AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 8, Number 10, 1992 Mary Ann Liebert, Inc., Publishers

Mapping of Linear B-Cell Epitopes on the Major Core Protein p24 of Human Immunodeficiency Virus Type 1 (HIV-1) P.

KUSK, T.H. BUGGE, B.0. LINDHARDT, E.F. HULGAARD, and K. HOLMBÄCK

ABSTRACT Three murine monoclonal antibodies (MAbs) F5-2, F5-4, and F5-16 defining three different epitopes on the major core protein p24 of the human immunodeficiency virus type 1 (HIV-1) were epitope mapped using a random fragment expression library representing the pl7- and p24-encoding part of the gag open reading frame. F5-2 denned an epitope within amino acids (aa) 14-23 at the N-terminus of p24, and F5-4 defined an epitope within aa 153-174 in the C-terminus of p24. F5-16 did not recognize any of the fusion proteins produced by the expression library indicating that this M Ab defines a true conformational epitope on p24. Since the N-terminus of p24 has been reported to be immunosilent in humans, 356 HIV-1 antibody-positive serum samples were tested for reactivity against the region of p24 defined by F5-2. More than one third of the samples recognized this region indicating that it is immunoreactive and, further, the presence of antibodies against this region was associated with a reduced CD4 cell count.

INTRODUCTION MAJOR CORE PROTEIN p24 of human immunodeficiency virus type 1 (HIV-1 ) is produced by proteolytic cleavage of the gag-encoded precursor p55.' It aggregates to form a coneshaped nucleocapsid structure inside the virus particle.2 Epitope mapping of p24 using murine monoclonal antibodies (MAbs) has been extensively reported.3-" However, analyses of p24 epitopes recognized by human HIV-1 antibody-positive serum samples are scarce and have, apart from a report by lanvier and co-workers'2 only included few samples.3-71314 In the present study three murine anti-p24 Mabs were subjected to an epitope mapping using a random fragment expression library derived from the gag gene of HIVIIIb. One of these MAbs defined an epitope in the N-terminus of p24. This epitope has been previously recognized as an epitope defined by humans,13 but the serological analysis has hitherto been concentrated to only 30 sera. Thus, in order to further investigate the human humoral immune response against the N-terminus of p24, we performed an epitope-specific screening of 356 human serum samples. A decreasing level of antibodies against p24 is known as a prognostic marker of disease progression.15 However, it has not yet been recognized whether antibodies against different parts of

THE

Statens Serum Institut,

p24 decline at equal speed or antibodies defining certain epitopes decline more rapidly. Hence, the human humoral response against the N-terminus of p24 in correlation to disease progression was assessed using an epitope-specific capture enzyme-linked immunosorbent assay (ELISA). MATERIALS AND METHODS Bacterial strains and plasmids The bacterial strains used were SCS-1: F~, recAl, endAl, gyrA96, thi, hsdR17(rk", mk+), supE44, relAl, lambda" (Stratagene, La Jolla, CA), and pop2136: K12, M72F",

lac"am, SmR (lambda ci 857, N7, N53, A HI bio"). Plasmids

used were pUEXl-3 as earlier described,l617 and pBH10/R3 containing a replication defective provirus of the BH10.2 clone.

Serum

samples and murine monoclonal antibodies

From a collection of stored serum samples from HIV-1 antibody-positive homosexual males at various clinical stages, 356 samples were randomly selected and, further, 99 consecutive serum samples from 12 HIV-1 antibody-positive persons

Department of Virology, Building 87, Artillerivej 5, 1789

DK-2300

Copenhagen S,

Denmark

1790

KUSK ET AL. Table 1. Epitope Mapping

of the

Defined Binding Regions and F5-4

F5-2 Amino acids numbers of fusion-protein inserts

MAb

-19a

to 23 and 14 to 51 153 to 180 and ?b to 174

F5-2 F5-4

of

p24

by

MAbs

Amino acid sequence

14 to 23: AISPRTLNAW 153 to 174: IRQGPKEPFRDYVDRFYKTLRAE

"The minus indicate that the sequence extends 19 aa into the coding region of pl7. b?, undetermined N-terminus.

were used. For negative control experiments 82 sera from HIV-1 antibody-negative males were used. The production of the murine MAbs (F5-2, F5-4, and F5-16) used for epitope mapping has been previously described. '8

