1 Introduction

1.1  Structure of hantaviruses

▼ 1 

Hantaviruses are spherical, enveloped RNA viruses with a diameter of 78-210 nm and belong to the family of Bunyaviridae[reviewed in Schmaljohn, 96]. Their genome consists of three segments of negative-sense, single-stranded RNA that code for three proteins. The large (L) segment encodes the RNA-dependent RNA polymerase (RdRp) and the small (S) segment the nucleocapsid (N) protein. The medium (M) segment codes for the glycoprotein precursor which is cleaved by a cellular protease into the two glycoproteins G1 and G2.

In contrast to other genera of Bunyaviridae, nonstructural proteins have not been described for hantaviruses. However, almost all hantaviruses associated with rodents from the subfamilies Arvicolinae and Sigmodontinae have a second open reading frame (ORF-2) on the S segment [Ulrich, 02]. Murinae associated hantaviruses do not have a second ORF on the S segment. If this second ORF of the Arvicolinae and Sigmodontinae associated hantaviruses, which encodes for a putative 60 – 90 amino acid (aa) long protein, is expressed in infected cells remains to be elucidated. Presence or absence of this second ORF on the S segment does not seem connected to the virulence of the viruses to humans, but might be relevant for the adaptation of the hantavirus to its rodent hosts.

Maturation of the majority of viruses from genera in the family of Bunyaviridae occurs intracellular by budding into the Golgi cisternae [Kuismanen, 85; Ellis, 88; Hobman, 93; Rwambo, 96; Jantti, 97]. Budding virus particles were found in the Golgi compartment in endothelial cells of patients with an epidemic haemorrhagic fever in China that was most probably caused by members of the Bunyaviridae family [Wang, 97]. Therefore, maturation of the hantavirus virions was thought to take place mainly at the Golgi compartment. In contrast, Sin Nombre virus (SNV) and Black Creek Canal virus, both members of the New World hantaviruses, have been found to bud predominantly at the plasma membrane [Goldsmith, 95; Ravkov, 97]. These controversial findings show that further investigations are needed to precisely identify the site and mechanism of budding for the different hantaviruses [Spiropoulou, 01].

▼ 2 

FIGURE 1: Schematic drawing of a hantavirus particle. Hantaviruses are enveloped negative-strand RNA viruses. The virus particle consists of an RNA-dependent RNA-polymerase (RdRp), two glycoproteins (G1 and G2) and the nucleocapsid (N) protein encoded by the three RNA segment, the large (L), the medium (M) and the small (S) segment, respectively. The RNA segments are associated with the N protein.

1.2 Geographic distribution and natural hosts of hantaviruses

Hantaviruses, in contrast to the other, arthropod borne genera of the Bunyaviridae, are transmitted by rodents. In these rodents, their natural hosts, they establish a persistent infection without causing disease [Meyer, 00; Plyusnin, 01a; Plyusnin, 01b]. Hantaviruses show a strong host specificity and interspecific spill over seems to be a rare event. So far about 25 hantavirus species have been identified that are associated with different rodent species (for a selection see Tab. 1).

The transmission of hantaviruses from rodents to humans is thought to occur mainly through aerosols of infected animal excreta, i.e. saliva, urine and faeces. In contrast to the Old World hantaviruses, there are indications for person to person transmission during an ANDV outbreak in Argentina [Padula, 98] and a series of cases in Buenos Aires [Pinna, 04].

1.3 Diseases caused by hantaviruses

▼ 3 

Hantaviruses cause two diseases in humans. Haemorrhagic fever with renal syndrome (HFRS), with a case fatality rate of up to 15 % is caused by Old World hantaviruses. With a lower frequency then HFRS worldwide, hantavirus cardiopulmonary syndrome (HCPS), with a case fatality rate of up to 40 % is caused by New World hantaviruses [for reviews see Schmaljohn, 97; Krüger, 01; Ulrich, 02].

