key: cord-0007755-pwj10hnb
authors: Oldstone, Michael B.A.; Southern, Peter J.
title: Trafficking of activated cytotoxic T lymphocytes into the central nervous system: Use of a transgenic model
date: 2002-12-10
journal: J Neuroimmunol
DOI: 10.1016/0165-5728(93)90230-v
sha: 1c39a75d28b5ac49de32b9d48dad1923ea0bd8c3
doc_id: 7755
cord_uid: pwj10hnb

We have used cell or tissue-specific promoters to express lymphocytic choriomeningitis virus (LCMV) proteins in selected cells in independent lines of transgenic mice. Upon adoptive transfers into these mice, MHC-restricted LCMV-specific cytotoxic T lymphocytes homed specifically to either the choroid plexus (SV40 promoter) or β cells of the islets of Langerhans (rat insulin promoter). The availability of promoters specific for neurons, oligodendrocytes and astrocytes makes this approach compelling for evaluating T cell trafficking into the CNS and for analyzing antigen presentation in vivo in the CNS.

Damage to the brain during viral and/or autoimmune disease is often caused by cytotoxic T lymphocytes (CTL) or their products (Watanabe et al., 1983; Vandenbark et al., 1985; Chan et al., 1989; Klavinskis et al., 1989; Sinha et al., 1990) . Indeed, depletion of such cells prevents brain injury. Moreover, in a host that is T cell-depleted or deficient, reconstitution by adoptive transfer of activated antiviral or anti-self specific CTL leads to central nervous system (CNS) disease.

CTL recognize a unique amino acid sequence of a viral or self (or altered self) protein bound within the groove formed by the al and a2 arms of the major histocompatibility complex (MHC) glycoproteins (Zinkernagel and Doherty, 1974; Townsend et al., 1986; Oldstone et al., 1988; Van Bleek and Nathenson, 1990; Falk et aI., 1991) . CTL generated against lymphocytic choriomeningitis virus (LCMV) are CD8 ÷ and restricted by MHC class I glycoprotein molecules (Zinkernagel and Doherty, 1974; Oldstone, 1987 Oldstone, , 1991 . The peptides derived from LCMV Armstrong (ARM) proteins that act as 'CTL epitopes' and are restricted by MHC class I H-2 b, and H-2 d glycoproteins have been deciphered through the combined usage of genetic and biochemical techniques (Oldstone et al., 1988; Whitton et al., 1988a,b; Klavinskis et al., 1990; Yanagi et al., 1992) . For example, H-2 b mice have three known CTL epitopes. They map to LCMV glycoprotein (GP) amino acids (aa) 34-43 (KAVYNFATC), GP aa 276-286 (SGVENPGGYCL) and nucleoprotein (NP) aa 396-404 (FQPQNGQFI) (Fig. 1) . These peptides represent optimal binding and CTL recognition sequences (Gai/-in and . At the clonal level, of over 40 CTL clones analyzed, 85% react with the GP-2 epitope aa 276-286 and only 3% with the NP epitope. For H-2 d mice, LCMV NP aa 119-127 (PQASG-VYMG) is the optimal peptide, and > 96% of their CTL recognize this at both the clonal (analysis of over 50 clones) and primary immunization (7 days after initial inoculation) levels H. Lewicki, J.L. Whitton, M.B.A. Oldstone, unpublished observation, 1992) . The lack of a GP epitope in H-2 d mice, under normal conditions, conveniently segregates H-2 d from H-2 b CTL responses Fig. 1) . When CTL recognize the viral peptide pro-cessed by the appropriate MHC molecule at the target cell's surface, the result is target cell destruction. The brain was formerly considered a privileged site whose borders disallowed the entry of lymphocytes. However, it is now clear from work of the Wekerle, Lampert, and Hickey laboratories, that activated T lymphocytes easily and rapidly enter the brain (Wekerle et al., 1986; Hickey et al., 1991) , then pass through unless confronted by the appropriate peptide-MHC complex (Oldstone et al., 1986) . Unfortunately, analysis of cells' movement in the CNS is complicated, at least in the course of viral infections, because viral gene products also frequently replicate at multiple CNS sites as well as other tissues. Therefore, we devised a strategy to direct trafficking of activated T lymphocytes to a specific CNS site or cell and also to evaluate in vivo the ability of such cells to present and process self, antiself, or viral antigens. Cell-specific promoter(s) and transgenic technology Rail et al., 1992) were combined with adoptive transfer of activated T lymphocytes (Oldstone et al., 1986; Joly et al., 1991) to implant and express viral or 'self' genes of interest in specific CNS cells. Here we document the expression of LCMV protein containing a CTL epi-tope(s) in the choroid plexus using the SV40 promoter or in /3 cells of the islets of Langerhans using the rat insulin promoter (RIP). Upon adoptive transfer of MHC-restricted LCMV-specific CTL, T lymphocytes then homed directly to the choroid plexus or to/3 cells, respectively.

Transgenic mice were made as described earlier . The cDNA genes for the LCMV NP and GP coding regions were assembled from overlapping cDNA clones derived from the S RNA segment of LCMV (ARM clone 53b) (Dutko and Oldstone, 1983; Southern et al., 1987) . The NP and GP cDNA genes were cloned into the RIP or the pSV2B expression vector as BamHI-BgllI fragments. signal. The original upstream XhoI and the downstream HindIII cloning sites were converted to BglII site by linkers. The SV40 pSV2B vector is a derivative of the pSV243-globin vector. The upstream HindIII site, adjacent to the SV40 early promoter, was converted to BglII. pSV2B also contains the SV40 t-intron and early region polyadenylation signal. After cloning and amplification, the RIP NP and RIP GP transcription units were isolated from flanking plasmid sequences and purified on a high resolution sucrose gradient. The SV NP and SV GP cassettes were excised from the plasmids and purified by preparative agarose gel electrophoresis .

Transgenic mice were generated using C57BL/6

mice as a source of oocytes. Injected eggs were implanted in pseudopregnant CD1 females. Founder mice demonstrating integrated copies of the transgene were crossed to b x d or C57BL/6 mice for one generation to produce the murine lines described. Thereafter each line was bred to at least the F3 generation to confirm transmission of the transgene. Mice utilized came from the vivarium of The Scripps Research Institute.

Mice carrying the transgene were identified by hybridization of DNA extracted from tail biopsies using LCMV-specific GP and NP probes (Southern et al., 1987) . Mice were further characterized by Southern and Northern blot analysis and PCR (Southern et al., 1987; Oldstone et al., 1991; De la Torre and Oldstone, 1992) .

Expression of LCMV GP or NP proteins was determined by adoptive transfer of LCMV-specific MHC-restricted CTL (Oldstone et al., 1986 . For these studies, mice bearing the transgene received 5 x 10 7 LCMV-specific CTL i.p. The splenic CTL were obtained from H-2 b, H-2 d, or b x d donors immunized 45-70 days earlier with 1 x 105 pfu of LCMV i.p. In other instances, transgenic mice received 1-2 x 106 LCMV GP (H-2 b [D b] restricted) or 1-2 x 10 6 LCMV NP (H-2 d [L d] restricted) CTL clones given i.p. or i.e. Details of adoptive transfer, as well as the generation and use of LCMV GP and NP CTL clones, are given elsewhere (Oldstone et al., 1988; Whitton et al., 1988a Whitton et al., ,b, 1989 Joly et al., 1991; Yanagi et al., 1992) . At 1, 3, 7, and 10 days following transfer, pancreas, liver, and brain tissues were removed, fixed in Bouin's solution, and prepared for histological examination.

LCMV-specific CTL activity was determined by a 5-9-h 51Cr release assay (13-16). Balb Clone 7 (H-2a), MC57 (H-2 b) or SWR/J fibroblasts (H-2 q) cells were the targets (13-16). Uninfected target cells or target cells expressing viral epitopes were tested. The latter cells were infected 48 h earlier with LCMV ARM at moi 1 pfu/cell, or infected 10-14 h earlier with a 27 vaccinia virus recombinant expressing either LCMV GP or LCMV NP at moi 3 pfu/cell (Whitton et al., 1988b) . Cells were labeled with 5~Cr and resuspended at 105 ml-I in RPMI/5% FBS after which 104 target cells were added to each well of 96-well plates. All assays were done in triplicate and included controls for spontaneous and maximum 51Cr release from the targets. Percent specific CTL lysis was calculated at 100 x (experimental release-spontaneous release)/(maximum release-spontaneous release). The variance among triplicate samples was < 10%.

To generate LCMV-specific CTL, mice were inoculated i.p. with 1 x 105 pfu of LCMV ARM. Seven days later, their spleens were removed and single lymphoid cell suspensions, free from red blood cells, prepared as described (Oldstone et al., 1988; Whitton et al., 1988b) .

The antibody response to whole LCMV was assayed by using an ELISA with whole inactivated virus. Responses to NP or GP were evaluated with PAGE or immunofluorescence as described elsewhere .

Tissues taken for histological analysis were fixed in zinc formalin (10%) and stained with hematoxylin and eosin. In other instances, tissues were snap frozen in liquid nitrogen, sectioned on a cryomicrotome, and analyzed for expression of LCMV proteins (Joly et al., 1991; Oldstone et al., 1991) .

To direct expression of LCMV proteins to the choroid plexus, we fused the SV40 enhancer region and promoter to cDNA encoding the NP or GP of LCMV ARM and inoculated the product into the germline of b x d mice. This strategy was selected because of reports that the SV40 enhancer-promoter sequence fostered the expression of SV40 or human papilloma T antigen to the choroid plexus (Marks et al., 1988; Messing et al., 1988; Feigenbaum et al., 1992) . Of 32 pups born from eggs inoculated with SV40 LCMV GP, six had integrated the LCMV GP DNA as documented by Southern hybridization analysis. From these mice, two lines were established in which LCMV GP was passed into progeny mice as detected by tail DNA dot blot hybridization. These murine lines were designated SVGP-7 and SVGP-31. Northern blot analysis was inconsistent in detecting LCMV RNA in several tissues analyzed (brain, choroid plexus, thymus, spleen, liver, kidney, and heart). Further, no LCMV protein was observed in the choroid plexus by immunohistochemistry. In contrast, adoptively transferred LCMV-specific MHC restricted CTL clearly localized in the choroid plexus (Fig. 2) . However, neither nontransgenic littermates (data not shown) nor other transgenic mice with LCMV expressed in cells other than the choroid plexus showed any sign of LCMV-specific CTL in the choroid plexus (Fig. 2) . Three of seven pups born from eggs inoculated with SV40 LCMV NP DNA were DNA positive but the transgene was not transmitted to other progeny and transgenic line(s) were not developed from this construct. Figure 1 shows that SVGP-31 mice were unable to mount specific CTL responses to the LCMV GP but Fig. 2 . Trafficking of LCMV-specific MHC-restricted CTL to the choroid plexus of transgenic mice expressing LCMV GP under the control of the SV40 promoter/enhancer region. Panels 1 and 2 are photomicrographs of the choroid plexus from a b × d mouse of transgenic line SV GP-31 that was inoculated i.v. with 5 x 107 LCMV-specific splenic CTL 5 days earlier. CTLs were generated in a b x d mouse inoculated with 1 x 105 pfu of LCMV and adoptively transferred 45 days later. Panel 3 is a photomicrograph of the choroid plexus from a b x d mouse of transgenic line SV GP-7 that was inoculated i.c. with 2 × 10 6 LCMV GP H-2b-restricted CTL clone (228) (see Oldstone et al., 1988; Joly et al., 1991) 3 days earlier. Panel 4 is a photomicrograph of the pancreas from the mouse whose choroid plexus is shown in Panels 1 and 2. Note the absence of inflammatory cells in the islets of Langerhans. Inflammation did not occur when either this CTL clone (Panel 5) or 5 x 10 7 LCMV-specific splenic CTL were transferred i.c. or i.v., respectively. Trafficking of LCMV-specific MHC-restricted CTL to the islets of Langerhans cells occurred in mice from transgenic lines expressing LCMV NP or GP behind the RIP. Panel 6 is a photomicrograph from a RIP NP 25-19 b x d that received 2 x 10 6 CTL clones specific for LCMV NP. Similar results were obtained when 5 x 10 7 CTLs from spleens of b x d mice were transferred 45 days after a single inoculation of 1 × 105 pfu of LCMV. Fig. 3 . A cartoon sketching an experimental strategy for studying CNS pathophysiology using the transgenic approach. HA, hemagglutinin; IDDM, insulin-dependent diabetes mellitus.

readily generated CTL responses to the viral NP. The SVGP-7 line mounted a modest but consistent CTL response to LCMV GP; all five mice in this group made CTL responses (> 10% specific 51Cr release). Of over 32 individual SVGP mice tested, only one spontaneously made antibodies to LCMV. However, 28 days after challenge with 1 × 105 pfu of LCMV ARM i.p., ten of ten such mice made vigorous responses to LCMV. Of four randomly selected individual mice from each SV40 GP line, all made responses to both GP and NP of LCMV. The generation of RIP LCMV NP and GP transgenic lines has been reported elsewhere (Oldstone et al., 1991) . As with the SV40 LCMV transgenic mice, transmission of the LCMV gene was detected easily by Southern blot analysis. Although LCMV RNA was not consistently noted in Northern blot studies , the PCR method readily identified LCMV RNA in the pancreas (M. von Herrath et al., manuscript in preparation, 1993) . Again, viral protein was not observed by immunohistochemistry but was documented by CTL recognition Fig. 2) . Three RIP GP (GP-28, GP-64, and GP-70) and three RIP NP lines were selected for further study. All three RIP GP lines made LCMV-specific CTL to the viral GP and NP (Fig. 1) . Similarly, RIP NP25-20 and NP-54 lines made CTL responses to LCMV GP and NP expressed behind a vaccinia virus promoter (Fig. 1) . However, in contrast, as shown in Fig. 1 , RIP LCMV NP25-19 failed to generate CTL to NP but made a vigorous CTL response to LCMV GP.

