|
|
||||||||
CUTTING EDGE |



Department of Internal Medicine,
* Division of Rheumatology and
Division of Hematology, Washington University School of Medicine, St. Louis, MO 63110
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Two principle strategies can be envisioned to modulate the undesirable effects of this effector system. One approach is to block a component that is required in the activation process. For example, patients with PNH have been successfully treated by inhibiting C5 and thereby lysis of their regulator-deficient cells (5). Another promising concept has been described in mouse models of autoimmune disease and of ischemia/reperfusion injury, which can be favorably influenced by providing additional regulation (6). In both cases, the systemic C blockade has a downside due to reduced ability of being efficiently activated in the setting of an infection. Another strategy is to provide the host with the regulatory elements at the site(s) where damage is occurring. This approach centers on the generation of chimeras between a C regulator and either Abs or Ab fragments that bind a specific target where more inhibition is needed (7, 8) or between a regulator and protein domains that recognize an inflammatory milieu (9).
To provide patients with long-lasting, protective levels of targeted C regulators, retrovirus-mediated gene therapy is an attractive treatment option. Because of the limited size of retroviral vector genomes (10), we have constructed C regulator-scFv fusion proteins. These are comprised of single polypeptides and fit into commonly used retroviral vectors. The approach that we have developed involves the direct attachment of C regulators such as the decay-accelerating factor (DAF) (11) and C receptor 1-related gene/protein y (Crry) (12) to the RBC membrane to modulate C activation at this specific site.
This proof-of-concept study was designed to determine, in a mouse model of Crry deficiency, whether in vivo retroviral gene transfer with a vector encoding a membrane-targeted regulator could reconstitute the deficient RBCs with the missing protein.
| Materials and Methods |
|---|
|
|
|---|
The cDNAs of the RBC-targeted Crry forms (12) were excised with EcoRI/XbaI (Crry4-Ter, 1664 bp) and EcoRI/Asp718 (Crry5-Ter, 1930 bp). This was followed by blunt end modification of the restriction site overhangs. hAAT-WPRE-767 (13) was linearized with NotI (6001 bp), blunt ended, and ligated with the above-described fragments to generate RV-4 and RV-5, respectively.
Animals
C57BL/6 wild-type (WT) mice (2- to 4-mo old) were used as recipients for RBC transfusions and as a serum source. Mice deficient in both Crry and C3 (referred to herein as Crry knockout (KO) mice) have been previously described (14) and are on a mixed C57BL/6 (B6) and sv129 background (B6.129-Crrytm1HmoC3tm1Hrc). They were injected via the temporal vein 2 to 3 days after birth. RBC coating levels were determined by FACS and normalized to that of native Crry on RBCs of WT control mice. Blood samples were collected as described (12). Experiments were performed in accordance with the institutional regulations of Animal Welfare.
Cells, transfections, virus production, and assessment
The 293GPG packaging cell line was provided by Daniel Ory (Washington University School of Medicine). These cells were transiently transfected with the expression plasmids using LipofectAMINE (Invitrogen Life Technologies). The virus-containing supernatants were concentrated by ultracentifugation (15). The titer was determined by real-time PCR of genomic DNA (Qiagen) of infected 3521 mouse cells using WPRE-specific primers (13). Also, construct-specific proviral integration events were detected by PCR (from infected 3521 cells (Fig. 1C) or mouse organs (Fig. 4A)). Primer sequences (Integrated DNA Technologies) are available upon request.
|
|
To assess the clearance of mouse RBCs in vivo, cells from WT, nontreated Crry KO, and the experimental animals were labeled ex vivo with the membrane dye PKH-26 (Sigma-Aldrich) (12). They were introduced into WT mice via the tail vein in a 300 µl cell suspension. For FACS analysis, blood samples were collected by tail clipping. Donor RBC survival was calculated as reported (16).
Flow cytometry
RBCs were washed and then incubated with the rat anti-Crry mAb 1F2 followed by incubation with FITC-conjugated secondary goat anti-rat IgG pAb (BD Pharmingen). Cells were next washed and analyzed by flow cytometry (FACScan, BD Biosciences) as described (12). C activation was monitored by direct immunostaining with FITC-conjugated rabbit anti-mouse C3 polyclonal Ab (F(ab')2, ICN Pharmaceuticals). To assess for free fusion protein in plasma, naive Crry KO RBCs (2 x 107 cells/ml) were incubated with 110% sera from mice R3, L2, and L4, washed and then submitted for FACS analysis (12).