In 1978 it was proven that the hantavirus prototype HTNV was the causative agent of KHF, a severe form of HFRS [Lee, 78c]. The virus had been isolated from the lungs of an A. agrarius coreae captured in the rural endemic areas of Korean haemorrhagic fever (KHF) cases in 1976 [Lee, 78a]. Much earlier however, a war nephritits clinically very similar to the milder form of HFRS occuring in Scandinavia (nephropathia epidemica, NE, see Tab. 1) had been reported among British soldiers stationed in Flanders during World War I [reviewed in Lee, 82a].

TABLE 1: Natural reservoir and geographical distribution of selected hantaviruses and their associated diseases. [Krüger, 01; For a more complete summary of hantaviruses see Hooper, 01c; Khaiboullina, 02]

virus

humandisease

rodent host

(subfamily / species)

distribution

reference

  

Murinae

  

Hantaan (HTNV)

HFRS

Apodemus agrarius

(striped field mouse)

Asia

[Lee, 78b]

Dobrava (DOBV-Af)

(DOBV-Aa)

HFRS

Apodemus flavicollis

(yellow-necked mouse)

Apodemus agrarius

(striped field mouse)

Europe

[Avsic-Zupanc, 92a]

[Klempa, 03]

Seoul (SEOV)

HFRS

Rattus species

(rats)

Asia / worldwidea

[Lee, 82b]

  

Arvicolinae

  

Puumala (PUUV)

NE

Clethrionomys glareolus

(bank vole)

Europe

[Brummer-Korvenkontio, 80]

Tula (TULV)

-

Microtus species

(common voles)

Europe

[Plyusnin, 94]

[Sibold, 95]

  

Sigmodontinae

  

Sin Nombre (SNV)

HCPS

Peromyscus maniculatus

(deer mouse)

North America

[Nichol, 93]

Andes (ANDV)

HCPS

Oligoryzomys species

(rice rats)

Argentina, Chile

[Levis, 97]

HFRS haemorrhagic fever with renal syndrome; NE nephropathia epidemica; HCPS hantavirus cardiopulmonary syndrome
a SEOV has mostly been found in Asia, but occurs world wide

▼ 4 

In Europe mainly two hantaviruses have been found to cause HFRS of different severity in humans [Mustonen, 98b; Sibold, 99a; Plyusnin, 01a]. PUUV is known to cause NE [Brummer-Korvenkontio, 82], with a case fatality rate of up to 0.1 % [Lähdevirta, 82]. In south-east Europe DOBV carried by the yellow-necked field mouse A. flavicollis (DOBV-Af) is responsible for clinically severe HFRS cases with a case fatality rate of up to 12 % [Avsic-Zupanc, 92b; Avsic-Zupanc, 95a; Papa, 01]. Recently, DOBV-Af-like strains (Saaremaa and DOBV-Aa) have been found in the striped field mouse A. agrarius[Nemirov, 99; Sibold, 01a; Klempa, 03]. It has been proposed that mild clinical courses of DOBV infections in central and eastern Europe might be due to infections by those virus strains [Schütt, 01; Plyusnin, 01a; Golovljova, 02; Ulrich, 02; Klempa, 04a; Klempa, 04b].

The clinical features of HFRS are fever, headache, back and abdominal pain, drop in blood pressure, hypotension, and in severe cases haemorrhages, renal failure, shock and cardiovascular collapse. Some of these symptoms are thought to be caused by an increased capillary permeability and vascular leakage, a characteristic phenomenon of HFRS [Kanerva, 98b].

The reasons for the differences in severity of disease and case fatality rate caused by the different hantaviruses are not clear, but seem to be determined by virus- and host-specific factors. A major virulence factor is represented by the G1 protein, as a change of HTNV virulence was accompanied by a change of one amino acid in the G1 protein [Isegawa, 94; Ebihara, 00]. In another study, indications were found that a mutation in the noncoding region of the S segment might be responsible for the infectivity of PUUV in bank voles and cell culture [Lundkvist, 97b]. On the other hand, it has been shown that the human HLA alleles B8, DR3 and DQ2 are associated with a more severe outcome of PUUV infection, whereas HLA allele B27 is associated with a milder outcome of PUUV infection in humans [Mustonen, 98a; Vapalahti, 01].

1.4 Treatment of hantavirus infections

▼ 5 

Hantaviruses can cause severe infections, which in some cases can result in a lethal outcome. Except for treatment with ribavirin, curing hantavirus infections is restricted to a treatment of the symptoms caused by the infection. Ribavirin, which is used against a wide range of RNA viruses can help against hantavirus infections. There seems to be no benifit of a ribavirin treatment for patients infected with New World hantaviruses [Mertz, 04]. For Old World hantaviruses however, it has been shown that treatment with ribavirin decreases virus titres and increases surviving probabilities in suckling mice infected with HTNV [Huggins, 86]. Ribavirin is a guanosine analogue and its incorporation into mRNA is followed by a stop in transcription. Recently, Severson and colleagues have found a higher mutation rate in the S segment mRNA in HTNV infected VeroE6 cells in the presence of ribavirin. They consider this "error catastrophe" as the reason for the antiviral property of ribavirin [Severson, 03]. In vivo there might be an additional antiviral effect of ribavirin as it has been shown that ribavirin alters the T cell balance to the T helper 1 (Th1) subset in hepatitis B virus (HBV)- and hepatitis C virus (HCV)-specific immune response [Hultgren, 98]. If this alteration in T cell balance maybe relevant for ribavirin treatment of hantavirus infections remains to be elucidated.

1.5 Vaccine development

1.5.1  Whole virus vaccines

Because of the therapeutical limitations of infections, a prophylactic vaccine for hantavirus infections is needed. Several killed whole virus vaccines generated in mouse brains or cell culture are commercially produced and licensed for human use in Asia [reviewed by Krüger, 01]. HantavaxTM is a formalin-inactivated HTNV vaccine grown in suckling mouse brains and supplemented with alum gel as adjuvant [Lee, 99]. HantavaxTM seems to be efficient in preventing HFRS: In Korea, the number of hospitalised HFRS cases have dropped by half since HantavaxTM became available, from 1234 cases in 1991 to 687 cases in 1996. In Yugoslavia 2000 people became vaccinated with placebo or HantavaxTM. In the placebo group five cases of HFRS occurred while in the HantavaxTM group no cases of HFRS occurred [both trials reviewed in Lee, 99]. In Asia bivalent HTNV/SEOV vaccines have been developed [Krüger, 01; reviewed by Hooper, 01d]. However, the whole virus vaccines are not licensed outside Asia. There are certain disadvantages of whole virus vaccines. (i) handling of the hantaviruses requires level three safety facilities which complicates the production of a vaccine. (ii) Inactivation of the virus used for the vaccine could be inefficient which makes it obligatory to test each vaccine lot for infectivity. The need for biosafety level precautions and the danger due to ineffective inactivation can be circumvented by producing a subunit vaccine by recombinant technology.

1.5.2 Recombinant proteins as potential hantavirus vaccines

As has been outlined above, a recombinant subunit vaccine against hantaviruses is needed. Even though many antiviral subunit vaccines in clinical trial, so far there are only two on the market. One is the recombinant HBV surface antigen (HBsAg) expressed in yeast [McAleer, 92]. The fact that HBs particles can protect against HBV has been found out by immunisations with non-recombinant HBs particles purified from hyperimmune serum. The other is a non-recombinant influenza subunit vaccine [FluadTM, Chiron, Minutello, 99].

▼ 6 

Previously, hantavirus vaccine candidates have been generated on the basis of recombinant technologies, e.g. naked DNA vaccines, recombinant vaccinia and related poxviruses, and recombinant proteins expressed in transgenic plants, E. coli, yeast and insect or mammalian cells. Taken together, several of these recombinant vaccines based on N or the glycoproteins are able to induce protective immune responses in rodent animal models [Krüger, 01; for reviews see Hooper, 01e].

1.5.3 Recombinant virus-like particles

Non-infectious virus-like particles (VLPs) can be generated by heterologous expression of viral structural proteins and their spontaneous self-assembly. A variety of viral proteins have been used for the development of VLPs [for review see Pumpen, 03]. Bacteriophage coat proteins have been found to have a very limited insertion capacity for foreign protein segments [Pushko, 93; Voronkova, 02]. In contrast, bluetongue virus NS1 tubules, parvovirus B19 and yeast retrotransposon Ty-derived VLPs have been found to tolerate extended insertions of up to 100 - 200 foreign aa [Miyanohara, 86; Adams, 87; Mikhailov, 96].

The surface and the core antigen of the HBV have been used since the 1970s as carrier proteins for the generation of chimeric VLPs. The core protein of HBV (HBc) expressed in bacteria forms shells resembling those in HBV-infected liver cells [Cohen, 82]. Due to its advantageous features HBc has been extensively exploited as a carrier for foreign epitopes [for reviews see Ulrich, 98b; Pumpens, 01]. Indeed, on the basis of HBV core (HBc), highly promising vaccine candidates have been generated for influenza [Neirynck, 99] and malaria [Sällberg, 02].

▼ 7 

In vaccine development, HBc provides several advantages as a carrier for foreign epitopes: (i) The carboxy-terminal region of HBc responsible for nucleic acid binding can be deleted without disturbing the formation of HBc particles [Borisova, 89; Gallina, 89]. Particles formed by carboxy-terminally truncated HBc protein contain only traces of RNA [Birnbaum, 90; Ulrich, 93]. (ii) The structure of HBc particles has been resolved by cryoelectron microscopy and X-ray crystallography. In line with epitope mapping data the major immunodominant region (MIR) has been identified as the surface-exposed tip of spikes on the surface of HBc particles [Salfeld, 89; Böttcher, 97; Wynne, 99]. (iii) The MIR has been shown to be dispensable for particle assembly [Schödel, 92]. (iv) As expected for a highly repetitive antigen, HBc particles are highly immunogenic and improve the immunogenicity of per se low immunogenic foreign peptides presented on their surface [Clarke, 87; Francis, 90]. (v) In comparison to other VLP carriers, HBc has a favourable property in terms of vaccine development; it is not only a T cell dependent but also T cell independent antigen [Milich, 86b]. This T cell independence can be transferred to foreign segments presented on HBc particles [Fehr, 98].

In previous experiments, three potential insertion sites for foreign protein segments into HBc have been used: the amino-terminus, the MIR and different carboxy-terminal positions [for reviews see Ulrich, 98b; Pumpens, 01]. According to the three-dimensional structure [Böttcher, 97; Wynne, 99], epitope mapping data [Salfeld, 89] and empirical insertion data, the MIR represents the most preferential insertion site for foreign sequences [Schödel, 92; Borisova, 96; Lachmann, 99]. Therefore the MIR of the HBc protein has been chosen in this study as the place to insert parts of the DOBV N protein (see chapter 1.8).

1.5.4 The need of adjuvants in subunit vaccines

One of the problems of generating a protein subunit vaccine, e.g. based on VLPs is that proteins by themselves have a rather low immunogenicity. Therefore, adjuvants are needed to supplement the proteins to induce a strong protein-specific immune response. Adjuvants can either enhance or modify the immune response. Alternatively, they can also act as a depot so that the protein is released over a long period of time and thereby continuously stimulating a protein-specific immune response. Until recently only alum (aluminium hydroxide or aluminium phosphate) has been used as an adjuvant in vaccines for human use. In the year 2000 an influenza vaccine (FluadTM, Chiron) was introduced to the market in which MF59, a water in oil emulsion, is used as adjuvant [Podda, 01]. The development of new adjuvants helping in inducing a strong protective immune response is a crucial step in vaccine development. Ongoing studies investigate other adjuvants containing saponins, or small unmethylated DNA oligonucleotides, (CpG dinucleotides) to be used in human vaccines. Another approach is to use recombinant cytokines as adjuvants in subunit vaccines.

