In utero gene therapy rescues microcephaly caused by Pqbp1-hypofunction in neural stem progenitor cells.

Human mutations in PQBP1, a molecule involved in transcription and splicing, result in a reduced but architecturally normal brain. Examination of a conditional Pqbp1-knockout (cKO) mouse with microcephaly failed to reveal either abnormal centrosomes or mitotic spindles, increased neurogenesis from the neural stem progenitor cell (NSPC) pool or increased cell death in vivo. Instead, we observed an increase in the length of the cell cycle, particularly for the M phase in NSPCs. Corresponding to the developmental expression of Pqbp1, the stem cell pool in vivo was decreased at E10 and remained at a low level during neurogenesis (E15) in Pqbp1-cKO mice. The expression profiles of NSPCs derived from the cKO mouse revealed significant changes in gene groups that control the M phase, including anaphase-promoting complex genes, via aberrant transcription and RNA splicing. Exogenous Apc4, a hub protein in the network of affected genes, recovered the cell cycle, proliferation, and cell phenotypes of NSPCs caused by Pqbp1-cKO. These data reveal a mechanism of brain size control based on the simple reduction of the NSPC pool by cell cycle time elongation. Finally, we demonstrated that in utero gene therapy for Pqbp1-cKO mice by intraperitoneal injection of the PQBP1-AAV vector at E10 successfully rescued microcephaly with preserved cortical structures and improved behavioral abnormalities in Pqbp1-cKO mice, opening a new strategy for treating this intractable developmental disorder.

Rapid improvement of osseous sarcoidosis after the treatment of pulmonary aspergillosis by itraconazole.

Osseous sarcoidosis is relatively uncommon, and treatment with corticosteroids is not always effective. Moreover, patients with an advanced stage of pulmonary sarcoidosis are sometimes infected with aspergillus in the cavities of the pulmonary lesions, and long-term use of corticosteroids should be prohibited to prevent the development of fatal invasive pulmonary aspergillosis. Here, we described a unique case of osseous sarcoidosis with pulmonary aspergillosis, showing a rapid improvement of the osseous symptoms just after the administration of the antifungal agent, itraconazole. Itraconazole is likely to become a candidate among new therapeutic agents for osseous sarcoidosis.

The function of Ia+ dendritic cells and Ia- dendritic cell precursors in thymocyte mitogenesis to lectin and lectin plus interleukin 1.

The response of thymocytes to lectin is a standard tissue culture model for identifying cytokines such as IL-1 that are required for thymocyte mitogenesis. To study accessory cell requirements for these responses, it was necessary to deplete endogenous accessory cells with two techniques: anti-Ia and complement, and passage over nylon wool. Proliferation to Con A was then restored with 0.1-0.3% exogenous splenic dendritic cells, or 30-fold higher levels of peritoneal macrophages. The "costimulatory" action of IL-1, whereby responses to lectin were enhanced 3-10-fold, required the presence of dendritic cells. This effect of IL-1 could be reproduced by culturing the dendritic cells for 12 h in 1 U/ml human or murine rIL-1 alpha before addition to the thymocyte proliferation assay. The function of IL-1-treated dendritic cells was not blocked by a neutralizing anti-IL-1 antibody. The endogenous population of thymic accessory cells was partially characterized. A trace (0.1-0.3%) fraction of Ia+, Ig-, plastic nonadherent dendritic cells was visualized and enriched to a level of 1-10% by depleting CD4+,CD8+, and Ig+ lymphocytes. When this double-negative population was cultured with IL-1 and washed, the treated thymic dendritic cells were 10-fold more active as accessory cells. When the CD4-,CD8-, Ig- populations were depleted of dendritic cells with anti-Ia and complement, the subsequent addition of IL-1 had a second effect. Ia+ dendritic cells redeveloped over a 2-d interval, and they exhibited the same properties as resident dendritic cells in thymus and spleen. The majority were lysed by 33D1 anti-dendritic cell mAb and complement, lacked Fc receptors, and acted as powerful stimulators of the MLR and Con A mitogenesis. The development of dendritic cells did not occur with IL-2, -3, -4 or granulocyte/macrophage colony-stimulating factor or in nylon-nonadherent populations. The IL-1-dependent, Ia- precursor was not detectable in bone marrow. These results begin to analyze the endogenous accessory function of the thymus in culture. Dendritic cells actively stimulate thymocyte mitogenesis. The mitogenic action of IL-1 involves effects on resident Ia+ dendritic cells as well as a new population of thymic, Ia- precursors.

