Fig. 4. Legend on previous page.

T-I 051

GBV-B 0 W W W W B W El000 0 GBV-A 0 +w 0

7F

5 %

6 5

8 4 6

2a 3; 9

1

0 -50 0 50 100

Days Post Inoculation Fig. 5. Legend on previous page.

again of a unique origin not immunologically related to any of the known hepatitis viruses.

The inoculation of two macaques with neat T-1053 plasma did not result in any detectable evidence of dis- ease (ALT, RT/PCR or ELISA) in either animal (results not shown). The inability to passage either of these agents in macaques is consistent with earlier primate studies [Tabor et al., 19801 indicating the limited host range for the GB agent(s).

Study #3 Challenge With T-1053 The four surviving tamarins from studies #1 and #2

were challenged with T-1053 plasma at >4 x lo4 TID at 97 days PI (T-1048, T-1057, and T-1061) or 69 days PI

(T-1051). At the time of challenge, T-1048 and T-1051 were weakly positive by RT/PCR for GBV-B sequences (Figs. 2, 5). T-1057 was negative for GBV-B sequences at the time of and 62 days prior to the challenge (Fig. 3), while T-1061 was negative for GBV sequences at the time of and 6 days prior to the challenge (Fig. 4). Tama- rins T-1048, T-1057 and T-1061 had no evidence of de- tectable GBV-A sequences at any time prior to the chal- lenge. There were no significant elevations in liver enzyme values noted following challenge, suggesting that these tamarins were not susceptible to a second hepatitis infection following inoculation with high-ti- ter GB agent. All but one of the tamarins (T-1061) had a modest increase in the antibody response to one or more of the GBV-B recombinant antigens subsequent to the

Transmission of GB Hepatitis 87

T- I 044

GBV-B U m H U U B W U D 0 0 0 GBV-A 0 000000000 0 0 0 0

100 1 J

!i 60 a

2o t n l

0 50 100 u

-50 Days Post Inoculation

Fig. 6. Legend on page 85.

Fig. 7. Agarose gel analysis of RTlPCR products generated using c4 primers. Tamarin T-1061 serial samples, Lane 1: preinoculation; Lane 2 marker; Lanes 3-10 days 7, 14, 21, 49, 84, 105, and 112, postinoculation, respectively. A TBE-agarose gel stained with ethidium bromide. B: Phosphorimage after Southern blotting of gel to nitrocellulose and hybridization to radiolabeled probe.

viral challenge. Following the challenge with the T-1053 inoculum, tamarins T-1051 and T-1061 did not exhibit RT/PCR-detectable GBV-B sequences. Similar results were observed in T-1057 except for one sample obtained 190 days after challenge. "-1048 had been positive for GBV-B sequences before the challenge, and all but one sample tested positive after the challenge. No appreciable difference in PCR signal was observed prior to and after challenge. Seven days after the chal- lenge, GBV-A sequences were detected in all three of the tamarins that had not shown any previous evidence of GBV-A sequences (T-1048, T-1057 and T-1061).

GBV-A sequences alone were detected by RTiPCR in T-1057 for 77 days after the challenge and in T-1061 for the entire period from challenge to sacrifice (277 days). Tamarin T-1048 was positive for GBV-A sequences for 28 days after the challenge. Tamarin T-1051 was posi- tive for GBV-A sequences by RT/PCR following its first inoculation (with a dilution of T-1053) and re- mained positive before and after the challenge with

Thus, it appears that all four tamarins were pro- tected against reinfection with GBV-B, and that three of the tamarins (T-1048, T-1057 and T-1061) were sus-

T-1053.

TAB

LE I

. Su

mm

ary

of T

rans

mis

sion

Stu

dies

Dur

atio

n of

GB

V-B

E

leva

ted

liver

enz

ymes

an

tibod

y re

spon

se"

PCR

res

ults

St

udy

Inoc

ulum

T

amar

ins

Dur

atio

na

ALT

pea

k" (

Val

ue)b

Sa

crif

iced

" N

S3/4

N

S5

Cor

e G

BV

-B

GB

V-A

1

H20

5

2 T

-105

3 ne

at

dilu

ted

dilu

ted

dilu

ted

filte

red

3 T

-105

3 ne

at

neat

ne

at

neat

4 T-

1057

(7 d

pi)

T-10

57 (2

2 dp

ce)