▼ 8 

Complete Freund's adjuvant (CFA) due to its high content of mycobacterial cell wall components is a useful adjuvant in research only, but will not be certified for human use. But it represents a useful adjuvant to investigate if a protein is able to induce an immune response in animal models. In this study recombinant proteins were applied with CFA and incomplete Freund's adjuvant (IFA) to enable comparisons to the results of protection studies in an animal model [Lundkvist, 96; Dargeviciute, 02; de Carvalho Nicacio, 02].

1.5.5 Hantavirus proteins suitable as a subunit vaccine

An efficient hantavirus vaccine should protect against infections by all members of the genus Hantavirus. To develop a broadly protective vaccine against hantaviruses two alternative approaches could be followed: (i) generation of a bi- or multivalent vaccine consisting of antigens from different hantavirus species or (ii) identification of antigen(s)/epitope(s) providing cross-protection against a broad range of different hantaviruses.

So far, nothing is known about the potential of the RdRp as a vaccine. In the same line, research about RdRp is rather limited and until recently it had not been heterologously expressed as an entire protein [Jonsson, 01]. In a recent study a part of a recombinant RdRp expressed in E. coli has been shown to be immunogenic in rabbits [Kukkonen, 04]. Compared to the other proteins, the RdRp of different hantaviruses have the highest aa identity; there is a minimum of 70 % aa identity between the RdRp of the hantaviruses [Tab. 2 and Kanerva, 98a]. Therefore, one can expect a high cross-reactivity of the immune response induced by RdRp. This cross-reactivity would be highly favourable for vaccine development. However, further investigations are needed to characterise the potential of RdRp as a vaccine.

▼ 9 

TABLE 2: Amino-acid identities (in %) in the glycoprotein precursor protein and the RNA-dependent RNA polymerase of different hantaviruses after alignment by Clustal Method.

Glycoprotein precursor sequences: DOBV-Slovakia (Genbank accession number AY168578), HTNV 76-118 (M14627), PUUV-Vranica/Hällnäs (U14136), ANDV (AF324901), SNV (L37903). RNA-dependent RNA polymerase (RdRp) sequences: HTNV 76-118 (D25531), PUUV-Sotkamo (Z66548), ANDV (AF291704), SNV (L37902). Grey areas: no sequence data available for the glycoprotein precursor or the polymerase of the respective hantavirus.

In contrast to the RdRp, the glycoproteins are well known to have a potential as a vaccine. The proteins G1 and G2 can be expressed in baby hamster kidney (BHK) cells with the help of a alphavirus replicon [Kallio-Kokko, 01b] or by vaccinia virus (VACV) vectored expression [Schmaljohn, 90e]. Passive transfer of serum from SEOV infected rats protected rats from a subsequent SEOV challenge [Zhang, 89]. As the serum had high neutralising antibody titres, protectivity was thought to be mediated by G1/G2-specific antibodies. The specificity of the transfered antibodies, however, was not determined. Immunisation with a VACV vector expressing G1 and G2, as well as passive transfer of G1/G2-specific monoclonal antibodies protected hamsters [Schmaljohn, 90d] and suckling mice [Arikawa, 92] from a HTNV challenge.

The glycoprotein precursors of different hantaviruses have the lowest aa identity among the hantavirus proteins. There is an aa identity of 50 % to 70 % of the glycoprotein precursors [Tab. 2 and Kanerva, 98d]. Due to these large aa differences in the glycoproteins of different hantavirus strains it is improbable that a subunit vaccine based on the glycoprotein of one hantavirus strain will protect against heterologous hantaviruses [Schmaljohn, 90c; Ruo, 91]. In line, serum from SEOV infected rats protects newborn rats against SEOV but not HTNV [Zhang, 89]. As a neutralisation titre of 1:640 against SEOV was found in the passively transferred sera, protection was thought to be provided by G1/2 specific antibodies. Thus, G1/G2 specific antibodies might not be very cross-protective.

▼ 10 

It is thought that the G1/G2-specific immune response protecting against hantaviruses is based primarily on antibodies. However, the role of G1/G2-specific cellular immune response in protection against hantavirus infection has not been investigated so far.

TABLE 3: Amino acid (aa) identities (in %) of the entire nucleocapsid proteins and their amino terminal 120 aa of different hantavirus strains used in this study. The aa identities were determined by alignment using the Clustal method.

Sequences: DOBV-Slovenia, DOBV-Slovakia, HTNV 76-118, PUUV-Vranica/Hällnäs, PUUV-Sotkamo, PUUV-Kazan [For all sequences see Razanskiene, 04]; ANDV-AH1 and SNV-3H226 [Lopez, 97a] and SNV-3H226 .[Hjelle, 94b].

As the N proteins of different hantaviruses are more closely related to each other than the glycoproteins [Tab. 2 and 3, and see Kanerva, 98c] it has since long been speculated that a vaccine based on N protein might be more cross-protective than a vaccine based on G1 and G2 [Asada, 89]. The N protein of hantaviruses is highly immunogenic. Natural hantavirus infections of rodents and humans result in the induction of strong N-specific antibody and T cell responses [reviewed in Khaiboullina, 02]. In addition, immunisation of rodents with recombinant N (rN) protein induced N-specific B and T cell responses [Lundkvist, 97c; Ulrich, 98a; de Carvalho Nicacio, 01f; Dargeviciute, 02; de Carvalho Nicacio, 02]. In rodent animal models a protective immune response has been mediated by immunisation with vaccinia-vectored N-encoding vaccines [Schmaljohn, 90b; Xu, 92], E. coli-expressed chimeric HBc particles carrying amino-terminal parts of N protein [Ulrich, 98a] as well as with rN proteins expressed in E. coli[Lundkvist, 96; de Carvalho Nicacio, 02], yeast Saccharomyces cerevisiae[Dargeviciute, 02] or insect cells [Schmaljohn, 90a; Lundkvist, 96; Schmaljohn, 99]. However, in a hamster challenge model the protection mediated by a SEOV N-encoding DNA vaccine was found to be low [Kamrud, 99].

1.6 Animal models for hantavirus research

▼ 11 

As described above, in nature hantaviruses can persistently infect their natural host without showing signs of disease. Although groups have tried to infect various rodents with hantaviruses [Asada, 88a], there are only a few natural hantavirus rodent hosts that have been bred in captivity to investigate hantavirus infection and challenge. Bank voles can be infected experimentally with PUUV [Lundkvist, 96], rats with HTNV [Lee, 81] and deer mice with SNV [Botten, 00]. Moreover, PUUV and HTNV as well as SEOV can infect Syrian hamsters [Chu, 95; Hooper, 99; Hooper, 01a]. Additionally to hamsters, HTNV can infect Mongolian gerbils (Meriones unguiculatus) in the laboratory [Xu, 92]. However, the New World hantaviruses ANDV and Maporal virus are the only ones that cause a lethal infection resembling HCPS in Syrian hamsters [Hooper, 01b; Milazzo, 02a].

PUUV, which readily infects its natural host, the bank vole C. glareolus could not infect adult immunocompetent laboratory mice like BALB/c mice [Klingström, 02a]. Comparably, BALB/c and ICR have been reported to be susceptible to only transient infection with HTNV, with virus titres in the infected mice lasting for a maximum of five days [Asada, 87d] or ten days [Kariwa, 95] post infection. In contrast to the only transient infection of laboratory mice, it has recently been described that immunocompetent adult BALB/c, C57BL/6 and SJL/J mice were susceptible to HTNV [Wichmann, 02a]. C57BL/6 and BALB/c mice have even been reported to die eight to eleven days post infection [Wichmann, 02b]. These experiments were conducted with the same HTNV strain (76-118) used in studies where the virus could only transiently infect laboratory mice (see above). Thus the peculiarities of the HTNV used by Wichmann et al. have to be determined. DOBV infection experiments with BALB/c and NMRI mice resulted in almost all mice in the induction of DOBV N-specific antibodies. However, only some animals had S segment RNA and none of them had N-antigen in their lungs [Klingström, 02b]. Similar findings were obtained recently for DOBV infection of C57BL/6 mice, where infected mice developed G1/2-specific antibody response but N antigen and S segment RNA could not be detected in the lung [Klingström, 04]. Hence, additional investigations are needed to prove if DOBV can reproducibly infect laboratory mouse strains and if infection is transient or long lasting.

Athymic nude mice, which have only very limited amounts of T cells, died from an HTNV infection [Asada, 87c]. In the same line, SCID mice which lack the recombinant VDJ region, leading into T cell and B cell deficiencies, died from HTNV and SEOV infections [Yoshimatsu, 97]. Besides the adult rodent models, intracerebral or subcutaneous injection with HTNV [Nakamura, 85b; Yoshimatsu, 93] and injection with DOBV [Klingström, 03] has been shown to be lethal for suckling mice.

▼ 12 

Besides rodents, nonhuman primates can be experimentally infected with hantaviruses. PUUV and ANDV can both infect cynomolgous macaques leading to symptoms similar to those seen in human HFRS and HCPS patients, respectively [McElroy, 02; Klingström, 02c] .

1.7 Nucleocapsid protein specific immune response

1.7.1  Antibody response

The N protein is the major antigenic target in the early IgG response of NE patients [Lundkvist, 93b] whereby after disease, more and more G1/2-specific antibodies can be found [Lundkvist, 93b; Vapalahti, 95a]. N-specific antibodies were present already at the onset of disease, while G1/2-specific IgG antibodies were present in only 2 % of the acute sera compared to 87 % of old immune sera from NE patients [Kallio-Kokko, 01a]. Human sera contained IgG1 and IgG3 in acute sera and IgG1 and IgG4 in the sera 2 years after infection against all three structural proteins N, G1 and G2 [Lundkvist, 93a].

In rodents, the N-specific antibody response has been extensively studied, both after hantavirus infection as well as after immunisation with N protein constructs. Experimental PUUV infection as well as immunisation with PUUV rN protein induced a strong N-specific antibody response in bank voles [Lundkvist, 96; Dargeviciute, 02; de Carvalho Nicacio, 02] and laboratory mice [de Carvalho Nicacio, 01e].

▼ 13 

The immunodominant B cell epitope region has been located at the amino terminus of the N protein. Human IgG response in sera of HFRS and HCPS patients was mostly directed to the 119 amino terminal amino acids of hantavirus rN protein [Jenison, 94a; Lundkvist, 95; Vapalahti, 95b; Elgh, 96a; Gött, 97; Milazzo, 02b]. In sera from rodent, like PUUV infected bank voles as well as most monoclonal antibodies derived from PUUV infected animals reacted mostly with amino-terminal peptides of the N protein [Lundkvist, 96; Lundkvist, 02]. Sera of deer mice infected with SNV also reacted more strongly to the N-terminal aa 17-58 compared to the remaining portion of the protein [Yamada, 95].

There are several indications that N-specific antibodies may play a role in protecting against hantavirus infection. N-specific antibodies have been demonstrated to provide protection against a hantavirus infection in cell culture, in the suckling mouse model as well as in adult bank voles [Yoshimatsu, 93; Yoshimatsu, 96e; Lundkvist, 02]. The immunological mechanisms behind the protectivity induced by N-specific antibodies remains to be clarified. Antibody-dependent cytotoxicity (ADCC) has been discussed as one of the mechanisms by which the antibodies can confer protection by binding to infected cells and marking them for destruction [de Carvalho Nicacio, 01d]. An inhibition of transcription has been discussed as another mechanism by which N-specific antibodies can confer protection by binding to N protein, which is bound to the viral RNA and which subsequently inhibits transcription [Yoshimatsu, 96d].

1.7.2 Cellular immune response

In patients infected with HTNV and PUUV, the CD8+ cellular immune response was found to be at least partly directed against the N protein [Van Epps, 99; Van Epps, 02a; Terajima, 02a]. In PUUV infected NE patients G1/2-specific CD8+ cells have been detected [Terajima, 02b]. PUUV induced CD8+ memory cells in NE patients [Van Epps, 02b]. Moreover a milder NE course could be connected to HLA B27 [Mustonen, 98a]. Thus, cellular immune responses are likely beneficial to humans in terms of recovering from hantavirus disease.

▼ 14 

There is an ongoing controversy if the cellular immune response measured in acutely ill patients is connected to immunopathogenesis or protective immune response. It has been suggested that the N-specific CD4+ and CD8+ cells from SNV infected patients might be involved in immunopathogenesis [Ennis, 97]. In line with these findings, numerous cells producing cytokines as IL-1, IL-6, TNF-α, IFN-γ, IL-2, IL-4, and TNF-β were detected by immunohistochemical staining in autopsy tissues from HCPS patients [Morii, 99]. In another study the ratio of activated to non-activated lymphocytes was higher in acute HFRS patients than in convalescent patients [Huang, 94]. Whether these cytokine producing cells or the activated lymphocytes helped clearing the virus infection or were involved in immunopathology has yet to be elucidated.

Hantavirus specific T cell responses have been investigated in various rodent studies. Cytotoxic splenocytes from HTNV infected mice have been found to lyse HTNV and SEOV infected macrophages [Asada, 88b]. In another study, BALB/c mice infected with HTNV developed IFN-γ secreting CD8+ cells as well as HTNV-specific cytotoxic T cells (CTLs) [Araki, 03]. HTNV-specific CD4+ and CD8+ cells induced in mice after HTNV infection can protect mice against a HTNV challenge as shown in adoptive transfer experiments [Asada, 87b]. Most important in protection were CD5+ positive lymphocytes (T cells and subsets of B cells) as protection dropped most when these cells were lysed before transferring spleen cells from HTNV immunised mice into naive mice. As in the investigations outlined above, it remains to be investigated to which of the hantaviral proteins the protective immune response was directed against.

For the first time cytotoxic T cells with a proven N-specificity have been generated in C57BL/6 mice by infecting them with HTNV or immunising them with HTNV N protein-derived peptides [Park, 00]. In other studies, PUUV N-specific proliferation was found in splenocytes of PUUV rN protein immunised BALB/c [de Carvalho Nicacio, 01c] as well as of bank voles immunised with rN proteins of DOBV, ANDV or TOPV [de Carvalho Nicacio, 02].

▼ 15 

Taken together, it is not clear what kind of N-specific immune response is needed to protect against hantaviruses, but it can be deduced from the studies mentioned above, that N-specific antibodies as well as T cells can play a role in protecting against a hantavirus infection.

1.8 Objectives of the study

DOBV carried by the yellow necked field mouse A. flavicollis is a highly virulent virus responsible for clinically severe HFRS cases with a high case fatality rate [up to 12 %, Avsic-Zupanc, 99] and able to kill suckling mice [Klingström, 03]. The main objective of this study was to compare the immunogenicity of entire rN protein of DOBV (strain Slovenia) to the immunogenicity of HBc particles harbouring 120 amino-terminal aa of DOBV (strain Slovenia) N protein (HBcdDOB120). To allow comparison to earlier studies, the immunisation scheme used in this study has previously been used in challenge experiments [Lundkvist, 96].

As humoral, as well as cellular immune responses can be involved in protection against hantaviruses, these investigations included the characterisation of humoral and cellular immunity after immunisation with DOBV rN or HBcdDOB120 protein. The characterisation of humoral immunity should be obtained by analysis of the antibodies against homologous DOBV rN protein and heterologous rN proteins. The estimation of N-specific IgG subclasses should allow a first idea about the cytokine milieu created by lymphocytes involved in the N-specific immune response. In addition, the proliferation of N-specific lymphocytes and their secreted cytokines were analysed to characterise the N-specific T cell response after immunisation with the two proteins.

▼ 16 

A potential problem for the use of chimeric VLPs based on HBc might be a pre-existing immunity due to HBV infection. Therefore, an additional aim of the study was to investigate whether a pre-existing anti-HBV core immunity influences the subsequent immune response against a foreign protein sequence presented on HBc particles in mice.


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
XDiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
22.08.2006