In vitro C deposition assays
Ab-induced C3 deposition on mouse RBCs was performed as previously described (12). Briefly, RBCs from WT, Crry KO and the experimental mice were sensitized with an anti-mouse CD24 mAb J11d (rat IgM; Research Diagnostics) before exposure to 5% WT mouse serum.
Immunofluorescence analysis of liver sections
Following fixation of 5-µm liver cryosections with acetone, the sections were blocked with 5% normal goat serum. The sections were then double-stained for Crry (rat anti-Crry mAb followed by FITC-conjugated secondary Ab) and for CD16/CD32 (FcRII/III) with PE-conjugated rat mAb 2.4G2 (BD Pharmingen) (17).
| Results and Discussion |
|---|
|
|
|---|
3 x 107 IU/ml for both virus preparations. The 2- to 3-day-old Crry KO mice were injected via the temporal vein with 100 µl of the retrovirus preparations (3 x 106 IU/animal). Two mice received RV-4, two mice RV-5, and three mice a mixture of both viruses (to mimic a treatment with two different C regulators). The initial Crry transgene expression was detected by immunostaining of freshly isolated RBCs as early as 8 days postinfection and ranged from 4 to 60% relative to the native protein on WT RBCs with a homogeneous coating pattern (Fig. 2A). The levels increased for all mice during the initial 6 to 8 wk after which an equilibrium was established. Subsequently, the levels remained stable for >12 mo (Fig. 2B). The average coating level for each mouse ranged from 34 to 465% relative to the copy number of native Crry on wild-type cells, i.e., all but mouse L5 achieved levels on their RBCs greater than the endogenous level. Western blot analysis (12) confirmed the expected molecular weights of the host-derived C regulators (not shown). Of note, development of Abs against the expressed fusion protein using a FACS-based assay (12) were not detected (not shown). This likely reflects their immature immune system at the time of vector injection and has been reported for mice (18) and dogs (19). Furthermore, the inhibitors were not detectable in serum (data not shown). Thus, the downside of a globally impaired C system is avoided.
|
To demonstrate C regulation, experiments were performed (12) using the RBCs coated in vivo with the host-derived inhibitors. In a FACS-based C activation assay, deposition of C3 fragments on IgM-sensitized RBCs was highest in the complete absence of Crry (Fig. 3A, KO). However, for those mice in which 1.5-fold or higher levels of the regulator were attached, C deposition was reduced to or below that of wild-type RBCs. As little as 25% of the normal Crry copy number (L5) reduced the amount of deposited C3 by >50%. These results are in agreement with those obtained from earlier studies in which the Crry-scFv fusion proteins were generated by standard transfection procedures (12).
|
2 days (Fig. 3B, R2). These results are consistent with what has been shown for RBCs coated ex vivo with the same proteins. The reduced survival rate of Crry-coated KO RBCs relative to WT cells reflects the progressive transfer of regulators from the treated RBCs to the hosts RBC pool (not shown, but see Ref. 12). We next examined the preference of the virus for selected target organs. Genomic DNA from 4 organs of mice R3, L2, and L4 was analyzed by PCR using a primer combination that would specifically amplify the integrated Crry/scFv cDNA (Fig. 1, diamond-flanked primers). As a loading control, the same DNA was amplified with primers (Fig. 1, circle-flanked lines) specific for CCPs 1 and 2 (Ref. 21) and Fig. 4A, lower panel). No amplification product was observed from DNA of a nontreated Crry KO mouse (Fig. 4A, lanes 1, 5, 9, and 13). However, signals of 633 (RV-4) and 823 bp (RV-5) were generated primarily from the livers of the treated mice R3, L2, and L4. It also verifies the existence of two types of integrated proviral DNA, because mouse R3 was infected with a mixture of RV-4 and RV-5. The other two mice received either RV-4 (L2) or RV-5 (L4). The faint PCR signals from lung, kidney, and spleen of the treated mice indicates that these organs are not the primary target of the injected vectors. With such low PCR signals and the retroviral vector containing a liver-specific promoter, we do not expect a significant contribution of these organs to protein production.
To further demonstrate protein expression of targeted Crry from the primary target organ, immunofluorescence analysis of liver sections was performed. Liver sections from WT mice revealed a positive staining pattern, characteristic of a ubiquitously expressed plasma membrane protein (Fig. 4B, WT). As expected, mice lacking Crry were negative for the C regulator (Fig. 4B, KO). In the case of retrovirally transduced liver cells, a more scattered and/or clustered signal pattern was anticipated. Indeed, most of the cells were negative for Crry and the few positive signals exhibited a punctuate pattern (Fig. 4B, R3). This would be anticipated for vesicle localization of a secreted protein. In light of the hepatocyte-specific promoter used (hAAT) (22), this liver cell type would be the major source of the secreted fusion proteins. To verify this assumption, liver sections were also stained with an anti-CD16/CD32 mAb that recognizes Fc
receptors II and III (FcRII/III). They are primarily expressed by sinusoidal endothelial cells (SECs) and liver macrophages but not by hepatocytes and vascular endothelial cells (17). Using anti-Crry-FITC and anti-FcRII/III-PE Abs, we identified the hepatocyte as the major cell type of secreted RV-4/-5 (Fig. 4B, R3, inset).
To examine for adverse effects that might have occurred secondary to the infection process of neonatal liver cells, e.g., tumor formation via insertional mutagenesis (23), histological examination (H&E staining) did not reveal liver pathology (not shown). This may be explained by the relatively low amount of virus used, being 16-fold lower compared with a previous study in which a similar type of virus was neonatally injected (24).
In summary, these data establish that targeting of a C regulator to a specific site via fusion with a single chain Ab fragment results in production of host-derived proteins sufficient to protect from a C-mediated insult. Thus, liver-directed synthesis of scFv-targeted C regulators represents a treatment strategy for a long-term protein supply to block undesirable C activation at specific sites. It has the potential to deliver a broad range of other therapeutic effector molecules to their site of action.
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by National Institutes of Health Grants 5RO1AI37618 and RO1AI41592 (to J.P.A.). ![]()
2 Address correspondence and reprint requests to Dr. John P. Atkinson, Department of Medicine, Division of Rheumatology, Washington University School of Medicine, 660 South Euclid Avenue, Campus Box 8045, St. Louis, MO 63110; E-mail address: jatkinso{at}im.wustl.edu or Dr. Dirk Spitzer, Department of Medicine, Division of Rheumatology, Washington University School of Medicine, 660 South Euclid Avenue, Campus Box 8045, St. Louis, MO 63110; E-mail address: dspitzer{at}im.wustl.edu ![]()
3 Abbreviations used in this paper: aHUS, atypical hemolytic uremic syndrome; AMD, age-related macular degeneration; CCP, C control protein repeat; Crry, complement receptor 1-related gene/protein y; hAAT, human
1 anti-trypsin; PNH, paroxysmal nocturnal hemoglobinuria; scFv, single chain Ab fragment; SEC, sinusoidal endothelial cell; WPRE, woodchuck hepatitis virus posttranscriptional regulatory element; WT, wild type. ![]()
Received for publication June 16, 2006. Accepted for publication August 9, 2006.
| References |
|---|
|
|
|---|
-glucuronidase in mucopolysaccharidosis VII dogs. Mol. Ther. 5: 141-153. [Medline]
-glucuronidase expression in situ and correction of liver pathology in mucopolysaccharidosis type VII mice. Hum. Gene Ther. 10: 85-94. [Medline]This article has been cited by other articles:
![]() |
P. Sabbatini, J. L. Rowand, A. Groy, S. Korenchuk, Q. Liu, C. Atkins, M. Dumble, J. Yang, K. Anderson, B. J. Wilson, et al. Antitumor Activity of GSK1904529A, a Small-molecule Inhibitor of the Insulin-like Growth Factor-I Receptor Tyrosine Kinase Clin. Cancer Res., May 1, 2009; 15(9): 3058 - 3067. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |