- Email: [email protected]
Adenoviral-mediated expression of porphobilinogen deaminase in liver restores the metabolic defect in a mouse model of acute intermittent porphyria
ARTICLE
doi:10.1016/j.ymthe.2004.05.018
Adenoviral-Mediated Expression of Porphobilinogen Deaminase in Liver Restores the Metabolic Defect in a Mouse Model of Acute Intermittent Porphyria Annika Johansson,1,* Grzegorz Nowak,2 Christer Mo¨ller,3 Pontus Blomberg,4 and Pauline Harper1 1
2
Porphyria Centre Sweden, Division of Clinical Chemistry, Department of Laboratory Medicine, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden Division of Transplantation Surgery, Department of Laboratory Medicine, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden 3 HemeBiotech A/S, 181 70 Lidingo¨, Sweden 4 Vecura, Clinical Research Center, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden *To whom correspondence and reprint requests should be addressed. Fax: +46 8 585 827 60. E-mail: [email protected].
The aim of this study was to investigate the potential of gene therapy in the treatment of acute intermittent porphyria (AIP), a disorder caused by a partial deficiency of porphobilinogen deaminase (PBGD), the third enzyme in heme synthesis. The condition is biochemically characterized by accumulation of the porphyrin precursors 5-aminolevulinic acid (ALA) and porphobilinogen (PBG). Here we present the first experiments in vivo using adenoviral vectors to replace the deficient enzyme in the liver of an AIP mouse model. The use of adenoviral vector carrying the cDNA of luciferase in wild-type mice confirmed that transgene expression after intravenous administration was found mainly in liver. When PBGD-deficient mice were administered with adenoviral vector carrying the cDNA of mouse PBGD, the hepatic PBGD activity increased in a dose- and time-dependent manner. The highest activity was found 7 days after injection and remained high after 29 days. The expressed enzyme was shown to correct the metabolic defect in the PBGD-deficient mice as no accumulation of ALA or PBG occurred in plasma, liver, or kidney after induction of heme synthesis by phenobarbital. The study demonstrates that hepatic PBGD expression prevents the accumulation of porphyrin precursors, suggesting a future potential for gene therapy in AIP. Key Words: acute intermittent porphyria, 5-aminolevulinic acid, porphobilinogen, porphobilinogen deaminase, recombinant adenoviral vector
INTRODUCTION Acute intermittent porphyria (AIP) is the most common type of acute hepatic porphyria [1], affecting about 1 in 20,000 in the European population [2], with a prevalence of up to 1 in 10,000 in Sweden [3]. It is an autosomal dominant disorder caused by a deficiency of porphobilinogen deaminase (PBGD; EC 4.3.1.8), which is the third enzyme in the dedicated heme biosynthetic pathway. Clinical symptoms include acute episodes of abdominal pain often accompanied with peripheral neuropathy and mental disturbances [4]. Without treatment the disorder may be fatal [5]. The precipitating factors, such as certain drugs, alcohol, fasting, stress, and reproductive hormones, have the capacity to induce heme synthesis [6]. Under conditions of increased demand for heme the
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy 1525-0016/$30.00
deficient enzyme becomes rate limiting and as a consequence the heme precursors upstream of PBGD, 5-aminolevulinic acid (ALA) and porphobilinogen (PBG), accumulate and are excreted in urine [4]. Increased levels of ALA and PBG have been found mainly in plasma and liver from AIP patients with active disease [7,8], and the fact that recurrent acute attacks were prevented by liver transplantation in a young AIP patient [9] indicates that AIP is primarily a hepatic disorder. Current treatment protocols involve glucose and heme administration [10]. This approach has improved the prognosis of the disorder but may fail to reverse an established neuropathy [11]. In addition, adverse effects in the form of deteriorated peripheral venous access have been reported connected with repeated heme administration [12]. Thus, it would be desirable to find new
337
ARTICLE
therapeutic alternatives. Since AIP is caused by a defect in the PBGD gene, gene therapy is one obvious treatment strategy. Recently, we demonstrated in vitro that the use of nonviral vectors carrying the cDNA of PBGD gives rise to expression of active PBGD [13] that can correct the biochemical defect in PBGD-deficient cells [14]. Different strategies have been tested for introducing the PBGD cDNA into a mouse model of AIP using nonviral vectors, but due to the low efficiency of gene transfer no increase in PBGD expression has been found [15]. Another strategy for obtaining a high level of hepatic gene transfer involves the use of recombinant adenoviral vectors, which have been shown to target mainly the liver in rodents when administered systemically [16,17]. Adenoviral vectors have a number of advantages such as the ability to be produced at a high titer and to transduce postmitotic cells. The major drawbacks, however, are transient expression of the transgene and ability to trigger the host cellular immune response [18,19]. The objective of the present study was to evaluate the potential of gene therapy in AIP using recombinant adenoviral vectors in a mouse model for the disease, known to display biochemical features of AIP after administration of phenobarbital [20 – 22]. The AIP mouse model has been engineered with a residual hepatic PBGD activity of 30% of normal [22] and phenobarbital is a strong inducer of the first and rate-limiting step of heme synthesis, ALA synthase [20]. Here we present the first successful results using adenoviral-mediated gene transfer of PBGD cDNA to the PBGD-deficient mice, resulting in a high hepatic PBGD expression in vivo, and its abolishing effect on the accumulation of ALA and PBG during the phenobarbital induction of heme synthesis.
RESULTS Tissue Distribution of Luciferase Activity We determined the tissue distribution of transgene expression in wild-type (C57BL/6) mice following intrave-
doi:10.1016/j.ymthe.2004.05.018
nous administration of 1 109 infectious particles of recombinant adenoviral vector carrying the luciferase cDNA (Ad-EGFPLuc). Forty-eight hours after injection, more than 98% of the total luciferase expression was found in the liver tissue, with much lower levels in the other tissues (Fig. 1). Dose – Response Study on Hepatic PBGD Expression We determined the activity of PBGD in liver in PBGDdeficient mice 48 h after injection of increasing doses, from 1 105 to 5 108 infectious particles, of recombinant adenovirus coding for PBGD (Ad-PBGD). We observed endogenous hepatic PBGD activity in the PBGD-deficient mice to be about 30% of the level found in wild-type mice. Following injection of Ad-PBGD into the PBGD-deficient mice, we observed a dose-dependent increase in the activity of the enzyme. The highest PBGD activity was obtained using 5 108 infectious particles, which resulted in a 24-fold increase in the basal enzymatic activity found in wild-type mice (Fig. 2). No increase in PBGD activity was found in other tissues, i.e., spleen, kidney, lung, heart, and brain (data not shown). Time-Course Study on Hepatic PBGD Expression We determined the hepatic PBGD activity in PBGD-deficient mice 2, 7, 14, and 29 days following injection of 1 108 infectious particles of Ad-PBGD. We observed the highest activity after 7 days, corresponding to a 29-fold increase compared to the basal activity of wild-type mice (Fig. 3). The PBGD activity slowly decreased but remained 4-fold over the basal activity 29 days after the administration of Ad-PBGD. Effects of Hepatic PBGD Expression on Plasma and Tissue Levels of ALA and PBG We studied the metabolic effects of the increased hepatic PBGD activity in PBGD-deficient mice injected with 5 107 infectious particles of Ad-PBGD 5 days prior to the
FIG. 1. Luciferase activity in tissues from wild-type C57BL/6 mice 48 h after intravenous administration of 1 109 infectious particles of the recombinant adenoviral vector Ad-EGFPLuc. The luciferase activity, expressed as RLU/mg protein, was determined in liver, spleen, kidney, lung, heart, and brain tissue homogenates. Data are presented as means F SD of four animals.