T-10

44

T-10

44

T-1

053

T-1

061

T-1

057

T-1

048

T-1

055

T-1

038

T-1

051

T-1

049

T-1

048

T-10

57

T-1

061

T-1

051

T-1

044

T-1

058

T-10

47

T-10

56

7-1

1

14-5

6 14

-84

14-4

2

7-1

1

11-1

4 14

-63

11-1

4

none

c no

ne

none

no

ne

14-6

3 no

ne

42-5

6 42

-84

11 (1

72)

28 (

138)

63

(91)

14

(198

)

11 (

161)

14

(178

) 28

(30

8)

14 (1

23)

42 (

106)

49 (

104)

63

(85)

12

137

386 -

12

14

115 14

137

386

115 -

-

49-9

1

56-6

3 -

-

-

56-1

15

-

112

-

-

56-1

15

6343

4 N

D~

N

D

ND

56-8

4 84

-132

-

-

-

49-1

15

-

-

99-1

13

84-1

32

49-1

15

-

ND

N

D

ND

-

63

-

21-6

3

-

112-

126

-

2S11

5

-

ND

N

D

ND

+ + + + + + + + + - + + + -

+ t + + + + + +

"Day

s po

stin

ocul

atio

n, dp

i. hI

UIL

. 'E

leva

tion

in

live

r enz

vmes

not

not

ed a

fter

chal

leng

e.

I

dNot

dete

rmin

ed.

'Day

s po

stch

alle

nge,

dpc.

Transmission of GB Hepatitis 89

TABLE 11. Serologic Analysis of Tamarin Sera Serum Anti-HAV Anti-HBc Anti-HCV Anti-HEV T-1048 pre - - - -

- - - - T-1048 post T-1057 pre - - - T-1057 post - - -

T-1061 pre + + T-1061 post + + T-1051 pre - - -

+ +

- - - -

+ - - - - T-1051 DOSt

ceptible to infection with GBV-A. This may not be sur- prising because these tamarins did not have an ongoing infection with GBV-A, as indicated by the lack of PCR- detectable GBV-A, and therefore had not developed a detectable antibody response. Tamarin T-1051 re- mained positive for GBV-A. It is likely that the viremia in T-1051 was a result of the first inoculation with diluted T-1053 rather than the challenge with T-1053 since no appreciable difference in PCR signal was ob- served prior to and after challenge.

Study # 4 Separation of GBV-A and GBV-B Infectivity

At day 7 PI, serum from T-1057 was positive only for GBV-B sequences. Inoculation of tamarin T-1044 with this serum induced elevated liver enzyme values from days 14 to 63 PI (Fig. 6). GBV-B sequences were de- tected by RT/PCR from day 7 to day 77 PI. A transient antibody response to one of the GBV-B recombinant proteins was observed between 63 and 84 days PI. How- ever, neither GBV-A sequences as determined by RT/ PCR, nor antibodies against the GBV-A recombinant pro- teins as determined by ELISA, were detected. These RT/ PCR results were confrmed using additional primer sets specific for other regions of the GBV-A and GBV-B ge- nomes (results not shown). Initial data from a passage of GBV-B-only positive serum from T-1044 into two tama- rins also resulted only in GBV-B viremia from days 7 to 42 PI. However, the ALT levels did not elevate above the cutoff until day 42 PI. These data inhcate that GBV-B in the absence of GBV-A induces hepatitis in tamarins.

At 22 days after the challenge with T-1053 plasma, serum from T-1057 was positive only for GBV-A se- quences. This serum was used to inoculate tamarin T-1058. Initial data from this inoculation indicate that no significant elevations in ALT levels occurred during the early part of the study (41 days). GBV-B sequences have not been detected in any of the samples tested to date (days 14 though 41 PI). However, GBV-A se- quences were detected at days 14,35 and 41 PI, indicat- ing a transient GBV-A viremia.