Construction

of gag-derived random fragment library

The random fragment expression library was constructed from the open reading frame of the gag gene spanning from the SstI site to the Bglll site of the BH10.2 proviral clone. The library was made using the pUEX expression vectors containing the lacL gene transcriptionally controlled by the lambda PR promoter which is repressed by a temperature-sensitive repressor ci ts857 at 30°C and induced by shifting the temperature to 42°C. The library was constructed essentially as described,19 except that DNAse I degraded random fragments between 50 and 300 bp of size were blunt ended and ligated to a ready-made adaptor made from the following two adaptors: adaptor A: 5'-GATCCGGCAACGAAGGTACCATGG-3' and adaptor B: 5'-CCATGGTACCTTCGTTGCCG-3'. pUEXl was cut with BamHI and ligated to the above mentioned ready-made adaptor. The adapted random fragments were then ligated to the adapted pUEXl vector. The library was transformed into highly competent cells of the SCS-1 strain of Escherichia coli (Stratagene, La Jolla, CA) followed by incubation at 30°C overnight on L-broth agar plates containing 35 p.g/ml ampicillin. The constructed library gave rise to a total of approximately 6000 ampicillinresistant colonies.

Production

offusion proteins

in

large quantities

A bacterial strain producing the fusion protein of choice was inoculated in 100 ml L-broth, containing 100 u-g/ml ampicillin and grown at 30°C until OD60,,,,,,, 0.2. Profein synthesis was induced by raising the temperature to 42°C for 90 min. Cells were harvested and treated with lysozyme followed by lysis in 1% (v/v) Triton X-100. The cell lysate was centrifuged at 100,000 g for 30 min and the supernatant discarded. Next, 1 ml of 6 M urea, 50 mM Tris pH 7.6,0.5 mM dithiotreitol was added to the pellet containing inclusion bodies harboring large amounts of fusion protein, and the mixture was sonicated 2 x 20 s using an amplitude of 20 p.m. The mixture was clarified at 20,000 g for 30 min at room temperature. The supernatant normally contained more than 80% fusion protein of the total protein content in the preparations. =

Capture ELISA Microtiterplates (Maxisorb, Nunc, Roskilde, Denmark) were ng/well of a murine monoclonal anti-ßgalactosidase antibody. The plates were washed three times in 0.5 M NaCl, 2.7 mM KC1, 1.5 mM KH2P04, 6.5 mM coated with 27.5

Epitope Mapping of p24 using »urine »ab's

gag Protein induction and

fragment

libraries

screening

pw

of random

synthesis and screening of colonies was performed essentially described.20 The murine MAbs were Induction of protein

p24

plS

p24

as

diluted 50 times before incubation with the induced random fragment libraries.

sequencing of inserts positive colonies DNA

in

pUEX vectors from

Plasmid inserts prepared as plasmid minipreparations were subjected to double-stranded dideoxy sequencing using Sequenase (USB, Cleveland, OH) as described by the manufacturer. The sequencing primers for sequencing inserts in the pUEXvectors were:

5'-primer:

5'-GGGGATTGGTGGCGACGAC-

TCCTGG-3', 3'-primer: 5'-CTAGAGCCGGATCGATCCG-

GTC-3'.

153

I

f

Binding region 14

!

|

: 174 of F5-4

23

Binding region of F5-2

FIG. 1. Schematic presentation of the mapping of epitopes on p24 defined by F5-2 and F5-4. The position of the fusion protein inserts scored in the screenings are given relative to p24 as horizontal lines. The smallest binding regions defined by the MAbs are depicted by vertical lines. The numbers at each vertical line are amino acid position numbers of p24 derived from the HXB2-clone.

IMMUNOGENIC EPITOPES OF HIV-1

p24

1791

1.5

ANTIGENIC

1.0

0.5

0.0

-0.5

-1.0

F5-2

-1.5

F5-4 HYDROPHOB! 50

100

150

200

Hydrophilicity plot of p24 based on the method by Hopp and Woods.21 Hydrophilic regions lie above the 0-line. binding regions defined by F5-2 and F5-4 are symbolized by vertical bars. FIG. 2.

protein

diluted 800X in NT buffer was added (100 Fusion protein applied in the capture ELISA was produced as described above. After 90 min incubation, serum samples diluted lOOx in NT buffer were added (100 mm3/well). Incubation at room temperature confined for 2 h. The bound human antibodies were detected by a combination of biotinylated goat anti-human IgG (TAGO, Burlingame, CA) diluted Fusion

mm3/well).

200 x in NT buffer, horseradish peroxidase-coupled streptavidin (Medac, Hamburg, Germany) diluted 2000x in NT buffer, and orthophenylenediamine (Sigma, St. Louis, MO).

RESULTS AND DISCUSSION

Epitope mapping using

Reactivity

of human sera against the "amino acid 14-23" epitope of p24

0.2 0.4 0.6 APsorbance at 492 FIG. 3.

Summary

of the

serum

1.2

sample screening using

the

capture ELISA specific for the "amino acid 14-23" epitope of

p24. Each sample was tested against ß-galactosidase as a negative control. The displayed absorbance at 492 nm for each sample was calculated as the difference between A492nm (p24ß-galactosidase) and A492nm (ß-galactosidase). Group 1: HIV-1 antibody-positive serum samples. Group 2: HIV-1 antibodynegative serum samples. The dashed line represents

murine MAbs

Three murine anti-p24 MAbs (F5-2, F5-4, and F5-16) previshown to define nonoverlapping epitopes18 were used to screen the constructed random fragment library. It appears from the results of the epitope mapping that F5-2 and F5-4 define continuous epitopes (Table 1; Fig. 1). F5-2 defines an epitope located within amino acids (aa) 14-23 of p24. This epitope is located in an area of low hydrophilicity as shown in the Hopp and Woods hydrophilicity plot in Figure 2 and has not been included in any epitope predictions. However, Mathiesen et al. have identified an epitope surrounding the region within aa 14-23,13 but Langedijk and co-workers did not detect human antibodies to this region by testing 12 human serum samples against a panel of overlapping peptides representing p24.7 F5-4 defines an epitope located within aa 153-174 of p24 (Table 1; Fig. 1). The epitope is located within the most hydrophilic area in p24 (Fig. 2) and its presence would therefore be expected according to the epitope prediction method described by Hopp and Woods.2' Accordingly, Zvelebil and co-workers22 predicted this region to be the most antigenic on p24 using the Hopp and Woods algorithm. Furthermore, this region was also predicted as the most antigenic part of p24 according to a proposed tertiary structure of p24 based on sequence comparisons between foot-and-mouth disease virus VP2 coat protein, simian immunodeficiency virus p24, and HIV-1 p24 as well as secondary structure predictions.23 Hence, many groups have reported antibodies of both human and murine origin directed against this region3'5'711 demonstrating its immunodominant nature.

ously

nm

The

KUSK ET AL.

1792 Table 2. Alignment of Amino Acid Sequences in the NDifferent HIV-1 Isolates" AA #

10

HIVHXB2 HIVLAI HIVMN HIVJH31 HIVCDC41 HIVOYI HIVSF2 HIVHAN HIVRF HIVELI HIVZ2 HIVNDK HIVMAL

MVHQAISPRTLNAWVKVVEEKAF

20

30

and

C-Terminus

of

p24

160

150

from

170

SILDIRQGPKEPFRD Y VDRFYKTLR AEQ A

-K-

I -

"The underlined bold sequence in the N-terminus is the binding region defined underlined bold sequence in the C-terminus is the binding region defined by F5-4, and the HIVHXB2 isolate. Numbers above sequences indicate aa numbers.

by =

F5-2 and the aa

identical to



F5-16 did not

recognize any of the approximately 2,000 gag-ß-galactosidase fusion proteins represented in the expression library. However, F5-16 reacts with HIV-1-infected, fixed cells (data not shown) and recognizes p24 in Western blot analysis.18 Moreover, F5-16 does not recognize p24 from HIV-2 (as do F5-2 and F5-4) (B. 0. Lindhardt, unpublished observations). The fact that F5-16 did not recognize any of the approximately 2,000 fusion proteins from the random fragment library indicates that it possibly defines a conformational epitope on p24. On the other hand, the ß-galactosidase portion of the fusion protein might block the reactivity of this MAb. However, Anti-p24 reactivity of human sera related to CD4 count ("amino acid 14-23"-epitope of p24) >400

CD CJ

Mapping of linear B-cell epitopes on the major core protein p24 of human immunodeficiency virus type 1 (HIV-1).

Three murine monoclonal antibodies (MAbs) F5-2, F5-4, and F5-16 defining three different epitopes on the major core protein p24 of the human immunodef...
794KB Sizes 0 Downloads 0 Views