338
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
ARTICLE
doi:10.1016/j.ymthe.2004.05.018
FIG. 2. Dose response of hepatic PBGD activity in PBGD-deficient mice injected with recombinant adenovirus (Ad-PBGD). The PBGD-deficient mice were intravenously given increasing amounts of AdPBGD (1 105 to 5 108 infectious particles). Hepatic PBGD activity, expressed as pkat/g protein, was determined 48 h after Ad-PBGD administration. The basal hepatic enzyme activity in the PBGDdeficient mice was about 30% of that found in wildtype (C57BL/6) mice. Values are means F SD of three animals.
first phenobarbital injection. We chose this titer on basis of the results from the dose – response assay, determined 48 h after Ad-PBGD administration, in which it resulted in a PBGD activity that was 3.5-fold the wild-type level. The hepatic PBGD activity in the Ad-PBGD-treated mice, determined at the end of the experiment 9 days after AdPBGD administration, was 2121 F 725 pkat/g protein (n = 5), compared to 34 F 6 pkat/g protein (n = 8) in the PBGD-deficient control animals and 135 F 18 pkat/g protein (n = 3) in wild-type animals. We did not observe any accumulation of plasma ALA and PBG during the phenobarbital induction period in the mice subjected to gene therapy (Fig. 4). In contrast, in the PBGD-deficient mice given only phenobarbital, the plasma levels of ALA and PBG gradually increased during the 4-day phenobarbital-induction period (Fig. 4). We observed an analogous response in phenobarbital-induced PBGD-deficient mice that were mock transduced with 5 107 Ad-LacZ infectious particles (data not shown). The plasma levels of ALA and PBG in the controls, i.e., in the PBGD-deficient mice not treated with Ad-PBGD and phenobarbital, were low at all time points (Fig. 4). Six hours after the fourth phenobarbital injection we observed a similar pattern in liver and kidney, i.e., undetectable levels of ALA and PBG in
the PBGD-deficient mice treated with phenobarbital and Ad-PBGD and high levels in the PBGD-deficient mice treated only with phenobarbital (Table 1).
DISCUSSION The prevailing hypothesis of the pathogenesis of AIP symptoms is based on the assumption of a neurotoxic effect of the accumulated porphyrin precursors ALA and PBG and/or a lack of heme [4]. Here we present the first positive results from experiments in vivo using phenobarbital-induced PBGD-deficient mice as a model for AIP, in which the accumulation of ALA and PBG is prevented by increased hepatic expression of PBGD by use of recombinant adenoviral vectors. Recombinant adenoviral vectors can transduce nondividing cells with high efficiency and rapidly accumulate in the liver after systemic administration [17,23], making them a useful tool for liver-directed gene therapy. To confirm previous studies showing that recombinant adenoviral vectors is targeted mainly to liver [17,23], wildtype mice were intravenously given 1 109 Ad-EGFPLuc infectious particles and the expression of luciferase was determined in various tissues. After 48 h more than 98% FIG. 3. Time course of hepatic PBGD activity in PBGDdeficient mice, after intravenous administration of 1 108 infectious particles of Ad-PBGD. The PBGD activity, expressed as pkat/g protein, was measured in liver homogenates from animals sacrificed 2, 7, 14, and 29 days after Ad-PBGD administration. The dashed line represents the basal hepatic PBGD activity determined at one of the time points in wild-type animals. Values are means F SD of three animals.
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
339
ARTICLE
doi:10.1016/j.ymthe.2004.05.018
FIG. 4. Plasma levels of (A) ALA and (B) PBG in PBGDdeficient mice after phenobarbital-induction with and without administration of recombinant adenovirus AdPBGD. The PBGD-deficient mice (n = 5) were treated for 4 days with increasing doses (75, 80, 85, and 90 mg/kg body weight) of phenobarbital. One group of animals (n = 5) was pretreated with an intravenous injection containing 5 107 infectious particles of AdPBGD 5 days prior to the first phenobarbital injection (Phenobarbital + Ad-PBGD). The arrows indicate instances of phenobarbital (PB) administration. The control PBGD-deficient mice (n = 3) did not receive any phenobarbital. Blood was taken 6 – 8 h after each phenobarbital dose. The concentrations of ALA and PBG in plasma are expressed as Amol/L.
of the total luciferase activity was found in the liver (Fig. 1). Previous studies in rodents indicate that after intravenous injection of recombinant adenovirus, over 90% of the hepatocytes are transduced in vivo as determined by h-galactosidase staining [24,25]. The basal tissue enzymatic activity of PBGD from the deficient animals was about 30% of the level found in wild-type mice (Fig. 2), which confirms previous studies [20,22]. When Ad-PBGD was injected intravenously in the PBGD-deficient mice, a high expression of functional PBGD enzyme was obtained in the liver in a dose-dependent manner 48 h after the injection (Fig. 2). Any increase in PBGD activity was not found in the other tissues, i.e., spleen, kidney, lung, heart, and brain, showing that the gene transfer accomplished in these tissues was not sufficient to increase the endogenous PBGD activity. The hepatic PBGD activity reached the highest level 7 days after administration of 1 108 Ad-PBGD infectious particles. Then it declined, but remained high (four times higher compared to wild type) at 29 days after adminis-
340
tration (Fig. 3). It has been reported that the transient gene expression limited to about 4 weeks is due to immune response against the transgene and the vector [26,27], as well as down-regulation of the CMV promoter [28]. The time for transgene expression may vary, as it is dependent on the nature of the transgene and the mouse strain used, with the C57BL/6 mouse showing the most persistent expression [29]. In this study the immune response to Ad-PBGD was not studied, and whether the high expression of the transgene coding for the endogenous mouse PBGD elicits an immune response remains to be elucidated. The transient increase in hepatic PBGD activity resulting from intravenous injection of Ad-PBGD was, however, long enough to permit a study of the effect in the mouse model on the levels of porphyrin precursors during phenobarbital induction of heme synthesis. Without a triggering factor such as phenobarbital, the PBGDdeficient mice do not display the biochemical features seen during AIP crises, i.e., accumulation of ALA and PBG. The injection of 5 107 infectious particles resulted
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
ARTICLE
doi:10.1016/j.ymthe.2004.05.018
TABLE 1: 5-Aminolevulinic acid (ALA) and porphobilinogen (PBG) contents in liver and kidney from PBGD-deficient mice, with or without phenobarbital administration for 4 days, and from PBGD-deficient mice injected with 5 107 infectious particles of recombinant adenovirus coding for PBGD (Ad-PBGD) 5 days prior to the first phenobarbital injection Sample
PBGD-deficient mice (n = 3)
PBGD-deficient mice+ phenobarbital (n = 5)
PBGD-deficient mice+ phenobarbital + Ad-PBGD (n = 5)
ALA concentration (pmol/mg tissue) Liver UD 13 F 6.8 Kidney UD 4.2 F 2.0
UD UD
PBG concentration (pmol/mg tissue) Liver UD 206 F 29 Kidney UD 43 F 8.2
UD UD
Levels of ALA and PBG in liver and kidney are expressed as means F SD (UD, under the detection limit, i.e., ALA <1.0 and PBG <3.5 pmol/mg tissue). The porphyrin precursors were measured 6 h after the fourth injection of phenobarbital.
in an increase in PBGD activity from 3.5- to 16-fold compared to the wild-type level from day 2 to 9, respectively, after injection of Ad-PBGD. The PBGD expression 6 – 9 days after injection of Ad-PBGD was demonstrated to be sufficient to suppress the effect of phenobarbital induction on the plasma levels of ALA and PBG seen in PBGD-deficient mice given only phenobarbital (Fig. 4). This shows that gene administration can effectively repair the affected catalytic step in the PBGD-deficient liver and prevent the accumulation of porphyrin precursors after induction of heme synthesis. In a recent study based on enzyme-replacement therapy, we showed that plasma levels of PBG, but not ALA, could be reduced after a single intravenous injection of recombinant PBGD (rhPBGD) [30]. To obtain an effect on ALA plasma levels prolonged rhPBGD administration is probably needed. In the present study, the intracellular expression of PBGD lasted for at least 4 weeks and resulted in an efficient downstream metabolism of both ALA and PBG. In the phenobarbital-induced PBGD-deficient mice, high levels of ALA and PBG were found in liver as well as in kidney 6 h after the fourth phenobarbital injection, confirming previous observations [30]. The levels found in kidney could reflect either a site of high porphyrin precursor synthesis or merely an accumulation during the filtration process. The Ad-PBGD mediated increase in PBGD activity evidently restores ALA and PBG metabolism both in liver and in kidney, as no accumulation occurred during phenobarbital induction (Table 1). Since no increase in PBGD activity was found in kidney after treatment with Ad-PBGD, the present data indicate that the high levels of porphyrin precursors found in kidney
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
probably are due to accumulation during the filtration process. In conclusion, recombinant adenoviral vector coding for PBGD is able to express the active enzyme in liver of an AIP mouse model in a dose- and time-dependent manner. The hepatic PBGD expression is shown to prevent accumulation of porphyrin precursors in phenobarbital-induced PBGD-deficient mice, pointing to a potential for effective gene therapy in AIP.