DISCUSSION RTPCR was used to monitor viremia associated with

the two flavivirus-like genomes that have been identi-

fied in the H205, GB pass 11 inoculum. Transmissibil- ity of both agents was demonstrated in tamarins. The infectivity of both of these agents was shown to be fil- terable through a 0.1 pm filter. GBV-B viremia alone was sufficient to cause elevations in liver enzyme levels as seen in tamarins T-1048, T-1057, T-1061, T-1044, T-1047, and T-1056. No elevations of liver enzyme lev- els were observed during a GBV-A-only viremic stage as seen after the T-1053 challenge of tamarins T-1057, T-1061 and T-1051. These results strongly suggest that GBV-B is the probable causative agent for hepatitis in these tamarins. However, as shown in Table I, the aver- age peak ALT value for the five animals positive for both GBV-A and GBV-B sequences (188 IU/L) was higher than the average value for the six GBV-B-only animals (120 IU/L). In addition, the initial elevation and the peak value occurred earlier in animals positive for both GBV-A and GBV-B than for animals positive only for GBV-B. These results suggest that the severity of hepatitis is related to the presence of both agents at significant levels. The observation from the additional passage of GBV-B into tamarins T-1047 and T-1056 that only minor elevations in liver enzymes occurred with GBV-B viremia supports the assumption that the presence of GBV-A is necessary for major elevations in ALT levels to occur in tamarins. However, one cannot rule out the necessity for at least minimal levels of GBV-A, that is, a t levels below the limit of sensitivity of RT/PCR, for a GBV-B viremia to occur. The absence of detectable viremia may not necessarily indicate that the animal has cleared the virus as has been shown in studies monitoring interferon therapy in HCV [Shindo et al., 1991; Schlauder et al., 19923. Additional experi- ments using mixed inocula of GBV-A and GBV-B viral stocks are needed to address these questions.

ELISAs were developed employing recombinant an- tigens for the detection of antibodies to GBV-A and GBV-B. The absence of a detectable seroconversion to GBV-A in any of the infected tamarins may be due either to the selection of recombinant antigens that do not elicit an immune response, or to the possibility that GBV-A itself is not immunogenic in tamarins. The im- mune response to the GBV-B antigens used in these ELISAs appears to be of a short duration, at most 150 days PI. One explanation for the lack of a prolonged immune response is that the epitopes used in these ELISAs were not from the dominant epitopes to which

90

the immune response is generated. An alternative ex- planation is that the hepatic challenge in tamarins is insufficient to induce a long-lived response. Similar se- rologic responses of short duration have been observed in some patients with acute resolving infections of hep- atitis C [Tan et al., 19941 as well as in experimental hepatitis E infections in macaques [Tsarev et al., 19941.

The question of chronicity versus resolution of vire- mia remains unanswered. Four of five animals de- scribed in this study resolved GBV-B viremia after 84 days PI. In contrast, T-1048 remained viremic for 136 days and was found to be viremic at the time of sacrifice (137 days PI). Of the four animals that were positive for GBV-A sequences, two showed resolution by 77 days after the first appearance of GBV-A sequences. In con- trast, tamarin T-1061 was viremic for 245 days, up to the time the animal was sacrificed. In addition, tama- rin T-1051 was viremic up to the time of sacrifice (day 113 PI); however, it is unclear if this persistent viremia was due to the initial inoculation with T-1053 plasma or a result of the subsequent challenge with additional T-1053 plasma 69 days later.

The finding of spontaneous hepatitis in tamarins has previously led to speculation that the GB agent was of tamarin origin [Parks and Melnick, 19691. We have found no molecular or serologic evidence for a detect- able “natural GB” infection prior to experimental infec- tion in any of the animals used in these studies. The inability to detect, by infectivity or RT/PCR, one or both of these agents in the serum available from the surgeon GB does not permit a direct link to human hepatitis. However, this could be explained by the conditions un- der which this sample was stored and/or by the fact that the sample was obtained three weeks after the onset of disease, either of which may have compromised the infectivity of the sample. Acquisition of properly stored GB serum (if available), as well as screening of addi- tional non-A through E hepatitis samples is under way in order to define better the relationship between the GB agents and human hepatitis.

The data presented in this study indicate that an acute hepatitis infection is induced in tamarins inocu- lated with either the H205 or T-1053 plasma. Inocula- tion with either H205 or T-1053 resulted in GBV-A and GBV-B viremia, although only an antibody response to GBV-B was detected. Challenge of tamarins in which only GBV-B genome was detected resulted in the ap- pearance of GBV-A sequence. Passage of GBV-B alone was demonstrated and shown to result in a mild eleva- tion in liver enzyme levels. These results support the hypothesis that the GB agent actually represents two independent viral agents.

Schlauder et al.

Medicine and Surgery in Primates for the excellent animal care.

ACKNOWLEDGMENTS We thank Drs. Elizabeth Muchmore, Doug Cohn,

James Mahoney and the Laboratory of Experimental

REFERENCES

Almeida JD, Deinhardt F, Holmes AW, Peterson DA, Wolfe L, Zucker- man AJ (1976): Morphology of the GB hepatitis agent. Nature 261:608-609.

Dawson GJ, Chau KH, Cabal CM, Yarbough PO, Reyes GR, Mushah- war IK (1992): Solid-phase enzyme-linked immunosorbent assay for hepatitis E virus IgG and IgM antibodies utilizing recombinant antigens and synthetic peptides. Journal of Virological Methods 38:175-186.

Deinhardt F, Holmes AW, Capps RB, Popper H (1967): Studies on the transmission of disease of human viral hepatitis to marmoset mon- keys. I. Transmission of disease, serial passage and description of liver lesions. Journal of Experimental Medicine 125:673-687.

Deinhardt F, Peterson D, Cross G, Wolfe L, Holmes AW (1975): Hepa- titis in marmosets. American Journal of Medical Science 270: 73-80.

Karayiannis P, Petrovic LM, Fry M, Moore D, Enticott M, McGarvey MJ, Scheuer PJ, Thomas HC (1989): Studies of GB hepatitis agent in tamarins. Hepatology 9186192.

Parks WP, Melnick J L (1969): Attempted isolation of hepatitis viruses in marmosets. Journal of Infectious Diseases 120:539-547.

Parks WP, Melnick JL, Voss WR, Singer DB, Rosenberg HS, Alcott J , Casazza AM (1969): Characterization of marmoset hepatitis virus. Journal of Infectious Diseases 120:54&559.

Paul DA, Knigge MF, Ritter A, Gutierrez R, Pilot-Matias T, Chau KH, Dawson GJ (1994): Determination of hepatitis E virus seropreva- lence by using recombinant fusion proteins and synthetic peptides. Journal of Infectious Diseases 169:801-806.

Peters T, Schlayer HJ, Preisler S, Kopp B, Berthold H, Gerok W. Rasenack J (1993): Frequency of hepatitis C in acute post-transfu- sion hepatitis after open-heart surgery: A prospective study in 1,476 patients. Journal of Medical Virology 39139-145.

Pilot-Matias TJ, Pratt SD, Lane BC (1993): High-level synthesis ofthe 12-kDa human FK506-binding protein in Escherichia coli using translational coupling. Gene 128:219-225.

Sambrook J, Fritsch EF, Maniatis T (1989): “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor: Cold Spring Harbor Laboratory.

Schlauder GG, Leverenz GJ, Amann CW, Lesniewski RR, Peterson DA (1991): Detection of the hepatitis C virus genome in acute and chronic experimental infection in chimpanzees. Journal of Clinical Microbiology 29:2175-2179.

Schlauder GG, Leverenz GJ, Mattsson L, Weiland 0, Mushahwar IK (1992): Detection of hepatitis C viral RNA by the polymerase chain reaction in serum of patients with post-transfusion non-A, non-B hepatitis. Journal of Virological Methods 37: 189-200.

Shindo M, DiBisceglie AD, Cheung L, Shih W-K, Christian0 K, Fein- stone S, Hoofnagle J H (1991): Decrease in serum hepatitis C viral RNA during alpha-interferon therapy for chronic hepatitis C. An- nals of Internal Medicine 115:700-704.

Simons JN, Pilot-Matias TJ, Leary TP, Dawson GJ, Desai SM, Schlauder GG, Muerhoff AS, Erker JC, Buijk SL, Chalmers ML, VanSant CL, Mushahwar IK (1995): Identification of two flavivi- rus-like genomes in the GB hepatitis agent. Proceedings of the National Academy of Sciences USA in press.

Tabor E, Peterson DA, April M, Seeff LB, Gerety R J (1980): Transmis- sion of human non-A, non-B hepatitis to chimpanzees following failure to transmit GB agent hepatitis. Journal Medical Virology 5:103-108.

Tan D, Im SWK, Peng WW, Ng MH (1994): Follow-up study of acute hepatitis C. Archives of Virology 138:71434.

Thiers V, Lunel F, Valla D, Azar N, Fretz C, Frangeul L, Huraux J-M, Opolon P, Brechot C (1993): Post-transfusional anti-HCV-negative non-A non-B hepatitis (11) serological and polymerase chain reac- tion analysis for hepatitis C and hepatitis B viruses. Journal of Hepatology 18:3439.

Tsarev SA, Tsareva TS, Emerson SU, Yarbough PO, Legters LJ, Moskal T, Purcell RH (1994): Infectivity titration of a prototype strain of hepatitis E virus in cynomologus monkeys. Journal of Medical Virology 43:135142.