Publications

Publications

Publications

1. Khanna, N., AnandaRao, Rand Jana, A. M. Novel recombinant dengue

multiepitope (r-DME) proteins as diagnostic intermediates. 2004.

PCTIIN04100237.

2. AnandaRao, R, Swaminathan, S., Fernando, S., Jana, A. M. and Khanna,

N. A custom-designed recombinant multiepitope protein as a dengue

diagnostic reagent. Protein Exp. Purifi. 41 (2005) 136-147.

3. AnandaRao, R, Swaminathan, S. and Khanna, N. The Identification of

Immunodominant Linear Epitopes of Dengue Type 2 Capsid and NS4a

Proteins Using Pin-Bound Peptides. 2005. Virus Research (in press).

138

(U) INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT)

== -== --~ --iiiiii -iiiiii iiiiii --== ----iiiiii -== == --iiiiii iiiiii

--iiiiii -==

(19) World Intellectual Property Organization

International Bureau

(43) International Publication Date 17 February 2005 (17.02.2005) peT

(51) International Patent Classification7: C07K 14/18, GOIN 33/569

(21) International Application Number: PCT IIN2004/000237

(22) International Filing Date: 9 August 2004 (09.08.2004)

(25) Filing Language: English

(26) Publication Language: English

(30) Priority Data: 974IDel12003 7 August 2003 (07.08.2003) IN

(71) Applicants (for all designated States except US): INTER-NATIONAL CENTRE FOR GENETIC ENGINEER-ING AND BIOTECHNOLOGY [INIIN]; Amna Asaf Ali Marg, New Delhi 110 067 (IN). DIRECTOR GEN-ERAL, DEFENCE RESEARCH & DEVELOPMENT ORGANISATION [INIIN]; Ministry of Defence, Govt. of India, West B1ock-8, Wing-I, R K Puram, Scctor-l, New Delhi 110 066 (IN).

(72) Inventors; and (75) Inventors/Applicants (for US only): KHANNA, Navin

[INIIN]; International Centre [or Genetic Engineering and Biotechnology, Amna Asaf Ali Marg, New Delhi 110 067 (IN). ANANDA RAO, Ravulapalli [INIIN]; International Centre for Genetic Engineering and Biotechnology, Aruna Asaf Ali Marg, New Delhi 110 067 (IN). JANA, Asha, Mukul [INIIN]; Defence Research & Development Estab-lishment, Jhansi Road, Gwalior 474 002. Madhya Pradesh (IN).

(74) Agents: SUBRAMANIAM, Uariharan et al.; Subrama-niam, Nataraj & Associates. E-556 Greater Kailash II. New Delhi 110048 (IN).

~ (81) Designated States (unless otherwise indicated, for every

-< r-... M \C

kind of national protection available): AE. AG, AL, AM, AT, AU, AZ, BA. BB. BG. BR, BW, BY, BZ, CA, CR, CN,

1IIIIIIIIIImllllllllllllllllllllllllllllllllllllllllliliIIIIIIIII!IIIIIIIII DlII~ 111111111111

(10) International Publication Number

WO 2005/014627 A1 CO, CR, CU, CZ, DE, DK. DM. DZ. EC, EE, EG. ES, FI, GB. GD. GE, GR, GM, HR, HU. ro, IL, IN, IS, JP, KE, KG. KP, KR, KZ, LC, LK, LR, LS, LT. LU. LV. MA. MD. MG, MK, MN, MW, MX, MZ. NA, NI, NO, NZ, OM, PG, PH. PL, PT, RO, RU, SC, SD, SE, SG. SK, SL. SY. TJ. TM, TN, TR, TT. n, UA. UG, US, UZ, VC, 'fN, YU, ZA, ZM, Zw.

(84) DeSignated States (unless otherwise indicated, for every kind of regional protection available): ARIPO (BW, GR, GM, KE, LS, MW, MZ, NA, SD, SL, SZ, n, UG, ZM, ZW), Eurasian (AM, AZ. BY, KG, KZ. MD. RU. TJ, TM), European (AT, BE, BG, CU, CY, CZ, DE, DK, EE. ES. FI, FR, GB. GR, HU, IE. IT, LV. MC, NL, PL. PT, RO, SE, SI. SK, TR). OAPI (BF. BJ. CF. CG. CI. CM. GA. GN. GQ. GW, ML, MR, NE, SN, TD, TG).

Declaration under Rule 4_17: as to applicant:< entitlement to apply for and be granted a patent (Rule 4. 17(ii)) for thefolLowmg designations AE, AG, AL, AM, AT, AU, AZ, BA, BB, BG, BR, BW, BY, BZ, CA, CIl, CN, CO, CR, CU, CZ, DE, DK, DM, DZ, Ee, EE, EG, ES, FI, GB, GD, GE, Gil, GM, IlR, IlU, /D, IL, IN, IS, Ip, KE, KG, KP, KR, KZ, LC, 10K, LR, U;, LT, Uf, LV. MA, MD, MG, MK, MN, MW, MX, MZ, NA, NI, NO, NZ, OM, PG, I'll, PL, PI; RO, RU, SC, SD, SE, SG, SK, SL, SY, TI, TM, TN, TR, IT, 12, UA, UG, UZ, VC, VN, YU, ZA, ZM, zw, ARIPO patent (BW, Gil, GM, KE, LS, MW, MZ, NA, SD, SL, SZ, 12, UG, ZM, ZW), Eurasian patent (AM, AZ, BY, KG, KZ, MD, RlJ, TJ, TM), Eumpeanpatent (AT, BE, BG, CIl, CY, CZ, DE, DK, El:.~ ES, FI, FR, GB, GR, IlU, IE, IT, LU, MC, NL, PL, PT, RO, SE, SI, SK, TR), OAPI patent (BF; Bl, CF; CG, CI, CM, GA, GN, GQ, Gw, ML, MR, NE, SN, TD, TG)

Published: with international search report before the expiration of the time limit for amending the claims and to be republished in the event of receipt of amendments

For two-letter codes and other abbreviations, refer to the "Guid-ance Notes on Codes and Abbreviations" appearing at the begin-ning of each regular issue of the PCI' Gazette.

~------------------------------------------------------------------~ (54) Title: RECOMBINANT DENGUE MULfI EPfrOPE PROTEINS AS DIAGNOSTIC INTERMEDIATES -l(') = (57) Abstract: Novel recombinant protein antigens for use in the diagnosis of dengue are discclosed. These arc prepared by assem-= bling key immunodominant linear dengue-specific epitopes, chosen on the basis of pepscan analysis. phage display and computer M predictions. One of these developed to specifically detect dengue-specific IgM and the other to detect IgG. These novelrecombinant o dengue multiepitope proteins were expressed in E. coil, purified in a single step and used as capture antigens in ELISA. The ELISA :> results, using a large panel of suspected dengue patient sera. were in excellent agreement with those obtained using the commercially ~ available Dengue Duo IgM and IgG Rapid strip test (PanBio). Similarly, tests were carried out using a rapid strip test.

WO 2005/014627 PCTIIN2004/000237

RECOMBINANT DENGUE MULTI EPITOPE PROTEINS AS DIAGNOSTIC

INTERMEDIATES

Field of the invention

The presen~ invention relates to novel kits and reagents for diagnosis of Dengue

5 viral infections. In particular, the present invention relates to novel kits for diagnosis of

the four known closely related, antigenically distinct serotypes of Dengue virus. More

particularly. the present invention relates to multiepitope recombinant proteins and their

use in the diagnosis of dengue and other viral infections.

Background of the invention

10 Currently, dengue fever is the most important re-emerging mosquito-borne viral

disease, with the major proportion of the target population residing in the developing

countries of the world. This has prompted·the need to develop inexpensive, simple and

rapid diagnostic tests, without compromising on sensitivity and specificity.

Dengue fever and its more severe manifestations, namely, dengue haemorrhagic

15 fever and dengue shock syndrome, are caused by infection with the dengue viruses,

which are transmitted by mosquitoes of the genus Aedes. These viruses can produce a

spectrum of clinical symptoms in infected individuals, ranging from inapparent or mild

febrile illness to severe and fatal haemorrhagic disease. Epidemiological and laboratory.

evidences suggest that both viral and host immunologic factors are involved in the

20 pathogenesis of severe dengue disease. In recent decades, there has been a dramatic

increase in the incidence and clinical severity of dengue infections. According to the

World Health Organization's estimates, there may be currently as many as 100 million

cases of dengue fever every year. About 2.5 billion people in over a hundred tropical

and sub-tropical countries, representing ..... 40% of the world's population, are now at

25 risk from dengue. The global resurgence of dengue has been attributed to several

factors, such as . lack of effective vector control measures, uncontrolled urbanization

coupled to concurrent population growth and increased air travel. This, in conjunction

with the unavailability of a vaccine, has led to the current emergence of dengue as a

serious pUblic health threat (Gubler, D. J. (1998) Dengue and dengue haemorrhagic

30 fever. Clin. Microbiol. Rev. 11: 480-496.}.). There is no effective antiviral therapy.for

the treatment of dengue infections (Leyssen, P., De Clercq, E., and Neyts, J. (2000)

Perspectives for the treatment an.d injections with Flaviviridae. Clin. Microb. Rev. 13:

67-82.). Early di.agnosis, followe~ by supportive care, and symptomatic treatment

t~ough fluid replacement are the keys to survival in cases of severe dengue infection.

WO 2005/014627 PCTIIN2004/000237 2

There are four closely related, antigenically distinct, serotypes [(Gubler, D. J.

(J998) Dengue and dengue haemorrhagic fever. Clin. Microbiol. Rev. 11: 480-496.);

{Leyssen, P., De Clercq, E., and Neyts, J. (2000) Perspectives for the treatment and .

infections with Flaviviridae. Clin. Microb. Rev. 13: 67-82.}; (Lindenbach, B. D., and

5 Rice, C. M (2001) Flaviviridae: The viruses and their replication, pp.991-1041. In P-M. Knipe and. P. M Howley (eds.-in-chiej), Fields Virology lh ed Lippincot Williams

and Wilkins, Philadelphia.),' (Kuhn, R J., Zhang, w., Rossman, MG., Pletnev, S. v., Corver, J., Lenches, E.,)] of dengue viruses, each of which can cause disease. These

viruses are members of the faniily Flaviviridae; they have a common morphology,

10 genomic structure and antigenic determinants (Lindenbach, B. D., and Rice, C. M

(2001) Flaviviridae: The viruses and their replication, pp.991-104J. In D. M. Knipe

and P. M Howley (eds.-in-chiej), Fields Virology lh ed Lippincot Williams and

Wilkins, Philadelphia). A computer generated graphic representation of the dengue .

virion has been depicted in the art. (Kuhn, R J., Zhang, w., Rossman, MG., Pletnev, S.

15 v., Corver, J., Lenches, E., Jones, C. 'T., Mukhopadhyay, s., Chipman, P. R, Strauss, E.

G., Baker, T. s., and Strauss, J. H. (2002) Structure of dengue virus: implications for

j/avivirus organization, maturation, and fusion. Cell 108: 717-725.). :The major

structural protein covering almost the entire surface of the virion is the Envelope (E)

protein. The E protein is critical in the process of viral invasion by virtue of its

20 capacity to interact with host cell surface receptors (Chen, Y., Maguire, T., and Marks,

R M. (J996) Demonstration of binding of dengue envelope protein to target cells. J. .

Virol. 70: 8765-8772), and it is the major virus antigen capable of eliciting protective

and long-lasting immune responses against infection.[(Men, R, Wyatt, L.,. Tokimatsu,

I, Arakaki, S., Shameem, G., Elkins, R, Chanock, R, Moss, B., and Lai, c.-J. (2000)

25 Immunization of rhesus monkeys with a recombinant of modified vaccinia virus Ankara

expreSSing a trun9ated envelope glycoprotein of dengue type 2 virus induced resistance

to dengue type 2 virus challenge. Vaccine 18: 3113-3122.}; (Putnak, R, Feighny, R.,

Burrous, J., Cochran, M, Hackett, C., Smith, G., and Hoke, C. (l991) Dengue-/ virus

envelope glycoprotein gene expressed in recombinant baculovirus elicits virus-

30 neutralizing antibody in mice and protects them from virus challenge. Am. J. Trop.

Med Hyg. 45: 159-167.); (Churdboonchart, v., Bhamarapravati, N.,Peampramprecha,

S. and Srinavin, S. (1991) Antibodies against dengue viral proteins in primary cmd

secondary dengue hemorrhagicfever. Am. J. Trop. Med Hyg. 44: 481-493.)J

WO 2005/014627 PCTIIN2004/000237 3

The virion contains two other structural proteins, premembrane (prM),

implicated in the maintenance of the structural integrity of E, and capsid (C)" a hi~y .

basic protein, which interacts with the RNA genome. The -11 kilobase (Kb) RNA

genome of the virus has positive polarity and serves as the viral mRNA. Its 5' 'end is

5 capped, but lacks a 3' poly A tail. The 5' quarter of the genome encodes the structural

proteins C, prM and E, mentioned above. The rest of the genome encodes seven

nonstructural (NS) proteins. The genomic RNA contains a single open reading frame

(ORF) of over 10 Kb, flanked by 5' and 3' non coding regions. The order of proteins

encoded in the long ORF is 5' C-prMlM-E-NSI-NS2A-NS2B-NS3-NS4A-NS4B-NS5

10 3'. as· depicted in Figure 2B. This ORF is translated into a single polyprotein

pr,ecursor, which is co-tr~slationally and post-translationally processed by both host­

as well as virus-encoded proteolytic enzymes to give rise ~o 3 structural (C, prM and E) .

and 7 non-structural (NS) proteins. From the perspective of developing diagnostic

antigens, E [(Innis B., L., Thirawuth, V. and Hemachudha, C. (1989) Identification of

15 continuous epitopes of the envelope glycoprotein of dengue type 2 virus. Am. J Trop.

Med Hyg. 40: 676-687.); (I'rirawatanapong, T., Chandran, B., Putnak, Rand

Padmanabhan, R (1992) Mapping of a region of dengue virus type-2 glycoprotein

required for binding by a neutralizing monoclonal antibody: Gene 116: 139-150.),]

NSI (50. Wu, H C., Huang, Y. L., Chao, T. T., Jan, J. T., Huang, J. L, Chiang, H .Y,

20 King, C. C. and Shaio, M F. (2001) Identification of B-cell epitope oj dengue virus

type 1 and its application in diagnosis oj patients. J. Clin. Microbiol. 39: '977-982.);

. (Huang, J. H, Wey, J. J., Sun, Y. C., Chin, C., Chien, L. J. and Wu, .Y C. (1999)

Antibody responses 'to an immunodominant nonstructural 1 synthetic peptide in

patients with dengue fever and dengue hemo"hagic fever. J. Med Virol. 57: i-8.);

25 (Garcia, G., Vaughn, D. W. and Del Ange~ R M (1997) Recognition of synthetic

o/igopeptide jr-01Jl nonstructural proteins NS1 and NS3 oj DEN-4 virus by sera from

dengue injected children. Am. J. Trop. Med Hyg. 56: 466-470), (Fa/conar, A. K, Young, P. R and Miles, M. A. (1994) Precise location of sequential dengue virus

subcomplex and complex B cell epitopes on the nonstructura/-1 glycoprotein. Arch

30 Virol. 137: 315-326.)] and NS3 (Garcia, G., Vaughn, D. W. and Del Angel, R M .

(1997) Recognition of synthetic o/igopeptide from nonstructural proteins NS1 and NS3 ofDEN-4 virus by serafrom dengue irifected children. Am. J. Trop. Med Hyg. 56: 466-

470.) are important as they carry numerous immunodominant epitopes.

wo 2005/014627 PCTIIN2004/000237 4

Most often, the diagnosis of dengue infections in endemic regions is based on

clinical presentation, and it can be confused with other viral diseases (such as rubella,

enteroviruss and influenza) with similar clinical features. Further, clinical presentation

of dengue can vary, making accurate diagnosis extremely difficult. This underscores

5 the importance of laboratory-based diagnostic tests in providing timely medical

attention (George, R. and Lum, L. C. S. (1997) Clinical spectrum of dengue infection,

pp. 89-113. In D. J. Gubler and G. Kuno. G. (eds.), Dengue and Dengue Hemorrhagic

Fever, CAB International, Wallingford)

Four major laboratory criteria are used to confirm dengue virus infection,'

10 namely, the isolation of infectious virus, the demonstration of elevated virus-specific

antibody titers, the demonstration of dengue virus antigens andlor the detection of viral

RNA [(Guzman, M G. and Kouri, G. (1996) Advances in dengue diagnosis. Clin.

Diagn. Lab. Immunol. 3: 621-627.); (Vorndam, V. and Kuno, G. (1997) Laboratory

diagnosis of dengue virus,infections, pp. 313-333. In D. J. Gubler and G. Kuno. G.

15 (eds.), Dengue and Dengue Hemorrhagic Fever, CAB International, Wallingford.)].

Isolation of virus from clinical samples can be achieved by intracerebral inoculation of

1-2 day old suckling mice, intrathoracic inoculation of mosquitoe larvae, or by using

mammalian/insect cells in culture. Of these, the mosquito system is the most sensitive.

All these methods are slow and cumbersome. Viral RNA can be detected using.

20 coupled reverse transcription and polymerase chain reaction (RTPCR) or direct

molecular hybridisation. Viral antigens can be detected by immunohistochemistry or

immunofluorescence. However, the complexity of these assays and their bJgh cost

preclude their routine use. Serodiagnosis of dengue viruses is complicated by the

existence of cross-reactive antigenic determinants shared by members of. the

25 Flaviviridae family .. Commercial kits available for dengue diagnosis through detection

ofvirus-specifiG~tibodies use virus lysates as the coating antigen for antibody capture

and, consequently, suffer from poor sensitivity and specificity. Additionally, the

production of viral antigen using the mouse, mosquito or tissue culture-based system is

associated with high cost.

30 There is thus, currently a need for developing cost-effective, simple and rapid

diagnostics that combine sensitivity and specificity.

WO 2005/014627 PCTIIN2004/000237 5

Objects of the invention

Accordingly, it is one of the important objects of the pr~ent invention to

provide inexpensive, simple and rapid diagnostic kits and reagents for detecting dengue

infections.

S It is another object of the present invention to provide inexpensive, simple and

rapid diagnostic kits and reagents for detecting dengue infections without

compromising on sensitivity and specificity.

It is yet another object of the present invention to provide diagnostic kits and ..

methods for detecting dengue infections, which obviates many of the disadvantages of

10 the prior art.

It is a further object of the present invention to provide diagnostic kits and

methods capable of specifically detecting dengue-specific IgM (Immunoglobin M).

It is a further object of the present invention to provide diagnostic kits and

methods capable of specifically detecting dengue-specific IgG (Immunoglobin G).

1 S It is a further Qbject of the present invention to provide novel recombinant

protein antigens by assembling key immunodominant'linear dengue-specific epitopes.

Summary· of the invention

The above and other objects of the present invention are achieved by providing

two dengue multiepitope proteins, one designed to detect IgM and the other to detect

20 IgG antibodies in dengue patient sera. The present invention also discloses a simple and

rapid dengue spot test that are useful under field conditions.

An important aspect of the present invention resides in designing and

expressing two novel recombinant. protein. antigens by assembling key

immunodominant linear dengue-specific epitopes, chosen on the basis of pepscan

2S analysis, phage display and computer predictions. One of these developed to

specifically detest dengue-specific IgM and· the other to detect IgG. Th.ese novel

recombinant dengue multiepitope proteins were expressed in E. colt, purified in a

single step and used as capture antigens in EUSA The ELISA results, using a large

panel of suspected dengue patient sera, were in excellent agreement with those obtained

30 using the commercially available Dengue Duo IgM and IgG Rapid strip test (panBio).

The present invention also provides a simple and rapid spot test for dengue

detection using these recombinant multiepitope proteins. The high epitope density,

careful choice of epitopes and the use of E. coli system for expression, cou,pled to

WO 2005/014627 PCTIIN2004/000237 6

simple purification, jointly has resulted in inexpensive diagnostic tests kits and methods

with a high degree of sensitivity and specificity.

Detailed description

The present invention will now be described in greater detail with reference to

5 the accompanying drawings and the following examples.

In the drawings:

Figure 1 shows th~ global prevalence of dengue and its mosquito vee·tor (WHO

report 2000). The bar diagram depicts the remarkable rise in. incidence of dengue

infections in recent years.

10 Figure 2A depicts the Dengue virus structure and genome organization,

particularly, computer generated graphic representation of dengue virion structure ..

Figure 2B shows schematic representation of the ~ 11 KbRNA genome of

dengue viruses. The asterisks indicate the. locations of immunodominant epitopes

identified by pepscan, phage display and computer predictions as mentioned in the text.

15 NC denotes the non-coding regions of the genome.

Figure 3A· shows the design of the IgG-specific recombinant dengue

multiepitope protein (rDME-G), particularly, schematic representation of the synthetic

gene encoding linear immunodominant epitopes (shown by the boxes) derived from E,

NS 1 and NS3 interco~ected by triglycyllinkers. The gene was designed with 5' Bam

20 HI and 3' Bgl IT sites to facilitate cloning.

Figure 3B shows Map of the E. coli expressioh plasmid obtained by inserting

the rDME-G gene into the vector pQE60. Expression of the recombinant gene is under

the control of an IPTG inducible promoter. The vector provides a 6xHis Tag to

facilitate one-step purification.

25 Figure 3C shows a computer generated structure of the rDME-G protein based

on homology mQd.elling. The blue (dark) segments indicate the epitopes and the yellow

segments indicate the triglycyllinkers,

Figure 3D shows a schematic representation of the IgM multiepitope of the

present invention.

30 Figure 3E shows a schematic representation of the IgG multiepitope of the

present invention

Figure 4A depicts the design of the IgM-specific recombinant dengue

multiepitope protein (rDME-M). The gene was designed with 5' Bam HI and 3' Bgl II

sites to facilitate cloning. It schematic represents the synthetic gene encoding linear

WO 2005/014627 PCTIIN2004/000237 7

immunodominant epitopes (shown by the boxes) derived from NSI of all four dengue

virus serotypes, linked by tetraglycyllinkers

Figure 4B depicts the Map of the E. . coli expression plasmid' obtained' by

inserting two copies of the rDME-M gene sequentially into the vector pMAL-c2x.

5 Expression of the recombinant gene is under the control of an IPTG inducible

promoter. A 6xHisTag was incorporated at the 3' end of the recombinant molecule .

(prior to insertion into pMAL-c2x) to facilitate one-step purification.

~igure 4C shows a computer generated structure of the rDME-M protein based

on homology modelling.

10 Figure SA shows rapid spot tests for the detection of anti-dengue antibodies in

sera. The assay utiliies p~ed rDME-G (or rD'ME-M) at spot <CT' and normal

(dengue-negative) human serum at spot "C". The presence of IgG antibodies is

revealed by gold-conjugated protein G; IgM antibodies are detected with gold-labeled

rD'ME-M, instead.

15 Figure 5B depicts the design of the proposed dengue dual (IgM & IgG)

detection test.

Figure 5C shows a computer generated structure of the rDME-M protein based

on homology modelling.

Figure 6 shows the regions of linear immunodominant epitopes of dengue .

20 proteins.

Figure 7A depicts a map of plasmid pQE-60-IgG-rDME.

Figure 7B shows SDA-PAGE analysis ofIgG-rDME protein expreSsion.

Figure 8A shows map of plasmid [pMALc2x-rDME]

Figure SB shows SDA-P AGE Analysis IgG-rDME protein expression.

25 Figure 8C shows SDA-PAGE analysis single copy IGM-rDME protein

expression.

Figure 9 shows localizatio of single copy IgM-rDME protein expression.

Figures lOA, lOB and IOC show purofocation of rDME proteins by ni-NTA

affinity Chromatograhy.

30 Figures IIA and lIB ~ow the Western and Immunoblot analysis of rDME

proteins.

An important feature of the present invention resides in designing and

expressing synthetic protein antigens by assembling key immunodominant linear

5

WO 2005/014627 PCT/IN2004/000237 8

dengue-specific epitopes. This multiepitope approach to designing antigens has several

advantages as summarized in Table 1:

Table 1: The advantages offered by the recombinant multiepitope protein strategy

• Higher sensitivity (epitope density) • Higher specificity (cross-reactive epitopes) • Inexpensive (shake-flask culture: ~50, 000 testslL) • No multiple peptide synthesis

10 • No multiple protein expressions/purifications • No virus culture and viral antigen purification (eliminates biohazard) • Simultaneous detection of multiple infections

The high density of the epitopes in the recombinant dengUe multiepitope

15 (rDME) protein and the careful choice of only dengue-specific epitopes as its

components contribute to a high degree of sensitivity and specificity. Further, the

applicants' novel approach of using a recombinant multiepitope protein completely

obviates multiple peptide synthesis and mUltiple protein expression; it also avoids .

expensive and time-consuming virus culture (for antigen preparation) and t~e

20 associated biohazard risk. The design of the rDME protein and the' ease of its

expression and purification makes this a highly cost-effective approach to dengue

diagnosis. This approach has the potential for the simultaneous detection of multiple

infectious diseases.

According to the present invention, novel IgG- and IgM-specific dengue

25 multiepitope proteins were developed and were expressed in E. coli as described below

and purified to homogeneity by afftnity chromatography.

In brief, the novel IgM multiepitope and IgG multiepitope of the present

invention were synthesised as follows:

1. IgM multiepitope:

30 (a) The novel IgM multiepitope of the present invention, was synthesised having

immunodominal epitope from NS1 of DEN-I, DEN-2, DEN-3 and DEN-4. All

these epitopes were linked with four glycine linkers, and codon optimized for

expression in E. coli . The schematic representation of the IgM multiepitope of the

present invention is shown in Figure 3D .The novel IgM multiepitope of the

35 present invention has the nucleotide sequence as shown in Seq .. lD 1. Similarly, it

has the amino acid sequence as shown in Seq ID 2.

WO 2005/014627 PCTIIN2004/000237 9

Seq ID 1.

(a) IgM muJtiepitope sequence

5 GGA TCC GAT AGe GGC TGe GTG ATT AAC TOO AAA GGC CGT GAA CTG

10

AAATOC .

GGC (;GT GGC GGT GAT AGe GGe TGe GTG GTG AGe TOO AM AAC AAAGAACTG .

AAA TGC GGTGGC GGT GGC GAT ATG GGC TGe GTG ATT AAC TGG AAAGGCAAA

GAA CTG AAA TGC GGC GGT GGC GGT GAT ATG GGe TGe GTG GTG 15 AGC TOO AGe

GGC AAA GAA CTG AAA TGe GGT GGC GGT GGC AGA TCT

Seq ID2

20 gly ser asp ser gly cys val i/e asntrp lys gly arg glu lys cys

gly gly gly gly asp ser gly cys val val ser trp Iys asn 25 lys glu leu

/ys cys gly gly gly gly asp met gly cys val iIe asn trp ly~ gly lys

30 glu leu Iys cys gly gly gly gly asp met gly cys val val ser trp ser

gly lys glu leu lys cys gly g/y gly gly arg ser

35 (b) The synthesised gene is clon~ in-frame with maltose binding protein in

40

pMAL-c2X vectO! with MBP fusion and expressed in E.coli (DH5-a).

(c) The protein is purified under denaturing conditions by Ni-NTA affinity

chromatography.