Functional analyses of thymic CD5+ B cells. Responsiveness to major histocompatibility complex class II-restricted T blasts but not to lipopolysaccharide or anti-IgM plus interleukin 4.

The function of thymic B cells in several standard in vitro assays was investigated. Thymic B cells, 75% of which were CD5+, showed a poor responsiveness to the mitogens LPS or anti-mu plus IL-4. Both proliferation and antibody formation were much lower in thymic than splenic B cell cultures. However, CD5- B cells purified using a cell sorter responded well to B cell stimulants, whereas purified CD5+ thymic B cells did not, indicating that CD5+ thymic B cells were unresponsive to B cell growth factor or LPS. Thymic B cells could be activated polyclonally by direct interaction with alloreactive T blasts, as manifested by DNA synthesis and antibody formation. These findings indicate that CD5+ thymic B cells may not be stimulated via sIg and IL-4, but require instead direct interaction with T blasts.

Contrasting effect of alpha/beta- and gamma-interferons on expression of macrophage Ia antigens.

IFN-gamma is known to induce the expression of Ia antigens on macrophages. We found that murine IFN-alpha and -beta blocked the effects of IFN-gamma in a dose-dependent manner. The antagonistic effect of IFN-alpha and -beta was observed even when macrophages were prestimulated with IFN-gamma. These inhibitory effects of IFN-alpha or -beta were blocked by their respective antibodies. The block exerted by IFN-alpha/beta was similar whether Ia levels were monitored by immunofluorescence with anti-Ia mAb, or by stimulation of freshly sensitized, alloreactive T lymphoblasts. Adherent macrophage-rich populations from newborn mice were incapable of expressing Ia antigens following stimulation with IFN-gamma, and would inhibit the response of adult macrophages to this lymphokine. Addition of anti-IFN-beta mAb, but not anti-IFN-alpha allowed newborns' macrophages to express Ia in response to IFN-gamma, and ablated the suppressive activity toward adult cells. These results indicate that IFN-alpha and -beta, which can be produced in the course of self-defense responses and during ontogeny, may contribute to the down-regulation of macrophage Ia expression.

Unusual phenotype of B cells in the thymus of normal mice.

A small number of B cells are found in the thymus of normal mice. A population of B lymphocytes could be enriched to greater than 90% purity by isolating a low-density fraction on Percoll density gradients and then depleting T cells with a mixture of anti-Thy-1, CD4, and CD8 mAbs and complement. Enrichment was monitored by surface Ig staining and by functional studies (responsiveness to LPS, and to anti-mu plus IL-4). When the phenotype of these B cells was studied by flow cytometry, 60-80% had the phenotype Ly-1+ (CD5), Ia+, B220low (CD45R), and Mac-1+ (CD 11b). In contrast, splenic B cells lacked CD5 and CD11b and expressed higher levels of B220 and Ia antigens. These results indicate that most thymic B cells have the phenotype of the Ly-1 B cell subset, which was identified previously as a trace subpopulation in some peripheral tissues and is thought to play a role in autoantibody formation.

Neuraminidase-treated macrophages stimulate allogenic CD8+ T cells in the presence of exogenous interleukin 2.

Prior work has shown that purified, resident, and inflammatory peritoneal macrophages are weak stimulators of the allogeneic MLR. We have identified conditions whereby thioglycollate-elicited macrophages become stimulatory, but primarily for the CD8+ T cell subset. The conditions were to treat the macrophages with neuraminidase and to supplement the MLR with rIL-2. These treatments together led to proliferative and cytotoxic responses by isolated CD8+ but not CD4+ T cells. Likewise when MHC-congenic strains were evaluated, an MLR was observed across isolated class I but not class II MHC barriers. Pretreatment of the macrophages with IFN-gamma further enhanced expression of class I MHC products and stimulatory activity, but did not seem essential. While these treatments did not render macrophages stimulatory for an MLR in purified CD4+ cells, blastogenesis of CD4+ cells was observed when the MLR involved bulk T cells. Small allogeneic B lymphocytes behaved similarly to macrophages, in the pretreatment with neuraminidase and supplementation with rIL-2 rendered B cells stimulatory for allogeneic, enriched, CD8+, but not CD4+, T cells. Spleen adherent cells, which are mixtures of macrophages and dendritic cells, stimulated both CD4+ and CD8+ T cells, and neither neuraminidase nor exogenous IL-2 was required. We think that these data suggest that most macrophages and small B cells lack three important functions of dendritic cells: a T cell-binding function that can be remedied by neuraminidase treatment, a T cell growth factor-inducing function that can be bypassed with exogenous IL-2, and an IL-2 responsiveness function that is required by CD4+ lymphocytes.

Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor.

Antigen-presenting, major histocompatibility complex (MHC) class II-rich dendritic cells are known to arise from bone marrow. However, marrow lacks mature dendritic cells, and substantial numbers of proliferating less-mature cells have yet to be identified. The methodology for inducing dendritic cell growth that was recently described for mouse blood now has been modified to MHC class II-negative precursors in marrow. A key step is to remove the majority of nonadherent, newly formed granulocytes by gentle washes during the first 2-4 d of culture. This leaves behind proliferating clusters that are loosely attached to a more firmly adherent "stroma." At days 4-6 the clusters can be dislodged, isolated by 1-g sedimentation, and upon reculture, large numbers of dendritic cells are released. The latter are readily identified on the basis of their distinct cell shape, ultrastructure, and repertoire of antigens, as detected with a panel of monoclonal antibodies. The dendritic cells express high levels of MHC class II products and act as powerful accessory cells for initiating the mixed leukocyte reaction. Neither the clusters nor mature dendritic cells are generated if macrophage colony-stimulating factor rather than granulocyte/macrophage colony-stimulating factor (GM-CSF) is applied. Therefore, GM-CSF generates all three lineages of myeloid cells (granulocytes, macrophages, and dendritic cells). Since > 5 x 10(6) dendritic cells develop in 1 wk from precursors within the large hind limb bones of a single animal, marrow progenitors can act as a major source of dendritic cells. This feature should prove useful for future molecular and clinical studies of this otherwise trace cell type.

Distinct mechanisms of neonatal tolerance induced by dendritic cells and thymic B cells.

To assess the role of different types of antigen-presenting cells (APC) in the induction of tolerance, we isolated B cells, macrophages, and dendritic cells from thymus and spleen, and injected these into neonatal BALB/c mice across an Mls-1 antigenic barrier. One week after injection of APC from Mls-1-incompatible mice or from control syngeneic mice, we measured the number of thymic, Mls-1a-reactive, V beta 6+ T cells and the capacity of thymocytes to induce a graft-vs.-host (GVH) reaction in popliteal lymph nodes of Mls-1a mice. Injection of thymic but not spleen B cells deleted thymic, Mls-1a-reactive V beta 6+ T cells and induced tolerance in the GVH assay. The thymic B cells were primarily of the CD5+ type, and fluorescence-activated cell sorter-purified CD5+ thymic B cells were active. Injection of dendritic cells from spleen or thymus also induced tolerance, but the V beta 6 cells were anergized rather than deleted. Macrophages from thymus did not induce tolerance. Dendritic cells and thymic B cells were also effective in inducing tolerance even when injected into Mls-, major histocompatibility complex-incompatible, I-E- mice, but only thymic B cells depleted V beta 6-expressing T cells. Therefore, different types of bone marrow-derived APC have different capacities for inducing tolerance, and the active cell types (dendritic cells and CD5+ thymic B cells) can act by distinct mechanisms.

The tissue distribution of the B7-2 costimulator in mice: abundant expression on dendritic cells in situ and during maturation in vitro.