MATERIALS AND METHODS Construction of recombinant adenoviral DNA. Three recombinant E1and E3-deleted adenoviral vectors based on serotype 5 were constructed using the Adeno-X Expression System (Clontech, Palo Alto, CA, USA): one encoding a fusion protein of enhanced green fluorescent protein (EGFP) and luciferase (Ad-EGFPLuc), one encoding the mouse housekeeping isoform of PBGD (Ad-PBGD), and one encoding h-galactosidase (Ad-LacZ). All cDNAs were under the control of the human cytomegalovirus promoter. The coding sequence of the EGFP – luciferase fusion protein was amplified from pEGFPLuc (Clontech) by PCR using the upstream primer 5V-CGCGGGGCGGCCGCGCCACCATGGTGAGCAAG3V and downstream primer 5V-CCCGGGGGTACCTTACACGGCGATCTTTCC-3V. The PCR amplification resulted in an EGFPLuc cDNA fragment flanked by a NotI site and a consensus Kozak sequence [31] in the 5V end and a KpnI site in the 3V end. Following digestion of the EGFPLuc fragment with NotI and KpnI the cDNA was subcloned into the shuttle vector, resulting in pShuttle-EGFPLuc. The coding sequence of the mouse housekeeping isoform of PBGD was amplified from the plasmid p-mPBGDhou [13] by using 5V-CGCGGGGCGGCCGCGCCACC A T G T C C G G T A A C G G C G G C - 3 V a s f o r w a r d p r i m e r a n d 5 VCCCGGGGGTACCTTAGCGCACATCATTAAG-3V as reverse to generate a PBGD cDNA fragment flanked by the same sites as for EGFPLuc. The PBGD cDNA fragment was subcloned to generate pShuttle-PBGD. The expression cassettes in pShuttle-EGFPLuc, -PBGD, and -LacZ (Clontech) were excised by PI-SceI and I-CeuI and ligated into the Adeno-X-Viral DNA to generate pAd-EGFPLuc, pAd-PBGD, and pAd-LacZ. Adenoviral vector production and purification. The recombinant adenoviral plasmid DNA was linearized by digestion with PacI and transfected into the human embryonic kidney (HEK) 293 packaging cell line (Clontech) using polyethylenimine [32]. The HEK 293 cells were cultured in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum and 50 mg/L gentamicin (Gibco, Grand Island, NY, USA). Recombinant adenovirus was amplified in HEK 293 cells and purified by two centrifugation steps of cesium chloride gradients (CsCl) [33]. Viruses were purified from CsCl by desalting on NAP-10 columns (Amersham Biosciences, Uppsala, Sweden) and eluted with phosphate-buffered saline. Glycerol was added to a final concentration of 10% and the viruses were aliquoted and stored at 80jC. The number of viral particles was assessed by measurement of the optical density at 260 nm and the number of infectious particles was determined by end-point dilution assay on HEK 293 cells [34]. Vector titers were approximately 5 1012 viral particles/ml and 0.5 – 1 1011 infectious particles/ml. No evidence of replicationcompetent adenoviruses was found in observations of the cytopathic effect [35] on A-549 cells (kind gift from Professor Wadell at Umea˚ University, Umea˚, Sweden). Animals and injection procedures. Adult PBGD-deficient mice, generated as described by Lindberg et al. [22], and wild-type (C57BL/6) mice were
341
ARTICLE
used in the study. The local ethics committee approved all animal procedures and the mice were treated according to Swedish regulations and laws for laboratory animals. To overload the deficient enzymatic step, the PBGD-deficient mice were intraperitoneally administered with increasing doses (75, 80, 85, 90, mg/kg body weight) of phenobarbital (modified from [22]) for 4 consecutive days. Recombinant adenoviral vectors, between 1 105 and 1 109 of infectious particles per mouse, were injected intravenously in a volume of 200 Al. Specimen collection. Blood was collected from the tail vein in MiniCollect tubes with Li-heparin additive (Greiner Bio-One, Longwood, FL, USA) and centrifuged at 3000 g for 10 min. The plasma was stored at 80jC until analysis. Liver, spleen, kidney, lung, heart, and brain were removed from the animal under deep anesthesia by an overdose of isoflurane (Fluovac Unit; IMS, Cheshire, UK). The tissues were snap-frozen in liquid nitrogen and stored at 80jC until preparation. Tissue preparation. To measure luciferase activity, the tissues were immersed in 1 ml of 1 Cell Culture Lysis Reagent (Promega, Madison, WI, USA). Before homogenization, the liver from each animal was divided in four parts. Each sample was homogenized on ice, using a Potter – Elvehjelm glass homogenizer (inner diameter 8.0 mm, frosted walls) fitted with a Teflon pestle (diameter 7.8 mm), at a speed of 120 rpm. After centrifugation at 10,000 g for 10 min (4jC) the supernatants were pooled and stored at 80jC until analysis. To measure the PBGD activity, the tissues were immersed in 2 ml of a 50 mmol/L Tris – HCl buffer (pH 8.2) and homogenized using the same procedures as described above. For determination of ALA and PBG concentrations, a part of the frozen tissue sample (0.1 – 0.4 g) was homogenized in 0.5 ml 50 mmol/L Tris – HCl buffer (pH 8.2) containing 20 mmol/L citrate using the same procedure as described above. The sample was kept in the dark during the whole procedure. After centrifugation at 3000 g for 5 min at 4jC the supernatant was transferred to a new tube and centrifuged at 13,000 g for 10 min at 4jC. The supernatant was stored at 80jC until analysis of ALA and PBG concentrations. Determination of luciferase activity. The luciferase activity was measured using the Luciferase Assay System (Promega). Tissue homogenate (20 Al) was added to 100 Al of luciferase substrate and the relative light units (RLU) were measured during 10 s in a multiwell luminometer (1420 Victor2 Multilabel Counter; Wallac, Finland). The samples were analyzed in duplicate and the background signal (<400 RLU) determined in homogenates of untreated animals was subtracted. Purified recombinant luciferase (Promega) was used to produce a standard curve of RLU versus amount of enzyme. According to the luciferase standard used, 1 106 RLU correspond to approximately 1 ng luciferase. Protein content was determined using a Micro BCA Protein Assay Kit (Pierce, Rockford, IL, USA) and luciferase activity was expressed in terms of RLU/mg tissue protein. Determination of PBGD activity. The PBGD activity in tissue homogenates was determined according to the method of Magnussen et al. [36]. The protein content was determined using the Bio-Rad DC Protein Assay (Hercules, CA, USA) and the sample diluted to 0.5 g protein/L with 50 mmol/L Tris – HCl buffer (pH 8.2) before PBGD analysis, performed as previously described [13]. The PBGD activity was expressed in terms of pkat/g protein. One katal equals the conversion of 1 mol of substrate (PBG) per second and 1 unit/L corresponds to 16.67 nkat/L [37]. Determination of ALA and PBG concentrations. In tissue homogenates and plasma, ALA and PBG were quantified by using a LC-MS method
342
doi:10.1016/j.ymthe.2004.05.018
previously described by us [30]. The tissue concentrations of ALA and PBG were expressed as pmol/mg wet weight tissue and the plasma concentrations as Amol/L.