(d) The purified protein was evaluated with dengue specific ELISA

2. IgG multiepitope:

(a) The IgG multiepitope of the present invention was designed, which is having'

one epitope from core, one epitope from prM, eight epitopes from DEN-2

leu

WO 2005/014627 PCT/IN2004/000237 10

envelop'e, four epitopes from DEN-4 NS1, one epitope from DEN-l NS1, one

epitope from DEN-2 NSI and one epitope from DEN-4 NS3. All these epitopes

were linked with three glycine linkers, and codon optimized for expression in

E.coli. The schematic representation of the IgG multiepitope of the present'

5 invention is shown in Figure 3E. It has a nucleotide sequence as shown in Seq ID,

3 and amino acid sequence as shown in Seq ID 4.

Seq ID 3

IgG multiepitope sequence

10 GTG CTG CGT GGC TTT CGT AAA GAA ATT GGC GGC GGC GGC CTG ACCACC

15

CGT AAC GGe GAA CCG CAT ATG ATT GTG ATG CGT CAe} GAA AAA GGCAAA

TCC CTG CTG TIT AAA ACC GGC GAT GGC GTG OOC GGC GGC OOA TCCGAA

ACC CTG GTG ACC TIT AAA AAC CCG CAT GCG AAA AAA CAG GAT 20 GTGGTG .

GTG CTG GGC AGC GGC GGC GGC AAC CTG CTG ITT ACC GGC GGC GGCCCGTTT

25 GGC GAT AGC TAT ATT ATT ATT GGe GTG GAA GGC GGC GGC CAG CTGAAACTG

30

AAC TOO ITT AAA AAA OOC AGC AGC GGC GGC GGC ACC GeG TGG GATTTT

GGC AGC CTG GGe GGC GTG TIT ACC AGe ATT GGC GGC GGC GTG ATT

ATT ACC TGG _ATT GGC GGC GGC AGC ACC AGe CTG AGC GTG GGC 3S GGC GGC

GTG ACC CTG TAT CTG GGC GeG GGC GGC GGC GAA CAT AAA TAT AGCTGG

40 AAA AGC OOC GGC GGC GAT AGC GGC TGe GTG GTG AGC TGG AAA AACAAA

45

GAA CTG AAA TGC GGe GGC GGC AAA ITT CAG CCG GAA AGC CCG GCGCGT

WO 2005/014627 PCT1IN2004/000237 11

CTG GCG AGC GCG AIT, CTG'AAC GCG GGe GGC GGC CTG AAA TAT AGCTGG

AAA ACC TOO GGC AAA GCG AAA GGC GC.JC GGC ITT CTG ATT GAT 5GGCCCG

GAT ACC AGe GAA TGC CCG AAC GAA CGT CGT GCG GGC GGC GGC TGGTAT

GGC ATG GAA ATT CGT CCG CTG AGe GAA AAA GAA GAA AAC ATG 10 GTGOOC

GGC GGC ATT CTG GAA GAA AAC ATG GAA GTG GAA ATTTGG AGC CGTGAA

15 GGC GAA AAA AAG AAA CTG AGA TCT

SeqID 4

20 val leu arg gly phe arg Iys glu He gly gly gly gly leu thr thr

arg am gly glu pro his mel iIe val mel arg gIn glu lys gly lys

25 ser leu leu phe lys thr gly asp gly val gly gly gly gly

ser glu

thr leu val Ihr phe Iys am pro his ala Iys Iys gin asp 30 val val

val leu gly ser gly gly gly asn leu leu phe thr gly g/y gly pro phe

35 gly asp ser tyr ile tIe ile gly val glu gly gly gly gin leu Iys leu

asn trp pJie-Iys /ys gly ser ser gly gly gly thr ala trp asp phe

40 gly ser leu g/y gly val phe thr ser ile gly gly gly val ile

ile thr trp ile gly gly gly ser thr ser leu ser val gly gly gly

45 val thr leu tyr leu gly ala gly gly gly glu his Iys tyr ser trp

wo 2005/014627 PCTIlN2004/000237 12

lys ser gly gly gly asp ser gly cys val val ser trp lys asn lys

glu leu lys cys gly gly gly lys phe gin pro glu ser pro 5 ala arg

leu ala ser ala ile leu asn ala gly gly gly leu lys tyr sel' trp

10 lys thr trp gly lys ala lys gly gly gly phe leu ile asp gly pro

15

asp thr ser . glu cys pro asn glu arg arg ala gly gly gly tlp. tyr

gly met glu ile arg pro leu ser glu lys glu glu asn met val gly

gly gly ile leu glu glu asn met glu val glu ile trp thr 20 arg glu

gly glu lys lys lys leu arg ser

(b) The synthesised gene was cloned in pQE-60 and expressed In E. coli

25 (SGI3009).

(c) The protein was purified under denaturing conditions by ni-NTA affinity

chromatography ..

(d) The purified protein was evaluated with dengue specific ELISA.

The above procedures will be described in detail hereinafter:

30 IgG-specific epitopes were carefully identified by one or more of the following

techniques: pepscan, phage display or computer predictions. A synthetic gene (Figure

3A), that encodes a recombinant IgG-specific dengue multiepitopeprotein (rDME-G),

was first gener~ted by ligation of carefully designed oligonucleotides encoding the

selected epitopes. The resultant gene encoded 15 linear epitopes, of which 8 were from

35 the E protein and 7 from the nonstructutal proteins NSI and NS3. These epitopes range

from 6 to 20 amino acid (aa) residues in length. Adjacent epitopes are separated by

gly-gly-gly tripeptide linkers. The rDME-G gene was inserted in-frame with. the 6x

histidine tag-encoding sequence of the bacterial expression vector pQE60, under the

control of an inducible PT5 promoter. This expression vector is depicted in Figure 3B.

40 The rDME-G (-25 kDa) protein was expressed in E. coli and purified to homogeneity, .

using Ni-NTA affInity chromatography under denaturing conditions. The rDME-G

WO 2005/014627 PCT1IN2004f000237 13

protein, modeled in Figure 3C, shows iliat all the epitopes are well displayed and· free~y

acceSsible for antibody binding.

An IgM-specific 15 aa epitope of the NSI protein of dengue virus serotype 2,

identified on the basis of computer prediction and' ELISA with patients sera was

5 selected as the building unit for designing an IgM-specific dengue multiepitope

(rD:ME-M) protein. Corresponding NSI epitopes from dengue serotypes 1,'3 and 4

were identified by sequence homology. Again, as described for rDME-G, a sy~thetic

. gene, encoding each of these four NSI epitopes joined by (gIy)4 tetrapeptide linkers,

was created. Two copies of this synthetic gene were fuseP in tandem as shown in

10 Figure 4A This gene was then inserted in-frame with the malE 'gene of the vector .

pMAL-c2x under the Ptac promoter, as shown in Figure 4B. As problems of protein

insolubility during the expression and purification of rDME-G protein were

encountered, the applicants chose to express the rDME-M protein as a fusion derivative

of maltose binding protein (MBP) to enhance its solubility. The rDME-GIMBP fusion

15 protein was purified to near homogeneity by Ni-NTA affinity chromatography. The

rDME-M protein (without its fusion partner) is modeled in Figure 4C, showing once

again that its epito~es are freely accessible for binding as in the rDME-G protein.

Evaluation of the rDME proteins as dia~ostic reagents

Having created the rDME proteins, their utility as diagnostic reagents for

20 dengue detection was tested. To determine if the rDME proteins could recognize and

bind dengue virus-specific antibodies, the purified proteins were tested separately as

capture antigens in ELISA, using dengue . .wus type 2 hyperimmune murine serum as

the test sample. For comparison, a control experiment was performed in parallel, using

dengue type-2 virus (instead of the rDME proteins) as the capture antigen. Antibody

25 titers determined in the test and control experiments were comparable, indicating that

our synthetic rD]\-1E proteins were capable of efficiently recognizing serum dengue . .

antibodies. This suggested that these two rDME proteins might serve as potential

diagnostic reagents for the detection of dengue antibodies in patient sera.

In order to evaluate the feasibility of using these rDME pro.teins as diagnostic

30 reagents to detect IgG and IgM anti-dengue antibodies, the applicants developed in-

house ELISA protocols (ICGEB protocols). In these assays, either rDME-G or iDME­

M protein was used separately to capture either IgG or IgM class of anti-dengue,

antibodies, respectively, from patient sera. Captured IgG and IgM antibodies were

revealed using horseradish peroxidase conjugated anti-human IgG and anti-human

wo 2005/014627 peT /IN2004/00023 7 14

IsM, respectively. We analyzed a large panel (n=I72) of suspected dengue patient

sera, obtained from dengue endemic regions in Sri Lanka, for the presence of dengue

antibodies, using our rUME proteins in the ELISA format described above. We then

compared our results with those obtained using PanBio's Dengue Duo IgM and IgG

5 rapid strip test. All samples were also tested for the presence of infectious virus and

viral RNA. Based on the data obtained, the samples could be categorized into six

10'

15 '

20

groups as summarized in Table 2 below:

Table 2: Evaluation of rDME-M and rDME-G proteins

Groupa JgMlIgG statusb Total PanBioo/lCGEBd data Samples

la IgM/IgG- 16 16/16 Ib IgM"/IgG- 5 2/5 lc Ig~/IgG+ 1 III 2 Ig~/IgG+ 12 12/12 3 Ig~/IgG- 37 37/37 4 IgM/IgG+ 21 21/21 5 IgM'/IgG- 59 59/59 6 ? 21 ?

aGroup 1 samples tested positive for virus; groups 2-6 tested negative for virus

bpresence and absence of IgM or IgG is indicated by '+' and '-' superscripts

25 respectively

"Data generated using Dengue Duo IgM &. IgG Rapid Strip test purchased from PanBio

Pty., Australia.

dThe ICGEB data were obtained using .rDME-M and rDME-G separately as capture

antigens in ELISAs to detect IgM and Igri, respectively

30 ? All samples gaye inconsistent results in both tests

The first group (n=22) represents the infected sera, in which the viral RNA

could be detected by RTPCR; further infectious virus could be isolated from all but one

sample of this group. With regard to the serology, this group was heterogenous, as

reflected by its division into subgroups a, b and c. A total of6 samples (all 5 of lb + the

35 single one from lc) were found to contain IgM antibodies. Out of these 6, the PanBio

strip test identified 3 samples (2 from Ib and the single one from Ic) as IgMI-. This

observation suggests that the rDME-M based ICGEB IgM ELISA is more effective in

the early detection of dengUe infection in a slightly larger proportion of samples. With

WO 2005/014627 PCTIIN2004/000237 15

regard to IgG, only one sample (lc) was found to be positive; this was corroborated by

the PanBio test. The remaining samples (group Ia; n=16) were negative for both IgM

and IgG. Once again, these results were borne out by the PanBio test. The sampl~s

represented by groups 2-6 (collective 0=150), were all found to be virus-!RNA-. Of the

5 129 samples, represented collectively by groups 2-5, we could detect the presence of

IgM in 49 samples (12 samples of group 2 + 37 samples of grOU!? 3), and IgG in 33

samples (12 samples of group 2 + 21 sample-.8 of group 4), using the ICGEB ELISAs.

Samples in group 5 w~re all IgM"lIgG-, and are presumably 'normal', as all of them

tested negative in the virus isolation and RTPCR assays as well .. In regard to the'

10 PanBio test, 19 samples of group 1 (16 from group la, 2 from group Ib and the single

sample from group lc) and all samples of groups 2-5 (n=129) yielded results that were

identical to those obtained using the ICGEB IgM and IgG ELISAs. For s~ples in

Group 6 (n=21), the results of the ICGEB ELISAs did not agree with those of the

PanBio test. The reason for the observed discrepancy is unclear at present, but is

15 presumably related to differences in the nature of the capture antigen. It must be

pointed out that neither virus nor viral RNA could be detected in all .these

'indeterminate' samples. It is, therefore, essential at this juncture to establish the

clinical histories of the patients, from whom these samples were obtained, in order to

understand the reason for the observed discrepancy. The overall comparative analysis

20· of our data with the conventional PanBio results suggests that there is an excellent

agreement between the ICGEB and the PanBio tests; with -86% (148 out of 172) ~fthe

samples analyzed giving identical results. Further, the ICGEB IgM ELISA picked up

three additional samples (group Ib), which the PanBio test failed to identify.' The fact,

that these three samples were virus +!RNA +, suggests that the ICGEB IgM ELISA test is .

25 more effective in early diagnosis of dengue infection.

Design of kits for rapid detection of dengue specific IgM and IgG

The present invention particularly relates to the development of an inexpensive,

simple and rapid test to detect dengue specific IgM and IgG antibodies that will have

utility under· field conditions. Therefore, having demonstrated that the applicants'· .

30 recombinant multiepitope proteins can be used to identify dengUe specific antibodies ill the ELISA format, the applicants sought to adapt this to a spot test format, as shown in

Figure SA. The design of the test was simple. It consisted of a piece of nitrocellulose

membrane with two spots C (control) and T (test). Spot C contained normal human

serum and spot T contained either rDME-G (IgG-specific test) or rDME-M (lgM-

WO 2005/014627 PCTIIN2004/000237 16

specific test). The nitrocellulose membrane with the two spots (invisible in an unused

test module) was mounted on an adsorbent pad and enclosed in a shallow well of a

plastic module. To perform the spot test, a drop of the serum sample was added to the

well, washed and treated with gold conjugated protein-G for IgG-specific test, and gold

5 conjugated rDME-M for IgM-specific test. A positive test was indicated by two pink

spots (C and T) and a negative test shows a single pink spot at C with no visible spot at'

T. An invalid test was indicated by the absence of pink color in both Cand T. The , ,

present example was performed as two separate tests to detect IgM and IgG anti-

dengue antibodies but it is within the scope of the present invention to perform it as a

10 single test, that can simultaneously score both classes of antibodies, as shown in Figllre

5B. In this dual test, the rDME-G and rDME-M proteins are spotted separately on the

nitrocellulose membrane. As a control, normal human serum was applied' along the -1- ' shaped line. IgG and IgM antibodies, captured from a test serum sample, will then be

revealed, using a mixture of gold-labeled anti-human IgG/IgM.

15 Various tests were also carried out to compare the conventional PanBio Dengue

20

25

IgM kit with the' JgM kit of the present invention. The results, which are self­

expanatory, are shown in the Tables 3 to 8 below:

Comparison of Pan Bio Dengue IgM kit with ICGEB Dengue IgM kit

Cut off value is 0.450

Table 3. Virus isolation positive and PCR positive sera samples

S.NO ' Sample PCR Virus ICGEB PanBio ICGEB ID (Serotype isolation IgM IgM 'JgM

& +1-) (serotype O.D' (+1-) ( +1-) & +1-) (450nm)

1 D16 D2+ D2+ 0.466 + + 2 -027 D2+ D2+ 0.531 + + 3 75 D2+ D2+ 0.879 - + 4 D68 D2+ D2+ 1.514 - + 5 D65 D2+ D2+ 0.473 - + 6 D30 D2+ D2+ 0.309 - -7 D19 D2+ D2+ 0.236 - -8 D1l9 D2+ D2+ 0.289 - -9 D92 D3+ D3+ 0.221 - -10 D22 D2+ D2+ 0.372 - -11 D39 D3+ D3+ 0.17 - -

5

WO 2005/014627 PCTIIN2004/000237 17

Table 4. Virus isolation negative and PeR positive sera samples

S.NO

1

S.NO

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 -28 29 30 31

Sample PCR Virus ICGEB PanBio ID (Serotype isolation IgM IgM

&+1-) (serotype O.D (+1-) &+/-) (4S0nm)

011 02+ - 0.651 +

Table 5. Virus isolation negative and PCR negative sera samples (A. Both Pan bio and ICGEB positives)

Sample PCR Virus ICGEB PanBio ID (Serotype isolation IgM IgM

&+1-) (serotype O.D (+1-) &+1-) (450nm)

0109 - - 1.39i + 0148 - - 0.552 + 0129 - - 0.509 +

PO 171 - - 0.616 + 038 .- - 0.474 +

P0186 - - 0.524 ..., 76 - - 0.596 + 85 - - 0.46 +

054 - - 0.921 + 034 - - 0.845 +

P0194 - - 0.759 + 023 ~ - 0.537 + 018 - - 0.886 + 029 - - 0.639 + 025 - - 0.496 + 062 - - 0.47 + 059 - - 0.758 + 053 - - 0.685 + 0107 - - 0.639 + D110 - - 0.648 + 0111 - - 0.623 + 0114 - - 0.471 + 0116 - - 0.742 + 0118 - - 0.577 + D142 - - 0.462 + DI64 - - 0.722 + DI45 - - 0.862 + D160 - - 0.794 + 071 - - 0.689 + 0165 - - 0.517 + 090 - - 1.048 +

ICGEB IgM ( +1-)

+

ICGEB IgM ( +1-)

+ + + + + + + + + + + + +. + +. + -v + + + + + --+ ----=--+ . -----.:.. + ;-

~=--+ + +

.+

5

10

WO 2005/014627 PCTIIN2004/000237

32 33 34 35 36 37 38 39

- 40 41 42 43 44 45 46 47 48 49 50 51 52 53

S.NO

1 2 3 4 5 6

S.NO

1 2

18

D93 - - 0.574 + DI02 - - 0.522 + D147 - - 0.596 +

'D153 - - 0.805 + D140 - - 0.788 + 134 - - 1.228 +

D138 - - 1.489 + D139 - - 0.716 + D151 - - 1.091 + D159 - - 0.701 +

D2 - - 0.46 + D4 - - 0.472 + D6 - - 0.623 + D12 - - 0.822 + D13 - ~ 0.546 + D14 - - 0.549 + D36 I 0.625' + - -D42 - - 0.596 + D45 - - 0.641 + D47 - - 0.989 + D5l - - 0.469 . + D52 - - 0.565 +

Table 6. Virus isolation negativee and PCR negative sera samples ( B. panbio negative and ICGEB positive)

Sample PCR Virus ICGEB PanBio ID (Serotype isolation IgM IgM

& +/-) (serotype O.D (+/-) & +/-) (450nm)

PD199 - - 0.498 -78 - - 0.748 -79 - - 0.784 -

D156 - - 0.54 -D1l7 - - 0.634 -D50 - - 0.774 -

Table 7. Virus isolation negativee and PCR negative sera samples ( C. panbio positive and ICGEB negative)

Sample PCR Virus ICGEB PanBio ID (Serotype isolation IgM IgM

& +/-) (serotype O.D (+/-) & +/-) (450nm)

72 - - 0.373 + 73 - - 0.433 +

+ + + + +, + +

'+ + + + + + +

,+

+ + + + + + +'

ICGEB IgM ( +/-)

+ + + + + +

ICGEB IgM ( +/-)

-.:.

WO 2005/014627 PCTIIN2004/000237 19

3 P0196 - - 00408 + -4, P0178 - - 0.333 + -5 0143 - - 0.439 + -6 099 - - 0.37 + -7 049 - - 0.448 + -8 041 - - 0.443 + -

5 Table 8. Virus isolation negativee and PCR negative sera samples , (D. panbio negative and ICGEB negative)

S.NO Sample PCR Virus ICGEB . PanBio ICGEB ID (Serotype isolation IgM IgM IgM

&+1-) (serotype O.D (+1-) , (+1-) &+1-) (450nm)

1 0108 - - 0.342 - -2 P0180 - - 0.346 - ---=--3 P0169 - - 0.338 - -4 P0170 - - 0.218 - -5 POI72 - - 0.45 - ---6 037 - - 0.351 - -7 P0175 - - 0.278 - -8 P0176 - - 0.221 - -9 P0207 - - 0.27 - -10 P0174 - - 0.173 - -11 P0208· - - 0.317 - .-12 69 - - 0.146 - -13 70 - - 0.434 - -

-~

14 71 - - 0.148 - -15 74 - - 0.164 - --16 77 - - 0.146 - -17 80 - - 0.169 - -'

-~

18 81 - - 0.342 - -19 82 - - 0.153 - -20 83 - - 0.226 - -21 -84 - - 0.228 - -22 ' 195 - - 0.239 - -23 192 - - 0.208 - -24 190 - - 0.221 . - -25 055 - - 0.4 - -26 032 - - 0.209 - -27 033 - - 0.3 - -28 067 - - 0.383 - '--29 066 - - 0.249 - -30 064 - - 0.152 - --31 063 - - 0.352 - -32 P0167 - - 0.37 - -

WO 2005/014627 PCTIIN2004/000237 20

33 017 - - 0.204 - -34 020 - - 0.412 - -35 021 - - 0.175 - -36 035 - - 0.441 - -37 031 - - 0.2 - -38 ·024 - - 0.341 - -39 D26 - - 0.418 - -40 061 - - 0.36 - -41 060 - - 0.436 - -42 058 - - 0.41 - -43 057 - - 0.321 - -44 056 - - .0.37 - -45 0112 - - 0.264 - -46 0113 - - 0.384 . - -47 0115 - - 0.161 - -48 0141 - - 0.363 - -49 0154 - - 0.401 - -50 0166 - - 0.438 - -51 0157 - - 0.115 - -52 0158 - - 0.314 - -53 0149 - - 0.394 - -54 0150 - - 0.194 - ., 55 0181 - - 0.32 - -56 P0173 - - 0.246 - -57 0155 - - . 0.249 - -58 0205 - - 0.356 - -59 0206 - - Q.404 - -60 0203 - - 0.193 - -61 091 - - 0.198 - -62 0104 - - 0.356 - -63 0120 - - 0.26 - -64 0121 - - 0.399 - -65 0128 '- - 0.348 - -66 0163 - - 0.242 - -67 0152 - - 0.288 - -68 0161 - - 0.31 - -69 .0162 - - 0.188 - -70 01 - - 0.307 - -71 03 - - 0.288 - -72 05 - - 0.413 - -73 07 - - 0.264 '- -74 08 - - 0.155 - -75 09 - - 0.161 - -76 010 - - 0.165 - -77 015 - - 0.234 - -78 040 - - 0.276 - -79 043 - - 0.388 - -80 044 - - 0.194 - -

WO 2005/014627 PCTIJN2004/000237 21

81 046 0.346 82 048 0.431

Various tests were also carried out to compare the conventional PanBio Dengue

IgG kit with the IgG kit of the present invention. The results are shown in the Tables 9

5 to 14 below:

10 Cut off value is 0.450

Comparison of Pan Bio Dengue IgG kit witb ICGEB Dengue IgG kit

Table 9. Virus isolation positive and PCR positive sera samples

S.NO Sample PCR . Vb'us ICGEB PanBio ID (Serotype isolation IgG IgG.

. & +1-) (serotype O.D (+1-) &+I-} (450nm)

1 D68 02+ D2+ 0.47 + 2 022 02+ D2+ 0.293 -3 75 02+ D2+ 0.286 -4 D19 D2+ D2+ 0.193 -5 D65 D2+ D2+ 0.222 -6 D30 D2+ D2+ 0.216 -7 D27 D2+ D2+ 0.372 -8 D119 D2+ D2+ 0.312 -9 D92 D3+ D3+ . 0.371 -10 D39 D3+ D3+ 0.18 -11 D16 D2+ D2+ 0.291 -

15 Table 10. Virus isolation negative and PCR positive sera samples

S.NO Sample PCR Virus ICGEB PanBio ID (Serotype isolation IgG IgG

. &+1-) (serotype O.D (+1-) ,

& +1-) (450nml 1 Dll D2+ - 0.205 -

20

25

ICGEB IgG (+1-) .

+ ----------

ICGEB IgG ( +1-)

-

5

WO 2005/014627 PCTIIN2004/000237

S.NO

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

22

Table 11. Virus isolation negative and PCR negative sera samples lA. Both Pan bio and ICGEb positives)

Sample PCR Virus ICGEB PanBio ID (Serotype isolation IgG IgG ..

& +1-) (serotype O.D (+1-) &+1-) (450nm)

DI08. - - 0.994 + D129 ,.. - 1.514 +

PD180 - - 1.086 + PD171 - - 2.518 + PD175 - - 0.456 +

79 - - 1.378 + 84 - - 1.724 + 85 - - 0.669 + 195 - - 1.642 + D54 - - 0.692 + D66 - - 0.602 + D156 - - 1.04 + D17 - - 0.5 + D18 - - 1.183 + 026. - - 0.594 + 060 - - 0.688 + D53 ! - - 0.548 +

D114 - - 0.805 + D115 - - 0.775 + D158 - - 0.592 + D181 - - 0.578 + D155 - - 0.495 + D96 - - 1.243 + D128 - - 2.438 + D163 - - 0.766 + D140 - - 0.92 + D162 - - 1.497 +

D4 - - 0.824 + D5 - - 1.065 + D6 - - 0.471 + D8 - - 0.501 + D13 - - 2.284 + D14 - - 2.522 + 015 - - 0.921 .+ D43 - - 0.463 + D48 - - 0.518 + D49 - - 0.594 +

ICGEB IgG ( +1-)

+ + + + + + +. + + + + + + + + + + + + + + + + + + + + + + -+-+ + + + -+-+ +

5

10

15

I

WO 2005/014627 PCTIIN2004/000237 23

Table 12. Virus isolation negative and PCR negative sera samples ( B. panbio negative and ICGEB positive)

S.NO Sample PCR Vims ICGEB PanBio rCGEB·

1 2 3 4 5 6

S.NO

1 2 3 4 5 6 7

S.NO

1 2 3 4 5 6

·7

ID (Serotype isolation IgG IgG &+1-) (serotype O.D (+1-)

&+1-) (450nm) PD199 - - 0.656 -

83 - - 0.491 -034 - - 0.464 -064 - - 0.472 -D145 - - 0.453 -D151 - - 0.474 -

Table 13. Virus isolation negative and PCR negative sera samples . (C. panbio positive and leGEB negative)

Sample PCR VII11S ICGEB PanBio ID (Serotype isolation IgG IgG

& +1-) (serotype O.D (+1-) &+1-) (450nm)

73 - - 0.417 + . P0196 - - 0.307 + P0178 - - 0.235 + D117 - - 0.204 + 0166 - - 0.447 + D93 - - 0.435 + 050 - - 0.348 +

Table 14. Virus isolation negative and PCR negative sera samples ( D. panbio negative and ICGEB negative)

-Sample PCR Virus ICGEB PanBio

ID (Serotype isolation IgG IgG &+1-) (serotype O.D (+1-)

& +1-) (450nm) 0109 - - 0.33 -0148 - - 0.377 -

PD169 - - 0.348 -PD170 - - 0.25 -PDI72 - - 0.184 -

D37 - - 0.333 -D38 - - 0.42 -

IgG ( +1-)

+ + + + + +

ICGEB IgG

( +1-)

-------

ICGEB IgG ( +1-)

-------

WO 2005/014627 peT 1IN2004/00023 7 24

8 PD176 - - 0.285 - -9 PD207 - - 0.321 - -10 PD174 - - 0.264 - -11 PD186 - - 0.407 - -12 PD208 - - 0.294 - -13 69 - - 0.218 - '-14 70 - - 0.386 - -15 71 - - 0.277 - -16 72 - - 0.334 - -17 74 - - 0.387 - -18 76 - - 0.339 - -19 77 - - 0.239 - -20 78 - - 0.431 - -21 80 - - 0.353 - -22 81 - - 0.286 - -23 82 - - 0.338' - -24 192 - - 0.297 - -25 190 - - 0.235 - -26 D55 - - 0.408 - -27 D32 - - 0.293 - -28 D33 - - 0.447 - -29 D67 - - 0.272 - -30 D63 - - 0.287 - -31 PD194 - - 0.332 - -32 D23 - - 0.29 - -33 D20 - - 0.353 - -34 D21 - - 0.37 - -35 D35 - - 0.294 - -36 D31 - - 0.283 - -37 D29 - - 0.364 - -38 D24 - - 0.32 - -39 D25 - - 0.373 - -40 D62 - - 0.239 - -Al D61 - - 0.28 - -42 D59 - - 0.283 - -43 D58 - - 0.28 - -44 . D57 - - 0.213 - -45 D164 - - 0.224 - -46 DI57 - - 0.31 - -47 D149 - - 0.33 - -48 D150 - - 0.378 . - -49 D160 - - 0.386 - -50 D71 - - 0.438 - -51 DI65 - - 0.345 - -52 PDI73 - - 0.362 - -53 D143 - - 0.353 - -54 D95 - - 0.323 - -55 D205 - - 0.253 - -

wo 2005/014627 PCT1IN2004/000237 25

56 0206 - - 0.284 - -57 0203 - - 0.399 - -58 090 - - 0.303 - -59 091 - - 0.293 - -60 097 - - 0.311 - -61 D98 - - 0.366 - -62 099 - - 0.26 - -63 0102 - - 0.371 - -64 0104 - - 0.323 - .. 65 0120 - - , 0.277 - .. 66 0111 - - 0.219 - -67 0112 - - 0.268 - -68 0113 - - 0.359 - -69 0110 - - 0.333 - -70' 0107 - - 0.405 - -71 056 - - 0.427 - -72 0116 - - 0.426 - -73 0118 - - 0.348 - .. 74 0142 - - 0.3 - .. 75 0141 - - 0.311 - - "

76 0154 - - 0.25 - -77 0121 - - 0.384 - -78 0147 - - 0.238 .. -79 0153 - .. 0.293 - -80 0152 - - 0.199 - -81 134 - - 0.222 - -82 0138 - - 0.255 - -83 D139 - - 0.351 .. -84 0161 - - 0.226 - -85 0159 .. - 0.202 - -86 02 - - 0.361 - -87 03 - - 0.398 - -88 07 - - 0.271 - -89 09 - - 0.296 - -90 010 - .. 0.359 - -91 012 - .. 0.237 - -92 .036 - .. 0.257 - -93 040 - - 0.171 - -94 041 - ., 0.292 - -95 ' 042 - - 0.382 - -96 D44 - - 0.304 - .. 97 045 - - 0.375 .. -98 046 - - 0.439 - ..,

99 047 - - 0.285 - -100 051 - - 0.363 - -101 052' - - 0.259 - -

WO 2005/014627 PCT/IN2004/000237 26

The above results clearly establish the superiority of the dengue diagnosis kits

present invention.