B7-2 is a recently discovered, second ligand for the CTLA-4/CD28, T cell signaling system. Using the GL-1 rat monoclonal antibody (mAb), we monitored expression of B7-2 on mouse leukocytes with an emphasis on dendritic cells. By cytofluorography, little or no B7-2 was detected on most cell types isolated from spleen, thymus, peritoneal cavity, skin, marrow, and blood. However, expression of B7-2 could be upregulated in culture. In the case of epidermal and spleen dendritic cells, which become highly immunostimulatory for T cells during a short period of culture, the upregulation of B7-2 was dramatic and did not require added stimuli. Lipopolysaccharide did not upregulate B7-2 levels on dendritic cells, in contrast to macrophages and B cells. By indirect immunolabeling, the level of staining with GL-1 mAb exceeded that seen with rat mAbs to several other surface molecules including intercellular adhesion molecule 1, B7-1, CD44, and CD45, as well as new hamster mAbs to CD40, CD48, and B7-1/CD80. Of these accessory molecules, B7-2 was a major species that increased in culture, implying a key role for B7-2 in the functional maturation of dendritic cells. B7-2 was the main (> 90%) CTLA-4 ligand on mouse dendritic cells. When we applied GL-1 to tissue sections of a dozen different organs, clear-cut staining with B7-2 antigen was found in many. B7-2 staining was noted on liver Kupffer cells, interstitial cells of heart and lung, and profiles in the submucosa of the esophagus. B7-2 staining was minimal in the kidney and in the nonlymphoid regions of the gut, and was not observed at all in the brain. In the tongue, only rare dendritic cells in the oral epithelium were B7-2+, but reactive cells were scattered about the interstitial spaces of the muscle. In all lymphoid tissues, Gl-1 strongly stained certain distinct regions that are occupied by dendritic cells and by macrophages. For dendritic cells, these include the thymic medulla, splenic periarterial sheaths, and lymph node deep cortex; for macrophages, the B7-2-rich regions included the splenic marginal zone and lymph node subcapsular cortex. Splenic B7-2+ cells were accessible to labeling with GL-1 mAb given intravenously. Dendritic cell stimulation of T cells (DNA synthesis) during the mixed leukocyte reaction was significantly (35-65%) blocked by GL-1.(ABSTRACT TRUNCATED AT 400 WORDS)

Massive pleural effusion and bronchopleural fistula in Wegener's granulomatosis.

Wegener's granulomatosis (WG) is characterized by systemic granulomatous necrotizing vasculitis, primarily affecting the respiratory tract and kidneys. We describe a rare case in a 28-year-old woman with WG, presenting with a massive lateral pleural effusion, accompanied by an aseptic bronchopleural fistula formed during immunosuppressive treatment. Although any organ can be involved in WG, only left pleuritis and a purpuric lesion on the neck were detected in this case. The pleural effusion and bronchopleural fistula resolved following immunosuppressive treatment for six months. Thus, WG should be considered in the differential diagnosis of a massive pleural effusion, and fistula formation is a possible complication of treatment. Moreover, immunosuppressive treatment was sufficient to resolve the massive pleural effusion and fistula formation without infection (120 words).

Fibrinogen Lima: a homozygous dysfibrinogen with an A alpha-arginine-141 to serine substitution associated with extra N-glycosylation at A alpha-asparagine-139. Impaired fibrin gel formation but normal fibrin-facilitated plasminogen activation catalyzed by tissue-type plasminogen activator.

An A alpha-arginine-141 to serine substitution has been identified in a homozygous dysfibrinogen, fibrinogen Lima, associated with impaired fibrin polymerization. The point mutation created an asparagine-X-serine-type glycosylation sequence, and indeed, extra, mainly disialylated biantennary oligosaccharides have been isolated from A alpha asparagine-139 of the patient's fibrinogen. This type of glycosylation sequence is unique for human fibrinogen, because the sequences shown for normal and abnormal fibrinogens are all asparagine-X-threonine types. The terminal sialic acids of the extra oligosaccharides seem to have largely contributed to the impaired fibrin gel formation, as evidenced by its correction to a near normal level by desialylation. Nevertheless, the polymerizing fibrin facilitated tissue-type plasminogen activator-catalyzed plasmin formation in a normal fashion, indicating that the initial two-stranded fibrin protofibrils had been constructed normally. Thus the impaired fibrin gel formation could be attributed to the delay in their subsequent lateral association, most probably because of the repulsive forces generated by the negative electric charge of the extra sialic acids. The substitution of a basic residue arginine to a noncharged residue serine may also have contributed to the impaired function in a similar manner or by steric hindrance in association with bulky extra oligosaccharide chains.

Effects of recombinant alpha-interferon-gelatin conjugate on in vivo murine tumor cell growth.

A recombinant human alpha-interferon A/D (IFN), also known to be effective in mice, was conjugated to gelatin with a water-soluble carbodiimide. The IFN-gelatin conjugate was much more efficient than free IFN in activating mouse peritoneal macrophages (Mø) in an in vitro experiment to inhibit the growth of IFN-resistant subline cells (RR1) of murine fibrosarcoma. A single i.p. injection of the conjugate administered to normal mice was also more effective than one of free IFN in activating peritoneal Mø and natural killer cells in peritoneal exudate cell and spleen cell populations. In the investigation on body distribution of the IFN-gelatin conjugate, an enhanced affinity to Mø as well as a prolonged retention were observed in comparison with free IFN. An injection of the IFN-gelatin conjugate i.p. was more effective than one of free IFN in suppressing the in vivo growth of not only IFN-sensitive SS2 cells but also RR1 cells in the peritoneal cavity of mice, although RR1 cells were only susceptible to the indirect effect of IFN via host cells, in contrast to SS2 cells. In addition to an increased recruitment of Mø to the peritoneal cavity in RR1-bearing mice receiving i.p. injection of the IFN-gelatin conjugate, these Mø were activated to inhibit the in vitro growth of RR1 cells. These results indicate that the IFN-gelatin conjugate is a promising antitumor agent that is much more effective than free IFN. The dose of IFN in the conjugate required for exerting the antitumor effects is much lower than that of free IFN, which leads to a reduction of adverse effects of IFN.

Effect of recombinant human interferon-alpha A/D on in vivo murine tumor cell growth.

We investigated the effect of human recombinant interferon-alpha A/D A/D-IFN), which is known to delay the growth of murine tumor cells, on the growth of S1 and R1 subline cells of murine Meth A fibrosarcoma in the peritoneal cavity of mice. In vitro growth of S1 cells was sensitive to, and that of R1 cells was resistant to, the direct effect of A/D-IFN, as with murine natural IFN-alpha/beta, which was used originally to isolate these sublines. In vivo, however, the growth of not only S1 cells but also R1 cells was suppressed by the administration of A/D-IFN, and the survival time of tumor-bearing mice was prolonged. Although A/D-IFN had a direct effect on S1 cells in vivo, R1 cells were susceptible only to the indirect effect via the host cells. Macrophages (M phi) harvested from the peritoneal cavity of A/D-IFN-treated mice bearing ascitic R1 cells were very effective in suppressing the in vitro growth of R1 cells; those from non-R1-bearing A/D-IFN-treated mice were less effective. The results of in vitro experiments indicate that M phi are very probably activated by the synergism of A/D-IFN and M phi diameter-activating factor(s) produced by lymphoid cells in tumor-bearing mice.

Direct and indirect effects of interferon on in vivo murine tumor cell growth.

We cloned two sublines (S1 and R1) of murine Meth A fibrosarcoma cells with respect to their sensitivity to a murine alpha/beta-interferon (IFN) preparation. The growth of S1 cells was suppressed and that of R1 cells was hardly affected by IFN in vitro. This was also the case with cells enclosed in cell-impermeable diffusion chambers in peritoneal cavities. Nevertheless, IFN suppressed the growth of not only S1 cells but also R1 cells in mice inoculated i.p. with these cells, and the survival rates of both S1 cell recipients and R1 cell recipients were markedly improved. S1 cells were observed microscopically to be injured by the direct effect of IFN in vitro and in vivo, but R1 cells in in vitro culture with IFN and those surviving in vivo in the presence of IFN appeared to proliferate well. In the peritoneal cavity of R1 recipients treated daily with IFN, the recruitment of macrophages was enhanced in comparison with untreated R1 recipients. Adherent peritoneal exudate cells obtained from IFN-treated, R1-bearing mice were highly suppressive for the in vitro growth of not only R1 cells but also allogeneic and human cells. The role of macrophages in the indirect effect of IFN on tumor cell growth is discussed.

Regional pharmacokinetics of doxorubicin following hepatic arterial and portal venous administration: evaluation with hepatic venous isolation and charcoal hemoperfusion.

