hCTLA4-Ig transgene expression in keratocytes modulates rejection of corneal xenografts in a pig to non-human primate anterior lamellar keratoplasty model.

BACKGROUND: Human corneal allografting is an established procedure to cure corneal blindness. However, a shortage of human donor corneas as well as compounding economic, cultural, and organizational reasons in many countries limit its widespread use. Artificial corneas as well as porcine corneal xenografts have been considered as possible alternatives. To date, all preclinical studies using de-cellularized pig corneas have shown encouraging graft survival results; however, relatively few studies have been conducted in pig to non-human primate (NHP) models, and particularly using genetically engineered donors. METHODS: In this study, we assessed the potential benefit of using either hCTLA4-Ig transgenic or α1,3-Galactosyl Transferase (GT) Knock-Out (KO) plus transgenic hCD39/hCD55/hCD59/fucosyl-transferase pig lines in an anterior lamellar keratoplasty pig to NHP model. RESULTS: Corneas from transgenic animals expressing hCTLA4-Ig under the transcriptional control of a neuron-specific enolase promoter showed transgene expression in corneal keratocytes of the stroma and expression was maintained after transplantation. Although a first acute rejection episode occurred in all animals during the second week post-keratoplasty, the median final rejection time was 70 days in the hCTLA4-Ig group vs. 21 days in the wild-type (WT) control group. In contrast, no benefit for corneal xenograft survival from the GTKO/transgenic pig line was found. At rejection, cell infiltration in hCTLA4Ig transgenic grafts was mainly composed of macrophages with fewer CD3+ CD4+ and CD79+ cells than in other types of grafts. Anti-donor xenoantibodies increased dramatically between days 9 and 14 post-surgery in all animals. CONCLUSIONS: Local expression of the hCTLA4-Ig transgene dampens rejection of xenogeneic corneal grafts in this pig-to-NHP lamellar keratoplasty model. The hCTLA4-Ig transgene seems to target T-cell responses without impacting humoral responses, the control of which would presumably require additional peripheral immunosuppression.

Ectopic expression of TrKA in the adult rat basal ganglia induces both nerve growth factor-dependent and -independent neuronal responses.

Ectopic expression of tropomyosin-related kinase A (TrkA), the high-affinity receptor of nerve growth factor (NGF), has been widely used in cell culture systems to uncover its role in cell survival or death events. In contrast, little is known about the consequences of its expression in vivo. To address this question, adeno-associated virus (AAV) vectors were used to express TrkA in the substantia nigra (SN) and striatum of adult rats. Nine weeks after transfer, tyrosine hydroxylase (TH) and dopamine transporter (DAT) mRNAs were slightly decreased in the ipsilateral SN. This decrease was no longer significant when NGF was delivered into the striatum. There was no change of DAT binding sites or D1 or D2 receptor mRNAs and binding sites in the striatum, suggesting that ectopic TrkA exerts a limited effect on the pool of TH and DAT transcripts, without affecting overall dopamine signaling. When transferred into the striatum, TrkA transgene had no effect on the size of the cholinergic interneurons, but it exerted typical neurotrophic effects, as shown by an enlargement of the projection neurons and nitric oxide synthase (nNOS)-expressing interneurons. This trophic action was amplified by a delivery of NGF. No toxic effect of the transgene was noted. These data indicate that ectopic expression of TrkA may result in the promotion of neurotrophic effects or can influence neuronal plasticity in the absence of exogenous NGF in neuronal populations that naturally fail to respond to this factor.

Immunoregulatory properties of neural stem cells.

Transplantation of neural cells provides an interesting form of therapy for certain CNS disorders. Although the brain has a special immune status, xenografts of fetal porcine neuroblasts are ultimately rejected after a lag of several weeks. Various strategies have been proposed to prevent this process. These include the design of transgenic pigs whose neurons have an increased immunosuppressive potential. An interesting alternative is provided by the use of neural stem/progenitor cells, which are multipotent cells found in the fetal or adult CNS. These cells are known to be poorly immunogenic. However, pig or rat neural stem/progenitor cells are highly immunosuppressive, as shown by their ability to block the proliferation of activated T lymphocytes. This effect is mediated by cell secreted factor(s), whose nature is discussed.

γ-Aminobutyric acid type B receptor changes in the rat striatum and substantia nigra following intrastriatal quinolinic acid lesions.