Example

Materials and methods

5 Materials

Escherichia coli strains, DfI5a. was ,purchased from invitrogen' life

technologies, USA and SG~3009 was purchased from Qiagen, Germany. The plasmid

pQE-60, Ni-NT A supertlow resin a~d anti-His mAb were purchased from Qiagen,

Germany. The plasmid pMALc2x was purchased from New England Biolabs, USA

10 Urea was purchased from Serva Electrophoresis GmbH, Heidelberg. All other

chemicals were from sigma Chemical, St. Louis, Missouri, USA.

Designing ojrecombinant dengue multiepitope proteins (r-DME)

Two recombinant dengue multiepitope proteins (r-DME), which are IgG and

IgM specific, were designed from dengue virus- specific linear imrnunodominant

15 epitopes.

IgG specific r-DME

IgG specific r-DME was designed with eight epitopes from Envelope, SIX

epitopes from NS1 and one epitope from NS3. The location is shown in Figure 6. As

can be seen from this figure, it depicts regions of linear imrnunodominant epitope.s of

20 dengue proteins. Linear immunodominant epitopes numbered as 1-8 from envelope, 9-

14 from NSI and 15 from NS3 were included in IgGspecific recombinant dengue

multiepitope protein, and In IgM specific recombinant dengue multiepitope protein ,

epitope numbered as 10 from DEN-2 and the corresponding region of other three

dengue serotypes were included. The details of the epitopes are shown in Table 15

25 below:

Table. 15

Epitope Peptide Sequence Peptide Location Antibody Dengue Reference No. response serotype

1 ETLVTFKNPHA Envelope, 235-255 IgG 2 1 KKQDVVVLGS' ... aa

SeqID 5

2 NLLFTG ... SeqID 6 Envelope, 276-281 IgG 2 1 aa

3 PFGDSYIIIGVE ... Envelope, 372-383 IgG 2 1

SeqID 7 aa

WO 2005/014627 PCTIIN2004/000237 27

4 QLKLNWFKKGSS Envelope, 386-397 IgG ·2 2 ... Seq1O 8 aa

5 TAWDFGSLGGV Envelope, 418-433 IgG 2 . 1 FfSIG ... Seq ID 9 aa

6 VIITWIG ... Seq 10 Envelope, 461-46i . IgG 2· I 10 aa

7 STSLSV .............. Envelope, 472-477 IgG 2 . 1 SeqID 11 aa

8 V1L YLGA. .. Seq Envelope, 485-491 IgG 2 1 ID12' aa

9 EHKYSWKS ... Seq NSl, 110-117 IgG . 1 3 IDS

10 DSGCVVSWKN NSl, 1-15 IgM, low 2 4 KELKC ... Seq 10 14 levels of IgG

11 KFQPESP ARLAS NSl, 33-49 IgG 4 5 AILNA-... SeqID

15

12 LKYSWKTWGK NSI, 111-121 IgG 4 6 AK ... Seq1O 16 .'

13 FLIDGPDTSECP NS1, 133-149 IgG 4 5 NERRA ... Seq ID 17

14 WYGMEIRPLSE NSI, 330-346 IgG 4 5 KEENMV ... Seq 10

18

IS ILEENMEVEIWT NS3, 572-591 IgG 4 '5 REGEKKKL ... Seq

ID 19

Epitope numbered as I is spans from 235-255 amino acids in the DE'N-2

envelope protein, which contains two linear epitopes of amino acid 235-242 and 248-

255. First linear epitope reacted with 4 out of7 DEN-2 patients sera, 3 out of 8 DEN-I - . 5 patients sera, 1 out of 5 DEN-3 patients sera, 2 out of 6 DEN-4 patients sera, 0 out of 6

Japanese Encephalitis VIrUS (JEV) patients sera. Second linear epitope reacts 5 out of7

DEN-2 patie.qts sera, 3 out of 8 DEN-l patients sera, 3 out of 5 DEN:3 patients sera, 2.

out of 6 DEN-4 patients sera, 0 out of 6 JEV patients sera. Epitope numbered as 2, 6; 7

and 8 are spans from 276-281, 461-467, 472-477 and 485-491 amino acids respectively

10 in the DEN-2 envelope protein. These are proposed DEN-2 specific epitopes. Epitope

numbered as 3 is spans from 372-383 amino acids in the DEN-2 envelope protein. It

reacted with 7 out of 7 DEN-2 patients sera, 6 out of 8 DEN-l patients sera, 5 out of 5

WO 2005/014627 peT IIN2004/000237 28

DEN-3 patients sera, 2 out of 6 DEN-4 patiet;lts sera, 2 out of 6 mv patients sera. This

s~owed strong reaction with DENl, 2,.3 and weak with DEN-4 and JEV. Epitope'

numbered as 4 is spans from 386-39.7 amino acids in the DEN-2 envelope protein. This

binds specifically to the 3H5 mAb. Epitope numbered as 5 is spans .from 418-433

5 amino acids in the DEN-2 envelope protein. It reacted with 6 out of 7 DEN-2 patients'

sera, 5 out of 8 DEN-l patients sera, 4 out of 5 DEN-3 patients sera, 1 out of 6 DEN-4

patients sera, 1 out of 6 mv patients sera. This showed strong reaction with DEN 1 , 2, 3

and weak with DEN-4 and mv. Epitope numbered as 9 is spans from 110-117 .amino

acids in the DEN-l NSI protein. It reacted 20 out of 21 DEN-l patients sera and did

10 not react with DEN2, 3 and 4 patients sera. This epitope sequence is quite different

from mv, and mv hyperimmune sera did not shown any reaction with this epitope.

Ep'itope numbered as 10 is spans from 1-15 amino acids in the DEN-2 NSI protein. It

showed significant reaction with dengue fever patients sera but not with JEV patients

sera. It showed more IgM antibody response than IgG antibody response. Epitope

15 numbered as 11, 13 and 14 are spans from 33-49, 133-149, 330-346 amino acids

respectively in the DEN-4 NS 1 protein. These were reacted significantly with dengue

patients sera. Epitope numbered as 12 is spans from 111-121 amino acids in the DEN-4

NS 1 protein. This epitope was identified by using the mAb raised against DEN-2 NS 1

and this was found to be a dengue complex epitope. Epitope numbered as 15 is spans

20 from 572-591 amino acids in the DEN-4 NS3 protein. This epitope reacted significantly

with dengue patients sera.

All the 15 epitopes were linked with tri-glycyllinkers (if the glycine is coming

from epitope, where only two glycyllinkers were added) for the flexibility of the each

epitope . The nucleotide sequence of rDME gene was codon optimised to E. coli, and

25 got synthesized withBam III at 31 site and Bgl n at 51 site from Geneart, Regensburg,

Germany.

IgM specific r-DME IgM specific r-DME was designed with 15 amino acid long immunodominant

epitopes from nonstructural protein 1 (NSl) of DEN-I, 2, 3 and 4. Epitope numbered as

30 10 (Fig. 60 and Table. 15) is spans from 1-15 amino acids in the DEN-2 NSI protein. It

showed significant reaction with dengue fever patients sera but not with JEV patients

sera. It showed more IgM antibody response than IgG antibody response, Due to its

low IgG response it was included in IgG sfecific rDME. This sequence is quite similar

with DENl, 3 and 4, because of that, sequence 1-15 amino acids of corresponding

WO 2005/014627 PCT11N2004/000237 29

DEN-I, 2 and 3 NSI'were taken. These four epitopes were linked with tetra-glycyl

linkers for the flexibility of the each epitope . The nucleotide sequence of rDME gene . .

was codon optimised to E coli, and got synthesized with BamID at 3'site. and Bgill at

5'site from Geneart, Regensburg, Germany.

5 Construction ofpQFr60-IgG and IgM- rDME pJasmids Synthesized genes (IgG rDI\4.E is 708bp and IgM rDME is 240bp) got cloned in

peR-Script at KpnI and SacI sites with BamID at 3' Site and Bgill at 5'site. The genes

retrieved from peR-script by .restriction digestion with BamIll and Bgill and ligated

into pQE-60 vector (m which starting codon and 6X His tag is coming), which wa~

10 precut with the same enzymes and treated with calf intestine alkaline phosphatase to

prevent self ligation. The ligated product was transformed into Eeoli DHSa. cells and

selected on kanamycin plates. Screened the clones by restriction digestion with BamHI and Bgill .enzyIPes. One of the selected clone plasmid DNA was transformed into

E.coli SG13009 cells and selected on kanamycin and ampicillin plates and screened the

15 clones by restriction digestion with BamHI and Bgill enzymes.

Construction of pMALc2x- IgG and IgM- rDME plasmids Genes got retrieved from pQE-60-IgG rDME and' pQE-60-IgM rDME plasmids

by restriction digestion with BamHI and Hindfll enzymes, Which releases the gene .

with 6X His tag and terminator codon (IgG rDME is 732bp and IgM rDME is 264bp)

20 and ligated into pMALc2x vector that was precut with the same enzymes. The ligated

product was transformed into E. coli DH5a cells and selected on ampicillin plates ..

Screened the clones by restriction digestion with BamID and Hindill enzymes. In

pMALc2x rDME plasmid, maltose-binding protein (MBP) will expresse along . with

rD:MEs.

25 Construction of pMALc2x- IgM rDME double copy plasmid Gene got.retrieved from pQE-6O-IgM rDME single copy plasmid by restriction.

digestion with BamHI and BglII enzymes and ligated into pQE-60-IgM rDME single

copy plasmid. that had been precut with BgllI. The ligated product was transformed into

E.coli DHSa cells and selected on kanamycin plates. Screened the clones by restriction

30 digestion with BamID and Bgill enzymes. Positive clone upon restriction digestion

with Bamm and BgJII enzymes, releases double the size of the single copy (PQE-6.0-

IgM rDME double copy plasmid, 474bp).

wo 2005/014627 PCTIIN2004/000237 30

Double copy of gene got retrieved from pQE-60-IgM rDME double copy

plasmid by restriction digestion with BamM and HindIll enzymes and ligated into

pMALc2x vector that had been precut with BamM and HindIII. The ligated product

was transformed into E.coli DHSa. cells and selected on ampicillin plates. Screened the,

5 clones by restriction .digestion with Bamfll and HindIII enzymes (releases 50:4bp

fragment).

Expression screening

Positive clones conftrmed by restriction digestion were inoculated into 3ml test­

tube cultures and allowed to grow at 37°C in a shaker at 200rpm. Cultures were

10 induced with ImM isopropylthiogalactoside (IPTG) at logarithmic phase (at OD of

~0.5 at 600nm) for 4 hours. After induction, equal number of cells from various

induced cultures and respective un-induced cultures were lysed in sample buffer and

analysed by SDS-PAGE . The clone, which expressed maximum levels of the expected

recombinant protein was chosen for further experiments.

15 Localization oj rDME proteins in induced cells

Test-tube cultures of 5m1 were induced with ImM IPTG at log phase for 4 h.

Pelletted I m1 of culture and resuspended in 0.2m1 Tris-EDTA pH8.0 (10mM Tris and

ImM EDTA) with lysozyme (lmglml). Vortexed gently for lmin, and kept at 37°C for

lh. The lysate was centifuged at IS, OOOg for 5 min and the supernatant was transferred

20 into a fresh tube. The pellet was resuspended in 0.2m1 SDS-PAGE sample buffer and

solubilized by boiling for 10 min. Both pellets ~d supernatants were analysed by SDS­

PAGE.

Purification ojproteins by Ni-NTA affinity chromatogrhaphy

25 Pre-culture was prepared by inoculating IOIlI glycerol stock of pQE-60-IgG

rDME into 20mLLB medium with IOOf!g of ampicillin and 251lg ofkanamycin/ml, and'

10lll glycerol stock ofpMALc2x- rDME (IgG, IgM, IgM double copy) into 20ml LB

medium with 100llg of ampicillin. Cultures were grown overnight in a shaker at 37°C,

at 200rpm. Cultures were inoculated into I L LB flasks with respective antibiotics and

30 incubated at 37°C, at 200rpm for 2-3h. Cultures were induced with ImM

isopropylthiogalactoside (IPTG) at logarithmic phase (at OD of ~O.5 at 600nm) for 4

hours. Prior to puriftcation small aliquots of culture were analysed by SDA-P AGE.

. "

WO 2005/014627 PCTIIN2004/000237 31

The induced cultures were harvested by centrifugation in a Sorvall GS3 ro~ at

6000 rpm for 20min at 4DC. The cell pellet 03.5g wet weight) was resuspended in'

20m! buffer B (10mM NaH:z}'04, 10 mM Tris-Cl, 8 M urea, pH 8.0) and kept the cells

on stirring for 60min at room temperature. Lysate was centrifuged at 10,000 rpm in .

5 Sorvall GSA rotar at room temperature for 30 min. The supernatant was removed and

analysed by SDS-P AGE.

The supernatant was bound to Ni-NTA Superflow matrix (~3ml packed

volume) th8t is pre washed with buffer B for neutralization of matrix and'kept on flip­

flop shaker for Ih at r~m temperature (R.T'). The mix was packed into a column and

10 flow-through was collected. The column was washed twice with 15m! buffer C (10mM

NaH:z}'04, 10 mM Tris-Cl, 8M urea, pH 6.3) and eluted four times with 3ml of buffer D

(lOmM NaH:z}'04, 10 mM Tris-Cl, 8 M urea, pH 5.9) and E (lOmM NaH2P04, 10 roM

Tris-Cl, 8 M urea, pH 4.5). All the fractions were analysed by SDS-PAGE and the pure

eluted fractions were pooled and flash-frozen for storage at -80DC with 50llg/ml

15 gentamycin. Purified proteins were dialysed against different urea concentration (8 - 0,

M) and different buffers with out urea (sodium acetate buffer pH 4 & 5, sodium.

phosphate buffer pH 11.5 & 12, glycine buffer pH 10& 11, sodium carbonate and

bicarbonate buffer pH 10.8 & 10).

Western blotting 20 Purified protein was run on. 15% SDS-PAGE, along with pre-stain~d marker

and transferred electrophoretically to nitrocellulose membrane. The membrane 'was

blocked with 1 % PVP made in IX PBS for 2h at RT. The membrane was washed three

times with lXPBS-T (IX PBS containing 0.1% Tween 20) and incubated with anti-His

mAbs at a dilution of 1: 2000 for 90min at RT. The membrane was washed three times

25 with IX PBS-Tan incubated with anti-mouse IgG-a1kaline p1iospha~se at a dilution of

1:5000 for 90min~at RT. The membrane was washed three times with IX PBS-Tand

incubated with substrate (5-brom0-4-ch1oro-3-indolyl phosphate with nitroblue '

tetrazolium) for 30 min at RT .

Immunoblotting 30 Purified protein was run with fused wells on 15% SDS-PAGE, along with p~e,-

stained marker and transferred electrophoretica1ly to nitrocellulose membrane. The

membrane was cut into small vertical strips and was blocked With IX PBS containing

2% Tween 20 for Ih at RT. The strips were washed thrice with IX PBS-T (IX PBS

containing 0.1% Tween 20) and incubated with dengue patients sera (Division of

WO 2005/014627 PCTIIN2004/000237 32

Virology, Defense Research and Development Establishment, Gwalior, INDIA) at.

I: 100 dilution, and for contro~ negative human sera was used with same conditions for

30min at RT. The strips were washed thrice with IX PBS-T and incubated with anti-'

human IgG-peroxidase at a dilution of 1:10,000 for 30min at RT. The strips were

5 washed thrice with IX PBS-T and incubated with substrate (3,3', 5,5'­

Tetramethylbenzidine) for 30 min at RT.

Results

Expression of /gG and /gM- rDME proteins B. Expression of /gG- rDME in pQE-60 vector

10 The protein IgG-rDME was expressed in-frame with the translation initiator

codon and the 6x His tag of pQE-60 vector at BamIll arid Bgill sites, the vector map is

shown in Figure SA. The predicted protem is of 244 aa, of which methionine and

glycine, and 6 histidines are coming from the vector sequence and molecular weight is

~25 kDa. The protein was expressed at ~120mg/ml, under the control of IPTG

15 inducible phage T7 promoter and the repressor is supplied by pREP-4 plasmid in the '

E.coli strain SG13009. Positive clones were screened by restriction digestion with

BamIll and Bglll enzymes. Selected positive clone was induced with ImM IPTG for 4h

at 37°C, and the expected ~25kDa protein was observed by SDS-PAGE analysis

(Fig.3B). The selected positive clone was used for further studies.

20 B. Expression of /gM- rDME in pMALc2x vector

The gene IgM-rDME was cloned in pQE-60 vector and the protein did not

express in the host. Then the gene was cloned in-frame with lY.1BP in pMALc2x vector

and expressed at ~60mg/ml (Fig. SA), under the control of the inducible tac promoter in

E.coli DH5a.. The expressed protein molecular weight is ~52kDa with S6aa of IgM-

25 rDME with his tag and the complete lY.1BP. Positive clones were screened by restriction

digestion with HamIll and HindlIII enzymes. Selected positive clone was induced with

0.4 roM IPTG for 4h at 37°C, and the expected ~52kDa protein was observed by ~DS­

PAGE analysis (Fig.SC). The selected positive clone was used for further studies.

To increase the epitope density ofIgM-rDME, two copies of the same gene was

30 cloned in pQE-60 and sub-cloned in-frame with lY.1BP in pMALc2x vector. The same

procedure was followed as above for single copy to express the double copy and. the

expected ~60kDa protein was expressed at ~30mg/ml and observed by SDS-PAGE

analysis (Fig. SD).

WO 2005/014627 peT 1IN2004/000237 33

Localization rDME proteins

To determine the solubility of proteins under native conditions (with the

absence of denaturing reagents), all the three clones culture (PQE60-IgG-rDME and

pMALc2x-IgM rDMEsingle copy and double copy clones) were induced with lrriM '

5 lPTG for 4 h. Culture pellet was resuspended in Tris-EDT A pH8.0 containing

lysozyme. Vortexed gently for lmin, and incubated at 37'C for lh. The lysate was

,centifuged at IS, 0008 for 5 min and the supernatant was transferred into' a fresh tu?e,

The pellet was resuspended in 0.2m! SDS-PAGE sample buffer and solubilized by

boiling for 10 min. Both pellets and supernatants were analysed by SDS-·PAGE (7).

10 In case of pQE-60-IgG protein was not found in supernatant fraction and observed in

the pellet. It clearly indicates that the protein is associated with peUet, so it is not

soluble under native condition. To OVercome this insolubility of the protein,' we

expressed the protein at low IPTG concentrations (0.005, 0.06, 0.03 and 0,02 mM) and

low tempera1ures (27 and 32°C). These did not help in the solubility of the protein. .

15 Then the protein was expressed in-frame with MBP in pMALc2x vector (Fig. 9). Here

expression levels are high but did not help in solubility.

For further experiments, the IgG-rDME protein, which was expressed .in pQE-

60 was used. Due to insolubility ofIgG-rDME protein, it was purified under denaturing.

condition in the presence of 8M urea.

20 In case of IgM-rDME, single copy and double copy clones, which wer~

expressed with MBP fusion, were Checked for their solubility under native Condition. ,

Single copy protein is slightly soluble under native condition, as it is shown in Figure

10, where single copy protein was observed both in supernatant and pellet. The same

procedure was repeated for'double copy, where the protein is associated with pel1~t, so

25 it is not soluble under native condition. Both single and double copy proteins were

. purified under dejlaturing condition in the presence of 8M urea, as urea do not interfere

in ELISA and immunoblots.

Purification ofrDMEs

Purification of rDMEs were done under denaturing conditions using 8M urea

30 using Ni-NTA Superf10w matrix. Induced culture was resuspended in buffer B of

pH7.5 and kept the cells on stirring for 60min at RT. The centrifuged lysate was bound

to using'Ni-NTA Superflow matrix and incubated at RT on flip-flop shaker for lh. The

mix was packed into a column and flow-through was collected. The column was

washed with buffer C of pH 6.3 to remove nonspecific bound protdns and eluted the·

WO 2005/014627 PCTIIN2004/000237 34

protein with of buffer D of pH 5.9 and followed by buffer E of pH 4.5. The fractions

were analysed by SDS-PAGE (Figure 11) and eluted fractions, which are pure enough,

were pooled and flash-frozen for storage at -80°C with 50Ilg/ml gentamycin for further

experiments.

5 Purified proteins were' subjected to electrophoresis on a denaturing gel and

electro-transferred on to a nitrocellulose membrane and subjected to western analysis.

The membrane was blocked with 1% PVP for 2h at RT. The membrane was washed

with IX PBS-T and incubated with anti~His mAbs for 90 min at RT. The membrane

was washed again and incubated with anti-mouse IgG-alkaline phosphatase for 90min

10 at RT. The membrane was washed and protein bands were visualized by incubating in

substrate (5-bromo-4-chloro-3-indolyl phosphate with nitroblue tetrazolium) for 30 min

atRT.

To check the solubility, purified proteins were dialysed against different urea

concentrations and different buffers of pH ranging from 4.0 to 12. The results (not

15 shown) indicate that IgG rDME protein is soluble in sodium acetate buffer at pH4.0 and

5.0 and the precipitation was least in comparison with other buffers tested and it need

minimum of 3M urea to keep soluble protein of2mg/ml.

As IgM single copy is quite soluble under native conditions, the protein is quite

soluble in an the above buffers and it need 3M urea to keep 5mg/ml.

20 Preliminary analysis of /gG- rDME with dengue sera

To check the reactivity of IgG-rDME to dengue patients sera, protein was

electro- transferred on to nitrocellulose membrane and reacted with dengue patients

sera for an hour at RT. Then the membrane was washed with IX PBST and incubated

with anti-human IgG-peroxidase for 3 o min at RT. After washing the membranewas

25 developed by incubating with substrate (3,3',5,5'- Tetramethylbenzidine) for 30 min at

RT. Control experiment was done with dengue negative human sera, the results clearly

indicates that the IgG-rDME is recognizing the antibodies in dengue patients sera.

Rapid strip test of M-rDME

Having demonstrated that recombinant multiepitope proteins of the present

30 invention can be used to identify dengue specific antibodies in the ELISA format, the

applicants listed the invention to a strip test format. The design of the test was simple,

It consisted of a piece of nitrocellulose' membrane and coated with M -rDME protein (a

thin streak) and dried at room temeperature, for 10min. The strip was then dipped in a

, drop of the dengue human positive ~erum sample which was mixed with gold

5

10

15

20

25

30

35

WO 2005/014627 PCTIIN2004/000237 35

conjugated M-rDME. A positive test was indicated by a pink line. A parallel control

experiment was canied out with the same conditions with negative human sera where

no pink line was seen.

WO 2005/014627 peT IIN2004/00023 7 36

Claims:

1. A recombinant dengue tnultiepitope (r-DME) for use in the diagnosis' and/or

detection of any or all of dengue speCific Immunoglobin M (IgM), comprising of

5 immunodominal epitope from nonstructural protein -1 (NS 1) of Dengue-virus type-

1, Dengue-virus type-2, Dengue-virus type-3 and Dengue-virus type-4, said

epitopes being linked with a predetermined number of glycine linkers, a?-d codon

optimized for expression in E. coli.

10 2. A recombinant dengue multiepitope as claimed in claim I having the following

schematic structure:

15 '----_D_l_N_S_l---lH'-_D_2_N_S_1-----lH'-_D_3_N_S_l--lW4 NSI I 3. A recombinant dengue multiepitope as claimed in claim 1 or 2 wherein said

20 multiepitope has a nucleotide sequence as shown Seq ID 1.

4. A recombinant dengue multiepitope as claimed in any preceding claim wherein

said multiepitope has a amino 'acid sequence as shown Seq ID 2.

25 5. A recombinant dengue multiepitope ,as claimed in any preceding claim wherein

said epitopes are linked with four glycine linkers.

6. A method for the manufacture of a recombinant dengue multiepitope (r-Dl\1E) for

use in the diagnosis and/or detection of any or all of dengue ,specific

30 Immunoglobin M (IgM), which comprises

35

(a) synthesising a gene comprising of immunodominal epitope from nonstructural

protein -1 (NS1) of Dengue-virus type-I, Dengue-virus type-2, Dengue-virus type-3

and Dengue-virus type-4, said epitopes being linked with a predetermined number

of glycine linkers; .

WO 2005/014627 PCTIIN2004/000237 37

(b) cloning said synthesised gene in-frame with maltose binding protein in

p~-c2X vector with MBP fusion and expressed it in E.coli (DHS-cx.).

(c) purifYing the protein so obtained under denaturing conditions.

5 7. A method as claimed in claim. 6 wherein said E.coli comprises strain DH5-cx..

8. A method as claimed in claim. 6 or 7 wherein said purification is carried out

under denaturing conditions by Ni-NTA affinity chromatography.