Adoptive transfer of LCMV MHC-restricted CTL into SVGP transgenic mice resulted in cellular infiltration in the choroid plexus. As seen in Fig. 2 , splenic CTL harvested from MHC compatible mice immunized 45-65 days earlier (panels 1, 2, 3) infiltrated the choroid 29 plexus but not the islets of Langerhans (panel 4) or liver (data not shown) of transfer recipients. These observations were recorded in mice from SVGP-7 and SVGP-31 lines (three to five mice per line). The responses were virus-specific, since CTL transferred into non-transgenic littermates or into transgenic mice in which the LCMV GP or NP was expressed in/3 cells of the islets of Langerhans did not infiltrate the choroid plexus (panel 5). However, these or similar LCMVspecific CTL transferred into RIP GP or NP transgenics panel 6 ) migrated into the islets of Langerhans.

CTL clone 228 recognizes GP amino acid residues 276-286 of LCMV ARM and is restricted by the D end of H-2 b glycoprotein (Oldstone et al., 1988) . Intracerebral transfer of this clone into SVGP-31 mice (Fig. 2, panel 3) led to an accumulation of lymphoid cells in the choroid plexus (three of four mice). Yet, inoculation i.p. with clone 228 did not yield lymphocytes in the choroid, presumably because of the IL-2 dependency of this cultured clone and the time required for its transit. When clone 228 was inoculated i.c. into non-transgenic littermates (four per group) or into RIP GP-64 mice (three per group) no lymphoid cells appeared in the choroid plexus. However, CTL clone 228 accumulated in the islets of Langerhans of RIP GP-64 mice (three of four).

Here, using transgenic technology, we caused the expression of a foreign protein (LCMV GP) in a prespecified compartment of the CNS and observed the trafficking to and accumulation of antiviral specific T cells at that CNS site. For these studies, we utilized the promoter and regulatory region of SV40, which has been shown to express its large T antigen and the T antigen of JC virus in the choroid plexus (Marks et al., 1988; Messing et al., 1988; Feigenbaum et al., 1992) . We adapted the enhancer-promoter region of SV40 to express LCMV GP because of the specific GP CTL reagents available and knowledge of GP and NP LCMV CTL responses (Oldstone et al., 1988; Whitton et al., 1988a Whitton et al., ,b, 1989 Klavinskis et al., 1990; Gairin and Oldstone, 1992; Yanagi et al., 1992) . After transferring LCMV-specific memory cells or cloned LCMV CTL into the transgenic mice, we demonstrated the accumulation of such cells specifically in the choroid plexus. These studies with SV40 LCMV GP transgenics extended observations by ourselves Von Herrath et al., 1992) and others (Ohashi et al., 1991) that the trafficking of lymphocytes to the islets of Langerhans is involved in the etiology and pathogenesis of autoimmune insulin-dependent diabetes mellitus. Analysis of the pathophysiology of SV40 LCMV trans-genics after LCMV challenge is to be reported elsewhere (P.S., unpublished observations).

A range of CNS-specific promoter-enhancer sequences is now available so that molecules representing modified self, self, or viral genes can be targeted to neurons, astrocytes, or oligodendrocytes. With this device, investigators in the Viral-Immunobiology Laboratory at Scripps are studying the expression of recorder genes like /3-galactosidase, viral genes like LCMV GP or NP, rabies GP or HIV GP and mutated or self genes like prion protein, /3-amyloid and cytokines (Campbell et al., 1992; Mucke et al., 1992; Rail et al., 1992 Rail et al., , 1993 Toggas et al., 1993; Mucke, 1993; Mucke and Rockenstein, 1993) (Fig. 3) . This approach should allow in vivo dissection of normal development as well as injury and disease of the CNS.

The SV40 LCMV GP line GP-31, generated CTL to LCMV NP but not GP. Additionally, one line, RIP LCMV NP25-19, generated CTL to LCMV GP but not NP. The RIP NP25-19 line expresses its transgene in the thymus as well as the /3 cells of the islets, suggesting that potentially NP reactive CTL are actively deleted (M. von Herrath and M.B.A. Oldstone, manuscript in preparation) . The SV40 lines are under evaluation, but the observation of thymic hyperplasia and thymic tumors in transgenic mice in which SV40 early regulatory sequences express T antigen (Marks et al., 1988; Messing et al., 1988; Feigenbaum et al., 1992) suggests a similar scenario in which the thymuses of such mice express LCMV GP and delete LCMV GPreactive CTL.

Neurologic disease induced in transgenic mice by the astrocyte-specific expression of interteukin-6

Infiltration of immune T cells in the brain of mice with herpes simplex virus-induced encephalitis

Selective disruption of growth hormone transcription machinery by viral infection

Genomic and biological variation among commonly used lymphocytic choriomeningitis virus strains

Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules

JC virus and simian virus 40 enhancers and transforming proteins: Role in determining tissue specificity and pathogenicity in transgenic mice

Design of high-affinity major histocompatibility complex-specific antagonist peptides that inhibit cytotoxic T lymphocyte activity: Implications for control of viral disease

T-lymphocyte entry into the central nervous system

Viral persistence in neurons explained by lack of major histocompatibility complex class I expression

Efficiency and effectiveness of cloned virus specific cytotoxic T lymphocytes in vivo

Vaccination and protection from a lethal viral infection: Identification, incorporation and use of a cytotoxic T lymphocyte glycoprotein epitope

The expression of viral and cellular genes in papillomas of the choroid plexus induced in transgenic mice

Developmental study of SV40 large T antigen expression in transgenic mice

Rapid activation of astrocyte-specific expression of GFAP-IacZ transgene by focal injury

Transgenic models to study the pathogenic role of mutated and non-mutated forms of human amyloid proteins in the development of Alzheimer's disease (AD)

Transgenic models to study diseases of the nervous system: An in vivo approach to dissect complex pathogenetic networks

Prolonged delivery of transgene products to specific brain regions by migratory astrocyte grafts

Ablation of "tolerance" and induction of diabetes by virus infection in viral antigen transgenic mice

Cytoimmunotherapy for persistent virus infection: Unique clearance pattern from the central nervous system

Arenaviruses: Biology and Immunotherapy, Current Topics in Microbiology and Immunology

Fine dissection of a nine amino acid glycoprotein epitope, a major determinant recognized by lymphocytic choriomeningitis virus specific class I restricted H-2D b cytotoxic T lymphocytes

Molecular anatomy of viral persistence

Virus infection triggers insulin dependent diabetes mellitus in a transgenic model: Role of anti-self (virus) immune response

Transgenic mice which express histocompatibility molecules targeted to unique cells of the central nervous system

Targeting specific host and viral genes to cells of the central nervous system

Autoimmune diseases: The failure of self tolerance

Molecular characterization of the genome S RNA segment from lymphocytic choriomeningitis virus

The effects of cerebral HIV-1 gpl20 expression in transgenic mice

The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides

Isolation of an endogenously processed immunodominant viral peptide from the class I H-2K b molecule

Specificity of T lymphocyte lines for peptides of myelin basic protein

Virus induced autoimmunity: Ability of various LCMV strains to cause insulin dependent diabetes mellitus in a RIP-LCMV-ARM transgenic mouse model

Adoptive transfer of EAE-like lesions from rats with coronavirus-induced demyelinating encephalomyelitis

Cellular immune reactivity within the CNS

Molecular definition of a major cytotoxic T lymphocyte epitope in the glycoprotein of lymphocytic choriomeningitis virus

Analyses of the cytotoxic T lymphocyte responses to glycoprotein and nucleoprotein components of lymphocytic choriomeningitis virus

Molecular analyses of a five amino acid cytotoxic T lymphocyte (CTL) epitope: An immunodominant region which induces nonreciprocal CTL cross-reactivity

Diversity of T-cell receptors in virus-specific cytotoxic T lymphocytes recognizing three distinct viral epitopes restricted by a single major histocompatibility complex molecule

Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system

This is Publication Number 7798-NP from the Department of Neuropharmacology, The Scripps Research Institute, La Jolla, CA. This work was supported in part by USPHS grants NS-12428 and AI-09484. Dale McFarlin was a friend, scientific collaborator and associate (of MBAO) for over the last two decades.