Changes in the regional distribution of the metabotropic GABA type B receptors (GABA(B)) were investigated in a rat model of Huntington's disease. Animals received a unilateral intrastriatal injection of quinolinic acid (QA), and GABA(B) immunoreactivity was monitored 3, 11, and 21 days postinjection in the striatum and substantia nigra (SN). Two antibodies, recognizing either the GABA(B1) or the GABA(B2) receptor subtypes, were used. QA injection rapidly induced a protracted increase in GABA(B1) or GABA(B2) immunoreactivity in the lesioned striatum, despite the neuronal loss. In the SN, a continuous increase in GABA(B1) and GABA(B2) immunoreactivity was observed at all time points in the ipsilateral pars reticulata (SNr), whereas the pars compacta (SNc) was unaffected by this phenomenon. This increase was supported by a densitometric analysis. At day 21 postlesion induction, intensely labeled stellate cells and processes were found in the ipsilateral SNr, in addition to immunoreactive neurons. Double labeling of GABA(B1) and glial fibrillary acidic protein (GFAP) showed that the stellate cells were reactive astrocytes. Hence, part of the sustained increase in GABA(B) immunoreactivity that takes place in the SNr and possibly the striatum may be ascribed to reactive astrocytes. It is suggested that GABA(B) receptors are up-regulated in these reactive astrocytes and that agonists might influence the extent of this astroglial reaction.

Trophic and immunoregulatory properties of neural precursor cells: benefit for intracerebral transplantation.

Intracerebral xenotransplantation of porcine fetal neuroblasts (pNB) is considered as an alternative to human neuroblasts for the treatment of neurodegenerative diseases. However, pNB are systematically rejected, even in an immunoprivileged site such as the brain. Within this context, neural stem/precursor cells (NSPC), which were suggested as exhibiting low immunogenicity, appeared as a useful source of xenogeneic cells. To determine the advantage of using porcine NSPC (pNSPC) in xenotransplantation, pNB and pNSPC were grafted into the striatum of rats without immunosuppression. At day 63, all the pNB were rejected while 40% of the rats transplanted with pNSPC exhibited large and healthy grafts with numerous pNF70-positive cells. The absence of inflammation at day 63 and the occasional presence of T cells in pNSPC grafts evoked a weak host immune response which might be partly due to the immunosuppressive properties of the transplanted cells. T cell proliferation assays confirmed such a hypothesis by revealing an inhibitory effect of pNSPC on T cells through a soluble factor. In addition to their immunosuppressive effect, in contrast to pNB, very few pNSPC differentiated into tyrosine hydroxylase-positive neurons but the cells triggered an intense innervation of the striatum by rat dopaminergic fibers coming from the substantia nigra. Further experiments will be required to optimize the use of pNSPC in regenerative medicine but here we show that their immunomodulatory and trophic activities might be of great interest for restorative strategies. This article is part of a Special Issue entitled "Interaction between repair, disease, & inflammation."

Minocycline promotes long-term survival of neuronal transplant in the brain by inhibiting late microglial activation and T-cell recruitment.

BACKGROUND: Cell therapy in the brain is limited by the requirement of high doses of immunosuppressors that have harmful side effects, and often, it cannot prevent the ultimate rejection of the transplanted cells. Alternative treatments that replace or enable a reduction in the doses of usual immunosuppressors have to be found. In this regard, minocycline shows potential as therapeutic agent. This drug crosses the blood-brain barrier, has good safety records, and exhibits strong antiinflammatory effects. METHODS: To study the impact of minocycline on the survival of intracerebral transplant, 400,000 porcine fetal neurons were transplanted into the striatum of rats treated daily with minocycline until sacrifice. Graft survival and immunologic reaction were evaluated by immunohistochemistry. RESULTS: In the control groups, all the grafts were rejected at day 63, whereas healthy grafts exhibiting tyrosine hydroxylase neurons were observed in 40% of the treated rats. The low immunoreactivity for ED1 and R73 in treated rats when compared with the control groups suggests that minocycline promotes long-term survival of neuronal xenograft by inhibiting microglial activation and T-cell recruitment. CONCLUSIONS: Our present data provide the first evidence of an effect of minocycline on the host immune response after neuronal transplantation into the brain. This observation raises new perspectives concerning the use of minocycline and provides basis for the development of safe and efficient immunosuppressive protocols for intracerebral transplantation.