10 9. A method as claimed in claim. anyone of claims 6 to 8 wherein aid epitopes

linked with four glycine linkers.

10. Use of the recom~inant dengue multiepitope (r-D~) as claimed in anyone

of claims 1 to 5 in the diagnosis and/or detection of any or all of dengue·

15 specific Immunoglobin M .(ImG).

11. A recombinant dengue multiepitope (r-DME) for use in the diagnosis and/or

detection of any or all of dengue specific Immunoglobin G (IgG), comprising

of one epitope from core, one epitope from prM, eight epitopes from DEN-2,

20 envelope, four epitopes from DEN-4 NSI, one epitope from DEN-I NSI, one

epitope from DEN-2 NS 1 and one epitope from DEN-4 NS3, said being 'linked

with a predetermined number of glycine linkers, and codon optimized for

expression in E.coli.

25 12. A recombinant dengue multiepitope (r-DME) as claimed in claim II having.

the following schematic representation:

I D2e U D2pr U D2E! U ~2E2 U D2E3 ,II D2E4

30

D4NS .~ D4NS H D4NS

35

WO 2005/014627 PCT/IN2004/000237 38

13. A recombinant dengue multiepitope (r-DME) as claimed in claims 11 or 12

wherein said multiepitope has a nucleotide sequence as shown in Seq ID 3.

14. A recombinant dengue multiepitope (r-DME) as claimed in anyone 6f claims

5 lIto 13 wherein said multiepitope has an amino acide sequence as shown in

SeqID 4.

10

15. A rec.ombinant dengue multiepitope (r-DME) as claimed in anyone of claims 11

to 14 said epitopes are linked with three glycine linkers.

16. A method for the manufacture of a recombinant dengue multiepitope (r-DME) for

use in the diagnosis and/or detection of any or all of dengue specific Immunoglobin

G (IgG), which comprise:

(a) synthesising a gene ~omprising of one epitope from core, one epitope from· , .

15 prM, eight epitopes from DEN-2 envelope, four epitopes from DEN-4 NSl,

20

one epitope from DEN-l NSl, one epitope from DEN-2 NSI and one epitope

from DEN-4 NS3, linking said epitopes with glycine linkers;

(b) cloning said synthesised gene in pQE-60 and. expressed it in E. coli.

(c) purifying the protein so obtained under denaturing conditions.

17. A method as claimed in claim ·16 wherein said epiotpes are linked with three

glycine linkers.

18. A method as claimed in claim 16 or 17 wherein said E.coli comprises strain

25 SG13009.

19. A method as claimed in anyone of claims 16 to 18 wherein said protein is

purified under denaturing conditions by ni-NT A affinity chromatography.

30 20. Use of the recombinant dengue multiepitope (r-DME) as claimed in anyone of

claims 11 to 15 in the diagnosis and/or detection of any or all of dengue

specific Inimunoglobin G (IgG).

WO 2005/014627 PCT1IN2004/000237 39

21. A method for detecting IgM anti-dengue anitbodies in a human sera which

comrpises subjecting human sera to ELISA in the presence of a recombinant

dengue mUltiepitope (r-DME) as claimed in anyone of claims 1 to 5.

5 22. A method for detecting IgG anti-dengue anitbodies in a human sera· which

comrpises subjecting human sera to ELISA in the presence of a recombinant

dengue multiepitope (r-DME) as claimed in anyone of claims 11 to 15.

23. A metht>d for detecting IgM anti-dengue anitbodies in a human sera wh~ch

10 comrpises subjecting said lulman sera to a rapid strip test using a nitoceHulose

membrane, in the prescence of a a recombinant dengue multiepitope (r-DME)

15

. . . as claimed in a1I.y one of claims 1 to 5, the appe~ce of a pink line indicating

that the sera is dengue positive and the absence of a pink line indicating that

the sera is degue negative.

24. A method for detecting IgG anti-dengue anitbodies in a human sera Which

comrpises subjecting said human sera to a rapid strip test using a nitocelh.ilose

membrane, in the prescence of a a recombinant dengue multiepitope (r-DME)

as claimed in anyone of claims 11 to 15, the appearance of a pink line

20 indicating that the sera is dengue positive and the absence 9f a pink line·

. indicating that the sera is degue negative.

25

30

WO 2005/014627 peT IIN2004/00023 7 I112

Figure 1

WO 2005/014627 peT lIN 2004/000237 2/12

A

NC ns2a ns4a NC

B 5'

Figure 2

WO 2005/014627 peT 1IN2004/00023 7 3/12

A

Bgfll

BamHI BellI

B

c

Figure 3

WO 2005/014627 PCT 1IN2004/00023 7 4112

L---_D_l_N_S_l_," --JHL-_D_2_" N_S_l----'H'-_D_3_N_S_l_" --1H'--_D_4_N_S_l....--..

Figure 3D

D2C 1--4 D2prM H D2 El W D2 E2 W D2 E3

Figure 3E

WO 2005/014627

BamHl' A

Bglll ... .

B

c

Figure 4

peT IIN2004/000237 5/12

rDME-M + rDME-M

, .. :. : :~:::.:: ::::~ . ~~~~ ': .: .

::::::::04W:: :'::::? ' ;I~. :~:::.::. rrn8 .' : : . : : : : : : : : : . : : : : : : . : : : : : : : : . : : : : : : : : : . : : .: ' . terminator . .. . . . ...... ......... . . . . .. .. . .. " ..

P~~/ } ~M~-c2XU· > Ampr .. . ,,': .... '.: .: : .: .... : : :: :.:: . : . ::':', : . : - . . . . . :~BcI :~ :: : -: . : : : : : : : : : : : : : ; . : : : :: : .. :: . : : . : .';:-' . . : .. . . . . .. ... . . . . .. . .. . .. . . . . , . . . . . . .... .. . .. " ... . . ...... .... . " .

: : :;:;::.;'; . ;;:: "'.~': : : :~:~:'o~i' : . : PI;'lR~2;!q~I · : :.: . :: · . .. ...

WO 2005/014627 PCTI1N2004/000237 6/12

A

B

Figure 5

100

1

!2 ~llQ· !l I ~ 11

1\ 100 200 \ , \ \ \\ ,

\\

E NSI

\\ \\

\ \

\

14 I -leOOH !

300 j 354 !

/ ! ! i

i NS2A ;

·NS3

B

500

200 300

\\ , 400 '\500

1--1111 I - 1 --I - -- JeOOH

1 2 345678

NS4A

-15

NS4B

623 COOH

NS5

Fig.6 . Regions of linear immunodominant epitopes of dengue proteins. Linear immunodominant epitopes numbered as 1-8 from envelope, 9-14 from NS1 and 15 from NS3 were included in IgG specific recombinant dengue multiepitope protein, and In IgM specific recombinant dengue multiepitope protein, epitope numbered as 10 from DEN-2 and the corresponding region of other three dengue s~rotypes were included . .

A B Bam HI

67.0·

6xHis Tag 43.0 •

ToTT 30.0·

20.0· Ori

14.0 •

Fig. 7. (A) Map of plasmid pQE-60-IgG-rDME. This is the plasmid which expresses IgG-rDME (rDMErepresents IgG-rDME in theabove plasmid map). Cloned IgG-rDME in-framw with initiator codon and 6xHis·Tagat Bam HI & BgI IT sites of pQE-60 vector. Abbreviations· of pT5, To TT, Ori, AropR are phage T5 promoter, lambda to transcriptional terminator, replication origin Sequence, Ampicillin marker. (B) SDA-PAGE analysis IgG-rDME protein expression. This is the coomassie stained protein gel picture of E.coli lysates of IPTG mduced (lane3) and uninduced Oane 2) and protein low molecular weight marker (lane 1, their sizes are shown in kDa at the left side of thegel) .·The arrow on right side of the gel picture indicates the position of the IgG':rDME-protein.

~ 0 N 0 0 UI as .... ~ ~ N -..I

~ N

1 2 3 1 2 3 1 2 3 ... ,lacI

94.0

~.~ 67.0 4

• • •

43.0 • Ptac

30.0 • •

20.0 • •

14~0 •

A B c D

. Fig. 8 . (A) Map of plasmid . [PMALc2x--rDME. This is the plasmid which expresses IgG-rnME, IgM-rDME single copy and double copy (rDMErepresents position ofIgG-rDME, IgM-rDME single copy and double copy in theabove plasmid map). Cloned­rDME genes in-frame with Mal E gene(which expresses the maltose binding protein) at Bam ill & hind III sites of pMALc2x vector. Abbreviations of Ptac, TT, AmpR, Ori and lac I are phage tac promoter, transcriptional terminator, Ampicillin marker, replication origin Sequence, lac repressor gene. (B) SDA-PAGE analysis IgG-rDME protein expression. This is the coomassie stained protein gel picture of E.coli lysates of IPTG induced ( lane 2) and uninduced (lane 3) and protein low molecular weight marker (lane 1, their sizes are shown in kDa at the left side of the gel). The arrow on right side of the gel picture indicates the position of the IgG-rDME protein.(C) SDA-PAGE analysis single copy IgM-rDME protein expression. This is the coomassie stained protein gel picture of E.coli lysates of IPTG induced ( lane 2) and uninduced (lane 3) and protein low molecular weight marker (lane. 1, theii sizes are shown in kDa at the left side of the gel). The arrow on right side of the gel picture indicates the position of the IgG-rDME protein. (D) SDA-PAGE analysis double copy IgM-rDME protein expression. This is the coomassie stained protein gel picture of E.coli iysate-s of IPTG induced (lane 3) and uninduced (lane 2) and protein low molecular weight marker (lane 1, their sizes are shown in kDa at the left side of the gel). The arrow on right side of the gel picture indicates the position of the IgG-rDME protein. .

~ 0 .... =-=-U1

= ~

"'" e'\ .... --.I

94.0e 67.0·

43.0·

30.0 •

20.0·

14.0.

1 2 3

Fig.l G Localization of single copy IgM-rOME protein · expression. Induced culture was lysed under native conditions and analysed the supernatant Oane 2) and pellet (lane 3) by SOS-PAGE and stained with coomassie dye. Protein low molecular weight marker ( lane 1) their sizes in kDa are shown at the left side of the gel picture. The arrow on right side of the gel picture indicates the position of the single copy IgM-rOME protein.

123456789 10

123456789 12345678

Fig: .ll·. Purification of rDME proteins by Ni-NTA affinity chromatography. Induced culture pellet was solubilized in 8M urea containg buffer and purified by using Ni-NTA affinity matrix an the fractions were analyzed by SDA-PAGE followed by coomassie staining. (A) Purification fractions of IgG-rDME with out :MBP, flow through (lane 2), wash (lane 3), elution fractions at pH5.9 (lane 4,5 and 6) and elution fractions at pH4.5 (lanes 7, 8,9 and 10) were analyzed by SDS-PAGE. (B)

. Purification fractions of single copy IgM-rDME with MBP fusion, wash (lane 2 and 3), elution fractions at pH5.9 (lane 4, and 5) and elution fractions at pH4.5 (lanes 6,7, 8, and 9) were analyzed by SDS-PAGE. (D) Purification fractions of double copy IgM-rDME with MBP fusion, flow through (lane 2), wash (lane 3), elution fractions at pH5.9 ( lane 4, and 5 ) and elution fractions at pH4.5 (lanes 6, 7, and 8) were analyzed by SDS-PAGE. In all the gel pictures, protein low molecular ~eight marker (lane I) with their sizes are shown in kDa at the left side of the gel. The arrow ' on right side of the gel picture indicates the positiop of the DME protein.

-----t-.I

113.0· 92.0 •

52.3·

35.0 •

28.9.

21.0.

1 2 3 4 5 6

Fig. rL Western and Iinmunoblot analysis of rDME proteins. (A) Western blot of rDME proteins with anti His mAbs. Purified proteins by Ni-NTA were subjected to electrophoresis on a denaturing gel and electro-transferred on to a nitrocellulose membrane and subjected to western analysis ofIgG-rDME (lane 2), single copy ofIgM-rDME (lane 3) and double copy of IgM-rDME (lane 4) proteins with anti His mAbs( Only IgG-rDME was subjected to immunoblot analysis ·with dengue patients sera (lane 6) and as a control it is reacted with· human negative· sera (lane 5). Pre-stained protein molecular weight marker (lanel) with their.sizes are shown in kDa at the left side of the gel. The arrow on right side 'of the gel picture in~ca,tes the position of the DME protein. .

WO 2005/014627 PCTIIN2004/000237 1/9

SEQUENCE LISTING

<110} International Centre for Genetic Engineering and Biotechnology Director General. Defence Research & Deuelopment Organisation Khanna. Nauin Ananda Rao, Rauulapalli Jana. Asha Mukul

<120} RECOMBINANT DENGUE MULTI EPITOPE PROTEINS AS DIAGNOSTIC INTERMEDIATES

<130} ICGEB-DRDO

<160} 19

<170} PatentIn uersion 3.1

<210} 1 <211} 2110 <212} DNA <213} Artificial Sequence

<220} <223} This DNA sequence was artificially synthesised from immunodominal

epitopes from Dengue Uiruses. type-1, type-2, type-3 and type-II

<400} 1 ggatccgata gcggctgcgt gattaactgg aaaggccgtg aactgaaatg cggcggtggc 60

ggtgatagcg gctgcgtggt gagctggaaa aacaaagaac tgaaatgcgg tggcggtggc 120

gatatgggct gcgtgattaa ctggaaagg~ aaagaactga aatgcggcgg tggcggtgat 180

atgggctgcg tggtgagctg gagcggcaaa gaactgaaat gcggtggcgg tggcagatct 240

<210} 2 <211} 80 <212} PRT <213} Artificial sequence

<220} <223}

This protein was artificially synthesised from ill\ll\unodominal epi topes from Dengue uirus type-1. type-2. type-3 and type-4

<400} 2

Gly Ser Asp Ser Gly Cys Ual lIe Asn Trp Lys Gly Arg Glu Leu Lys 1 5 10 15

Cys Gly Gly Gly Gly Asp Ser Gly Cys Ual Ual Ser Trp Lys Asn Lys 20 25 30

Glu Leu Lys Cys Gly Gly Gly Gly Asp Met Gly Cys Val lIe Asn Trp 35 40 45

WO 2005/014627 PCTIIN2004/0002J7 2/9

Lys &ly Lys ~lu Leu Lys Cys Gly Gly Gly Gly Asp Net ely Cys Ual 50 55 66

Ual Ser Trp Ser Gly Lys Glu Leu Lys Cys Gly Gly Gly Gly Rrg Ser 65 70. 15 80

<210> 3 <211> 843 <212> DHR <213> Rrtificial sequence

<220> <223> This nucleotide was artificially synthesised frolll iIImunod'olllinal'

epitopes fro" Dengue virus type-1, type.-2, type-3 and type-II

<.!l00> 3 gtgctgcg~g gctttcgtaa agaaattggc ggcggcggcc tgaccacccg taacggcgaa 60

ccgcatatga ttgtgatgcg tcaggaaaaa ggcaaatccc tgctgtttaa aaccggcgat 120

. ggcgtgggcg gcggcggatc cgaaaccctg gtgaccttta aaaacccgca tgcgaaaaaa 180

'caggatgtgg tggtgctggg cagcggcggc ggcaacctgc tgtttaccgg cggcggcccg 240

tttggcgata gctatattat tattggcgtg gaaggcggcg gccagctgaa actgaactgg 300 #

tttaaaaaag gcagcagcgg cggcggcacc gcgtgggatt ttggcagcct gggcggcgtg 360

tttaccagca ttggcggcgg cgtgattatt acctgg~ttg gcggcggcag caccagcctg

agcgtgggcg gcggcgtgac cctgtatctg ggcgcgggcg gcggcgaaca taaatatagc 480

tggaaaagcg,gcggcggcga tagcggctgc gtggtgagct ggaaaaacaa agaactgaaa SilO.

tgcggcggcg gcaaatttca gccggaaagc ccggcgcgtc tggcgagcgc gattctgaac 6·00

gcgggcggcg gcctgaaata tagctggaaa acctggggca aagcgaaagg cggcggcttt 660

ctgattgatg gcccggatac cagcgaatgc ccgaacgaac gtcgtgcggg cggcggctgg 720

t,atggcatgg aaattcgtcc gctgagcgaa aaagaagaaa acatggtggg cggcggcatt 1189

ctggaagaaa acatggaagt gga~atttgg acccgtgaag gcgaaaaaaa gaaactgaga 840

tct 843

<210> .. <211> 281 <212> PRT <213> Artificial sequence

(220) <223> This protein vas artificially synthesised fro A iRAUnodoNinal epit

opes fro~ Dengue virus type-1. type-2, type-3 and type-Il

<1l00> 4

WO 2005/014627 PCTIIN2004/000237 3/9

Ual Leu Arg Gly Phe Arg Lys Glu Ile Gly Gly Gly Gly Leu Thr Thr 1 5 10 15

Arg Asn Giy Glu Pt'o His Met Ile Ual Met At'g GIn Glu Lys Gly Lys 20 25 30

Ser Leu Leu Phe Lys Tht' Gly Asp Gly Ual Gly Gly Gly Gly Ser Glu 35 40 45

Thr Leu Ual Thr Phe Lys Asn Pro His Ala Lys Lys GIn Asp Ual Ual 50 55 60

Ual Leu Gly Ser Gly Gly Gly Asn Leu Leu Phe Thr Gly Gly Gly Pro 65 70 75 80

Phe Gly Asp Ser Tyr Ile Ile Ile Gly Ual Glu Gly Gly Gly GIn Leu 85 .90 95

Lys Leu Asn Tt'p Phe Lys Lys Gly Ser Ser ely Gly ely Thr Ala Trp 100 105 110

Asp Phe Gly Ser leu ely Gly Ual Phe Thr Ser lIe ely ely Gly Ual 115 120 125

lIe lIe Thr Trp lIe Gly Gly ely Ser Tht' Set' leu Ser Ual Gly Gly 130 135 140

Gly Ual Tht' Leu Tyr Leu Gly Ala ely Gly Gly Glu His Lys Tyt' Set' 145 150 155 160

Trp lys Set' Gly Gly Gly Asp Ser Gly Cys Ual Ual Ser Tt'p lys Asn 165 170 175

lys Glu Leu Lys Cys ely Gly ely lys Phe GIn Pt'o Glu Ser Pt'o Ala . 180 185 190

Arg leu Ala Ser Ala lIe leu Asn Ala Gly ely Gly leu lys Tyr Ser 195 200 205

Trp Lys Thr Trp ely Lys Ala lys Gly Gly ely Phe Leu lIe Asp ely 210 215 220

Pro Asp Thr Ser Glu Cys Pro Asn Glu Arg Arg Ala Gly Gly Gly Trp 225 230 235 240

WO 2005/014627 PCTIIN2004/000237 4/9

Jyr Gly Met GIu" lIe Arg Pro Leu Ser Glu Lys Glu Giu Asn Met Ual 21&5 256 255

Gly Gly Gly lIe Leu Giu Glu Asn ~t Glu Ual Glu lIe Trp Thr Arg 260 265 270

Glu Gly Giu Lys Lys Lys Leu Arg Ser 275 280

<210) 5 <211) 21 <212> PRJ <213> Artificial sequence

<220) <223) This peptide was artificially synthesised

<1&00> 5

Glu Thr Leu Ual Thr Phe Lys Asn Pro His Ala Lys Lys GIn Asp Ual 1 5 10 15

Val Val Leu Gly Ser 20

<210) 6 <211> 6 <212> PRT <213> Artificial sequence

<220> <223> This protein .as artificially synthesised

<400> 6

Asn Leu leu Phe Thr Gly 1 5

<210> 7 <211) 12 <212) PRJ <213> Rrtificia~ sequence

<220> <223> This protein .as artificially synthesised

<400) 7

Pro Phe Gly Asp Ser Tyr lIe lIe lIe Gly Val Glu 1 5 10

WO 2005/014627

<210) 8 <211) 12 <212) PRT <213) Artificial sequence

<220>

5/9

<223) Thsi protein was artificially synthesised

<400) 8

GIn Leu Lys Leu Asn Trp Phe Lys Lys Gly Ser Ser 1 5 10

<210) 9 <211) 16 <212) PRT <213) Artificial sequence

<220) <223) This protein was artificially synthesised

<400) 9

PCTIIN2004/000237

Thr Ala Trp Asp Phe Gly Ser Leu Gly Gly Ual Phe Thr Ser lIe Cly 1 5 10 15

<210) 10 <211) 7 (212) PRT <213) Artificial sequence

<220) <223) This sequence was artificially synthesised

<400) 10

Ual lIe lIe Thr Trp lIe Cly 1 5

(210) 11 <211) 6 <212) PRT <213) Artificial Sequence

<220) <223) This protein was artificially synthesised

<400) 11

Ser Thr Ser Leu Ser Ual 1 5

(210) 12 <211) 7

WO 2005/014627 6/9

(212) PRr (213) Rrtificial Sequence

(220) (223) This protein was artificially synthesised

(400) 12

Ual Thr Leu Tyr Leu Gly Ala 1 5

<210> 13 (211) 8 (212) PRT (213) Rrtificial ~equence

<220> <223> This protein was artificially synthesised

<400> 13

Glu His Lys Tyr Ser Trp Lys Ser 1 5

<210> ill <211> 15 <212) PRT <213} Artificial sequence

<220} <223} This protein was artificially ~ynthesised

<400} ill

Rsp Ser Gly Cys Ual Ual Ser Trp Lys Asn Lys Glu Leu Lys Cys 1 5 16 15

<210> 15 <211} 17 <212> PRT (213) Artificial sequence

<220) <223> This protein was artificially synthesised

<400> 15

PCTIIN2004/000237

Lys Phe Gln Pro Glu Ser Pro Rla Rrg Leu Ala Ser Ala lIe Leu Asn 1 5 10 15

Rla

(210) 16

WO 2005/014627

<211) 12 <212) PRT <213) A~tificial sequence

<220)

7/9

<223) This peptide was a~tificially synthesised

<400) 16

Leu Lys Ty~ Se~ T~p Lys Th~ T~p Gly Lys Ala Lys 1 5 10

<210) 17 <211) 17 <212;1 PRY <213) A~tificial sequence

<220) <223) This p~otein was a~tificially synthesised

<400) 17

peT 1IN2004/00023 7

Phe Leu lIe Asp Gly P~o Asp Yh~ Se~ Glu Cys P~o Asn Glu A~g A~g 1 5 10 15

Ala

<210) 18 <211) 17 <212) PRY <213) A~tificial sequence

<220) <223) This p~otein was a~tificially synthesised

<400) 18

Y~p Ty~ Gly Met Glu lIe A~g P~o Leu Se~ Glu Lys Glu Glu Asn Met 1 5 10 15

Ual

<210) 19 <211) 20 <212) PRY <213) A~tificial sequence

<220) <223) Yhis protein was artificially synthesised

<400) 19

wo 2005/014627 PCT1IN2004/000237

8/9

He LeuGlu Glu flsn Met Glu Ual Glu Ile Typ Thy fir!! Glu Gly Glu 1 5 16 15

tys Lys Lys teu 20

WO 2005/014627 9/9

lIe Leu Glu Glu Asn Met Glu Ual Glu lIe T~p Th~ A~g Glu Gly Glu 1 5 10 15

Lys Lys Lys Leu 20

PCTIIN2004/000237

INTERNATIONAL SEARCH REPORT lint - Application No

PCT/IN2004/000237 A. CLASSlFlCA nON OF ~BJECT MA ncR IPC 7 C07K14 18 GOIN33/569

According 10 In1emalional Patent CIassIIlcaIIon (!PC) 01' to both natlOnaI class!fJCa1lon and !PC

B. AELDS SEARCHED Milimum documentalion searched (claasIIi::aIIon sysIam fobved by classlllcation symbols)

IPC 7 C07K GOIN

Documentation searched other than mII*num ~tlon to the extent that such documents _ incIUCIGd In lhe IiekS searched

Emtronrc data base consulted duri1g the InlIIrmtional saarch (n""", of data base and, Where paclia:I, search "'""" used)

EPO-Internal, Sequence Search, WPI Data, BIOSIS, EMBASE

C. DOCUMENTS CONSIDERED TO BE RELEVANT

Calegc)lY' C.alion of doaJmenl. wIIh In<Ica1lon, where appropriate, of the relevanl ~ Relevant to claim No.

A US 6 749 857 81 (COLLER BETH-ANN S ET AL) 15 June 2004 (2004-06-15)

---A CUZZUBBO ANDREA J ET AL: "Use of

recombinant envelope proteins for serological diagnosis of dengue virus infection in an immunochromatographic assay" CLINICAL AND DIAGNOSTIC LABORATORY IMMUNOLOGY, vol. 8, no. 6, November 2001 (2001-11), pages 1150-1155, XP002312032 ISSN: 1071-412X

-----/--

[]] Further documents Bra listed in III" "'*InuaIIon of box C. [] Pat8111 family IIMI'IIbcIrs am IiGIBd In annex.

• Special categories of clad documen18 : '1" _ doaInInt publl!tled !Iller the IntemaIionaI fllIng dale

'A' document del!nilg the general SIIlIe of tile! art Wh!dJ Is not 01' poIorIy d:!ta and IlOl In oonIlIcI WiIIl the l!pJIiII3ion but Geld 10 undelll\ancl the pri1cIpIe (l'theory underlying the considered to be of particular reIeYmce in-*'"

'E" eerier document but p!bIIshed on (I' I1118r!he international fling date 'X' documant of particular relevance; lIIe clai'ned invention

cannot be con!!ldered IKl\IeI or cannot ba considered 10 'L' document WhIch may throw doobts on p!IorIty claim(s) or irwoIv9 an inventMI sIep when the doaJment is taken alone

which Is eKed 10 establish the pu~ OllIe of another 'Y' 00cumIInt of partIcUlar raIaYance; the claimed invention citation or other special reason (as specIIed) cam!OI be considered to ImIotw an Inwntive step when the '0' document r9lerrtng to an oral dIoIdoEmm. lOSe, exhiblion or c:IDctmwd Is c:ombinad wIIb one or more other st.Ich_

other means .-lis, such cornblnol!ion belng obIIIous 10" pen!Or1 sIcIIIed 'po document published prior 10 the IrMmIiIIaeaI ftl!ng da'" but "'the'"

later than the prlorlly dale claimed .&. doctJrnoIIt membQr althe same petenl family

Date of the actual corf1llellon of the InIemdIonaI saarcll oa 0/ "",ling of the btamatlonal search raport

27 December 2004 07/01/2005 Name and mallng addn!ss of the !SA AuthorizIId ollicer

European Patent omce, p.8. 5818 Patentlaan 2 fII..-2280HV~

Tel (+31-70) ~o--ro4O, Tx. 31 651 epo 01, Fotaki, M Fax: (+31-70) 340-31)16

Form PeT JISI\I2'IO (second __ 1 (.htn:Jary 2!lO4)

INTERNATIONAL SEARCH REPORT

C.(Conlinuatlon) DOCUMENTS CONSIDERED TO BE RELEVANT

Category· Cnatlon of documen1, WIth Indicallon, where appropriate, of the relevant passages

A FONSECA B A L ET AL: "RECOMBINANT VACCINIA VIRUSES CO-EXPRESSING DENGUE-1 GLYCOPROTEINS PRM AND E INDUCE NEUTRALIZING ANTIBODIES IN MICE" VACCINE, BUTTERWORTH SCIENTIFIC. GUILDFORD, GB, vol. 12, no. 3, 1994, pages 279-285, XP009018695 ISSN: 0264-410X

A US 5 824 506 A (CHAN LILY ET AL) 20 October 1998 (1998-10-20)

A WO 99/09414 A (DEUBEL VINCENT ; PASTEUR INSTITUT (FR) 25 February 1999 (1999-02-25)

A US 6 074 817 A (LANDINI MARIA P ET AL) 13 June 2000 (2000-06-13)

Form PCTllSAl210 (contmuatlon of .econd sheet) (January 2004)

lint( ,lanai Application No

PCT/IN2004/000237

Relevant 10 claim No.

INTERNATIONAL SEARCH REPORT 11m, Ional Application No II1form.aon on patent IsmAy members

PCT/IN2004/000237 Patent docLment I Publication I Pa1Btlt family I PublICatIOn

cited In search report date m8:nber(S) date

US 6749857 Bl 15-06-2004 US 2003175304 Al 18-09-2003 AU 752191 B2 12-09-2002 AU 8590598 A 22-02-1999 BR 9815551 A 31-10-2000 CA 2298538 Al 11-02-1999 EP 1005363 A2 07-06-2000 JP 2001511459 T 14-08-2001 WO 9906068 A2 11-02-1999

---------------US 5824506 A 20-10-1998 NONE --------------- ----------WO 9909414 A 25-02-1999 FR 2767324 Al 19-02-1999

AU 8735798 A 08-03-1999 . BR 9811161 A . 25""'07-2000

WO 9909414 Al 25-02-1999 --------------US 6074817 A 13-06-2000 IT T0940543 Al 02-01-1996

AT 275201 T 15-09-2004 AU 2418895 A 25-01-1996 CA 2194135 Al 18-01-1996 DE 69533442 01 07-10-2004 EP 0769057 Al 23-04-1997 WO 9601321 Al 18-01-1996 JP 10502253 T 03-03-1998

-----

INTERNATIONAL SEARCH REPORT national application No.

PCT/IN2004/000237

Box No.1 Nucleotide and/or amino acid sequence(s) (Continuation of Item 1.b of the first sheet)

I. With regard to any nueleoMe and/or amino acid sequence disclosed In the international application and necessary to the claimed invention, the international search was carried out on the basis of:

2.

a. type of material o a seQUence listing

D table(s) related to the sequence listing

b. format of material

III In written fOrmat

III In computer readable form

c. time of flling/fumishlng o contained In the International application as filed o filed together with the International application In computer readable fOrm

III furnished subsequently to this Authority for the purpose of search

In addiHon, In the case that mora than one version or copy of a sequence listing and/or table relaHng thereto has been flied or fUrnished, the required statements that the Information In the subsequent or additional copies Is Identical to that In the application as filed or does not go beyond the application as flied, as appropriate, were furnished.

3. Additional comments;

Form PCTIISA/210 (continuation of first Sheet( I)) (January 2004)

I Available online at www.sciencedirect.com

8CIENCE@~IRECT. Virus Research

ELSEVIER Virus Research xxx (2005) xxx-xxx www.elsevier.comllocate/virusres

The identification of immunodominant linear epitopes of dengue type 2 virus capsid and NS4a proteins using pin-bound peptides Ravulapalli AnandaRao, Sathyamangalam Swaminathan, Navin Khanna *

RGP Laboratory, International Centre Jor Genetic Engineering and Biotechnology, PO Box 10504, Aruna AmJ Ali Marg, New Delhi 110067, India

Received 6 December 2004; received in revised form 8 March 2005; accepted 8 March 2005

Abstract

We have used multi-pin peptide synthesis strategy to identify B-cell epitopes on two sm!!ll dengue virus proteins, capsid and NS4a. We 10 have identified several linear, immunodominant epitopes on both these proteins. Almost all these epitopes mapped to regions predicted to be 11 hydrophilic based on Kyte and Doolittle profiles. Of the capsid epitopes identified in this study, the most immunogenic ones mapped to the 12 C-terminal a4 helix, which lies on the solvent-exposed surface of the capsid dimer. The capsid epitopes were dengue-specific in that they 13 could recognize antibodies in dengue virus-, but not yellow fever virus (YFV)- or Japanese encephalitis virus (JEV)-immune sera. This study 14 has demonstrated the presence of anti-NS4a antibodies in dengue-patient sera definitively, for the first time, using authentic NS4a-derived 15 pin-bound peptides as capture antigens. All the NS4a epitopes mapped to the amino-terminal third of the NS4a molecule. Our study suggests 16 that the immunodominant epitopes of these two dengue proteins might have the potential to be used as a part of a recombinant mUlti-epitope 17 protein containing carefully chosen E and NS 1 epitopes for the detection of dengue infections with a high degree of sensitivity and specificity. 18 © 2005 Published by Elsevier B.Y.

19 Keywords: Dengue virus; Capsid; NS4a; Multi-pin peptide synthesis; Immunodominant'epitopes

20

21 1. Introduction

Dengue infecti ons generall y occur in areas where other ft a­~,23 viviral infections (such as Japanese encephalitis and yellow

24 fever) are also common. While preventive vaccines are avail­able for these other ftaviviral infections (reviewed in Monath, 1999; Tsai et aI., 1999), there is none available for the pre­vention of dengue (Lai and Monath, 2003; Saluzzo, 2003). Coupled with the lack of a specific antiviral therapy to treat dengue infections, early and accurate diagnosis of dengue in-

22

25

26

27

28

29

30 fection is of paramount importance in cJinic.al management of dengue patients (WHO, 1997). Increasingly, dengue diag­nosis has begun to depend on the detection of dengu~'~specific serum antibodies (Groen et aI., 2000; Cuzzubbo etal., 2001). Commercial ELISA kits that use whole. dengue virus prepa­rations, obtained either from infected cell cultures or animal

31

32

33

34

35

* Corresponding author. Tel.: +91 11 26l77357x2?2; fax: +91 11 26162316. ,1)'1

E-mail address:navin@icgeb.res.in (N~Khanmi). /' , "-...":::',/

0168-1702/$ - see front matter © 2005 Published by Elsevier B. V. doi: 10.1016/j.virusres.2005.03.022

tissues, as antigens, can lead to misdiagnosis due to the cross- 36

reactivity of other non-dengue ftavivirus-specific antibodies 37

towards dengue antigens. Replacing the whole viral antigens 38

will eliminate the cross-reactivity and help in the develop- 39

ment of improved diagnostic assays in terms of sensitivity 40

and specificity. One approach to address this would be to 41

detect dengue infections through the use of multiple dengue 42

virus-specific peptides to capture dengue-specific antibodies 43

from patient sera. 44

Dengue viruses encode and express three structural (cap- 45

sid, C; premembrane prM and envelope E) and seven non- 46

structural (NS) proteins (NS 1, 2a, 2b, 3, 4a, 4b, and 5) 47

(Lindenbach and Rice, 2001). Of these ten proteins, anti- 48

bodies to C, prM, E, NSI and NS3 have been detected in sera 49

of dengue-infected patients (Churdboonchart et aI., 1991; 50

Se-Thoe et aI., 1999; Valdes et aI., 2000; Cardosa et aI., 2002). 51

A dengue viral protein with an electrophoretic mobility sim- 52

ilar to that of NS4a has been observed in western blots using 53

dengue-patient sera, suggesting that dengue infection elicits 54

anti-NS4a antibodies as well (Churdboonchart et aI., 1991; 55

VIRUS 94016 1-9

-------------------------------

2 R. AnandaRao et at.! Virus Research xu (2005) xu-xu

56 Se-Thoe et aI., 1999). It has been reported that the preva-57 lence of these anti-NS4a antibodies which is "'5% in pri-58 mary infections, increases to "'50% in secondary infections 59 (Se-Thoe et aI., 1999). Obviously, peptides corresponding to 60 the antigenic determinants of the proteins that elicit anti bod-61 ies in dengue-infected patients would be useful as diagnostic 62 reagents. While the epitopes on the major antigens such as E 63 (A askov et aI., 1989; Innis et aI., 1989; Roehrig et aI., 1990; 64 Megret et aI., 1992; Jianmin et aI., 1995) and NS1 (Huang 65 et aI., 1999; Garcia et aI., 1997; Falconar et aI., 1994; Wu 66 et aI., 2001) have been mapped extensively, not much infor-67 mation is available in the literature regarding the epitopes of 68 the remaining proteins. In this study, we have focused on two 69 small proteins, the capsid protein (consisting of 100 amino 70 acid (aa) residues) and NS4a (ISO aa residues) for several 71 reasons. One, both proteins are reported to elicit antibodies 72 in dengue-infected patients; therefore, peptides carrying im-73 munodominant epitopes of these proteins could be potentially 74 useful as diagnostic agents. Two, not much is known regard-75 ing the antigenic structure of these proteins. While there is a 76 report in the literature identifying an epitope on the dengue n capsid recognized by a panel of six monoclonal antibodies 78 (Bulich and Aaskov, 1992), there is virtually no information 79 regarding NS4a epitopes. Finally, both these proteins are rel-80 atively small in size and amenable to rapid epitope mapping 81 using the multi-pin peptide synthesis approach (Rodda and 82 Tribbick, 1996). In this work, we present data on the linear 83 immunodominant epitopes of dengue type 2 virus (DEN-2) 84 capsid and NS4a proteins. We also show that these epitopes 85 can be used to develop recombinant multi-epitope proteins 86 with enhanced sensitivity of anti-dengue antibody detection 87 in patient serum.

68 2. Materials and methods

89 2.1. Computer analysis

90 The predicted protein sequences of DEN-2 capsid (DEN-91 2 NGC, Genbank AF038403) and NS4a (DEN-2 Jamaica, 92 GenBank M20SS8) were analyzed using MacVector soft-93 ware. Putative hydrophobic and hydrophilic regions along 94 the length of these molecules were predicted using the Kyte 95 and Doolittle method.

96 2.2. Peptide synthesis

97 A total of 90 peptides were synthesized in this study, of 98 which 37 were derived from the capsid and the rest from 99 NS4a. For convenience, these peptides have been designated

100 serial numbers with the prefix 'c' to denote capsid peptides and the prefix 'n' to denote NS4a peptides. All peptides were

102 decamers with the exoeption of the last NS4a peptide, nS3, 103 which was an octamer. Solid-phase peptide synthesis, based 104 on Fmoc (9-ftuorenylmethoxycarbonyl) chemistry, was per-105 formed using the multi-pin peptide synthesis kit purchased

...................... "

101

......

from Chi ron Mimitopes Pty Ud, Australia. Peptides were synthesized on polyethylene pins arranged in an 8 x 12 ma­trix (of the same dimensions as well in a 96-well microtiter plate). These pin-bound peptides were used in ELlSAs (be­low).

2.3. Murine hyperimmune sera and dengue-patient sera

Hyperimmune murine sera specific for DEN-2 and yel­low fever virus (YFV) were obtained from Dr. A.M. Jana, Defense Research Development Establishment, Gwalior, In­dia; Japanese encephalitis virus (JEV)-specific murine hy­perimmune serum was from Dr. S. Vrati, National Institute of Immunology, New Delhi, India. Hyperimmune sera were generated by repeated boosting to obtain high antibody titers. The dengue-patient sera were obtained from Dr. R. Agarwal, of Sanjay Gandhi Post Graduate Institute of Medical Sci­ences, Lucknow, India and Prof. S. Fernando, University of Sri Jayewardenepura, Sri Lanka. All sera were from DEN-2 infected patients.

2.4. Enzyme-linked immuno sorbent assay (ELISA) for epitope identification

ELISA was carried out using the pin-bound capsid and NS4a peptides. All incubations were done using the 96-welI microtiter format. Briefly, the assay was as follows. Pins were immersed into 200 ~I blocking buffer (5% skimmed milk powder (SMP) in 1 x phosphate buffert'AI saline (PBS), pH 7.2) at 37 DC for 4 h. The pins were washed (1 x PBS/0.1 % Tween 20) and then immersed in 200 ~I of 1: 100 diluted (dBu­ent: 1 x PBS/S% SMP/l % sodium deoxy cholate (SOC» murine hyperimmune serum (dengue type 2-, YFV- or JEV­specific) or patient serum. After incubation at 37°C for an appropriate time (15 min for human sera; 1 h for murine sera), the pins were washed (1 x PBS/O.1 % Tween-201l.% SDC) four times (5 min/wash) and incubated with 200 ~I of 1:7500 diluted secondary antibody-horseradish peroxidase (HRPO) conjugate purchased from Calbiochem. Depending on the assay, three different secondary antibody-HRPO con­jugates were used. Anti-mouse IgG-HRPO was used (incuba­tion time: 1 h) when the peptides were scanned with murine hyperimrnune serum; either anti-human IgM-HRPO or anti­human IgG-HRPO was used (incubation time: 15 min) in as­says designed to scan the peptides with dengue-patient sera. Each of these three HRPO conjugated secondary antibodies was pre-tested to ensure that they do :not bind to the pin­bound peptides in the absence of the primary antibody. Pins were washed four times as before and incubated with 200 ~I of 3,3',5,S'-tetramethylbenzidine (TMB) substrate solution at 37°C (15 min for assays with dengue-patient sera and 45 min for those with murine hyperimmune sera). Reactions were stopped by removal of the pins followed by the addition of 1 00 ~I 1 M H2S04. Optical densities (ODs) were mea­sured at 450 nm in a tecan microtiter plate reader. For the purpose of assigning linear epitopes, the mean 00 obtained

VIRUS 940161-9

106

10~

106

109

110

111

112

113

,,. 115

116

117

118

119

120

121

122

123

124

125

125

127

128

129

130

131

133

134

135

136

137

136 ~

139 '

140

141

142

143

144

145

146

147

148

1'49

150

151

152

153

154 ~

155

,56

157

158

Jfi9 1l;()

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

J'l77

178

179

180

181

182

183

184

185

R. AnandaRao et al. / Virus Research xxx (2005) xxx-xxx 3

using normal/pre-immune sera plus twice the standard de- drophilicity plot of the capsid protein, two major regions viation (S.D.) of the mean was used as the cut-off value. of hydrophilicity were discernible, one at the amino (N)-The pin-bound peptides were re-used after stripping off of terminal end spanning aa residues 1-25, and the other at the the bound antibody and conjugate by sonication in hot deter- carboxy (C)-terminal end (aa residues 70-100). The central gent (l % SDS at 60°C) under reducing conditions (0.1 % 13- domain, sandwiched between these two hydrophilic regions, mercaptoethanol), and verifying antibody conjugate removal, was largely hydrophobic and therefore not likely to be po-as per the manufacturer's instructions. Control ELISAs were tentially antigenic. In contrast, the NS4a Kyte and Doolittle run in parallel, in which DEN-2 virus was used as the cap- plot predicted the N-terminal one-third of the molecule to be ture antigen (instead of the pin-bound peptides) to detect hydrophilic and the C-terminal two-thirds of the molecule to anti-dengue antibodies in the test sera. Each ELISA was per- be largely hydrophobic. Since regions of hydrophilicity are formed twice. likely to be surface-exposed and therefore potentially anti-

genic, we focused on these regions of the two molecules for

2.5. In-house ELISA for the detection of serum peps can analysis. Accordingly, we synthesized overlapping

anti-dengue antibodies decapeptides offset by one residue corresponding to the hy-drophilic regions of capsid and NS4a proteins. In addition,

The assay was essentially similar to that described above, we also designed sequential, non-overlapping decapeptides

except that serum antibodies were captured in microtiter wells spanning the entire hydrophobic regions in the two proteins.

coated with different recombinant protein antigens or DEN-2 All these peptides were synthesized on pin surfaces using

virus Iysates, instead of the pin-bound peptides. The recom- Fmoc chemistry as described in Section 2.

binant protein antigens used were r-DME-G (AnandaRao et l·; aI., in press), its second-generation version, r-DME-G2 (this 3.2. Identification of capsid epitopes study) or DEN-2 E protein (Bisht et aI., 2002).

For m~ppingjmmunodominant epitopes on the DEN-2 capsid, a library of 37 peptides (designated as c1-c37) was

3. Results genenited.Of these, peptides c1-c13 and c19-c37 repre-sented overlapping peptides corresponding to the amino- and

3.1. Epitope selection carboxy-terminal regions of the capsid protein. Five peptides, c14-c18; were non-overlapping peptides spanning the cen-

The protein sequences of capsid (100 aa) and NS4a (~150 tral hydrophobic region of the capsid molecule (see Fig. 2). aa) of DEN-2 were predicted from nucleotide sequences re- These peptides (pin-bound) were scanned in ELISAs, against trieved from the GenBank database (capsid was from DEN-2 murine hyperimmune sera to identify potential B-ce11 epi-NGC; NS4a was from DEN-2 Jamaica). Kyte and Doolit- topes. We used three different hyperimmune sera (obtained tie hydrophilicity plots of these proteins generated using from mice immunized with DEN-2, JEV or YFV) as the MacVector software are shown in Fig. I. From the hy- source of primary antibodies and anti-mouse IgG-HRPO as

6.0.,..-------------------------...,

4.0

2.0

0.0

-0.2

~ -0.4 -L..:~-,---._-__r--._-__r--..,__-___r--...._-__,--_l :c 10 20 30 40 50 60 70 80 90 100 c. e 4.0.,..-------------------------...,

~ 3.0 == 2.0

1.0 0.0 -/i;;;-'~-__ _'--~

-1.0 -2.0

-3.0

20 40 60 80

Amino acid number 100 120 140

~-!l Fig. I. Kyte and Doolittle hydropathy'(iidfil"";sof DEN-2 virus capsid (A) and NS4a (B) proteins generated using MacVector software. The horizontal axis indicates amino acid residue number'and the ~~rtical axis indicates the hydropathy score. Positive scores indicate hydrophilicity and negative scores indicate hydrophobicity. /

!:

VIRUS 940161-9

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

4 R. AnandaRao et ai.1 Virus Research = (2005) =-=

MNNQRKKARN TPFNMLKRER NRVSTVQQL T

I~~ 2 3 4

5 6 7 8

9 10 II 1213 14-----

KRFSLGMLQG RGPLKLFMAL VA FLRFLTIPPT - 15 16 17-------

AGILKRWGTI KKSKAINVLR GFRKEIGRML NILNRRRR 18-------

Fig. 2. Schematic represenlation of capsid peptides synthesized for epitope mapping. The predicted amino acid sequence of DEN-2 virus capsid is shown on the top. A total of 37 decapeptides corresponding to the capsid were synthesized using multi-pin peptide synthesis kit (chiron). Overlapping peptides were synthesized corresponding to the pulative hydrophilic regions and contiguous (non-overiapping) peptides for the putative hydrophobic regions. The numbered lines below the protein sequences denote the peptides synthesized. These lines have been positioned exactly below the corresponding regions of the proteins from where their sequences were derived. (In the text, the capsid peptides have been given the prefix 'c').

218 the secondary antibody. In this experiment, several reactive 219 peptides along the entire length of the capsid molecule were 220 picked up. These results are presented in the main panel of 221 Fig. 3. The inset shows a control experiment, run in paral-222 leI, wherein DEN-2 virus was used as the capture antigen, 223 to detect anti-dengue antibodies in DEN-2 murine hyperim-224 mune serum. There were three clusters of overlapping pep-225 tides of which one cluster (c1-c3) mapped to the N-terminal 226 region and two clusters (c25-c28 and c33-c36) mapped to 227 the carboxy-terminal region of the capsid protein. The se-228 quences of these peptides and their relative epitope activi-229 ties are summarized in Table l. The first cluster, defined by 230 the core sequence 'NQRKKARN' mapped to the N-terminus

2.0

E 1.5 .:

=> on ~ - 1.0 co Q

0 0.5

4 7 10 13 16 19 22 25 28 31 34 31 Peptide Dumber

Fig. 3. IgG-specific epitope activities of synthetic capsid peptides of DEN-2 virus determined using murine hyperimmune sera. A total of 37 pin-botmd peptides (shown in Fig. 2) were scanned using DEN-2 (grey bars), YFV (black bars) and lEV (open bars) hyperimmune mice sera by EUSA, as described in materials and methods. (The inset shows a control experiment in which DEN-2 virus was used as the capture antigen to detect IgG antibodies in DEN-2 hyperimmune murine serum (grey bar) and normal murine serum (open bar». The dala represent the average of two separate experiments.

(;~, ./

(aa residues 3-10). Interestingly, this epitope overlaps (by 2 231

aa residues) with the N-terminal capsid epitope reported by 232

Bulich and Aaskov (1992). It is pertinent that the latter epi- 233

tope was mapped with monoclonal antibodies, while in this 234

Table 1 Immunoreactive dengue type 2 virus capsid peptides

ELISA (OD4SO)a Peptide # Peptide sequence (aa residue Us) Murineb Humane

c1 MNNQRKKARN (1-10) 0,29 c2 NNQRKKARNT (2-11) 0.38 0,10 c3 NQRKKARNTP (3-12) 0.49 c13 FNMLKRERNR (13-22) 0.13 c14d VSTVQQLTKR (23-32) 0.43 c15d FSLGMLQGRG (33-42) 0.10 cJ6d PLKLfMALVA (43-52) 0.11 cJ-rJ FLRFLTIPPT (53~2) 0.27 0.13 c19 KKSKAINVLR (73~2) 0.11 c22 KAINVLRGFR (76-85) 0.14 e23 AJNVLRGFRK (77~6) 0.14 c24 INVLRGFRKE (78-87) 0.13 c25 NVLRGFRKEI (79-88) 0,26 0.38 c26 VLRGFRKEIG (80-89) 1.50 0.14 c27 LRGFRKEIGR (81-90) 0.12 0.12 c28 RGFRKEIGRM (82-91) 0.61 c33 EIGRMLNILN (87-96) 0.68 c34 IGRMLNILNR (88-97) 1.10 c35 GRMLNILNRR (89-98) 1.17 c36 RMLNILI'I<'RRR (90-99) 0.20 0.19 c37 ~~RRR(91-100) 0.28

(-) indicates low «0.1) or no reactivity. • ELISA ODs represent corrected values obtained after subtracting cut-off

OD (0.495 for murine and 0,664 for human sera). b Hyperimmune mouse serum. C Dengue-patient serum (lgG- and IgM-positive using Dengue Duo Test). d Non-overlapping peptides.

VIRUS 94016 1-9

235

237

238

239

240

241

R. AnandaRao ef al. / Virus Research xxx (2005) xxx-xxx

study the N-terminal epitope has been mapped using poly-clonal serum. The second and third cluster of peptides identi­fied regions with the core sequences 'RGFRKEI' (aa residues 82-88) and 'RMLNILN' (aa residues 90-96) towards the C­terminus of the capsid molecule. In all instances, flanking residues did seem to have a role in the overall antigenicity as evidenced by the ELISA ODs of individual peptides in each

5

242 cluster. Two peptides (c14 and cl7) corresponding to the pu­tative hydrophic region of the capsid were identified with reactivities comparable to that of the N-terminal epitope. Of the entire panel of 37 peptides assayed, peptides c26 and

243

244

245

clusters recognized by the murine anti-dengue hyperimmune serum, but with some subtle differences (see Table J). For ex­ample, in the first cluster the dengue-infected human serum pool showed reactivity towards peptides cl and c2 and did not react with peptide c3. Similarly, in the second cluster (c25-c28), the human serum reacted with three (c25-c27) out of four peptides. In addition, the pooled human serum also recognized peptides c22-c24, which were not picked up by the murine serum. Of the four peptides in the third cluster (c33-c36), the human serum reacted with just a sin­gle peptide, c36. Interestingly, the dengue-infected human serum pool recognized peptide c37, which failed to be iden­tified using the murine serum.

246 c35 represented the most immunogenic regions of the capsid 247 molecule. Interestingly, none of these peptides reacted with 248 JEV hyperimmune murine serum. This was essentially true 249 for YFV hyperimmune serum as well with the single excep-250 tion of peptide c37 corresponding to the carboxy-terminal 10

ELISAs were also carried out using the pin-bound capsid peptides to identify IgM-specific anti-dengue antibodies in the dengue-patient serum pool. This experiment was similar to the one described above, except that the secondary anti­body used was anti-human IgM-HRPO conjugate. In contrast to the experiment above, only 7 peptides (cI6, c17, c18, c28, c30, c32, and'c33) displayed detectable reactivity towards IgM class of anti~dengue anti bodies, of the 37 tested. Of these, peptides cl7 &lci'c33, displayed relatively higher ELISA re­activitie~: with OD450 values of 0.17 and 0.19, respectively. The remai~inggve peptides showed relatively weaker reac­tivities «0.1 OD450).

251

252

253

aa residues of the capsid protein. This peptide showed a slight reactivity to YFV hyperimmune serum (Fig. 3).

We then tested these peptides against a pool of human 254 dengue-patient sera. To obtain this pool, we first screened sera

of suspected dengue patients, using a commercially available 255

258

dengue diagnostic kit, the Dengue Duo Test kit (PanBio Pty Ltd., Australia). This kit uses a mixture of purified recom­binant E proteins of all four DEN serotypes to simultane-

259 ously detect the presence of both IgM and IgG classes of 260 anti-dengue antibodies, in a single sample (Cuzzubbo et aI., 261 2001). Sera (n= 15) that were positive for anti-dengue anti-262 bodies (both IgM and IgG classes) were pooled together for

3.3. Identification of NS4a epitopes

263 screening the capsid peptides. Serum antibodies that were _ .. We synthesized a panel of 53 peptides, designated nl-n53, 264 captured by the pin-bound capsid peptides were detected us- fo'r }.JS4a epitope analysis (Fig. 5). Forty-three overlapping 265 ing anti-human IgG-HRPO. For comparison, these peptides p~ptides (offset by one residue) spanned the putative hy-266 were screened in parallel using a control serum pool, gen- . 'drophilic amino-terminal region of the NS4a molecule (aa 267 erated using sera that were seronegative for both IgM and residues I-50). The rest of the molecule, predicted to be 268

269

270

271

"'"

IgG classes of anti-dengue antibodies. The data are shown~ '.:-Iargely hydrophobic, was analyzed using a set of 10 con­in Fig. 4A. Data from parallel ELISAs done using DEN~2. 'tiguous peptides (peptides n44-n53). None of the 53 NS4a virus coated microtiter wells are depicted in Fig. 4B. Th--e::,',':/peptides produced a detectable signal in ELISA using DEN­human serum pool also reacted with the same three peptide ,," 2-specific hyperimmune murine serum as the source of pri­

,,·,J~ .. I mary antibodies, indicating that NS4a-specific antibodies are -- ."','" 'i either absent or present at very low undetectable titers in the

io1

'.2s'·-U" murine serum (data not shown). In contrast to this however,

we could discern several NS4a peptides that reacted with dengue-infected patient serum (lgG+ IIgM+ based on PanBio

0.4 Dengue Duo test). In this experiment, antibodies captured by 0.0 I the pin-bound NS4a peptides were detected using anti-human

4 7 10 13 16 19 22 25 28 31 34 37 M G IgG-HRPO conjugate. The results are shown in Fig. 6. The Peptide number . (B) .Den-2 data revealed three adjacent clusters of overlapping peptides

Il~''''',~ Pi 4 I G ·ft· . . . f h' /f 'd1f ~d fDEN 2 spanning the region of NS4a between aa residues 13 and 50.

Ig. . g -speci c epltope aclIvllIes 0 synt elIc caps I peplI es 0 - The sequences of these NS4a peptides and their relative reac-virus detennined using human sera. (A) The experim?nt'is ·the same as in Fig. 3, except that the capsid peptides were scanned'using normal human tivities are summarized in Table 2. The first cluster of peptides (black bars) and dengue-patient sera (open bars).:Thihoriiontal axis denotes (nI3-nI7) mapped to the region ofNS4a spanning aa residues

(A)

the peptide number and the vertical axis denotes~absorbance measured at 13-26. All these peptides shared the common core sequence 450 nm. The cut-off absorbance used to ideJltify riacti;e peptides is shown "MTQKAR'. The second cluster of peptides (n25-27 and by the dotted line. (B) Control experime~t done iIi parallel using DEN-2 virus as the capture antigen, in place of th~pin-bO\fn'd peptides. Antibodies n29) defined by the core sequence 'AVLHT A' mapped to aa captured from normal human (black ~.aTS) a~«(d~gue-patient (open bars) 25-38 of NS4a, A third cluster consisting of peptides n37, sera were revealed using either anti-h~inan Ig~ (M) or anti-human IgG (G) n39 and n41 could be discerned mapping to a region of NS4a conjugate. The data represent the average oLt~o separate experiments. corresponding to aa residues 37--50. All these three peptides t:-::::::.... ---.,.,-

~>-..

~:-.:-::::;-, ":--.-

(' ."-~ ~<~

<:-,

VIRUS 94016 1-9

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

6 R. AnandaRao et al. / Virus Research.ux (2005) =-=

SLTLNLITEM GRLPTFMTQK ARDALDNLAV 12

3 4 5 6

78 9

10 11 J2

LHT AEAGGRA YNHALSELPE 131415 16

1718

31 n - 33

34 35 36

37 3839

1920

40 41 42

43

TLETLLLL TL LATVTGGIFL FLMSGRGIGK MTLGMCCIIT - 44 45 46 47-----

ASILLWYAQI QPHWIAASn LEFFLIVLLI PEPEKQRTPQDN -48 49 50 51 --=----=---QLTYVVIAILTVV AATMA 52 53

Fig. 5. Schematic representation of NS4a peptides synthesized for epitope mapping. The predicted amino acid sequence of DEN-2 virus NS4a protein is shown on the top. A total of 53 peptides corresponding to NS4a were synthesized using mUlti-pin peptide symhesis kit (chi ron). Overlapping peptides were synthesized corresponding to the putative hydrophilic regions and contiguous (non-overlapping) peptides for the putative. hydrophobic regions. The numbered lines below the protein sequences denote the peptides synthesized. These lines have been positioned exactly below the corresponding regions of the proteins from where their sequences were derived. (In the text, the NS4a peptides have been given the prefix 'n ').

326 shared the core sequence 'YNHALS'. In addition, we also 327 found a single amino-terminal peptide (n2), corresponding 328 to aa residues 2-11, also displaying significant reactivity to-329 wards dengue-patient serum. Of the NS4a peptides tested, 330 n14 and n25 were the most immunogenic peptides. Next, we 331 sought to determine if our NS4a peptides could react with 332 dengue-specific IgM antibodies. To this end, we repeated the 333 above scanning experiment, using anti-human IgM-HRPO 334 to detect IgM class of NS4a-reactive antibodies in dengue-

];S,-------~I.5~---.

U

lS

~ 1.0 U . ~ ~~ ~ .. .. ~ 0.5

17 21 25 49 53

Peptide number

Fig. 6. IgG-specific epitope activities of synthetic NS4a peptides of DEN-2 virus determined using dengue-patient serum. Fifty-three pin-bound pep­tides (shown in Fig. 5) were scanned with normal human (black bars) dengue­patient sera (open bars) in an ELISA. The horizontal axis denotes the peptide number and the vertical axis denotes absorbance measured at 450 nm. The dotted line indicates the cut-off absorbance, used to identify reactive NS4a peptides. (The inset shows a control experiment in which DEN-2 virus was used as the capture antigen to detect IgG antibodies in nonnal human (black bar) and dengue-patient (open bar) serum). The data represent the average of two separate experiments.

.~ ,

patient sera. The results showed that none of the 53 NS4a 335

peptides displayed any measurable reactivity in this ELISA 336

(data not shown), suggesting that NS4a does not evoke a de- 337

tectable IgM response during dengue infection. 338

3.4. Capsid and NS4a epitopes enhance the sensitivity of 339

a first generation diagnostic antigen 340

We recently designed a novel recombinant protein by as- 341

sembling together IgG-specific, linear immunodominant epi- 342

topes ofE and NSI (AnandaRao et aI., in press). We showed 343

that this protein, r-DME-G (recombinant dengue multi- 344'1 epitope protein, IgG-specific), could detect anti-dengue anti- 345

Table 2 Immunoreactive dengue type 2 virus NS4a peptides

Peptide # Peptide sequence (aa residue Us) ELISA (OD4SO)a

n2 LTLNLITEMG (2-11) 0.56 n13 LPTFMTQKAR (13-22) 0.45 n14 PfFMTQKARD (14-23) 0.89 n15 TFMTQKARDA (15-24) 0.26 n16 FMTQKARDAL (16-25) 0.51 nl7 MTQKARDALD (17-26) 0.27 n25 illNLAVLHTA (25-34) 0.75 n26 DNLAVLHTAE (26-35) 0.31 n27 NLAVLHTAEA (27-36) 0.44 n29 AVLHTAEAGG (29-38) 0.29 n37 GGRAYNHALS (37-46) 0.23 n39 RAYNHALSEL (39-48) 0.21 n4l 0.37 YNHALSELPE(41-50) -----.------------~--~------~----------

• ELISA ODs represent corrected values obtained after. subtracting cut-off OD (0.35).

VIRUS 940161-9

-------1

R. AnandaRao ef al. / Virus Research xxx (2005) xxx-xxx 7

--i.. e 1.2

= -= .,., ~ 0.8 .. Q

0 0.4

0.0 2 3 4

Fig. 7. In-house ELISA for the detection of antibodies in dengue-patient serum. Anti-dengue antibodies in normal human (black bars) and dengue­patient (open bars) sera were detected using different proteins (1: no capture antigen; 2: r-DME-G2; 3: r-DME-G; and 4: r-envelope) as capture antigens. Captured antibodies were visualized using anti-human IgG-HRPO conju­gate. Data shown represent the average of triplicate assays.

346 bodies in a panel of dengue-patient sera, pre-confinned to be 347 seropositive for anti-dengue antibodies using a commercially 346 available dengue diagnostic kit. To address the question if the 349 C and NS4a epitopes identified above would be potentially 350 useful in developing diagnostic intermediates, we developed

'351 a second-generation r-DME-G protein (r-DME-G2). To this 352 end, the gene encoding the r-DME-G protein was re-designed 353 to include sequences encoding the capsid and NS4a epitopes 354 identified in our pepscan analysis. The re-designed gene was 355 expressed in Escherichia coli, and the expressed protein puri-356 fied by affinity chromatography (data not shown). The capac-357 ity of the r-DME-G2 protein to detect anti-dengue antibodies 356 in pooled dengue-patient sera was then compared with that of 359 the first generation r-DME-G protein, in an in-house ELISA. 360 In this assay, microtiter wells were coated with the differ-361 ent antigens and then incubated with an appropriately diluted 362 aliquot of pooled dengue-patient serum. Captured antibodies 363 were visualized using anti-human IgG-HRPO. Controls were 364 run in parallel, in which recombinant E protein was used as 365 the capture antigen. The results are depicted in Fig. 7. The ~66 data clearly show that, the r-DME-G2 protein (containing 367 epitopes from E, NS1, C and NS4a) is far more superior to 366 the r-DME-G protein (containing only E and NSI epitopes).

369 4. Discussion

Several dengue diagnostic kits are commercially available (Groen et aI., 2000; Cuzzubboet aI., 2001). Many of these use whole dengue virus preparations as antigens to detect anti-

373 dengue antibodies. The use of the whole dengue virus antigen is expensive and prone to serological cross-reactivity due to similarity with other f1aviviruses such as JEV and YFV. To

376 eliminate these shortcomings, the whole virus antigen must be replaced with a more appropriat~' antigen in dengue diag­nostics. In this context, knowledg~'of imrriimogenic epitopes

379 encoded by pathogens is a pre-req\!isiti for developing di-

370

371

372

374

375

377

376

F,' ...

360 agnostic peptides for the detection of. infections with a high 361 degree of sensitivity and spe~ificity, While the immunogenic

epitopes on the E and NSI proteins have been well docu- 362

mented (Aaskov et aI., 1989; Innis et aI., 1989; Roehrig et 363

aI., 1990; Megret et aI., 1992; Jianmin et aI., 1995; Huang 364

et aI., 1999; Garcia et aI., 1997; Falconar et aI., 1994; Wu et 365

aI., 2001), very little infonnation is available regarding the 366

antigenicity of the remaining dengue viral proteins. The cap- 367

sid and NS4a are two small viral proteins involved in genome 366

packaging (Lindenbach and Rice, 2001) and in anti-host vi- 369

ral defense (Munoz-Jordan et aI., 2003). Importantly, from 390

the viewpoint of diagnostics, these two proteins are also 391

implicated in the induction of B-cell antibody responses 392

in dengue virus-infected individuals (Valdes et aI., 2000; 393

Churdboonchart et aI., 1991; Se-Thoe et aI., 1999). The pur- 394

pose of this study was to locate linear B-cell epitopes on 395

dengue virus capsid and NS4a proteins, which might be po- 396

tentially useful in diagnosis. We therefore perfonned pepscan 397

analyses using pooled dengue-patient serum, rather than sin- 396

gle sera. As DEN-2 is currently the most prevalent of the four 399

serotypes in India, we performed our pepscan analysis with 400

DEN-2 virus~infected patient sera. 401

On the capsid, we identified several IgG-specific antigenic 402

regions, defined by reactive overlapping peptide clusters, us- 403

ing dengue-specific murine hyperimmune serum as well as 404

dengue-patient sera. Both sera identified three major peptide 405

clusters, one in the amino-terminal, and two in the carboxy- 406

terminal regions. In addition, both sera picked up epitopes, 407

defin'ed by single, non-overlapping peptides, in the central 406

capsid region (Table 1). For example, the most immunogenic 409

peptides identified using the murine hyperimmune serum 410

were peptides c26, c35, and c3, belonging to clusters 2, 3 411

and 1, respectively. Using the dengue-patient serum pool, es- 412

senti ally similar data were obtained. However, rather than 413

peptide c26, it was peptide c25, that was picked up most ef- 414

ficiently by the human anti-dengue antibodies. Based on our 415

data, the core sequences of peptide clusters recognized by 416

dengue-patient sera are NNQRKKARN (defined by overlap- 417

ping peptides cl and c2), RGFR (defined by overlapping pep- 416

tides c22-c27) and MLNILNRRR (defined by overlapping 419

peptides c36 and c37). Interestingly, the carboxy-terminally 420

located immunodominant peptides are located on an a helix, 421

exposed on the surface of the capsid dimer (Ma et aI., 2004). 422

In general, the observed ELISA reactivities were higher us- 423

ing the murine hyperimmune serum, as expected. Aside from 424

these IgG-specific epitopes, our studies also identified two 425

peptides, c17 and c33, which appeared to recognize anti- 426

dengue human antibodies of the IgM class. While peptide 427

c17 interacted with IgG antibodies as well, peptide c33 did 426

not. Peptide 33 presumably represents a unique IgM-specific 429

epitope. 430

The results also showed that none of the 37 capsid p~p- 431

tides reacted with either YFV or JEV hyperimmune mice 432

sera. A comparison of the core sequences of each of these im- 433

munodominant regions (c1ustal analysis), showed that there 434

was a high degree of similarity amongst the four dengue 435

serotypes. However, peptide sequences in corresponding re- 436

gions of YFV and JEV capsids diverged considerably from 437

VIRUS 94016 1-9

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

8 R. AnandaRao et al. / Virus Research xxx (2005) xxx-xxx

2 10

NHl trNQRXKARN

D-1 NNQRlCltTAR D-3 NNQRlCltTGIt D-4 N -QRItItVVR JEV TnPGGPGlt YFV MSGRnQGlt

(A)

17 22 29 34 4]

82 85 91 99

I-EJ-1 HLNILNRRR H COOH

RGPK MLNDmRRX KGPX MLSIrHKRK IGl"R HLNIt.lfGRK TSFIt LIDAVNKRG RXVK LMRGLSSRX

46

NHl H MTQlCAR r-I AVLHTA H YNRALS H I ICOOH

D-1 LTQRAQ VKLHNS YRRAME D-3 LAHRTR VMLBTS YRHAVE D-4 LSSRAQ VMLHTT YQHALN

JEV F'IIGn'R YLVATA HRMALE (8) YFV LAl\l(.GG SVFLHS YlmALS

Fig. 8. Major IgG-specific OFN-2 epitopes identified in this study. Line maps ofOEN-2 capsid (A) and NS4a (B) 'core' epitope sequences identified by pepscan analysis. Shown below are the corresponding peptide sequences (identified by c1ustal analysis) of these two proteins of other flaviviruses (abbreviations: O-J, 0-3 and 0-4: dengue virus type I, 3, and 4, respectively; JEV: Japanese encephalitis virus; YFV: yellow fever virus).

those of the dengue capsids (Fig. 8A). This explains the lack performed in the course of generating the murine hyperim-of reactivity of the dengue capsid peptides when scanned mune serum. This would effectively preclude any translation with either YFV- or lEV-hyperimmune mUrine'sera. We con- and replication of the viral genome, which is a must before c1ude that the capsid epitopes identified in this study are truly non-stn:ctural proteins like NS4a are produced and perceived dengue virus-specific and therefore have the potential to be by the immune system. useful in the detection of dengue infections. In a recent study, we demonstrated that a recombinant

All NS4a peptides with detectable epitope activity were dengue multi-epitope protein, r-DME-G, cre,ated by splic-mapped to the amino-terminal putative hydrophilic region of ing together linear immunodominant epitopes of E and NSI the protein. As with the capsid, we found multiple adjacent can serve as a useful dengue diagnostic antigen (AnandaRao clusters of reactive peptides, using dengue-patient serum. A et aI., in press). To gauge the diagnostic utility of the C and comparison of the core sequences of these NS4a peptide c1us- NS4a epitopes identi fied in this study, we developed a second-ters amongst dengue and other flaviviruses such as YFV and generation molecule by incorporating these into the r-DME-lEV (Fig. 8B), showed that the core sequence AVLHTA de- G molecule. Preliminary data suggest that the modified anti-fined by the second cluster is unique to the dengue viruses gen, r-DME-G2, displays relatively enhanced sensitivity in alone. On the other hand, the core epitope sequence defined detecting anti-dengue antibodies. In dengue endemic areas, by the remaining two peptide clusters showed varying de- secondary infections are most common. Thus, it is very likely grees of similarity with the corresponding NS4a peptides of that many of the dengue-patient sera used to generate the YFV and lEY. The previous reports in the literature that de- pooled serum in our studies are from secondary infections. scribed the detection of NS4a-specific antibodies in dengue This is consistent with all these sera testing positive using patient were based on the observation of a protein with the the PanBio kit, which is designed to detect secondary infec-predicted electrophoretic mobility of NS4a in western blots tions (Cuzzubbo et aI., 2(01). The prevalence of antibodies (Churdboonchart et aI., 1991; Se-Thoe et aI., 1999). A more to various structural and non-structural dengue proteins is recent report demonstrated that a recombinant NS3INS4a fu- generally higher in secondary compared to primary dengue sion protein displayed ELISA reactivity using dengue-patient infections (Churdboonchart etal., 1991; Se-Thoeetal., 1999; sera (Dos Santos et aI., 2004). However, this study did not Valdes et aI., 2000). For example, the seroprevalence of anti-unambiguously identify NS4a-specific antibodies in the pa- NS4a antibodies is ",50% in secondary infections (Se-Thoe tient sera. Our current study provides definitive evidence for et aI., 1999). It is therefore conceivable that the recognition the occurrence of NS4a-specific antibodies in dengue-patient of these antibodies in the pooled dengue-patient serum by the serum using authentic dengue type 2 NS4a-derived peptides r-DME-G2 protein contributes to its improved sensitivity. as capture antigens. Further, our data show that these NS4a- In conclusion, our study has identified several immun-specific antibodies that we detected are IgG type antibodies. odominant IgG-specific epitopes on both the capsid and NS4a We were unable to detect any NS4a peptides that showed proteins using a multi-pin peptide scanning approach. The reactivity when tested against murine dengue-specific hyper- capsid epitopes are specific to dengue virus and do not react immune serum. This perhaps is a reflection of effective c1ear- with YFV- and lEV-specific antibodies. This work shows for ance of virus-infected cells during the booster immunizations

(::-.'~ ;/ the first time the existence of anti-NS4a antibodies in dengue-

, ....

VIRUS 94016 1-9

473

mt 476

476

4n 478

479

480

481

482

483

-484

486

486

4117

488

~1

400

«31

<192

493

~94

4G5

496

«37

~98

<199

500

501

502

503

.504 )

505

506

507

508

~509 510

511

512

513

514

515

516

517

518

519

520

521

522

~ 523

,524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

. 541

~42 543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

l.,

TD5

R. AnandaRao et al. / Virus Research xxx (2005) xxx-xxx 9

infected human serum. The immunodominant peptides of capsid and NS4a, in conjunction with the well-documented dengue-specific epitopes of E and NSl, could be developed into diagnostic reagents for the detection of dengue infections with a high degree of sensitivity and specificity.

Acknowledgements

We thank Drs. A.M. Jana and S.Vrati for providing us the hyperimmune mice sera used in this study. We gratefully acknowledge Prof. S. Fernando and Dr. R Agarwal for the dengue-patient sera. The work was supported by ICGEB core funds. RA. is supported by a senior research fellowship from the Council of Scientific and Industrial Research, India.

References

Aaskov, J.G., Geysen, H.M., Mason, T1., 1989. Serologically defined linear epitopes in the envelope protein of dengue 2 (Jamaica strain 1409). Arch. Virol. lOS, 209-221.

AnandaRao, R., Swaminathan, S., Fernando, S, Jana, A.M. Khanna, N., in press. A custom-designed recombinant multiepitope pro­tein as a dengue diagnostic reagent. Protein Exp. Purif., doi: 10.1 0 16/j.pep.200S.01.009.

Bisht, H., Chugh, D.A., Raje, M., Swami nathan, S., Khanna, N., 2002. Recombinant dengue virus type 2 envelope/hepatitis B surface antigen hybrid protein expressed in Pichia pastoris can function as a bivalent immunogen. J. Biotechnol. 99, 97-110.

Bulich, R, Aaskov, J.G., 1992. Nuclear localization of dengue 2 virus core protein detected with monoclonal antibodies. J. Gen. Virol. 73, 2999-3003.

Cardosa, MJ., Wang, S.M., Sum, M.S.H., Tio, PH., 2002. Antibod­ies against prM protein distinguish between previous infection with dengue and Japanese encephalitis viruses. BMC Microbiol. 2,9, URL:,. http://www.biomedcentral.comlI471-2180/2/9.

Churdboonchart, v., Bhamarapravati, N., Peampramprecha, S., Sirinavin" S, 1991. Antibodies against dengue viral proteins in primary, and secondary dengue hemorrhagic fever. Am. J. Trop. Med. Hyg. 44, 481-493.

Cuzzubbo, AJ., Endy, TP, Nisalak, A., Kalyanarooj, S., Vaughn, D.W, Ogata, S.A., Clements, D.E., Devine, PL., 2001. Use of recombinant envelope proteins for serological diagnosis of dengue virus infection in an immunochromatographic assay. Clin. Diagn. Lab. Immunol. 8, 1150-1155.'

Dos Santos, FB., Miagostovich, M.P, Nogueira, RM., Schatzmayr, H.G., Riley, L.W, Harris, E., 2004. Analysis of recombinant dengue virus polypeptides for dengue diagnosis and evaluation of the humoral im­mune response. Am. J. Trop. Med. Hyg. 71, 1~IS2.

Fa1conar, A.K.I., Young, P.R, Miles, M.A., 1994. Precise location of sequential dengue virus subcomplex and complex B cell epitopes on the nonstructural-I glycoprotein. Arch. Virol. '137,315-326.

Garcia, G., Vaughn, D.W, Del Angel, RM., 1997. Recognition of syn­thetic oligopeptides from nonstructural pioteinsNSI and NS3 of dengue-4 virus by sera from dengue virus-infected children. Am. J. Trop. Med. Hyg. 56, 466-470.

i

Groen, 1., Koraka, P., Velzing, J., Copra, e., Osterhaus, A.D.M.E., 2000. Evaluation of six immunoassays for detection of dengue virus-specific immunoglobulin M and G antibodies. Clin. Diagn. Lab. Immunol. 7, 867-871.

Huang, 1.-H., Wey, 1.-J., Sun, Y.-e., Chin, C., Chien, L.-1., Wu, Y.-e., 1999. Antibody responses to an immunodominant nonstructural I syn­thetic peptide in patients with dengue fever and dengue hemorrhagic fever. 1. Med. Virol. 57, 1-8.

Innis, B.L., Thirawuth, v., Hemachudha, e., 1989. Identification of con­tinuous epitopes of the envelope glycoprotein of dengue type 2 virus. Am. 1. Trop. Med. Hyg. 40, 676-687.

Jianmin, Z., Linn, M.L., Bulich, R., Gentry, M.K., Aaskov, 1.G., 1995. Analysis of functional epitopes on the dengue 2 envelope (E) protein using monoclonal IgM antibodies. Arch. Virol. 140, 899-913.

Lai, e.-J., Monath, TP., 2003. Chimeric ftaviviruses: novel vaccines against dengue fever, tick-borne encephalitis, and Japanese encephali­tis. In: Chambers, TJ., Monath, TP (Eds.), The Aaviviruses: Detec­tion, Diagnosis and Vaccine Development, Elsevier Academic Press, Adv. Virus Res., vol. 61, pp. 469-509.

Lindenbach, B.D., Rice, e.M., 2001. Aaviviridae: the viruses and their replication. In: Knipe, D.M., Howley, PM. (Eds.), Fields Virology, fourth ed. Lippincott Williams & Wilkins, Philadelphia, pp. 991-1041.

Ma, L., Jones, e.T., Groesch, TD., Kuhn, RJ., Post, C.B., 2004. Solution structure of' dengue virus capsid protein reveals another fold. Proc. Natl. Acad. Sci: U.S.A. 101,3414-3419.

Megret, F, Hugnot, J.P, Fa1conar, A., Gentry, M.K., Morens, D.M., Mur­ray, J.M., Schlesinger, 1.1., Wright, PJ., Young, P., Van Regenmor­tel, M.H.V., Deubel, v., 1992. Use of recombinant fusion proteins and monoclonal antibodies to define linear and discontinuous anti­genic sites on the dengue virus envelope glycoprotein. Virology 187, 480-491.

Monath, TP, i999. Yellow fever. In: Plotkin, WA., Orenstein, S.A. (Eds.'),' Vaccines, third ed. WB. Saunders, Philadelphia, pp. 815-879.

Munoz-Jordan, J.L., Sanchez-Burgos, G.G., Laurente-Rolle, M., Garcfa­Sastre, A., 2003. Inhibition of interferon signaling by dengue virus.

,Proc. Natl. Acad. Sci. U.S.A. 100, 14333-14338. Rodda, SJ., Tribbick, G., 1996. Antibody-defined epitope mapping using

the multipin method of peptide synthesis. Methods: Companion to Methods in Enzymology 9, 473-481.

Roehrig, 1.T, Johnson, AJ., Hunt, A.R, Bolin, RA., Chu, M.e., 1990. Antibodies to dengue 2 virus E-glycoprotein synthetic peptides iden­tify antigenic conformation. Virology 177,668-675.

Saluzzo, J.-F, 2003. Empirically derived live-attenuated vaccines against dengue and Japanese encephalitis, in: Chambers, TJ., Monath, T.P . (Eds.), The Aaviviruses: Detection, Diagnosis and Vaccine Develop­ment, Elsevier Academic Press, Adv. Virus Res. vol. 61, pp. 419-443.

Se-Thoe, S.Y., Ng, M.M.L., Ling, A.E., 1999. Retrospective study of western blot profiles in immune sera of natural dengue virus infec­tions. 1. Med. Virol. 57, 322-330.

Tsai, TF, Chang, G.J., Yu, Y.X., 1999. Japanese encephalitis vaccines. In: Plotkin, W.A., Orenstein, S.A. (Eds.), Vaccines, third ed. W.B. Saunders, Philadelphia, pp. 672-710.

Valdes, K., Alvarez, M., Pupo, M., Vazquez, S., Rodriguez, R, Guzman, M.G., 2000. Human dengue antibodies against structural and non­structural proteins. Clin. Diagn. Lab. Immunol. 7, 856-8S7.

World Health Organization, 1997. Dengue hemorrhagic fever: diagnosis, treatment, prevention and control, second ed. Geneva, Switzerland.

Wu, H.-e., Huaung, Y.-L., Chao, T-T, Jan, J.-T, Huaung, 1.-L., Chiang, H.-Y., King, e.-e., Shaio, M.-F, 2001. Identification of B-cell epitope of dengue virus type I and its application in diagnosis of patients. J. Clin. Miocrobiol. 39, 977-982.

VIRUS 94016 1-9

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

"" 599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

Available online at www.sciencedirect.com

SCIENCE@DIRECT® Protein ..Expression & PUrification

ELSEVIER Protein Expression and Purification 41 (2005) 136-147

www.elsevier.comllocate/yprep

A custom-designed recombinant multiepitope protein as a dengue diagnostic reagent

Ravulapalli AnandaRaoa, Sathyamangalam Swaminathana, Sirimali Fernandob,

Asha M. Janac, Navin Khannaa,*

a International Centrefor Genetic Engineering and Biotechnology, New Delhi-ll0067, India b Faculty of Medical Sciences, University of Sri Jayewardenepura, Gangodawilla, Nugegoda, Sri Lanka

C Division of Virology, Defense Research and Development Establishment, Ministry of Defense, Government of India, Jhansi Road, Gwalior-474002, India

Received 17 November 2004, and in revised form 10 January 2005 Available online 26 January 2005

Abstract

Currently, dengue fever is the most important re-emerging mosquito-borne viral disease, with the major proportion of the target population residing in the developing countries of the world. In endemic areas, potentially fatal secondary dengue infections, charac­terized by high anti-dengue IgG antibody titers, are most common. Most currently available commercial dengue diagnostic kits rely on the use of whole virus antigens and are consequently associated with false positives due to serologic cross-reactivity, high cost of \,<;.1 antigen production, and biohazard risk. This has prompted the need to develop an alternate antigen to replace the whole virus anti­gen in diagnostic tests. We have designed and expressed a novel recombinant protein antigen by assembling key immunodominant linear IgG-specific dengue virus epitopes, chosen on the basis of pepscan analysis, phage display, and computer predictions. The recombinant dengue multiepitope protein was expressed to high levels in Escherichifl coli, purified in a single step, yielding >25 mg pure protein per liter culture. We developed an in-house enzyme-linked immunosorbent assay (ELISA) to detect anti-dengue anti­bodies in a panel of 20 patient sera using the purified recombinant dengue muItiepitope protein as the capture antigen. The ELISA results were in excellent agreement with those obtained using a commercially available diagnostic test, Dengue Duo rapid strip test from PanBio, Australia. The high epitope density, careful choice of epitopes, and the use of E. coli system for expression, coupled to simple purification, jointly have the potential to lead to the development of an inexpensive diagnostic test with a high degree of sensi­tivity and specificity. © 2005 Elsevier Inc. All rights reserved.