We evaluated the regional pharmacokinetics of doxorubicin after hepatic arterial infusion (HAI) and portal venous infusion (PVI) using a novel system for hepatic venous isolation and charcoal hemoperfusion (HVI-CHP). The HVI-CHP system was used to determine directly the doxorubicin plasma concentration in the hepatic vein and the hepatic venous flow rate, and simultaneously, to eliminate hepatic re-entry of the drug. Beagles received doxorubicin (1 mg/kg) through either the hepatic artery (HAI group, n = 6) or the portal vein (PVI group, n = 6). In both groups, hepatic venous blood was completely isolated and directed to the CHP filter. The filtered blood was returned through the left jugular vein. During HVI-CHP, the hepatic venous flow rate was monitored and plasma doxorubicin concentrations were serially measured in prefilter (= hepatic venous), postfilter, and systemic blood. The hepatic tissue uptake of doxorubicin was determined based on the blood flow rate and doxorubicin level in the hepatic vein. The hepatic extraction ratio of doxorubicin was defined as the percentage hepatic tissue uptake to the amount of drug administered. During drug infusion, similarly in either group, HVI-CHP produced a 66-87% reduction of the postfilter doxorubicin level as compared with the prefilter level. The prefilter drug level was significantly lower in HAI group than in PVI group (P < 0.01). Thus, the area under the time concentration curve for the prefilter drug level in the HAI group (6.90+/-0.96 microg min/ml) was significantly lower than that in the PVI group (18.10+/-2.90 microg min/ml, P < 0.01). Conversely, the hepatic extraction ratio in the HAI group (84.6+/-2.9%) was significantly higher than that in the PVI group (58.1+/-3.4%, P < 0.01). We conclude that in the beagle, doxorubicin is more effectively extracted by the liver when administered via the hepatic artery than when administered via the portal vein. These results indicate that HAI of doxorubicin is superior to PVI in terms of reduction of systemic drug exposure and systemic toxicity.

Competitive proliferation in the hematopoietic tissues of irradiated hybrid mice engrafted with parental bone marrow and spleen.

The kinetics of growth and differentiation of hematopoietic stem cells differ markedly according to their origin. A study of the ability of CFU from bone marrow (BM) or spleen to repopulate hemopoietic organs has been carried out in lethally irradiated mice restored with BM cells admixed with spleen cells bearing different chromosomal markers. Hemopoietic cells originating from AKR (40 acrocentrics) and AKR/T1ALD (36 acrocentrics + 2 metacentrics) mice were engrafted into lethally irradiated (AKR X AKR/T1ALD)F1 or (C3H X AKR/T1ALD)F1 hybrid recipients. Within 10 days, the BM-derived elements outnumbered the spleen-derived population in BM and spleen. This held even when the number of injected spleen-CFU was twice that of BM-CFU. This difference of growth rate subsided within 20 days. The first cells to reappear in the thymus bore the recipient karyotype (endoregeneration); they were later replaced by BM-derived elements but spleen-derived cells were never present in thymus in the case of competitive engraftment. In contrast, the lymph node cells bore the BM karyotype as well as the spleen karyotype. Injecting the spleen cells 3 days prior to the BM cells partially counterbalanced the over-growth of the BM-derived elements in the BM and spleen but did not affect the thymic repopulation which remained strictly derived from BM-CFU. When mice were injected only with BM-CFU or only with spleen-CFU, BM-derived cells were found in the thymus as early as 10-12 days after engraftment whereas the spleen-derived cells did not appear in the thymus until days 18-20.

Triple transduction with adeno-associated virus vectors expressing tyrosine hydroxylase, aromatic-L-amino-acid decarboxylase, and GTP cyclohydrolase I for gene therapy of Parkinson's disease.

Parkinson's disease (PD), a neurological disease suited to gene therapy, is biochemically characterized by a severe decrease in the dopamine content of the striatum. One current strategy for gene therapy of PD involves local production of dopamine in the striatum achieved by inducing the expression of enzymes involved in the biosynthetic pathway for dopamine. We previously showed that the coexpression of tyrosine hydroxylase (TH) and aromatic-L-amino-acid decarboxylase (AADC), using two separate adeno-associated virus (AAV) vectors, resulted in more effective dopamine production and more remarkable behavioral recovery in 6-hydroxydopamine-lesioned parkinsonian rats, compared with the expression of TH alone. Not only levels of TH and AADC but also levels of tetrahydrobiopterin (BH4), a cofactor of TH, and GTP cyclohydrolase I (GCH), a rate-limiting enzymes for BH4 biosynthesis, are reduced in parkinsonian striatum. In the present study, we investigated whether transduction with separate AAV vectors expressing TH, AADC, and GCH was effective for gene therapy of PD. In vitro experiments showed that triple transduction with AAV-TH, AAV-AADC, and AAV-GCH resulted in greater dopamine production than double transduction with AAV-TH and AAV-AADC in 293 cells. Furthermore, triple transduction enhanced BH4 and dopamine production in denervated striatum of parkinsonian rats and improved the rotational behavior of the rats more efficiently than did double transduction. Behavioral recovery persisted for at least 12 months after stereotaxic intrastriatal injection. These results suggest that GCH, in addition to TH and AADC, is important for effective gene therapy of PD.