Kinetics of microglial activation and degeneration of dopamine-containing neurons in a rat model of Parkinson disease induced by 6-hydroxydopamine.

In both Parkinson disease and in animal models of Parkinson disease, there is a microglial reaction in addition to the loss of dopaminergic neurons in the ventral midbrain. To determine the pathological role of this microglial reaction, we analyzed the kinetics of microglial activation and dopaminergic cell death induced in rats with the neurotoxin 6-hydroxydopamine. As early as Day 1 after the injection, there was a decline in the motor performance of the 6-hydroxydopamine-lesioned rats that correlated with a reduction of dopaminergic innervation of the contralateral striatum. Loss of dopaminergic neurons in the ventral midbrain developed a few days later and seemed to follow a specific temporospatial pattern. Degenerating neurons and activated microglia were seen only in areas in which dopaminergic cells were no longer observed, suggesting that the loss of the dopaminergic phenotype preceded the degenerative process. In sham-lesioned rats, there was a transient activation of microglia in the vicinity of the needle tract without any cell degeneration. This chronology of events supports the hypothesis that microglial activation is a secondary rather than primary phenomenon in dopaminergic cell degeneration induced by 6-hydroxydopamine.

Rat model of Parkinson's disease with bilateral motor abnormalities, reversible with levodopa, and dyskinesias.

Parkinson's disease (PD) is characterized by the bilateral degeneration of the midbrain dopamine-containing neurons with the most severe lesion in the posterolateral part of the substantia nigra pars compacta (SNpc). In humans, such lesions lead to specific motor abnormalities (i.e., akinesia, rigidity, and tremor) that are greatly improved by levodopa treatment. After a few years, the beneficial effect of the treatment is frequently offset by the development of dyskinesias. To improve treatment strategies, an animal model showing most of the histological and clinical characteristics of the human disease is mandatory. Ten rats received a bilateral injection of small doses of 6-OHDA in the medial forebrain bundle (MFB) and were compared with five sham-lesioned rats. The 6-OHDA-lesioned rats progressively developed abnormal motor behavior (assessed by the stepping test) compared with the sham-lesioned rats. The lesioned rats greatly improved under levodopa treatment, but developed concomitant dyskinesias. All 6-OHDA-lesioned animals had bilateral partial lesions of the SNpc, with the most severe lesion being in its posterolateral part. There was a significant correlation between the severity of the dopaminergic cell loss and the severity of the levodopa-induced dyskinesias. These rats constitute an interesting model of PD, sharing some of the main characteristics of the human disease.

Dendritic cell recruitment following xenografting of pig fetal mesencephalic cells into the rat brain.

Following transplantation into the rat brain, porcine neuroblasts differentiate and integrate host tissue, but due to their xenogeneic nature, these cells are generally rejected within several weeks. This rejection is accompanied by infiltration of the graft by macrophages and alphabetaT lymphocytes, but so far nothing is known about the potential role of dendritic cells (DCs) in this process. DCs are professional antigen presenting cells that have the unique ability to prime naive T cells, thereby initiating an antigen-directed immune response. Here, we provide evidence for DC recruitment following the transplantation of pig mesencephalic neural cells into the striatum of LEW.1A rats, as indicated by the high number of OX62+ cells in the rejecting graft and the absence of V65 staining. DCs were found as early as 3 and 8 days postimplantation together with ED1+ and OX42+ cells. This early recruitment, which is probably due to the surgical procedure, might be a critical step in the rejection process, enabling DCs to be loaded with xenoantigens. The number of intracerebral DCs subsequently decreased, being barely detectable in older non-infiltrated xenografts. However, DCs re-appeared as they were observed in grafts infiltrated by macrophages and T cells, a phenomenon that usually precedes graft rejection. Interestingly, we observed a tight correlation between the number of DCs and that of R7.3+ T cells infiltrating the graft. In addition, DCs were often found in close proximity to alphabetaT cells and most expressed MHCII. Taken together, these findings give credence to a role for infiltrating DCs in the mediation of T cell responses to intracerebral xenografting.

Beta1 integrin as a xenoantigen in fetal porcine mesencephalic cells transplanted into the rat brain.

Xenografts of fetal porcine mesencephalic cells implanted into the rat striatum are generally rejected within several weeks. The fetal donor mesencephalon predominantly consists of neurons, but also contains microglial and endothelial cells, which are more immunogenic. In the present work, we investigated the occurrence of donor endothelial cells in grafts of porcine mesencephalic cells implanted into the rat striatum. Pig endothelial cells were monitored by immunochemical methods, using a monoclonal antibody (mAb) that recognizes a peptidic epitope of the porcine beta1 integrin, and isolectin IB4, for the staining of the Galalpha1,3Gal epitope. The analysis also involved the detection of the pig hyaluronate receptor CD44, and the cell adhesion molecule CD31. The anti-beta1 integrin mAb revealed endothelial-like cells in grafts of porcine mesencephalic cells as soon as 1 week after implantation. A similar staining pattern was obtained with the IB4 lectin. Unlike aortic endothelial cells, these pig brain-derived endothelial-like cells were not recognized by the anti-CD44 antibody. They also failed to express the CD31 adhesion molecule, a fact which suggests that they remained poorly mature, even in grafts maintained during 45 days in immunosuppressed rats. Interestingly, a strong expression of beta1 integrin immunoreactivity was noticed in a large proportion (80%) of the cells freshly dissociated from the fetal pig mesencephalic tissue. The immunoreactivity decreased progressively after transplantation of the cells into the rat brain. This observation suggests that dissociated neuroblasts are capable of a temporary expression of beta1 integrin. This molecule is known to participate in the process of cell sorting and migration in the developing brain. Hence, its expression could be the hallmark of a rescue mechanism triggered by the disruption of the cell/matrix interactions during the dissociation of the fetal mesencephalon. This disruption might account for part of the dramatic cell death process that occurs during the manipulation of the donor tissue.

Transgenic expression of CTLA4-Ig by fetal pig neurons for xenotransplantation.

The transplantation of fetal porcine neurons is a potential therapeutic strategy for the treatment of human neurodegenerative disorders. A major obstacle to xenotransplantation, however, is the immune-mediated rejection that is resistant to conventional immunosuppression. To determine whether genetically modified donor pig neurons could be used to deliver immunosuppressive proteins locally in the brain, transgenic pigs were developed that express the human T cell inhibitory molecule hCTLA4-Ig under the control of the neuron-specific enolase promoter. Expression was found in various areas of the brain of transgenic pigs, including the mesencephalon, hippocampus and cortex. Neurons from 28-day old embryos secreted hCTLA4-Ig in vitro and this resulted in a 50% reduction of the proliferative response of human T lymphocytes in xenogenic proliferation assays. Transgenic embryonic neurons also secreted hCTLA4-Ig and had developed normally in vivo several weeks after transplantation into the striatum of immunosuppressed rats that were used here to study the engraftment in the absence of immunity. In conclusion, these data show that neurons from our transgenic pigs express hCTLA4-Ig in situ and support the use of this material in future pre-clinical trials in neuron xenotransplantation.

Compartmentalization of TCR repertoire alteration during rejection of an intrabrain xenograft.

Xenograft rejections of embryonic pig neural cells implanted into the adult rat striatum occurs within 3-4 weeks, following a dramatic T cell infiltration. Little is known about the cross-talk between the brain and peripheral lymphoid tissues which results in this recruitment and lymphocyte homing. To better characterize the dynamics of the T cell response against xenogeneic neural cells implanted into the brain parenchyma, we used both qualitative and quantitative methods to follow the alterations of the CDR3 length distribution (CDR3-LD) of the TCR (T cell receptor) beta chain in the transplanted striatum and compared this response to that observed in the deep cervical lymph nodes, spleen, and blood. Data showed that the T cell repertoire diversity was highly altered in the recipient brain during xenograft rejection. Comparison of the alterations of the CDR3-LD between several animals revealed a single public alteration in the Vbeta20 family, and many private alterations of the CDR3-LD which differed from one infiltrated brain to another. Alterations of the T cell repertoire were also observed in lymphocytes homed into the deep cervical lymph nodes. However, they differed from the alterations detected in the infiltrated brains. Conversely, no significant alteration of the CDR3-LD was detected in the spleen or in the blood. These data suggest that the deep cervical lymph nodes play an active role in the process of xenograft recognition or/and rejection. However, they also indicate that the fate of T cells homed in the brain and deep cervical lymph nodes differs.

Role of nigral lesion in the genesis of dyskinesias in a rat model of Parkinson's disease.

The pathogenesis of the motor fluctuations and dyskinesias that complicate levodopa treatment for Parkinson's disease (PD) remains uncertain. To evaluate the relationship between the degree of dopamine neuron loss and the severity of dyskinesias in a rodent model of PD, Sprague-Dawley rats were lesioned unilaterally using different doses of 6-hydroxydopamine injected into the substantia nigra pars compacta (SNc). All rats received two daily oral doses of levodopa for one month. In most of the animals chronic levodopa administration induced abnormal involuntary movements (AIMs), which were in some respects similar to human dyskinesias. We found that a minimum dopamine cell loss of around 95% was required for the development of dyskinesias after one-month of levodopa treatment. Moreover, we observed a positive relationship between the percentage dopaminergic cell loss in the SNc and the severity of levodopa-induced AIMs.

Blood T-cell receptor beta chain transcriptome in multiple sclerosis. Characterization of the T cells with altered CDR3 length distribution.

Multiple sclerosis is an inflammatory demyelinating disease of the CNS associated with T cells autoreactive for myelin components. In this study, we analysed the T-cell receptor (TCR) usage of the variable beta (Vbeta) chain transcriptome in the blood of multiple sclerosis patients at various stages of the disease using a global and quantitative comparison of the complementarity-determining region 3 length distribution (CDR3-LD) of transcripts of the 26 Vbeta genes. We investigated 35 patients: 12 with a high risk of multiple sclerosis, 10 with clinically definite multiple sclerosis, 13 with a relapsing-remitting worsening and active multiple sclerosis and 13 healthy individuals. Cells bearing the TCR transcripts with altered CDR3-LD were sorted and studied for CD4 or CD8 phenotype, cytokine transcript accumulation and response to human myelin basic protein (MBP). We show that patients from all the groups have a significantly skewed blood T-cell repertoire. Vbeta transcriptome patterns were more altered in patients from the clinically definite multiple sclerosis group and the worsening and active multiple sclerosis group than in the high risk group. The T cells sorted from Vbeta families with altered CDR3-LD concerned both CD4 and CD8 T cells, with a more pronounced skewing in the CD8 compartment. These cells displayed a significantly increased level of interferon-gamma, interleukin-2 and tumour necrosis factor-alpha transcripts compared with their counterparts from the healthy individual group. Furthermore, using interferon-gamma enzyme-linked immunospot (ELISPOT) assays, T cells from four out of seven altered Vbeta families tested from multiple sclerosis patients responded to human MBP, whereas no response was observed with human albumin or with altered Vbeta families from healthy individuals. Our data support the concept of an early autoimmune component in the disease and emphasize the possible involvement of CD8-positive T cells in multiple sclerosis.

Ectopic expression of the TrkA receptor in adult dopaminergic mesencephalic neurons promotes retrograde axonal NGF transport and NGF-dependent neuroprotection.

A recombinant adeno-associated virus (rAAV) was used to investigate the impact of an ectopic expression of the NGF high-affinity receptor in adult neurons. The rat TrkA cDNA cloned in a pCMX vector was first tagged with a human c-Myc sequence. The resulting vector was shown to encode a functional receptor which promoted the expression of TrkA immunoreactivity upon transfection of 293 fibroblasts or nnr5 cells, a TrkA-defective variant of PC12 cells. These cells also accumulate TrkA transcripts upon transfection and extended neurites in the presence of NGF. Therefore, the TrkA(myc) cassette was inserted into the pSSV9 plasmid. The new vectors shared properties similar to pCMX TrkA(myc) in 293 and nnr5 cells and enabled the preparation of rAAV TrkA(myc) viruses. Unilateral injection of this rAAV into the substantia nigra (SN) resulted in a protracted expression of TrkA (or c-Myc) immunoreactivity in numerous cell bodies, including tyrosine-hydroxylase (TH)-positive dopaminergic neurons. The presence of TrkA receptors in corresponding striatal dopaminergic endings was demonstrated by the advent of a striato-nigral retrograde axonal transport of (125)I-NGF. Likewise, ectopic expression of TrkA in neurons of the parafascicular thalamic nucleus promoted a striatofuge transport of NGF toward this structure. To investigate whether ectopic expression of TrkA in SN neurons may confer neuroprotection, lesions were induced by 6-hydroxydopamine in striata located ipsilateral to the virus injection site. NGF or vehicle were next delivered dorsally to the virus-treated SN for 2 weeks, before sacrifice and processing of brains for TH-immunohistochemistry. NGF treatment, in contrast to treatment with vehicle, significantly enhanced the number of dopaminergic neurons counted in the lesioned SN. These data suggest that ectopic TrkA can mediate the trophic actions of NGF and influence neuronal plasticity in vivo.

Lipopolysaccharide and TNFalpha regulate the expression of GDNF, neurturin and their receptors.

Inflammatory processes in the brain may trigger specific neuroprotective responses in glial cells. Here, we show that bacterial lipopolysaccharide strongly up-regulates glial derived neurotrophic factor (GDNF) mRNA while it down-regulates that of neurturin. Tumor necrosis factor alpha (TNFalpha) had different effects since it stimulated neurturin expression without enhancing GDNF mRNA. Interestingly, both lipopolysaccharide and TNFalpha triggered a significant decrease in the expression of the GDNF receptor, GFRalpha1, in glial cells. While the significance of such down-regulation during inflammatory processes remains to be characterised, the differential regulation of GDNF and neurturin following lipopolysaccharide and TNFalpha treatments suggest specific neuroprotective responses of glial cells in case of bacterial infection, trauma, transplantation or neurodegenerative diseases.

Treatment of experimental autoimmune encephalomyelitis in rat by 1,25-dihydroxyvitamin D3 leads to early effects within the central nervous system.

We report here that curative treatment of the multiple sclerosis paradigm, chronic relapsing experimental autoimmune encephalomyelitis (EAE) of the Lewis rat, by 1,25-dihydroxyvitamin D(3 )(1,25-D3) leads to a rapid clinical improvement accompanied by an inhibition of CD4, MHC class II and type II nitric oxide synthase (NOS II) expression in the posterior areas of the central nervous system (CNS). In contrast, the hormone has no effect on transforming growth factor-beta1 transcripts. Computer analysis of the NOS II promoter, expressed by microglia and astrocytes, reveals consensus sequence for vitamin D receptor binding, emphasizing the idea that 1,25-D3 may regulate some aspects of EAE by acting directly on CNS constituent cells. We also demonstrate that vitamin D deprivation leads to minimal effects on the kinetic profile of EAE accompanied by a moderate exacerbation of the clinical symptoms. Interestingly, curative treatment of vitamin D-deprived rats with a non-toxic-1,25-D3 analogue (MC1288) strongly inhibited EAE symptoms, thus promulgating the potential interest of such compounds in the management of multiple sclerosis.

Temporal analysis of cytokine gene expression during infiltration of porcine neuronal grafts implanted into the rat brain.

A large array of evidence supports the involvement of infiltrating T lymphocytes in the rejection process of intracerebral neuronal xenografts. Little is known, however, about the molecular mechanisms that drive the recruitment of this cell type. In the present work, we used real-time RT-PCR methodology to investigate the kinetics of cytokine gene expression during the infiltration of fetal porcine neurons (PNEU) implanted into the striatum of LEW.1A rats. T lymphocyte infiltration was followed by measuring the intracerebral levels of transcripts encoding the beta chain of the T cell receptor. These transcripts remained barely detectable until the fourth week (28 days) postimplantation, when a sudden accumulation occurred. Their kinetics, which support previous immunohistochemical observations, indicate that alphabetaT lymphocyte recruitment occurs rapidly after a delay of several weeks in this experimental model. Infiltration of PNEU grafts by T lymphocytes was accompanied by a concomitant, dramatic augmentation of transcripts coding for monocyte chemotactic protein-1 and RANTES (for regulated on activation, normal T cell expressed and secreted), two chemokines targeting this cell type, among others. Likewise, a sudden accumulation of transcripts of proinflammatory lymphokines [interleukin (IL)-1alpha, tumor necrosis factor-alpha, IL-6] as well as Th1 cytokines (IL-2, interferon-gamma) was also detected. In contrast, IL-4, -10, and -13 mRNA remained barely detectable at the different time points. No significant changes were noticed for IL-12 or transforming growth factor-beta transcripts. These data support the concept that T lymphocyte infiltration of PNEU grafts is actively promoted by a local production of chemokines and proinflammatory lymphokines and is based on a Th1 polarization.