Keywords: Dengue virus; Multiepitope protein; Dengue diagnostic test; Ni-NT A chromatography

Dengue fever (DF) and its more severe manifesta­tions, namely, dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS), are caused by infection with the mosquito-borne dengue viruses, which are members of the family Flaviviridae [1]. There are four closely related, antigenically distinct, serotypes (1-4) of dengue viruses, each of which can cause disease. In

• Corresponding author. E-mail address:navin@icgeb.res.in (N. Khanna).

1046-59281$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi: 1 0.1 016/j.pep.2005.01.009

recent decades, there has been a dramatic increase in the incidence dengue infections, with about 100 million cases of DF occurring each year [2]. Globally, about 2.5 billion people are estimated to be at r,isk from dengue. The lack of a licensed dengue vaccine, in conjunction with pre­dicted climatic changes and population growth is pro­jected to place 5-6 billion people at risk of dengue transmission in the coming decades [3].

Dengue infections may be clinically inapparent or may result in non-specific febrile illness, DF or DHF [4,5]. Severe plasma leakage can lead to fatal DSS and

R AnandaRao et al. I Protein Expression and Purification 41 (2005) 136-147 137

mortality rates for untreated patients can be as high as 40-50% [6]. Early diagnosis, followed by supportive care, and symptomatic treatment through fluid replacement are the keys to survival in cases of severe dengue infec­tion [4,7]. Definitive diagnosis of dengue infection depends on the identification of the virus, virus-encoded antigens, viral genomic RNA or the virus-induced anti­bodies [2,8]. Virus identification through its isolation can take several days and may not always be successful due to very small amounts of viable virus in the clinical sam­ples. Viral antigens can be detected by immunohisto­chemistry or immunofluorescence. However, the complexity of these assays and their high cost preclude their routine use. Viral RNA can be detected with a high degree of sensitivity, using coupled reverse transcription and polymerase chain reaction (RTPCR). Besides being expensive and requiring sophisticated equipment, the RTPCR assay is subject to wide variability, as demon­strated in a recent study conducted by the European Network for Diagnostics of Imported Viral Disease [9]. Further, a shortcoming common to all these methods is the narrow time window (~5 days), available for suc­cessful detection, which coincides with the febrile period during which patients are viremic [10,11). Often, dengue­infected patients do not seek immediate medical care as the initial manifestations are usually asymptomatic or mild fever. This precludes diagnostic tests based on the identification of the virus or its RNA genome because of the short duration of viremia. Thus, in a majority of cases the only feasible diagnostic test would have to be based on the identification of anti-dengue antibodies.

Numerous dengue diagnostic kits. in a multiplicity of formats, have become available recently (12). Most of these kits rely on the use of whole virus antigens (pro­duced in tissue culture or suckling mice brain) for the detection of anti-dengue antibodies in patient sera, and are consequently associated with an inherent biohazard risk. One kit, which has replaced the whole virus antigen with insect cell-expressed dengue envelope protein, elim­inates this risk [13]. However, all these kits are expensive due to the high costs associated with antigen production, making them unaffordable for use in the economically weaker countries where dengue is mostly prevalent. Apart from this, a major shortcoming of the commercial kits is that they do not differentiate between infections due to dengue and other f1aviviruses (such as Japanese encephalitis and yellow fever viruses). Additionally, sera from patients with typhoid, malaria, and leptospirosis also tend to score positive using these kits.

There is currently a need for developing cost-effective, safe, and simple diagnostics that combine sensitivity and specificity. In dengue endemic areas, secondary infec­tions (infection of individuals seropositive for dengue due to a prior exposure) are most common and often associated with DHF and DSS [10,14]. During second­ary infection with dengue virus, high levels of IgG serum

antibodies appear within 3-5 days after onset of illness, peak by about 2 weeks and then decline gradually during the following several months [4]. As these infections are characterized by significant serum IgG titers, we focused on developing a recombinant antigen designed to detect the IgG class of anti-dengue antibodies. To this end, we adopted a novel approach that entails the creation of a multiepitope protein consisting of several key IgG-spe­cific, immunodominant epitopes, encoded by the major structural envelope (E) protein and the non-structural (NS) proteins I and NS3 [1]. The E protein is the major structural component [15] and the most immunogenic of all the dengue viral proteins, eliciting the first and lon­gest-lasting antibodies [16,17]. Immunodominant epi­topes on the E protein are well-documented [18-20]. Amongst the non-structural proteins, NSI [21-24] and NS3 [16.22] are reported to elicit significant antibody responses, particularly in secondary infections [25], with the former being more immunogenic than the latter. Epi­topes on these proteins have been mapped using syn­thetic peptides spanning entire proteins [18,21] or defined regions based on computer predictions [22,23], phage displayed peptides [24,26] and recombinant frag­ments [19.20] on the basis of reactivity towards patient sera [18.22] or monoclonal antibodies [19-21,24,26].

This multiepitope protein, r-DME-G (recombinant dengue multiepitope protein, specific to 19G), was expressed to high levels in E. coli and purified efficiently in a single step by affinity chromatography. Using an in­house enzyme-linked immunosorbent assay (ELISA), we demonstrate that this recombinant synthetic protein is able to accurately identify patient sera that contain anti­dengue virus antibodies. In this paper, we describe the design of r-DME-G protein, its expression, purification, and a preliminary evaluation of its utility to serve as a diagnostic reagent in the detection of dengue infections.

Materials and methods

A{ aterials

Escherichia coli host strain DH5::x for routine recom­binant plasmid manipulations was purchased from Invitrogen, USA. E coli host strain SG 13009, harboring the lael repressor encoding plasmid pREP4 (kanf), for recombinant protein expression was from Qiagen, Ger­many. The plasmid pQE60 (ampf), Ni-NTA superftow resin, and anti-His monoclonal antibody (catalog 34660) were also from Qiagen. All secondary antibody-enzyme conjugates [anti-mouse IgG-alkaline phosphatase (AP), anti-mouse IgG-horseradish peroxidase (HRPO) and anti-human JgG-HRPO], and the AP substrate 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetra­zolium (BCIPINBT) were from Calbiochem, USA. The HRPO substrate 3,3' ,5,5'-tetramethylbenzidine (TMB)

138 R AnandaRao et al.I Protein Expression and Purification 41 (2005) 136-147

was from Kirkegaard and Perry laboratories, USA. The Dengue Duo Rapid Strip Test (referred to as Dengue Duo test here onwards) was purchased from PanBio Pty, Australia.

Construction of the r-DME-gene expression vector

A synthetic gene, codon-optimized for E. coli expres­sion, encoding a r-DME-G protein was first generated by ligation (GeneArt, Germany) of carefully designed oligonucleotides encoding 15 linear immunodominant epitopes, of which eight were from the E protein, six from NSI, and one from NS3. These epitopes range in length from 6 to 20 amino acid (aa) residues with adja­cent residues being joined together by triglycyl linkers. The r-D M E-G gene ("-'0.7 k b) was ligated into the Bam HI and BglII sites of the bacterial expression vector pQE-60, under the control of the isopropylthiogalacto­side (IPTG) inducible PT5 promoter, to generate plas­mid pQ-r-DME-G. In this plasmid, the r-DME-G gene is inserted in-frame with the ATG codon (at the 5' end) and 6x-His tag-encoding sequence (at the 3' end) pro­vided by the pQE-60 vector. Recombinant clones were selected on ampicillin plates and subjected to direct col­ony polymerase chain reaction (PCR) screening, using insert-specific primers, to identify recombinants harbor­ing the synthetic r-DME-G gene. Orientation of the insert was verified by restriction analysis of plasmid minipreps.

Expression screening

Plasmid miniprep DNA obtained from the DH5cr recombinants above was used to transform the strain SG 13009 (E. coli expression host), containing the pREP4 plasmid. The pREP4 plasmid encodes the lael repressor (required for regulated recombinant gene expression) and the kanamycin marker. Double recombinants har­boring both the pQ-r-DME-G and pREP4 constructs were selected in the presence of ampicillin and kanamy­cin. Several of these clones were inoculated into 3 ml test tube cultures and allowed to grow at 37°C in a shaker at 200 rpm. When the cultures were in logarithmic growth phase [corresponding to an optical density (OD) of "-'0.6 at 600nm], they were induced with 1 mM IPTG for "-'4h. After induction, equivalent number of cells from the different cultures (normalized on the basis of OD600 val­ues) were lysed in sample buffer and analyzed by SDS­PAGE [27]. Uninduced controls were analyzed in paral­lel. One clone that expressed maximal levels of the recombinant protein was chosen for further work.

Purification of r-DME-G protein

A pre-culture was set up by inoculating 20ml LB medium containing ampicillin (100 J.lg/ml) and kanamy-

cin (25 J.lg/ml), with 10 J.l1 glycerol stock of SG 13009 cells transformed with the pQ-r-DME-G expression vector. The culture was grown overnight in a shaker at 37°C, at 200rpm and inoculated into 1 L LB (containing both ampicillin and kanamycin, at the concentrations indi­cated above) in a 4-L Haffkine flask, which was placed in the shaker at 37°C for about 2-3 hat 125 rpm. When the OD600 of the culture reached "-'0.6 (a small aliquot of the uninduced culture was set aside for subsequent SDS­PAGE analysis), it was induced by the addition ofIPTG to a final concentration of 0.1 mM (IPTG dose variation experiments showed that induction at 0.1 mM was as good as that at 1 mM IPTG). Induction was allowed to proceed for 4h before harvesting the cells. Aliquots of the induced and uninduced cell cultures were analyzed by SDS-PAGE prior to initiating purification.

The induced culture was centrifuged in a Sorvall GS3 rotor at 6000 rpm for 15 min at 4°C. About 1 g of induced cell pellet (corresponding to 250 ml E. coli cul­ture) was lysed by stirring in 10 ml of 8 M urea (dissolved in 100 mM sodium phosphate/IO mM Tris-HCl, pH 8), at room temperature (R T) for 1 h. The lysate was clari­fied by centrifugation (10,000 rpm in Sorval GSA rotor at RT for 30 min). The resultant slilpernatant was mixed with 2ml Ni-NTA Superflow resin, that had been pre­equilibrated with the same buffer used for cell lysis above. This suspension was gently rocked for 1 h at R T and then packed into a column. After collecting the flow­through, the column was washed extensively with buffers 1 (PH 6.3) and 2 (PH 5.9) and eluted with buffer 3 (pH 4.5). Fractions of 3 ml were collected for analysis. The wash and elution buffers had the same composition as the lysis buffer (8 M urea/lOOmM sodium phosphate/ IOmM Tris-HCI) except that they differed in their pH. All fractions obtained during the purification were ana­lyzed by SDS-PAGE. The pH 4.5 eluates, which were homogeneous were pooled together, mixed with genta­micin to a final concentration of 50 J.lg/ml, flash-frozen in liquid nitrogen, and stored at -80°C until use.

Western blot analysis

The purified r-DME-G proteill1 was run on a 15% denaturing gel (SDS-PAGE), along with appropriate controls and pre-stained protein markers, and trans­ferred electrophoretically to nitrocellulose membrane and probed with different primary antibodies. In the experiment shown in Fig. 4B, the membrane was blocked with 1% polyvinyl pyrrolidone in 1 x phosphate-buffered saline (PBS) for 2 h at R T. It was then washed three times with 1 x PBS containing 0.1 % Tween 20 (l x PBS + 0.1 % T) and incubated with a commercially avail­able murine penta-Ris mAb (at 1:2000 dilution) for 90 min at R T. The membrane was washed again, as above, and incubated with anti··mouse IgG-AP conju­gate (1:5000 dilution) for a further 90min at RT. The

R. AnandaRao el aJ. I Protein Expression and Purification 41 (2005) 136-147 139

blot was washed and developed by incubation in BCIPI NBT substrate solution for 15min at RT. The experi­ment shown in Fig. 5C was perfonned similarly with the following changes. The r-DME-G protein was run in a single wide well along with pre-stained protein markers in an adjacent narrow lane. After electrotransfer as above, the membrane was cut vertically into strips and blocked with 1 x PBS+ 2%T, for I hat RT. The blocked strips were washed three times with 1 x PBS + 0.1 % T and incubated separately with a 1:100 dilution of either a dengue seropositive or seronegative (based on the Den­gue Duo test) human serum. for 30 min at R T. The strips were washed as before and incubated with anti-human IgG-HRPO conjugate (1:10.000 dilution) for 30min at R T. The strips were washed once again and incubated in TMB substrate solution for 30min at RT to visualize the r-DME-G protein band.

In-house enzyme-linked immunosorbent assa)'for the detection of anti-dengue IgG antibodies

The purified r-DME-G protein was diluted to 10 Jlg/ ml in 0.1 M carbonate buffer. pH 9.5, and used for coat­ing 96-well microtiter plates (100 Jll/well) at 4°C over­night. The coated wells were washed once with I x PBS + 0.1 %T and blocked with 5% skimmed milk pow­der (SMP) in 1 x PBS (l x PBS + 5%SMP) for 4h at 37°C. The wells were washed once again with 1 x PBS+O.I%T and then incubated for 15min at 37°C with 100 JlI dengue patient sera [at 1:20 dilution in I x PBS + 5% SMP containing 1 % sodium deoxycholate (SOC)]. Wells were washed using I x PBS + 0.1 % T + I%SDC and incubated with anti-human IgG-HRPO conjugate (at 1:7500 dilution). The wells were washed once again as above and incubated with 100 Jli TMB substrate for 15 min at 37°C. Peroxidase reaction was tenninated with 100JlI of I M H~04 and the absor-

bances read at 450 nm. In some ELISA experiments, the source of primary antibodies was murine hyperimmune serum (dilutions ranged from 1: 100 to 1:4000). In these experiments, the secondary antibody used was an anti­mouse IgG-HRPO conjugate. The duration of incuba­tion (with primary as well as secondary antibodies) was I h. The rest of the assay details were the same as described above.

Results

Design of a nore! [gG-specific dengue mu/tiepitope protein (r-DME-G protein)

In an effort to create a muitiepitope protein that could be of diagnostic utility, we focused on short, linear immunodominant dengue epitopes known to primarily eiicitlbind JgG class of either monoclonal or polyclonal anti-dengue virus antibodies. The locations of these epi­topes on the dengue virus polyprotein and their sequences are shown in Fig. 1. We identified 15epitopes, eight from the E protein, six from NSl, and one from NS3. on the basis of the criteria above. These epitopes ranged in length from 6 to 20 aa residues, and were drawn from dengue serotypes I (epitope 9), 2 (epitopes 1-8 and 10). and 4 (epitopes 11-15). We were unable to include epitopes of dengue virus type 3, as these are not documented in the available literature. Several of the epitopes, for example, epitopes 1,3,5 [I8J, 10 (23J, and 12 [21J were reported to show cross-reactivity towards alI four serotyes, whereas some, such as epitopes 4 [20J and 9 (24] were specific to single serotypes. We designed a synthetic gene, codon-optimized for expression in E. coli, encoding these epitopes in a tandem array. This gene, designated rDME-G (recombinant dengue multiepitope, IgG-specific), is predicted to encode a ~25 kDa recombi-

..... 0-..., . ~ ..., t 15

* NS4A

Ie I prM E NSI I I I NS3 I I NS5 ] ** ****** 12 345678

/jjl\\~ NS2B NS4B

Fig. I. Location of the chosen epitopes on the dengue virus polyprotein. The picture is a schematic representation of the dengue virus polyprotein. The asterisks indicate the epitopes, which are numbered 1-15. chosen for incorporation into the r-DME·G protein. The arrows point to the precise amino acid residue numbers (shown in italics) corresponding to each of these epitopes. Abbreviations used are: E, envelope; NSI and NS3, non· structural proteins I and 3.

140 R AnandaRao et al. I Protein Expression and Purification 41 (2005) 136-147

A * MGGSETLVTFKNPHAKKQDVVVLGSGGGNLLFTGGGPFGDSYI * IIGVEGGGQLKLNWFKKGSSGGGTAWDFGSLGGVFTSIGGGVI * . .

I TWIGGGSTSLSVGGGVTLYLGAGGGEHKYSWKSGGGDSGCVV SWKNKELKCGGGKFQPESPARLASAILNAGGGLKYSWKTWGKA KGGGFLIDGPDTSECPNERRAGGGWYGMEI RPLSEKEENMVGG GI LEENMEVE IWTREGEKKKLRSHHHHHH

B'-===============================================~

Fig. 2. Design of the r-DME-G protein. (A) Complete amino acid sequence (single letter code) of the r-DME-G protein assembled from the peptides indicated in Fig. 1. The epitope amino acids are shown in normal font and the triglycyllinkers are shown in bold font. In three epitopes, the carboxy terminal residue (indicated by asterisks) was G and formed part of the triglycyllinker. The amino-terminal residues 1-4 and the carboxy-terminal 8 residues, inclusive of the 6x-His tag, shown in italics, are encoded by plasmid-derived sequences. (B) A computer-generated representation of the r-DME-G protein in solution. The protein aa sequence was submitted to the 3D-PSSM web server and visualized using ViewerLite software. The epi­topes are shown in blue and the triglycyllinkers in yellow. Red indicates at-helical regions and white indicates unstructured regions.

nant protein contammg the 15 dengue epitopes described above, with adjacent epitopes separated by tri­glycyllinkers. In some of the epitopes (epitopes 2, 5, and 6) the carboxy-terminal residue was a glycyl residue. In such instances, we inserted two additional glycyl residues to generate the triglycyl linker. The predicted aa sequence of the r-DME~G protein is shown in Fig. 2A. A computer modeling analyses of the structure of the r-DME-G protein is depicted in Fig. 2B. This analysis suggests that the proposed design of the muItiepitope protein would be consistent with a structure that permits easy accessibility of all the constituent epitopes. Based on this it is highly likely that all the epitopes of the r-DME-G protein would be freely available for interac­tion with their cognate antibodies and would therefore contribute to its overall sensitivity and specificity in terms of its reactivity to patient sera.

Expression o/the r-DME-protein

The rDME-G gene was inserted in-frame with the ini­tiator codon and a carboxy-terminal 6x-His tag-encod­ing sequence of the bacterial expression vector pQE60, under the control of an inducible PT5 promoter. This expression vector is depicted in Fig. 3A. This vector was introduced into the E coli host SG 13009 (which harbors the pREP4 plasmid encoding the lad repressor) and transformants were selected in the presence of both ampicillin and kanamycin. Recombinant clones were then analyzed by expression screening, as described (in Materials and methods). In this experiment, induced

cells were lysed by directly boiling in SDS-PAGE load­ing buffer. Fig. 3B shows the induction profile of a typical clone. It is evident from this that induction of rDME-G gene expression using IPTG was accompanied by the appearance of a new "-'25 kDa. band, consistent with the predicted size of the r-DME-G protein. When induced cells were lysed by sonication in a native buffer, separated into supernatant and penet fractions as described before [28], and analyzed by SDS-PAGE, the protein was found to be exclusively associated with the pellet fraction.

Purification o/the r-DME-protein

As the r-DME-G protein was insoluble, we used Ni­NT A affinity chromatography under denaturing condi­tions to purify the protein. A lysate prepared from induced cells was chromatographed on a Ni-NTA resin column under denaturing conditions (in the pres­ence of 8 M urea throughout). Fractions collected dur­ing different steps of the purification were analyzed by SDS-PAGE as shown in Fig. 4A. Under the experi­mental conditions employed, all of the induced protein bound to the column, as evident from a comparison of the polypeptide profile of the initial lysate loaded on the affinity matrix (lane 2) with that of the ftow­through material (lane 3). An extensive wash at pH 6.3 eliminated most of the contaminating proteins (lanes 4-6). A further washing step (at pH 5.9) ensured that there was no further discernible non-specific contami­nants bound to the column (lane 7). In fact, we could

R. Anomia Rao et af. I Protein Expression and PurificaTion 41 f 2005) 136-147 141

A B 1 2 3 pT5 BamID 94.0. L 6;.0 • .. - ,,"

~~gl1I 43.0 • \ rDME , 6x-His tag t

II To TT 30.0 • \

"'---..;;,.# \\p. 20.0 •

14.0 •

Fig. 3. Expression of r-DME-G protein in E. coli. (A) Map of the expression plasmid. The synthetic r-DME-G gene (rDME) was inserted into the BamHI and Bgm sites of plasmid pQE60 in-frame with the vector provided initiator codon and the hexa-histidine tag (6 x-His tag). Other abbrevia­tions are as follows: pT5. phage T5 promoter: ToTT. A to transcriptional terminator: Ori and AmpR, plasmid replication origin and ampicillin resis­tance marker. respectively. (B) Coomassie stained denaturing gel showing the polypeptide profiles of E. coli harboring the plasmid in A, before (lane 2) and after (lane 3) IPTG induction. Protein molecular weight markers were run in lane I; their sizes in kilodalton shown to the left. The arrow at the right indicates the position of the rDME-G protein.

A 1 2 3 4 5 6 7 8 9 10 1112 13 14 15 16 B 113.0 • 1 2 3 4

~I' ' .. = 67.0. 92.0· .'"!

43.0. 1 IR 52.3 •

30.0 • ", Gil •• ----.. /IIri, i 35.0· ....

20.0 • \) , 28.9- .. .-

14.0 • \t ~! 21.0. ... .-l .;

Fig. 4. Purification and characterization of the r-DME-G protein. (A) SDS-PAGE analysis of Ni-NT A affinity column fractions obtained during the purification of r-DME-G protein. Samples analyzed in the gel are load (lane 2), flow through (lane 3). buffer 1 wash (lanes 4-6). buffer 2 wash (lane 7), and pH 4.5 eluates (lanes 8-16). Protein molecular weight markers were run in lane I: their sizes in kilodalton are shown to the left. The arrow at the right indicates the position of the rDME-G protein. (B) Western blot analysis of the purified r-DME-G protein using penta-His mAb. An aliquot of the purified protein (taken from a pool corresponding to lanes 7-15 in panel A) was run in lane 2. Negative (a protein v.;thout His tag. BSA) and pos­itive (rED2-m. a protein with 6x-His tag: [28]) controls were run in lanes 3 and 4, respectively. Protein molecular weight markers were run in lane I; their sizes in kilodalton are shown to the left. The arrows at the right indicate the positions of the rDME-G protein and the rD2E-IIT protein.

discern a trace amount of the recombinant protein emerging in the pH 5.9 wash. Elution with a pH 4.5 buffer resulted in the emergence of highly purified recombinant protein from the column (lanes 8-15). From a comparison of the protein profiles of the eluted materiaJ and the crude lysate (lane 2), it is clearly evi­dent that >95% purity has been achieved. Starting from ~ 1 g induced cell pellet (equivalent to 250 ml E. coli culture), we obtained ~7mg purified recombinant pro­tein (pooled fractions in lanes 8-15). This corresponds to a recovery of 58% as the crude lysate was estimated to contain ~ 12 mg of the recombinant protein, based on densitometric analysis (Table 1).

Next, we tested the purified protein in a Western blot assay, to confirm its identity as the r-DME-G protein. In this blot, the affinity purified protein was probed with a commercially available murine penta His mAb, specific to the engineered 6x-His tag on the recombinant pro­tein. In this blot, we also included another 6x-His tagged recombinant protein, described before [28], as a positive control. As a negative control (a protein lacking the tag),

Table 1 Summary ofr-DME-G purification from I L of induced culture

Purification step Total protein Purity Recovery (mg)' (%)b (%t

Crude IYS3ted 308 0 100 Ni-NT A affinity 28 95 58

chromatography

a Protein content was determined by the Bio-Rad method. using BSA as standard.

b Purity was assessed by SDS-PAGE analysis. C The amount of r-DME-G protein in the crude lysate (which was

estimated to be ~ 12 mg by densitometric analysis) this was designated as 1000/0.

d The purification was done using a 250 mI culture (wet cell weight of~l g).

we used bovine serum aJbumin (BSA). It is evident from the data shown in Fig. 4B that, the penta His mAb which specifically recognizes the 6 x -His tagged positive con­trol protein (compare lanes 3 and 4), also recognized ~25 kDa recombinant protein purified as described in the experiment shown in Fig. 3A.

142 R. AnandaRao et at. I Protein Expression and Purification 41 (2005) 136-147

We undertook a systematic investigation of the effect of urea and pH on the solubility of the r-DME-G pro­tein. To this end aliquots of the purified protein were dialyzed against two series of buffers. In the first series, we used pH 8 buffers with urea concentrations ranging from 0 to 8 M. The dialyzed material was clarified by centrifugation and both supernatant and pellet fractions (after re-suspension in initial volumes) were analyzed by SDS-PAGE. In the second series, we tested buffers of different pH (ranging from 4 to 12) in the absence of urea. Results from these experiments showed that at pro­tein concentrations of ~0.25mg/ml at pH 8, it was neces­sary to have a minimum urea concentration of 1 M to maintain the protein in soluble form. In the absence of urea, a significant proportion of the protein was found in the soluble phase in acidic (pH 4-5: up to ~80%) and alkaline (pH 9-11: up to ~50%) buffers (data not shown).

Murine and human anti-dengue antibodies recognize the r-DME-G protein

The r-DME-G protein was expressed and purified with the objective of developing it into a possible diag­nostic reagent for the detection of dengue infections. Therefore, to obtain proof of principle we tested the protein against murine dengue-specific hyperimmune serum and human patient sera. The murine hyperim­mune serum was tested in an ELISA format. We first ascertained that the murine serum does indeed contain dengue virus-specific antibodies by testing it first in an ELISA using infectious dengue type 2 as the capture antigen. This is shown in Fig. 5A. In this experiment, microtiter wells were coated with dengue type 2 virions (tissue culture supernatant obtained from dengue type 2-infected mosquito cells) and then incubated with serial dilutions of murine hyperimmune serum. The captured antibodies were then revealed with an anti-

A B

mouse IgG-HRPO conjugate. The data clearly proved the presence of dengue type 2-specific antibodies in the murine hyperimmune serum, but not in normal (un­immunized) mouse serum. Having confirmed the pres­ence of dengue-specific antibodies in the hyperimmune serum, we repeated the ELISA using r-DME-G as the coating antigen instead of dengue virions. The results of this experiment, shown in Fig. 5B, clearly mirrored the results in Fig. 5A. Absorbance values in general were somewhat higher, in the latter experiment, due to rela­tively higher background. It is likely that this is presum­ably a reflection of the use of crude tissue culture lysate, in the experiment shown in Fig. 5A, as the source of dengue virions for coating the microtiter wells. The ELISA absorbance values after correcting for the back­ground (i.e., after subtracting absorbance value obtained using normal mouse serum) are very similar .~.

regardless of whether the capture antigen is the virion or our recombinant protein, leading us to the conclu­sion that the r-DME-G protein is as efficient as the den­gue virion in detecting anti-dengue antibodies (with the added advantage that the assay background is much lesser).