ACKNOWLEDGMENTS We thank Associate Professor Stig Thunell for reviewing the manuscript and Helena Reuterwall at HemeBiotech A/S for technical assistance with the LC-MS method. The work was supported by grants from Clas Groschinsky’s Memorial Fund and the Karolinska Institute, Stockholm. RECEIVED FOR PUBLICATION MARCH 25, 2004; ACCEPTED MAY 11, 2004.
REFERENCES 1. Elder, G. H., Hift, R. J., and Meissner, P. N. (1997). The acute porphyrias. Lancet 349: 1613 – 1617. 2. Moore, M. R., McColl, K. E., Rimington, C., and Goldberg, A. (1987). Disorders of Porphyrin Metabolism. Plenum: New York. 3. Floderus, Y., Shoolingin-Jordan, P., and Harper, P. (2002). Acute intermittent porphyria in Sweden: molecular, functional and clinical consequences of some new mutations found in the porphobilinogen deaminase gene. Clin. Genet. 62: 288 – 297. 4. Anderson, K. E., Sassa, S., Bishop, D. F., and Desnick, R. J. (2001). In The Metabolic and Molecular Bases of Inherited Disease. (C. R. Scriver, A. L. Beaudet, D. Valle, and W. S. Sly, Eds.), Vol. 2, pp. 2991 – 3062. McGraw – Hill, New York. 5. Oomman, A., and Gurtoo, A. (2002). Acute intermittent porphyria as a cause of acute respiratory failure. J. Indian Med. Assoc. 100: 44 – 46. 6. Granick, S. (1966). The induction in vitro of the synthesis of delta-aminolevulinic acid synthetase in chemical porphyria: a response to certain drugs, sex hormones, and foreign chemicals. J. Biol. Chem. 241: 1359 – 1375. 7. Miyagi, K., Cardinal, R., Bossenmaier, I., and Watson, C. J. (1971). The serum porphobilinogen and hepatic porphobilinogen deaminase in normal and porphyric individuals. J. Lab. Clin. Med. 78: 683 – 695. 8. Schmid, R., Schwartz, S., and Watson, C. J. (1954). Porphyrin content of bone marrow and liver in the various forms of porphyria. AMA Arch. Intern. Med. 93: 167 – 190. 9. Soonawalla, Z. F., et al. (2004). Liver transplantation as a cure for acute intermittent porphyria. Lancet 363: 705 – 706. 10. Kauppinen, R. (1998). Management of the acute porphyrias. Photodermatol. Photoimmunol. Photomed. 14: 48 – 51. 11. Tenhunen, R., and Mustajoki, P. (1998). Acute porphyria: treatment with heme. Semin. Liver Dis. 18: 53 – 55. 12. Badminton, M. N., and Elder, G. H. (2002). Management of acute and cutaneous porphyrias. Int. J. Clin. Pract. 56: 272 – 278. 13. Johansson, A., Mo¨ller, C., Gellerfors, P., and Harper, P. (2002). Non-viral mediated gene transfer of porphobilinogen deaminase into mammalian cells. Scand. J. Clin. Lab. Invest. 62: 105 – 114. 14. Johansson, A., Mo ¨ ller, C., and Harper, P. (2003). Correction of the biochemical defect in porphobilinogen deaminase deficient cells by non-viral gene delivery. Mol. Cell. Biochem. 250: 65 – 71. 15. Johansson, A., Nowak, G., Mo ¨ ller, C., and Harper, P. (2004). Non-viral delivery of the porphobilinogen deaminase cDNA into a mouse model of acute intermittent porphyria. Mol. Genet. Metab. 82: 20 – 26. 16. Lefesvre, P., Attema, J., Lemckert, A., Havenga, M., and Bekkum Dv., D. (2003). Genetic heterogeneity in response to adenovirus gene therapy. BMC Mol. Biol. 4: 4. 17. Wood, M., et al. (1999). Biodistribution of an adenoviral vector carrying the luciferase reporter gene following intravesical or intravenous administration to a mouse. Cancer Gene Ther. 6: 367 – 372. 18. Russell, W. C. (2000). Update on adenovirus and its vectors. J. Gen. Virol. 81: 2573 – 2604. 19. Liu, Q., and Muruve, D. A. (2003). Molecular basis of the inflammatory response to adenovirus vectors. Gene Ther. 10: 935 – 940. 20. Jover, R., Hoffmann, F., Scheffler-Koch, V., and Lindberg, R. L. (2000). Limited heme synthesis in porphobilinogen deaminase-deficient mice impairs transcriptional activation of specific cytochrome P450 genes by phenobarbital. Eur. J. Biochem. 267: 7128 – 7137. 21. Lindberg, R. L., et al. (1999). Motor neuropathy in porphobilinogen deaminase-deficient mice imitates the peripheral neuropathy of human acute porphyria. J. Clin. Invest. 103: 1127 – 1134. 22. Lindberg, R. L., et al. (1996). Porphobilinogen deaminase deficiency in mice causes a neuropathy resembling that of human hepatic porphyria. Nat. Genet. 12: 195 – 199. 23. Zinn, K. R., et al. (1998). Imaging and tissue biodistribution of 99mTc-labeled adenovirus knob (serotype 5). Gene Ther. 5: 798 – 808. 24. Li, Q., Kay, M. A., Finegold, M., Stratford-Perricaudet, L. D., and Woo, S. L. (1993). Assessment of recombinant adenoviral vectors for hepatic gene therapy. Hum. Gene Ther. 4: 403 – 409. 25. Herz, J., and Gerard, R. D. (1993). Adenovirus-mediated transfer of low density
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
doi:10.1016/j.ymthe.2004.05.018
26.
27.
28.
29.
30.
lipoprotein receptor gene acutely accelerates cholesterol clearance in normal mice. Proc. Natl. Acad. Sci. U S A 90: 2812 – 2816. Yang, Y., Ertl, H. C., and Wilson, J. M. (1994). MHC class I-restricted cytotoxic T lymphocytes to viral antigens destroy hepatocytes in mice infected with E1-deleted recombinant adenoviruses. Immunity 1: 433 – 442. Yang, Y., Jooss, K. U., Su, Q., Ertl, H. C., and Wilson, J. M. (1996). Immune responses to viral antigens versus transgene product in the elimination of recombinant adenovirusinfected hepatocytes in vivo. Gene Ther. 3: 137 – 144. Loser, P., Jennings, G. S., Strauss, M., and Sandig, V. (1998). Reactivation of the previously silenced cytomegalovirus major immediate-early promoter in the mouse liver: involvement of NFkappaB. J. Virol. 72: 180 – 190. Michou, A. I., et al. (1997). Adenovirus-mediated gene transfer: influence of transgene, mouse strain and type of immune response on persistence of transgene expression. Gene Ther. 4: 473 – 482. Johansson, A., Mo ¨ ller, C., Fogh, J., and Harper, P. (2004). Biochemical characterization of porphobilinogen deaminase deficient mice during phenobarbital induction of heme synthesis and the effect of enzyme replacement. Mol. Med 9: 193 – 199.
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
ARTICLE
31. Kozak, M. (1987). An analysis of 5V-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15: 8125 – 8148. 32. Boussif, O., et al. (1995). A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc. Natl. Acad. Sci. U S A 92: 7297 – 7301. 33. Kremer, E. J., Boutin, S., Chillon, M., and Danos, O. (2000). Canine adenovirus vectors: an alternative for adenovirus-mediated gene transfer. J. Virol. 74: 505 – 512. 34. Nyberg-Hoffman, C., Shabram, P., Li, W., Giroux, D., and Aguilar-Cordova, E. (1997). Sensitivity and reproducibility in adenoviral infectious titer determination. Nat. Med. 3: 808 – 811. 35. Zhu, J., et al. (1999). Characterization of replication-competent adenovirus isolates from large-scale production of a recombinant adenoviral vector. Hum. Gene Ther. 10: 113 – 121. 36. Magnussen, C. R., Levine, J. B., Doherty, J. M., Cheesman, J. O., and Tschudy, D. P. (1974). A red cell enzyme method for the diagnosis of acute intermittent porphyria. Blood 44: 857 – 868. 37. Lentner, C. (1984). In Geigy Scientific Tables, Vol. 3, pp. 166 – 167. Ciba – Geigy, Basel.
343
doi:10.1016/j.ymthe.2004.05.018
Adenoviral-Mediated Expression of Porphobilinogen Deaminase in Liver Restores the Metabolic Defect in a Mouse Model of Acute Intermittent Porphyria Annika Johansson,1,* Grzegorz Nowak,2 Christer Mo¨ller,3 Pontus Blomberg,4 and Pauline Harper1 1
2
Porphyria Centre Sweden, Division of Clinical Chemistry, Department of Laboratory Medicine, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden Division of Transplantation Surgery, Department of Laboratory Medicine, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden 3 HemeBiotech A/S, 181 70 Lidingo¨, Sweden 4 Vecura, Clinical Research Center, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden *To whom correspondence and reprint requests should be addressed. Fax: +46 8 585 827 60. E-mail: [email protected].
The aim of this study was to investigate the potential of gene therapy in the treatment of acute intermittent porphyria (AIP), a disorder caused by a partial deficiency of porphobilinogen deaminase (PBGD), the third enzyme in heme synthesis. The condition is biochemically characterized by accumulation of the porphyrin precursors 5-aminolevulinic acid (ALA) and porphobilinogen (PBG). Here we present the first experiments in vivo using adenoviral vectors to replace the deficient enzyme in the liver of an AIP mouse model. The use of adenoviral vector carrying the cDNA of luciferase in wild-type mice confirmed that transgene expression after intravenous administration was found mainly in liver. When PBGD-deficient mice were administered with adenoviral vector carrying the cDNA of mouse PBGD, the hepatic PBGD activity increased in a dose- and time-dependent manner. The highest activity was found 7 days after injection and remained high after 29 days. The expressed enzyme was shown to correct the metabolic defect in the PBGD-deficient mice as no accumulation of ALA or PBG occurred in plasma, liver, or kidney after induction of heme synthesis by phenobarbital. The study demonstrates that hepatic PBGD expression prevents the accumulation of porphyrin precursors, suggesting a future potential for gene therapy in AIP. Key Words: acute intermittent porphyria, 5-aminolevulinic acid, porphobilinogen, porphobilinogen deaminase, recombinant adenoviral vector
INTRODUCTION Acute intermittent porphyria (AIP) is the most common type of acute hepatic porphyria [1], affecting about 1 in 20,000 in the European population [2], with a prevalence of up to 1 in 10,000 in Sweden [3]. It is an autosomal dominant disorder caused by a deficiency of porphobilinogen deaminase (PBGD; EC 4.3.1.8), which is the third enzyme in the dedicated heme biosynthetic pathway. Clinical symptoms include acute episodes of abdominal pain often accompanied with peripheral neuropathy and mental disturbances [4]. Without treatment the disorder may be fatal [5]. The precipitating factors, such as certain drugs, alcohol, fasting, stress, and reproductive hormones, have the capacity to induce heme synthesis [6]. Under conditions of increased demand for heme the
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy 1525-0016/$30.00
deficient enzyme becomes rate limiting and as a consequence the heme precursors upstream of PBGD, 5-aminolevulinic acid (ALA) and porphobilinogen (PBG), accumulate and are excreted in urine [4]. Increased levels of ALA and PBG have been found mainly in plasma and liver from AIP patients with active disease [7,8], and the fact that recurrent acute attacks were prevented by liver transplantation in a young AIP patient [9] indicates that AIP is primarily a hepatic disorder. Current treatment protocols involve glucose and heme administration [10]. This approach has improved the prognosis of the disorder but may fail to reverse an established neuropathy [11]. In addition, adverse effects in the form of deteriorated peripheral venous access have been reported connected with repeated heme administration [12]. Thus, it would be desirable to find new
337
ARTICLE
therapeutic alternatives. Since AIP is caused by a defect in the PBGD gene, gene therapy is one obvious treatment strategy. Recently, we demonstrated in vitro that the use of nonviral vectors carrying the cDNA of PBGD gives rise to expression of active PBGD [13] that can correct the biochemical defect in PBGD-deficient cells [14]. Different strategies have been tested for introducing the PBGD cDNA into a mouse model of AIP using nonviral vectors, but due to the low efficiency of gene transfer no increase in PBGD expression has been found [15]. Another strategy for obtaining a high level of hepatic gene transfer involves the use of recombinant adenoviral vectors, which have been shown to target mainly the liver in rodents when administered systemically [16,17]. Adenoviral vectors have a number of advantages such as the ability to be produced at a high titer and to transduce postmitotic cells. The major drawbacks, however, are transient expression of the transgene and ability to trigger the host cellular immune response [18,19]. The objective of the present study was to evaluate the potential of gene therapy in AIP using recombinant adenoviral vectors in a mouse model for the disease, known to display biochemical features of AIP after administration of phenobarbital [20 – 22]. The AIP mouse model has been engineered with a residual hepatic PBGD activity of 30% of normal [22] and phenobarbital is a strong inducer of the first and rate-limiting step of heme synthesis, ALA synthase [20]. Here we present the first successful results using adenoviral-mediated gene transfer of PBGD cDNA to the PBGD-deficient mice, resulting in a high hepatic PBGD expression in vivo, and its abolishing effect on the accumulation of ALA and PBG during the phenobarbital induction of heme synthesis.
RESULTS Tissue Distribution of Luciferase Activity We determined the tissue distribution of transgene expression in wild-type (C57BL/6) mice following intrave-
doi:10.1016/j.ymthe.2004.05.018
nous administration of 1 109 infectious particles of recombinant adenoviral vector carrying the luciferase cDNA (Ad-EGFPLuc). Forty-eight hours after injection, more than 98% of the total luciferase expression was found in the liver tissue, with much lower levels in the other tissues (Fig. 1). Dose – Response Study on Hepatic PBGD Expression We determined the activity of PBGD in liver in PBGDdeficient mice 48 h after injection of increasing doses, from 1 105 to 5 108 infectious particles, of recombinant adenovirus coding for PBGD (Ad-PBGD). We observed endogenous hepatic PBGD activity in the PBGD-deficient mice to be about 30% of the level found in wild-type mice. Following injection of Ad-PBGD into the PBGD-deficient mice, we observed a dose-dependent increase in the activity of the enzyme. The highest PBGD activity was obtained using 5 108 infectious particles, which resulted in a 24-fold increase in the basal enzymatic activity found in wild-type mice (Fig. 2). No increase in PBGD activity was found in other tissues, i.e., spleen, kidney, lung, heart, and brain (data not shown). Time-Course Study on Hepatic PBGD Expression We determined the hepatic PBGD activity in PBGD-deficient mice 2, 7, 14, and 29 days following injection of 1 108 infectious particles of Ad-PBGD. We observed the highest activity after 7 days, corresponding to a 29-fold increase compared to the basal activity of wild-type mice (Fig. 3). The PBGD activity slowly decreased but remained 4-fold over the basal activity 29 days after the administration of Ad-PBGD. Effects of Hepatic PBGD Expression on Plasma and Tissue Levels of ALA and PBG We studied the metabolic effects of the increased hepatic PBGD activity in PBGD-deficient mice injected with 5 107 infectious particles of Ad-PBGD 5 days prior to the
FIG. 1. Luciferase activity in tissues from wild-type C57BL/6 mice 48 h after intravenous administration of 1 109 infectious particles of the recombinant adenoviral vector Ad-EGFPLuc. The luciferase activity, expressed as RLU/mg protein, was determined in liver, spleen, kidney, lung, heart, and brain tissue homogenates. Data are presented as means F SD of four animals.
338
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
ARTICLE
doi:10.1016/j.ymthe.2004.05.018
FIG. 2. Dose response of hepatic PBGD activity in PBGD-deficient mice injected with recombinant adenovirus (Ad-PBGD). The PBGD-deficient mice were intravenously given increasing amounts of AdPBGD (1 105 to 5 108 infectious particles). Hepatic PBGD activity, expressed as pkat/g protein, was determined 48 h after Ad-PBGD administration. The basal hepatic enzyme activity in the PBGDdeficient mice was about 30% of that found in wildtype (C57BL/6) mice. Values are means F SD of three animals.
first phenobarbital injection. We chose this titer on basis of the results from the dose – response assay, determined 48 h after Ad-PBGD administration, in which it resulted in a PBGD activity that was 3.5-fold the wild-type level. The hepatic PBGD activity in the Ad-PBGD-treated mice, determined at the end of the experiment 9 days after AdPBGD administration, was 2121 F 725 pkat/g protein (n = 5), compared to 34 F 6 pkat/g protein (n = 8) in the PBGD-deficient control animals and 135 F 18 pkat/g protein (n = 3) in wild-type animals. We did not observe any accumulation of plasma ALA and PBG during the phenobarbital induction period in the mice subjected to gene therapy (Fig. 4). In contrast, in the PBGD-deficient mice given only phenobarbital, the plasma levels of ALA and PBG gradually increased during the 4-day phenobarbital-induction period (Fig. 4). We observed an analogous response in phenobarbital-induced PBGD-deficient mice that were mock transduced with 5 107 Ad-LacZ infectious particles (data not shown). The plasma levels of ALA and PBG in the controls, i.e., in the PBGD-deficient mice not treated with Ad-PBGD and phenobarbital, were low at all time points (Fig. 4). Six hours after the fourth phenobarbital injection we observed a similar pattern in liver and kidney, i.e., undetectable levels of ALA and PBG in
the PBGD-deficient mice treated with phenobarbital and Ad-PBGD and high levels in the PBGD-deficient mice treated only with phenobarbital (Table 1).
DISCUSSION The prevailing hypothesis of the pathogenesis of AIP symptoms is based on the assumption of a neurotoxic effect of the accumulated porphyrin precursors ALA and PBG and/or a lack of heme [4]. Here we present the first positive results from experiments in vivo using phenobarbital-induced PBGD-deficient mice as a model for AIP, in which the accumulation of ALA and PBG is prevented by increased hepatic expression of PBGD by use of recombinant adenoviral vectors. Recombinant adenoviral vectors can transduce nondividing cells with high efficiency and rapidly accumulate in the liver after systemic administration [17,23], making them a useful tool for liver-directed gene therapy. To confirm previous studies showing that recombinant adenoviral vectors is targeted mainly to liver [17,23], wildtype mice were intravenously given 1 109 Ad-EGFPLuc infectious particles and the expression of luciferase was determined in various tissues. After 48 h more than 98% FIG. 3. Time course of hepatic PBGD activity in PBGDdeficient mice, after intravenous administration of 1 108 infectious particles of Ad-PBGD. The PBGD activity, expressed as pkat/g protein, was measured in liver homogenates from animals sacrificed 2, 7, 14, and 29 days after Ad-PBGD administration. The dashed line represents the basal hepatic PBGD activity determined at one of the time points in wild-type animals. Values are means F SD of three animals.
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
339
ARTICLE
doi:10.1016/j.ymthe.2004.05.018
FIG. 4. Plasma levels of (A) ALA and (B) PBG in PBGDdeficient mice after phenobarbital-induction with and without administration of recombinant adenovirus AdPBGD. The PBGD-deficient mice (n = 5) were treated for 4 days with increasing doses (75, 80, 85, and 90 mg/kg body weight) of phenobarbital. One group of animals (n = 5) was pretreated with an intravenous injection containing 5 107 infectious particles of AdPBGD 5 days prior to the first phenobarbital injection (Phenobarbital + Ad-PBGD). The arrows indicate instances of phenobarbital (PB) administration. The control PBGD-deficient mice (n = 3) did not receive any phenobarbital. Blood was taken 6 – 8 h after each phenobarbital dose. The concentrations of ALA and PBG in plasma are expressed as Amol/L.
of the total luciferase activity was found in the liver (Fig. 1). Previous studies in rodents indicate that after intravenous injection of recombinant adenovirus, over 90% of the hepatocytes are transduced in vivo as determined by h-galactosidase staining [24,25]. The basal tissue enzymatic activity of PBGD from the deficient animals was about 30% of the level found in wild-type mice (Fig. 2), which confirms previous studies [20,22]. When Ad-PBGD was injected intravenously in the PBGD-deficient mice, a high expression of functional PBGD enzyme was obtained in the liver in a dose-dependent manner 48 h after the injection (Fig. 2). Any increase in PBGD activity was not found in the other tissues, i.e., spleen, kidney, lung, heart, and brain, showing that the gene transfer accomplished in these tissues was not sufficient to increase the endogenous PBGD activity. The hepatic PBGD activity reached the highest level 7 days after administration of 1 108 Ad-PBGD infectious particles. Then it declined, but remained high (four times higher compared to wild type) at 29 days after adminis-
340
tration (Fig. 3). It has been reported that the transient gene expression limited to about 4 weeks is due to immune response against the transgene and the vector [26,27], as well as down-regulation of the CMV promoter [28]. The time for transgene expression may vary, as it is dependent on the nature of the transgene and the mouse strain used, with the C57BL/6 mouse showing the most persistent expression [29]. In this study the immune response to Ad-PBGD was not studied, and whether the high expression of the transgene coding for the endogenous mouse PBGD elicits an immune response remains to be elucidated. The transient increase in hepatic PBGD activity resulting from intravenous injection of Ad-PBGD was, however, long enough to permit a study of the effect in the mouse model on the levels of porphyrin precursors during phenobarbital induction of heme synthesis. Without a triggering factor such as phenobarbital, the PBGDdeficient mice do not display the biochemical features seen during AIP crises, i.e., accumulation of ALA and PBG. The injection of 5 107 infectious particles resulted
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
ARTICLE
doi:10.1016/j.ymthe.2004.05.018
TABLE 1: 5-Aminolevulinic acid (ALA) and porphobilinogen (PBG) contents in liver and kidney from PBGD-deficient mice, with or without phenobarbital administration for 4 days, and from PBGD-deficient mice injected with 5 107 infectious particles of recombinant adenovirus coding for PBGD (Ad-PBGD) 5 days prior to the first phenobarbital injection Sample
PBGD-deficient mice (n = 3)
PBGD-deficient mice+ phenobarbital (n = 5)
PBGD-deficient mice+ phenobarbital + Ad-PBGD (n = 5)
ALA concentration (pmol/mg tissue) Liver UD 13 F 6.8 Kidney UD 4.2 F 2.0
UD UD
PBG concentration (pmol/mg tissue) Liver UD 206 F 29 Kidney UD 43 F 8.2
UD UD
Levels of ALA and PBG in liver and kidney are expressed as means F SD (UD, under the detection limit, i.e., ALA <1.0 and PBG <3.5 pmol/mg tissue). The porphyrin precursors were measured 6 h after the fourth injection of phenobarbital.
in an increase in PBGD activity from 3.5- to 16-fold compared to the wild-type level from day 2 to 9, respectively, after injection of Ad-PBGD. The PBGD expression 6 – 9 days after injection of Ad-PBGD was demonstrated to be sufficient to suppress the effect of phenobarbital induction on the plasma levels of ALA and PBG seen in PBGD-deficient mice given only phenobarbital (Fig. 4). This shows that gene administration can effectively repair the affected catalytic step in the PBGD-deficient liver and prevent the accumulation of porphyrin precursors after induction of heme synthesis. In a recent study based on enzyme-replacement therapy, we showed that plasma levels of PBG, but not ALA, could be reduced after a single intravenous injection of recombinant PBGD (rhPBGD) [30]. To obtain an effect on ALA plasma levels prolonged rhPBGD administration is probably needed. In the present study, the intracellular expression of PBGD lasted for at least 4 weeks and resulted in an efficient downstream metabolism of both ALA and PBG. In the phenobarbital-induced PBGD-deficient mice, high levels of ALA and PBG were found in liver as well as in kidney 6 h after the fourth phenobarbital injection, confirming previous observations [30]. The levels found in kidney could reflect either a site of high porphyrin precursor synthesis or merely an accumulation during the filtration process. The Ad-PBGD mediated increase in PBGD activity evidently restores ALA and PBG metabolism both in liver and in kidney, as no accumulation occurred during phenobarbital induction (Table 1). Since no increase in PBGD activity was found in kidney after treatment with Ad-PBGD, the present data indicate that the high levels of porphyrin precursors found in kidney
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
probably are due to accumulation during the filtration process. In conclusion, recombinant adenoviral vector coding for PBGD is able to express the active enzyme in liver of an AIP mouse model in a dose- and time-dependent manner. The hepatic PBGD expression is shown to prevent accumulation of porphyrin precursors in phenobarbital-induced PBGD-deficient mice, pointing to a potential for effective gene therapy in AIP.
MATERIALS AND METHODS Construction of recombinant adenoviral DNA. Three recombinant E1and E3-deleted adenoviral vectors based on serotype 5 were constructed using the Adeno-X Expression System (Clontech, Palo Alto, CA, USA): one encoding a fusion protein of enhanced green fluorescent protein (EGFP) and luciferase (Ad-EGFPLuc), one encoding the mouse housekeeping isoform of PBGD (Ad-PBGD), and one encoding h-galactosidase (Ad-LacZ). All cDNAs were under the control of the human cytomegalovirus promoter. The coding sequence of the EGFP – luciferase fusion protein was amplified from pEGFPLuc (Clontech) by PCR using the upstream primer 5V-CGCGGGGCGGCCGCGCCACCATGGTGAGCAAG3V and downstream primer 5V-CCCGGGGGTACCTTACACGGCGATCTTTCC-3V. The PCR amplification resulted in an EGFPLuc cDNA fragment flanked by a NotI site and a consensus Kozak sequence [31] in the 5V end and a KpnI site in the 3V end. Following digestion of the EGFPLuc fragment with NotI and KpnI the cDNA was subcloned into the shuttle vector, resulting in pShuttle-EGFPLuc. The coding sequence of the mouse housekeeping isoform of PBGD was amplified from the plasmid p-mPBGDhou [13] by using 5V-CGCGGGGCGGCCGCGCCACC A T G T C C G G T A A C G G C G G C - 3 V a s f o r w a r d p r i m e r a n d 5 VCCCGGGGGTACCTTAGCGCACATCATTAAG-3V as reverse to generate a PBGD cDNA fragment flanked by the same sites as for EGFPLuc. The PBGD cDNA fragment was subcloned to generate pShuttle-PBGD. The expression cassettes in pShuttle-EGFPLuc, -PBGD, and -LacZ (Clontech) were excised by PI-SceI and I-CeuI and ligated into the Adeno-X-Viral DNA to generate pAd-EGFPLuc, pAd-PBGD, and pAd-LacZ. Adenoviral vector production and purification. The recombinant adenoviral plasmid DNA was linearized by digestion with PacI and transfected into the human embryonic kidney (HEK) 293 packaging cell line (Clontech) using polyethylenimine [32]. The HEK 293 cells were cultured in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum and 50 mg/L gentamicin (Gibco, Grand Island, NY, USA). Recombinant adenovirus was amplified in HEK 293 cells and purified by two centrifugation steps of cesium chloride gradients (CsCl) [33]. Viruses were purified from CsCl by desalting on NAP-10 columns (Amersham Biosciences, Uppsala, Sweden) and eluted with phosphate-buffered saline. Glycerol was added to a final concentration of 10% and the viruses were aliquoted and stored at 80jC. The number of viral particles was assessed by measurement of the optical density at 260 nm and the number of infectious particles was determined by end-point dilution assay on HEK 293 cells [34]. Vector titers were approximately 5 1012 viral particles/ml and 0.5 – 1 1011 infectious particles/ml. No evidence of replicationcompetent adenoviruses was found in observations of the cytopathic effect [35] on A-549 cells (kind gift from Professor Wadell at Umea˚ University, Umea˚, Sweden). Animals and injection procedures. Adult PBGD-deficient mice, generated as described by Lindberg et al. [22], and wild-type (C57BL/6) mice were
341
ARTICLE
used in the study. The local ethics committee approved all animal procedures and the mice were treated according to Swedish regulations and laws for laboratory animals. To overload the deficient enzymatic step, the PBGD-deficient mice were intraperitoneally administered with increasing doses (75, 80, 85, 90, mg/kg body weight) of phenobarbital (modified from [22]) for 4 consecutive days. Recombinant adenoviral vectors, between 1 105 and 1 109 of infectious particles per mouse, were injected intravenously in a volume of 200 Al. Specimen collection. Blood was collected from the tail vein in MiniCollect tubes with Li-heparin additive (Greiner Bio-One, Longwood, FL, USA) and centrifuged at 3000 g for 10 min. The plasma was stored at 80jC until analysis. Liver, spleen, kidney, lung, heart, and brain were removed from the animal under deep anesthesia by an overdose of isoflurane (Fluovac Unit; IMS, Cheshire, UK). The tissues were snap-frozen in liquid nitrogen and stored at 80jC until preparation. Tissue preparation. To measure luciferase activity, the tissues were immersed in 1 ml of 1 Cell Culture Lysis Reagent (Promega, Madison, WI, USA). Before homogenization, the liver from each animal was divided in four parts. Each sample was homogenized on ice, using a Potter – Elvehjelm glass homogenizer (inner diameter 8.0 mm, frosted walls) fitted with a Teflon pestle (diameter 7.8 mm), at a speed of 120 rpm. After centrifugation at 10,000 g for 10 min (4jC) the supernatants were pooled and stored at 80jC until analysis. To measure the PBGD activity, the tissues were immersed in 2 ml of a 50 mmol/L Tris – HCl buffer (pH 8.2) and homogenized using the same procedures as described above. For determination of ALA and PBG concentrations, a part of the frozen tissue sample (0.1 – 0.4 g) was homogenized in 0.5 ml 50 mmol/L Tris – HCl buffer (pH 8.2) containing 20 mmol/L citrate using the same procedure as described above. The sample was kept in the dark during the whole procedure. After centrifugation at 3000 g for 5 min at 4jC the supernatant was transferred to a new tube and centrifuged at 13,000 g for 10 min at 4jC. The supernatant was stored at 80jC until analysis of ALA and PBG concentrations. Determination of luciferase activity. The luciferase activity was measured using the Luciferase Assay System (Promega). Tissue homogenate (20 Al) was added to 100 Al of luciferase substrate and the relative light units (RLU) were measured during 10 s in a multiwell luminometer (1420 Victor2 Multilabel Counter; Wallac, Finland). The samples were analyzed in duplicate and the background signal (<400 RLU) determined in homogenates of untreated animals was subtracted. Purified recombinant luciferase (Promega) was used to produce a standard curve of RLU versus amount of enzyme. According to the luciferase standard used, 1 106 RLU correspond to approximately 1 ng luciferase. Protein content was determined using a Micro BCA Protein Assay Kit (Pierce, Rockford, IL, USA) and luciferase activity was expressed in terms of RLU/mg tissue protein. Determination of PBGD activity. The PBGD activity in tissue homogenates was determined according to the method of Magnussen et al. [36]. The protein content was determined using the Bio-Rad DC Protein Assay (Hercules, CA, USA) and the sample diluted to 0.5 g protein/L with 50 mmol/L Tris – HCl buffer (pH 8.2) before PBGD analysis, performed as previously described [13]. The PBGD activity was expressed in terms of pkat/g protein. One katal equals the conversion of 1 mol of substrate (PBG) per second and 1 unit/L corresponds to 16.67 nkat/L [37]. Determination of ALA and PBG concentrations. In tissue homogenates and plasma, ALA and PBG were quantified by using a LC-MS method
342
doi:10.1016/j.ymthe.2004.05.018
previously described by us [30]. The tissue concentrations of ALA and PBG were expressed as pmol/mg wet weight tissue and the plasma concentrations as Amol/L.
ACKNOWLEDGMENTS We thank Associate Professor Stig Thunell for reviewing the manuscript and Helena Reuterwall at HemeBiotech A/S for technical assistance with the LC-MS method. The work was supported by grants from Clas Groschinsky’s Memorial Fund and the Karolinska Institute, Stockholm. RECEIVED FOR PUBLICATION MARCH 25, 2004; ACCEPTED MAY 11, 2004.
REFERENCES 1. Elder, G. H., Hift, R. J., and Meissner, P. N. (1997). The acute porphyrias. Lancet 349: 1613 – 1617. 2. Moore, M. R., McColl, K. E., Rimington, C., and Goldberg, A. (1987). Disorders of Porphyrin Metabolism. Plenum: New York. 3. Floderus, Y., Shoolingin-Jordan, P., and Harper, P. (2002). Acute intermittent porphyria in Sweden: molecular, functional and clinical consequences of some new mutations found in the porphobilinogen deaminase gene. Clin. Genet. 62: 288 – 297. 4. Anderson, K. E., Sassa, S., Bishop, D. F., and Desnick, R. J. (2001). In The Metabolic and Molecular Bases of Inherited Disease. (C. R. Scriver, A. L. Beaudet, D. Valle, and W. S. Sly, Eds.), Vol. 2, pp. 2991 – 3062. McGraw – Hill, New York. 5. Oomman, A., and Gurtoo, A. (2002). Acute intermittent porphyria as a cause of acute respiratory failure. J. Indian Med. Assoc. 100: 44 – 46. 6. Granick, S. (1966). The induction in vitro of the synthesis of delta-aminolevulinic acid synthetase in chemical porphyria: a response to certain drugs, sex hormones, and foreign chemicals. J. Biol. Chem. 241: 1359 – 1375. 7. Miyagi, K., Cardinal, R., Bossenmaier, I., and Watson, C. J. (1971). The serum porphobilinogen and hepatic porphobilinogen deaminase in normal and porphyric individuals. J. Lab. Clin. Med. 78: 683 – 695. 8. Schmid, R., Schwartz, S., and Watson, C. J. (1954). Porphyrin content of bone marrow and liver in the various forms of porphyria. AMA Arch. Intern. Med. 93: 167 – 190. 9. Soonawalla, Z. F., et al. (2004). Liver transplantation as a cure for acute intermittent porphyria. Lancet 363: 705 – 706. 10. Kauppinen, R. (1998). Management of the acute porphyrias. Photodermatol. Photoimmunol. Photomed. 14: 48 – 51. 11. Tenhunen, R., and Mustajoki, P. (1998). Acute porphyria: treatment with heme. Semin. Liver Dis. 18: 53 – 55. 12. Badminton, M. N., and Elder, G. H. (2002). Management of acute and cutaneous porphyrias. Int. J. Clin. Pract. 56: 272 – 278. 13. Johansson, A., Mo¨ller, C., Gellerfors, P., and Harper, P. (2002). Non-viral mediated gene transfer of porphobilinogen deaminase into mammalian cells. Scand. J. Clin. Lab. Invest. 62: 105 – 114. 14. Johansson, A., Mo ¨ ller, C., and Harper, P. (2003). Correction of the biochemical defect in porphobilinogen deaminase deficient cells by non-viral gene delivery. Mol. Cell. Biochem. 250: 65 – 71. 15. Johansson, A., Nowak, G., Mo ¨ ller, C., and Harper, P. (2004). Non-viral delivery of the porphobilinogen deaminase cDNA into a mouse model of acute intermittent porphyria. Mol. Genet. Metab. 82: 20 – 26. 16. Lefesvre, P., Attema, J., Lemckert, A., Havenga, M., and Bekkum Dv., D. (2003). Genetic heterogeneity in response to adenovirus gene therapy. BMC Mol. Biol. 4: 4. 17. Wood, M., et al. (1999). Biodistribution of an adenoviral vector carrying the luciferase reporter gene following intravesical or intravenous administration to a mouse. Cancer Gene Ther. 6: 367 – 372. 18. Russell, W. C. (2000). Update on adenovirus and its vectors. J. Gen. Virol. 81: 2573 – 2604. 19. Liu, Q., and Muruve, D. A. (2003). Molecular basis of the inflammatory response to adenovirus vectors. Gene Ther. 10: 935 – 940. 20. Jover, R., Hoffmann, F., Scheffler-Koch, V., and Lindberg, R. L. (2000). Limited heme synthesis in porphobilinogen deaminase-deficient mice impairs transcriptional activation of specific cytochrome P450 genes by phenobarbital. Eur. J. Biochem. 267: 7128 – 7137. 21. Lindberg, R. L., et al. (1999). Motor neuropathy in porphobilinogen deaminase-deficient mice imitates the peripheral neuropathy of human acute porphyria. J. Clin. Invest. 103: 1127 – 1134. 22. Lindberg, R. L., et al. (1996). Porphobilinogen deaminase deficiency in mice causes a neuropathy resembling that of human hepatic porphyria. Nat. Genet. 12: 195 – 199. 23. Zinn, K. R., et al. (1998). Imaging and tissue biodistribution of 99mTc-labeled adenovirus knob (serotype 5). Gene Ther. 5: 798 – 808. 24. Li, Q., Kay, M. A., Finegold, M., Stratford-Perricaudet, L. D., and Woo, S. L. (1993). Assessment of recombinant adenoviral vectors for hepatic gene therapy. Hum. Gene Ther. 4: 403 – 409. 25. Herz, J., and Gerard, R. D. (1993). Adenovirus-mediated transfer of low density
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
doi:10.1016/j.ymthe.2004.05.018
26.
27.
28.
29.
30.
lipoprotein receptor gene acutely accelerates cholesterol clearance in normal mice. Proc. Natl. Acad. Sci. U S A 90: 2812 – 2816. Yang, Y., Ertl, H. C., and Wilson, J. M. (1994). MHC class I-restricted cytotoxic T lymphocytes to viral antigens destroy hepatocytes in mice infected with E1-deleted recombinant adenoviruses. Immunity 1: 433 – 442. Yang, Y., Jooss, K. U., Su, Q., Ertl, H. C., and Wilson, J. M. (1996). Immune responses to viral antigens versus transgene product in the elimination of recombinant adenovirusinfected hepatocytes in vivo. Gene Ther. 3: 137 – 144. Loser, P., Jennings, G. S., Strauss, M., and Sandig, V. (1998). Reactivation of the previously silenced cytomegalovirus major immediate-early promoter in the mouse liver: involvement of NFkappaB. J. Virol. 72: 180 – 190. Michou, A. I., et al. (1997). Adenovirus-mediated gene transfer: influence of transgene, mouse strain and type of immune response on persistence of transgene expression. Gene Ther. 4: 473 – 482. Johansson, A., Mo ¨ ller, C., Fogh, J., and Harper, P. (2004). Biochemical characterization of porphobilinogen deaminase deficient mice during phenobarbital induction of heme synthesis and the effect of enzyme replacement. Mol. Med 9: 193 – 199.
MOLECULAR THERAPY Vol. 10, No. 2, August 2004 Copyright B The American Society of Gene Therapy
ARTICLE
31. Kozak, M. (1987). An analysis of 5V-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15: 8125 – 8148. 32. Boussif, O., et al. (1995). A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc. Natl. Acad. Sci. U S A 92: 7297 – 7301. 33. Kremer, E. J., Boutin, S., Chillon, M., and Danos, O. (2000). Canine adenovirus vectors: an alternative for adenovirus-mediated gene transfer. J. Virol. 74: 505 – 512. 34. Nyberg-Hoffman, C., Shabram, P., Li, W., Giroux, D., and Aguilar-Cordova, E. (1997). Sensitivity and reproducibility in adenoviral infectious titer determination. Nat. Med. 3: 808 – 811. 35. Zhu, J., et al. (1999). Characterization of replication-competent adenovirus isolates from large-scale production of a recombinant adenoviral vector. Hum. Gene Ther. 10: 113 – 121. 36. Magnussen, C. R., Levine, J. B., Doherty, J. M., Cheesman, J. O., and Tschudy, D. P. (1974). A red cell enzyme method for the diagnosis of acute intermittent porphyria. Blood 44: 857 – 868. 37. Lentner, C. (1984). In Geigy Scientific Tables, Vol. 3, pp. 166 – 167. Ciba – Geigy, Basel.
343