Having shown that the r-DME-Gprotein can indeed recognize and bind anti-dengue antibodies present in murine sera, we next tested human sera. For this pur­pose, we carried out an immunoblot experiment. In this assay, the r-DME-G protein electrophoresed and elec­trotransferred onto a nitrocellulose membrane, was probed separately with two different human serum sam­ples. Of these two, one tested negative and the other pos­itive using the commercially available Dengue Duo Rapid test (PanBio). This test detects the presence of anti-dengue antibodies in serum (see below). These results, presented in Fig. 5C clearly show that the den­gue-positive patient serum (lane 2) recognized and bound to the r-DME-G protein in the blot, while the dengue-negative serum (lane 1) did not. From this it is

c 1.8 1.0 ,-------------,

e = = 1.2 l/') ...,. - 0.5 = ~ 0.6

0 0.0

0.1 1 f).f)

2 4 0.1 1 2 4 1 2

Serum dilution (103)

Fig. 5. Anti-dengue virus antibodies recognize the r-DME-G protein. (A) Detection of dengue virus-specific antibodies by ELISA in hyperimmune murine serum using infectious dengue virus type-2 as the capture antigen. Antibodies captured from murine hyperimmune serum (black bars) and normal mouse serum (gray bars) were revealed using anti-mouse IgG-HRPO conjugate. (B) The same ELISA as in panel A was carried out with the single difference that, the r-DME-G protein was used as the capture antigen instead of infectious dengue virus. (C) Western blot analysis of r-DME­G protein. The r-DME-G protein was probed using human sera that were either negative (lane I) or positive (lane 2) for the presence of dengue-spe­cific antibodies (as determined by the Dengue Duo test). Antigen/antibody complexes on the blot were visualized using anti-human IgG-peroxidase conjugate.

R AnandaRao et al.1 Protein Expression and Purification 41 (2005) 136-147 143

evident that the r-DME-G protein displays specific reac­tivity towards anti-dengue antibody-containing human sera. This suggests that the r-DME-G protein could indeed be used to detect the presence of anti-dengue antibodies in patient sera and could therefore be exploited as a diagnostic tool.

The r-DME-G protein as a dengue diagnostic reagent

We next examined if this protein could indeed serve to detect dengue infections in patient sera. To this end, we obtained human sera from suspected dengue patients in Sri Lanka. These sera were first tested for the presence of dengue antibodies using the Dengue Duo commercial test, mentioned above. This test uses a mixture of puri­fied recombinant E proteins, of the four dengue virus serotypes, complexed to gold-labeled anti-dengue virus monoclonal mAb, to detect polyclonal dengue virus-spe­cific antibodies of both IgM and IgG classes, in patient sera. The 20 sera tested could be differentiated into two groups of 10 samples each, with one group being IgG­negative (sera# 1-10) and the other, IgG-positive (sera# 11-20). However, within each of these groups, the IgM status varied, with some being positive and the rest being negative (Table 2).

Table 2 Detection of dengue virus-specific IgG in patient sera

Serum #

I 2 3 4 5 6 7 8 9

10 11 12 I3 14 15 16 17 18 19 20

IgG status

PanBio·

+

+ + + + + + +

+ +

In-house ELISA b

0.29 0.32 0.25 0.20 0.28 0.21 0.38 0.29 0.36 0.29 1.51 252 1.19 2.28 2.52 0.92 0.99 1.09 2.44 1.72

IgMa

+ + + +

+ + + + + +

a The presence of dengue virus-specific IgG determined using the commercially available Dengue Due Test from PanBio Pty, Australia. This test detects the presence of both IgG and IgM (shown in the fourth column) antibodies in dengue patient sera, simultaneously. The "+" and "-" signs indicate the presence and absence. respectively. of dengue virus-specific antibodies as determined by the PanBio Test.

b Dengue virus-specific IgG levels were determined by an in-house ELISA in which the purified r-DME-G protein was used as a capture antigen. The values, shown in this column, indicate ELISA absor­bances measured at 450 nm.

Using the Dengue Duo test results as reference, we next proceeded to test if the r-DME-G protein could be used to detect the presence of dengue virus-specific anti­bodies in the panel of patient sera. To this end, we designed an in-house ELISA wherein microtiter wells were coated with r-DME-G protein to capture dengue virus-specific antibodies in patient sera. Captured antibodies if any, in the sera, were revealed using anti­human IgG-HRPO conjugate. The results, in absor­bance values (OD450nm) are presented alongside the Dengue Duo test results in Table 2. The 10 samples (sera# 1-10), which were IgG-negative in the PanBio test, displayed absorbance values in the range of 0.20-0.38 OD450nm• These values were quite comparable to the negative controls run in parallel in which we omitted the addition of the primary antibody. This leads us to con­clude that these 10 sera samples do not contain dengue­specific IgG antibodies. However, of these 10, 4 samples (sera# 6-9) tested IgM-positive, using the Dengue Duo test.

The remaining 10 samples (sera# 11-20) displayed significantly higher reactivities to the r-DME-G protein as evidenced by the relatively high ELISA absorbance values (ranging from 0.92 to 2.52 OD450nm), suggesting that these sera do contain dengue-specific (IgG) antibod­ies. That this indeed is the case is borne out by the Den­gue Duo test data, which show that all these 10 samples scored positive for dengue-specific IgG. Of these 10 IgG­positive samples, six (sera# 11-16) were also positive for dengue IgM antibodies (Dengue Duo test data). It is conceivable that the high ELISA reactivities of these six samples may reflect the outcome of IgM interference. However, this appears to be unlikely based on the obser­vation that the four IgM-positive serum samples (sera# 6-9), referred to above, did not show any ELISA reactiv­ity to the r-DME-G protein, suggesting that it is indeed IgG-specific.

Discussion

DF is a mosquito-borne viral disease posing a rapidly expanding public health threat in many areas of the world. The high mortality associated with DHF and DSS can be significantly minimized through timely med­ical care, which in turn depends on accurate diagnosis of dengue infections. Several dengue diagnostic kits are commercially available [12,13]. However, their wide­spread use is hampered by several factors, stemming from the use of whole virus preparations as antigens to detect anti-dengue antibodies in patient sera. First, the use of whole virus antigens in these commercial kits poses a biohazard risk. Second, the vimses are usually produced in tissue culture or suckling mice brain. The high costs associated with virus antigen production make these kits unatfordable, particularly in the

144 R. AnandaRao et al.! Protein Expression and Purification 41 (2005) 136-147

economically weaker dengue endemic countries. Finally, the whole virus antigens invariably pick up antibodies against other flaviviruses such as yellow fever and Japa­nese encephalitis viruses, leading to ambiguity in diagno­sis. Further, false positives are also known to arise with sera from malaria, typhoid, and leptospirosis patients. Thus, it is apparent that an ideal antigen for use in a den­gue diagnostic test must not only be free of the virus associated biohazard risk, it must also be inexpensive to produce, and possess a high degree of specificity, to facil­itate unequivocal diagnosis of dengue infections.

The present work is based on the premise that the use of a synthetic antigen, designed to be dengue-specific, and expressed to high levels in an E. coli-based expres­sion system could effectively address the issues of bio­hazard risk, cost, and specificity, associated with whole virus antigen-based diagnostic assays. Accordingly, we designed a synthetic antigen by splicing together dengue virus-specific epitopes using simple peptide linkers. To circumvent serologic cross-reactivity associated with whole virus or viral envelope proteins [13], we elimi­nated epitopes known to cross-react with sera from patients with diseases such as yellow fever and Japanese encephalitis. To develop this synthetic antigen, we focused on three proteins expressed by dengue viruses, namely, E, NSI, and NS3. The ability of these three pro­teins to elicit humoral immune responses is well-docu­mented and their antigenic determinants identified using a variety of approaches [16-26]. Using the information available from these studies, we chose epitopes for incor­poration into a synthetic antigen. The choice of epitopes was based on three criteria, namely, they had to be (i) immunodominant, (ii) specific to IgG class of anti-den­gue antibodies, and (iii) linear. These criteria were based on the following rationale. First, in order for the syn­thetic antigen to be capable of efficiently recognizing dengue virus-specific antibodies, it is necessary that its constituent epitopes exhibit significant reactivity to den­gue patient sera/anti-dengue mAbs. Second, the IgG­specificity of the epitopes was a selection criterion so that the r-DME-G protein would have a high sensitivity to pick up anti-dengue IgGs (which are recognized markers of secondary infections), common in dengue­endemic regions. Also, to ensure specificity to dengue alone, epitopes that displayed cross-reactivity towards other members of Flaviviridae were eliminated. Finally, it was necessary to work with linear epitopes so that when incorporated into the synthetic protein, they would presumably retain their immunoreactivity. On the other hand, if conformational epitopes were to be used in constructing the synthetic antigen, it is highly unlikely that they would retain their conformational integrity, and therefore their immunoreactivity, in the synthetic antigen.

We selected a total of 15 epitopes in accordance with the criteria discussed above. These ranged in length from

6 to 20 aa residues; eight were from the E protein and the remaining from the non-structural proteins NSI and NS3. These were from dengue virus serotypes 1,2, and 4. Next, we created the synthetic gene, r-DME-G, encoding all 15 epitopes in which adjacent epitopes were separated by triglycyl linkers. We used glycine in the linkers as it can provide flexibility stemming from its lack of a ~-car­bon, and is considered as one of the preferred linker resi­dues while designing chimeric proteins [29]. Computer modeling analyses showed that the protein encoded by the r-DME-G gene adopts a structure in which all the chosen epitopes are freely accessible. This suggested that all the 15 chosen epitopes would be able to collectively contribute to the overall specificity of the molecule. We engineered a 6 x -His tag-encoding sequence at the 3' end of this gene to facilitate one-step affinity purification of the expressed protein. The recombinant protein, r-DME­G, was expressed to high levels in E. coli under the con­trol of an IPTG-inducible promoter. As seen often with overexpressed proteins in E. coli, the r-DME-G protein was present exclusively in the insoluble phase of the cell lysate. As it is known that fusion to the maltose-binding protein (MBP) can promote the solubility of some pro­teins [30], we constructed an r-DME-G-MBP fusion. This strategy failed to produce soluble protein (data not shown). Therefore, we purified the r-DME-G protein on a Ni-NTA matrix, under denaturing conditions. Analy­sis by SDS-PAGE showed that a high degree of purity, >95%, had been achieved in a single step. Our data show that starting from a one liter culture of induced cells, >25 mg of purified protein, representing a ~58% recov­ery from the crude lysate, can be obtained. Purified r­DME-G protein could be detected in a Western blot using anti-His mAb, consistent with its ability to bind Ni-NT A matrix.

As proof of principle, we tested the r-DME-G protein in an in-house ELISA. In this assay, we used our recom­binant protein as the capture antigen and dengue virus type-2 hyperimmune mouse serum as the test sample. We then used anti-mouse-HRPO/TMB substrate to examine if the r-DME-G protein had successfully cap­tured anti-dengue virus antibodies from the murine serum. The results showed that indeed our synthetic antigen could recognize and bind anti-dengue antibodies as evidenced by the ELISA readouts. Further, the r-DME-G protein could interact with the murine anti-dengue antibodies as efficiently as the whole virus, justifying our design of the recombinant protein. Impor­tantly, the capacity of the r-DME-G protein to manifest immunoreactivity towards dengue virus-specific anti­bodies was corroborated using dengue patient sera in a Western blot experiment. This suggested that indeed the r-DME-G protein could have a potential use as a diag­nostic reagent.

Having ascertained the ability of the r-DME-G pro­tein to specifically detect anti-dengue virus antibodies,

R AnandaRao et al. I Protein Expression and Purification 41 (2005) 136··147 145

we next proceeded to evaluate its potential utility as a dengue diagnostic reagent. To this end, we used 20 sera samples drawn from patients in Sri Lanka suspected of dengue infection in the in-house ELISA described above. Based on the resultant ELISA absorbances, we could distinguish two groups. The first group, consisting of 10 sera, was characterized by low reactivity with 00450 val­ues in the range 020-0.38. Negative controls, run in par­allel, in which we either omitted adding any patient serum or those in which we coated the ELISA wells with unrelated proteins, resulted in quite similar absorbance values. On the basis of this, we conclude that these 10 samples are seronegative for anti-dengue virus antibod­ies. This conclusion is substantiated by the observation that each one of these samples tested seronegative using the commercially available Dengue Duo test. In striking contrast, the second group of 10 samples displayed sig­nificantly high 00450 values (range: 0.92-2.52). Clearly,

A 1

these samples are seropositive for anti-dengue antibod­ies. Once again this conclusion was in excellent agree­ment with the PanBio Test results. We conclude that, with reference to the PanBio Test, the in-house ELISA displayed 100% sensitivity, scoring 20 out of 20 sera accurately.

We had designed the r-DME-G protein using IgG­specific epitopes. An analysis of our ELISA results presented below suggests that, indeed. the r-DME-G protein specifically recognizes IgG antibodies, even in the presence of detectable IgM antibodies. As mentioned earlier, the PanBio Test scores for both IgG and 19M antibodies. Examination of the IgM status of our panel of patient revealed that 4 out of 10 sera were determined to be IgG-seronegative by the PanBio Test scored posi­tive for IgM. This suggests that the r-DME-G protein does not recognizelbind to IgM antibodies. Further, 6 out of 10 sera, which manifested significant ELISA

1 3

02 DI 03 D4 ~

T L V T F K N P H A K K Q," D V V V L G S G G G N L L F G1 D' S Y 1'1 ! I IJ ,DLLVTFKTAHAKK~EIVVVLGSGGGTTI FG,E'SYI V ELL V T F K N A H A K,K. Q. E I v V V L G S G G G T S I F G1 E I S lliJl V E R V T F K V P H A K; R D V T V L G S G G G N H .• "F:....:.A~G'-G"'-'G ........ P-=-F_G""I'-'D:o.li-=s......:..Y-'I'--''''-'·

D2 DI 03 D4

02 DI 03 D4

B

D4 DI 02 D3

D4 DI 02 D3

D4 Dl D2 03

4 5

. Gt V EGG Glq: L K L 1\' W F. KfK G S

vj~ A!G G G GAL K L S W F.: KIK G S I Gi I G G G GAL I K ~RIK G S [ ol V G G G GAL: T iL H W F R:K G S

SGGGTAWDF6SLGGVFTSI GGG S G G G TAW D F G SII IG G V F T S V G G G S G G G T A WD F G S. v;G G l N S L G G G S G G G TAW D F G S! V'G G L F T S L G G G

6 7 8

11 12 13

WKTWGKAKGGGFLI DG WK[ilWGKAKGGGFI I DG WKTWGKAKGGGfLI DG

~~:::::.....:~!....!:::-!!o-!...-",:-,..:.:.-'::.......:......:.w,-=G L A K G G G F I I D G

PID~TlsjE C P'iN E R RAGGGWYGMEf P'I\ITP\E C P'D N Q RAGGGWYGME I

pIE!T0E C P'N pI S T.' P [E C P: s

TN RAGGGWYGMEI AS RAGGGWYGME

W Tl~E G E' KKK L WTKEGf:RKKL WT KEG E,R K K l

I W T K I: G E. K. K K l

f

R P R P R P R P

15

L S EKE E N MVGGGl L E E N V K EKE E N L V G G G(YlL E E N L K EKE E N LVGGGI L E f N I NEKEE" ;M V G G G I lEE K

Fig. 6. Comparison of r-DME-G protein epitopes a(.TOSS all four dengue virus serotypes. (A) Multiple sequence alignment of the amino-terminal half of the r-DME-G protein. The sequence on top corresponds to the amino-tenninal half of the r-DME-G protein (created in this study) comprised of eight epitopes taken from the E protein of dengue virus type 2 (02). Shown below this are the predicted sequences of putative multiepitope proteins in which, the D2-derived E epitopes are replaced by those from dengue serotypes 1(01),3 (03) or 4 (D4). (B) Multiple sequence ali!1cnment of the car­boxy-terminal portion of the r-DME-G protein. The sequence on top corresponds to the carboxy-terminal portion of the r-DME-G protein (created in this study) comprised of five epitopes taken from the NS1 protein of dengue virus type 4 (04). Shown below this are the predicted sequences of putative multiepitope proteins in which, the D4-derived NSI epitopes are replaced by those from dengue serotypes 1(01),2 (02) or 3 (D3). In both panels, the epitopes are identified by the lines above and are numbered in italics. Epitope numbers correspond to those shown in Fig. 1. Identical res­

idues are shown in dark grey and similar residues in lig.ht grey boxes.

146 R. AnandaRao et al. / Protein Expression and Purification 41 (2005) 136-147

reactivity in the in-house assay, tested positive for both IgM and IgG using the PanBio Test. This observation leads to the conclusion that the ELISA reactivity is attributable to the recognition of IgG antibodies by the r-DME-G protein. In addition, the r-DME-G protein was designed to eliminate false positives by incorporat­ing only dengue virus-specific epitopes that did not man­ifest any cross-reactivity with antibodies specific to other flaviviruses. Consistent with this design, ELISAs carried out with r-DME-G as the coating antigen did not result in any measurable reactivity towards sera from typhoid and yellow fever patients. In contrast, these sera were scored positive in an ELISA test using dengue virus as the capture antigen (data not shown).

In designing the r-DME-G protein, most of the epi­topes were drawn from the E protein of dengue virus type 2 (eight epitopes) and the NSI protein of dengue virus type 4 (five epitopes). This is because E and NSI are the most immunogenic of the dengue virus proteins and information available in the literature pertains to only serotypes 2 and 4. The question this raises is: will the r-DME-G protein predominantly recognize antibod­ies to only serotypes represented by its constituent epi­topes or would it be specific to all four serotypes? The non-availability of patient sera representing all four serotype-specific infections precluded direct experimen­tal verification. However, we compared corresponding epitopes of other serotypes with those incorporated in the r-DME-G protein to predict a solution to this ques­tion. Epitopes 1 through 8 in the r-DME-G protein, occupying the amino terminal half of the molecule, are taken from the E protein of dengue virus type 2. If we were to replace these 8 epitopes of dengue virus type 2 with the corresponding epitopes taken from the E pro­teins of the dengue serotypes 1, 3, and 4, and perform a Clustal analysis of all four sequences, it is immediately apparent that there is a high degree of homology amongst the E epitopes of the four serotypes (Fig. 6A). A similar analysis of the NSI epitopes at the carboxy­terminal region of the r-DME-G protein, leads to the same conclusion, that is, the r-DME-G protein is highly representative of all four serotypes. Based on this we anticipate that our multiepitope protein would be equally specific towards all four serotypes. This is enforced by our observation that one of the patient sera which tested positive for dengue type 3 infection by R TPCR also showed significant reactivity towards the r­DME-G protein (data not shown).

In conclusion, the high density of the epitopes in the recombinant dengue multiepitope (rDME) protein and the careful choice of only dengue-specific epitopes as its components contribute to a high degree of sensitivity and specificity. Further, our strategy of using a recombinant multiepitope protein completely obviates multiple peptide synthesis and multiple protein expression; it also avoids expensive and time-consuming virus culture (for antigen

preparation) and the associated biohazard risk. The design of the rD ME protein and the ease of its expression and purification have the potential to make this a highly cost-effective approach to dengue diagnosis. In this regard, Huang et al. [31] have reported that recombinant NSI protein, produced by a genetic engineering approach, also may be valuable in developing a safe and cost-effective dengue diagnostic test. In addition to ELISA format we have used in this study, the r-DME-G protein can be adapted for use in alternative formats. For example, anti­dengue antibodies can be captured using unlabeled r­DME-G and revealed using gold-labeled r-DME-G. This could be incorporated into a spot test or a lateral flow test which are rapid and amenable to field use. We are devel­oping a similar approa(;h for the detection of dengue-spe­cific IgM in primary infections. This approach has the potential for the simultaneous detection of multiple infec­tious diseases. Lastly, this approach has general applica­bility to the detection of infection by pathogens.

Acknowledgments

This work has been supported by institutional core funds and a grant from the Ministry of Defense, Gov­ernment of India. R.A. is a Senior Research Fellow sup­ported by the Indian Council of Scientific and Industrial Research. .

References

[1] B.D. Lindenbach, CM. Rice, Flaviviridae: The viruses and their repli­cation, in: D.M. Knipe, P.M. Howley (Eds.), Fields Virology, fourth ed., Lippincott Williams & Wilkins, Philadelphia, 200 I, pp. 991-1041.

[2] DJ. Gubler, Dengue and dengue hemorrhagic fever, Clin. Micro­bioI. Rev. 11 (1998) 480-496.

[3] S. Hales, N. de Wet, J. Maindonald, A. Woodward, Potential effect of population and climate changes on global distribution of den­gue fever: An empirical model, Lancet 360 (2002) 830-834.

[4] World Health Organization, Dengue Haemorrhagic Fever: Diag­nosis, Treatment, Prevention and Control, second ed., Geneva, Switzerland, 1997.

[5] R. George, L.CS. Lum, Clinical spectrum of dengue infection, in: DJ. Gubler, G. Kuno (Eds.), Dengue and Dengue Hemorrhagic Fever, CAB International, Wallingford, UK, 1997, pp. 89-113.

[6] A. Igarashi, Impact of dengue virus infection and its control, FEMS Immunol. Med. Microbiol. 18 (1997) 291-300.

[7] S. Nimmannitya, Dengue hemon'hagic fever: Diagnosis and manage­ment, in: DJ. Gubler, G. Kuno (Eds.), Dengue and Dengue Hemor­rhagic Fever, CAB International, Wallingford, UK, 1997, pp. 133-145.

[8] P.Y. Shu, J.H. Huang, Current advances in dengue diagnosis, Clin. Diagn. Lab. Immunol. 11 (2004) 642-650.

[9] K. Lemmer, 0.0. Mantke, H.G.. Sae, J. Groen, C Drosten, M. Niedrig, External quality control assessment in PCR diagnostics of dengue virus infections, J. Clin. Virol. 30 (2004) 291-296.

[10] D.W. Vaughn, S. Green, S. Kalayanarooj, B.L. Innis, S. Nimman­nitya, S. Suntayakorn, T.P. Endy, B. Raengsakulrach, A.L. Roth­man, F.A. Ennis, A. Nisalak, Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease sever­ity, J. Infect. Dis. 181 (2000) 2-9.

R. AnandaRilo e/ al.I Pro/em Expression and Purification 41 (2005) 136-147 147

[11] K.I. Yamada, T. Takasaki. M. 1'awa, I. Kurane. Virus isolation as one of the diagnostic methods for dengue ,irus infection . .I. Clin. Viro!. 24 (2002) 203-209.

[12] 1. Groen, P. Koraka. 1. Velzing. C. Copra. A.D.M.E. Osterhaus. Evaluation of six immunoassays for detection of dengue ,;rus­specific immunoglobulin M and G antibodies. Clin. Diagn. Lab. Immuno!. 7 (2000) 867-871.

[13] AJ. Cuzzubbo, T.P. Endy. A Nisalak. S. Kalyanarooj, D.W. Vau­ghn, S.A Ogata. D.E. Clements. P.L. Devine. Use of recombinant envelope proteins for serological diagnosis of dengue virus infec­tion in an immunochromatographic assay. Clin. Diagn. Lab. Immuno!. 8 (2001) 1150-1155.

[14] D.W. Vaughn, lmited commentary: Dengue lessons from Cuba, Am. J. Epidemio!. 152 (2000) 800-S03.

[15] R.I. Kuhn, W. Zhang, M,G. Rossmann. S.V. Pletnev.l. Corver. E. Lenches, c.T. lones. S. Mukhopadhyay, P.R. Chipman. E.G. Strauss, T.S. Baker, 1.H. Strauss, Structure of dengue virus: Impli­cations for fla,;>irus organization. maturation. and fusion. Cell 108 (2002) 717-725.

[16] V. Churdboonchart. N. Bhamarapravati. S. Peampramprecha. S. Sirina>in, Antibodies against dengue ,iral proteins in primary and secondary dengue hemorrhagic fever. Am. J. Trop. Med. Hyg. 44 (1991) 481-493.

[17] B.L. Innis, Antibody responses to dengue >irus infection. in: DJ. Gubler, G. Kuno (Eds.), Dengue and Dengue Hemorrhagic Fever, CAB International, Wallingford, UK. 1997, pp. 221 243.

[18] B.L. Innis, V. Thirawuth, C. Hemachudha. Identification of con­tinuous epitopes of the envelope glycoprotein of dengue type 2 virus, Am. J. Trop. Med. Hyg. 40 (1989) 676-687.

[19] F. Megret, J.P. Hugnot, A Falconar, M.K. Gentry. D.M. Morens, 1.M. Murray, J.1. Schlesinger, P.J. Wright. P. Young. M.H.V. van Regenmortel. V. Deubel. Use of recombinant fusion proteins and monoclonal antibodies to define linear and discontinuous anti­genic sites on the dengue envelope glycoprotein. Virology 187 (1992) 480-491.

[20] T. Trirawatanapong. B. Chandran. R. Putnak. R. Padmanabhan. Mapping of a region of dengue virus type-2 glycoprotein required for binding by a neutralizing monoclonal antibody, Gene 116 (1992) 139-150.

[21] A.K.I. Falconar, P.R Young. M.A Miles. Precise location of sequential dengue ,irus subcomplex and complex B cell epitopes on the nonsttuctural-I glycoprotein. Arch. Viro!' 137 (1994) 315-326.

[22] G. Garcia, D.W. Vaughn. R.M. del Angel, Recognition of syn-thetic oligopeptides from nonstructural proteins l\'S I and NS3 of dengue-4 \irus by sera from dengue virus-infected children, Am. J. Trop. Med. Hyg. 56 (1997) 466-470.

[23] l.H. Huang, J.J. Wey, Y.c. Sun. C. Chin, L.J. Chien. Y.c. Wu, Antibody responses to an immunodominant nonstructural I syn­thetic peptide in patients with dengue fever and dengue hemor­rhagic fever, J. Med. Viro!. 57 (1999) I~.

[24] H.C. Wu. Y.L Huang, T.T. Chao, J.T. Jan, J.L. Huang. H.Y. Chiang. c.c.. King. M.F. Shaio. Identification of B-cell epitope of dengue \TIUS type I and its application in diagnosis of patients, J. Clin. Microbio!. 39 (2001) 977-982.

[25] K. Valdes. M. Alvarez. M. Pupo. S. Vazquez. R. Rodriguez, M.G. Guzman, Human dengue antibodies against structural and non­structural proteins. Clin. Diagn. Lab. Immuno!. 7 (2000) 856-857.

[26] ZJ. Yao. M.c..c.. Kao, K.c. Loh, M.C.M. Chung, A serotype-spe­cific epitope of dengue ,,;rus I identified by phage displayed ran­dom peptide library, FEMS Microbio!' Lett. 127 (1995) 93-98.

[27] u.K. Laemmli. Cleavage of structural proteins during the assem­bly of the head of bacteriophage T4, Nature 227 (1970) 680-685.

[28] S. laiswaL N. Khanna, S. Swaminathan. High-level expression and one-step purification of recombinant dengue virus type 2 envelope domain In protein in Escherichia coli, Protein Exp. Purif. 33 (2004) 80- 91.

[29] c..R. Robinson.. R.T. Sauer, Optimizing the stability of single­chain proteins by linker length and composition mutagenesis, Proc. Natl. Acad. Sci. USA 95 (1998) 5929-5934.

[30] R.B. Kapust. D.S. Waugh. Escherichia coli maltose-binding pro­t.ein is uncommonly effective at promoting the solubility of poly­peptides to which it is fused. Protein Sci. 8 (1999) 1668-1674.

[31] J.L. Hu.ang. J.H. Huang, R.H. Shyu, C.W. Teng. Y.L. Lin, M.D. Kuo, c.. W. Yao, M.F. Shaio. High-level expression (If recombinant dengue viral NS-I protein and its potential use a.s a diagnostic antigen.. J. Med. Virol. 65 (2001) 553-560.

~IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII