Adeno-Associated Virus D-Sequence-Mediated Suppression of Expression of a Human Major Histocompatibility Class II Gene: Implications in the Development of Adeno-Associated Virus Vectors for Modulating Humoral Immune Response

Abstract

A 20-nt long sequence, termed the D-sequence, in the adeno-associated virus (AAV) inverted terminal repeat was observed to share a partial sequence homology with the X-box in the regulatory region of the human leukocyte antigen DRA (HLA-DRA) promoter of the human major histocompatibility complex class II (MHC-II) genes. The D-sequence was also shown to specifically interact with the regulatory factor binding to the X-box (RFX), binding of which to the X-box is a critical step in the MHC-II gene expression, suggesting that D-sequence might compete for RFX transcription factor binding, thereby suppressing expression from the MHC-II promoter. In DNA-mediated transfection experiments, using a reporter gene under the control of the HLA-DRA promoter, D-sequence oligonucleotides were found to inhibit expression of the reporter gene expression in HeLa and 293 cells by ∼93% and 96%, respectively. No inhibition was observed when nonspecific synthetic oligonucleotides were used. D-sequence oligonucleotides had no effect on expression from the cytomegalovirus immediate-early gene promoter. Interferon-γ-mediated activation of MHC-II gene expression was also inhibited by D-sequence oligonucleotides as well as after infection with either the wild-type AAV or transduction with recombinant AAV vectors. These studies suggest that the D-sequence-mediated downregulation of the MHC-II gene expression may be exploited toward the development of novel AAV vectors capable of dampening the host humoral response, which has important implication in the optimal use of these vectors in human gene therapy.

Introduction

Recombinant adeno-associated virus (AAV) vectors have gained significant attention in the past decade as they have shown promise in gene therapy for a number of human diseases, such as Leber's congenital amaurosis (LCA),^1

–4 lipoprotein lipase deficiency,⁵ hemophilia B,^6

–11 aromatic L-amino acid decarboxylase deficiency,¹² choroideremia,¹³ Leber's hereditary optic neuropathy,¹⁴ hemophilia A,⁹ and spinal muscular atrophy (SMA).¹⁵ The first AAV “drug” (Glybera) was approved in 2012 for use in humans in Europe. In 2017, the FDA approved Luxturna as the first AAV drug in the United States for gene therapy of LCA, and in 2019, the FDA approved Zolgensma as the second AAV drug in the United States for gene therapy of SMA.

Despite these remarkable achievements, a significant obstacle remains in that the presence of pre-existing antibodies to AAV capsids precludes a large population of patients from benefitting from AAV vector-mediated gene therapy since even very low levels of pre-existing antibodies are capable of neutralizing the first generation of AAV vectors.¹⁶ Similarly, at the present time, repeat dosing with therapeutic AAV vectors is also not possible. Thus, there is a need to develop novel AAV vectors that are capable of either evading the host humoral immune response, or at the very least, dampening it.

Humoral immune response is mediated by antibody production after exposure to antigenic determinants in lymphatic organs, such as B lymphocytes, which are activated and differentiated to form plasma cells, which, in turn, synthesize and secrete antibodies specific for an antigen.¹⁷ Major histocompatibility complex class II (MHC-II) molecules are a class of MHC molecules that are present on professional antigen-presenting cells (APCs), such as dendritic cells, macrophages, and B lymphocytes.¹⁸ MHC-II gene expression is regulated by a master regulator, termed, class II major histocompatibility complex trans-activator (CIITA), expression of which is largely restricted to professional APCs.¹⁹

The AAV genome contains inverted terminal repeats (ITRs) that are 145 nt in length, the terminal 125 nt of which are palindromic and form T-shaped hairpin (HP) structures that serve as primers for AAV DNA replication.^20,21 The ITRs also contain an additional domain, designated the D-sequence, a stretch of 20 nt that are not involved in HP formation.²² We have previously documented that the D-sequence plays a crucial role in the life cycle of AAV in that (1) the D-sequence is the “packaging signal” for AAV,^23,24 (2) the first 10 nt in the D-sequence are indispensable for AAV DNA replication,²⁵ and (3) a distinct host cell protein interacts specifically with the double-stranded D-sequence, which we have designated as the double-stranded D-sequence-binding protein (dsD-BP).²⁵

Since dsD-BP forms a specific complex with the AAV D-sequence, we reasoned that DNA sequences with homology to the D-sequence must also exist in the human genome with which this protein might interact. A computer-based homology search revealed that a D-sequence-like sequence exists in the X-box in the regulatory region of the human leukocyte antigen DRA (HLA-DRA) promoter of the human MHC-II genes.

We report here that dsD-BP not only interacts with AAV D-sequence but also with X-box in the HLA-DRA promoter to which the regulatory factor binding to the X-box (RFX) transcription factor is known to bind,²⁶ suggesting that dsD-BP might be a putative RFX transcription factor. AAV D-sequence also strongly and specifically inhibits expression from the HLA-DRA promoter. Interferon-γ (IFN-γ)-mediated activation of MHC-II gene expression is also inhibited by D-sequence oligonucleotides and either the wild-type AAV or recombinant AAV vectors. These studies suggest that the D-sequence-mediated downregulation of the MHC-II gene expression could be exploited to generate novel AAV vectors to dampen the host humoral response, which has important implication in the optimal use of these vectors in human gene therapy.

Materials and Methods

Cell lines and cultures

Human embryonic kidney 293 (HEK293), and human cervical carcinoma, HeLa, cells were purchased from American Type Culture Collection (Manassas, VA). Cells were maintained in complete Dulbecco's modified Eagle's medium (DMEM; Mediatech, Manassas, VA) supplemented with 10% heat-inactivated fetal bovine serum (Sigma-Aldrich, St. Louis, MO) and 1% penicillin and streptomycin (Lonza, Walkersville, MD). Cells were grown as adherent cultures in a humidified atmosphere at 37°C in 5% CO₂, subcultured after treatment with trypsin-Versene mixture (Lonza) for 2–5 min at room temperature, washed, and resuspended in complete DMEM.

Preparation of whole-cell extracts

Whole-cell extracts (WCEs) were prepared as described previously.²⁷ Total protein concentration was determined with the Bio-Rad protein assay kit (Hercules, CA).

Electrophoretic mobility-shift assays

Electrophoretic mobility-shift assays (EMSAs) were performed essentially as described previously.²⁸ In brief, ³²P-labeled double-stranded oligonucleotides containing the AAV2 D-sequence (D), or the X-box sequence, were used as probes and incubated individually with WCE prepared from 293 cells for 20 min at 25°C and subjected to EMSA, in which the bound complexes were separated from the unbound probes on 6% polyacrylamide gels with 0.5′ TBE buffer (pH 8.0) containing 89 mM Tris, 89 mM boric acid, 1 mM ethylene diamine tetra-acetic acid (EDTA). Competition experiments using various unlabeled oligonucleotides were also performed. After electrophoresis, gels were dried in vacuo and autoradiographed with Kodak X-OMAT film.

Molecular cloning of human D-sequence-like sequences in the human genome

Human cellular D-sequence-like sequences were obtained by two consecutive rounds of PCR amplification using five different commercially available human genomic libraries (Clontech; now Takara Bio, Mountain View, CA) with adapter primers (AP-1, 5′-GTAATACGACTCACTATAGGGC-3′; AP-2, 5′-ACTATAGGGCACGCGTGGT-3′) and the AAV D(-) sequence primer (5′-AGGAACCCCTAGTGATGGAG-3′). The PCR amplification products were subjected to nucleotide sequencing. Four sets of sequences were obtained and were found to be homologous to the INK4A/ARF promoter, sorting nexin 1 (SNX1), lymphotoxin-α (LT-α) regulatory region, and X-box in the human leukocyte antigen DRA (HLA-DRA) promoter. These sequences were molecularly cloned in bacterial plasmids.

Recombinant plasmid vectors

A recombinant plasmid containing the HLA-DRA promoter-driven firefly luciferase (HLA-DRAp-FLuc) reporter gene was constructed as follows. In brief, a 467-bp DNA fragment (from −359 to +108) containing the HLA-DRA promoter was generated by PCR amplification using total human genomic DNA and the following primer pair: 5′-ACGCAAACTCTCCAACTGTCATTGC-3′ and 5′-TAGCACAGGGACTCCACTTATGGCC-3′.

The promoter sequence was cloned into a TA vector (Clontech). After digestion with NdeI and SacI, a 378-bp DNA fragment (from −329 to +49) was isolated and ligated upstream of the firefly luciferase (FLuc) reporter gene in the pGL3 cloning vector (Boehringer Mannheim, Indianapolis, IN). A recombinant plasmid expressing the LacZ reporter gene under the control of the cytomegalovirus promoter (CMV-LacZ) has been described previously.²⁹

DNA-mediated transfections and quantitation of FLuc expression in vitro

Two hundred ninety-three or HeLa cells (1 × 10⁵ cells/well) in 12-well plates were transfected with pGL3 vector or HLA-DRAp-FLuc reporter plasmids, with or without synthetic oligonucleotides (10–200 ng) or AAV DNAs (4 ∝ g) using the lipofection method as described previously.³⁰ Transfected cells were washed with phosphate-buffered saline (PBS), and cell lysates were prepared using 1 × passive lysis buffer. FLuc activity was determined 48 h post-transfections using an injector-equipped luminometer (BMG Labtech; FLUOstar Optima, Cary, NC), as recommended by the manufacturer.

Recombinant AAV vectors

Self-complementary AAV2 vectors expressing the enhanced green fluorescence protein (scAAV2-EGFP) or the mCherry (scAAV2-mCherry) reporter genes, and those containing the human X-box sequences (scAAV2-X-box-EGFP) were generated using the polyethyleneimine-mediated triple-plasmid transfection protocol.³¹ In brief, 293 cells were cotransfected with three plasmids using polyethyleneimine (linear, MW 25000; Polysciences, Inc., Warrington, PA), and medium was replaced 6 h post-transfections. Cells were harvested 72 h post-transfection, lysed by three rounds of freeze-thaw, and digested with Benzonase (EMD Millipore, Darmstadt, Germany) at 37°C for 1 h. AAV vectors were purified by iodixanol (Sigma, St. Louis, MO) gradient ultracentrifugation followed by ion exchange column chromatography (HiTrap Q HP; 5 mL, GE Healthcare, Piscataway, NJ), washed with PBS, and concentrated by centrifugation using centrifugal spin concentrators (Apollo; 150 kDa cutoff, 20 mL capacity, Orbital Biosciences, Topsfield, MA).

The physical genomic titers of highly purified scAAV vector stocks were determined by quantitative DNA slot blot analysis as described previously,³² with modification. In brief, 10 μL of vector stock was digested with Benzonase (EMD Millipore) at 37°C for 1 h. An equal volume of 100 mM NaOH was added, followed by incubation at 65°C for 30 min. A known quantity of plasmid DNA was denatured in the same manner for use as a reference standard for quantitation. Denatured DNA samples were loaded in twofold serial dilutions onto Immobilon-NY⁺ membranes (Millipore, Bedford, MA). After UV cross-linking, the membranes were prehybridized for 1 h at 42°C in a hybridization solution containing 6 × SSC, 100 g/mL denatured herring sperm DNA, 0.5% sodium dodecyl sulfate (SDS), and 5 × Denhardt's reagent. Subsequently, the membranes were hybridized with ³²P-labled DNA probes specific for EGFP or mCherry sequences in a hybridization solution at 42°C for 18–20 h. Membranes were washed twice with wash solution I (2 × SSC, 0.1% SDS) at room temperature for 15 min, twice with wash solution II (0.5 × SSC, 0.1% SDS) at 42°C for 15 min, and then exposed to Amersham Typhoon RGB Biomolecular Imager (GE Healthcare, Chicago, IL) at room temperature. Autoradiography was performed as described above. Some vectors were also custom packaged by SAB Tech, Inc. (Philadelphia, PA) and PackGene (Worcester, MA). The vector titers and potencies were not significantly different from those produced by in-house methods.

Intravascular and intramuscular injections in mice

Six- to 10-week-old C57BL6/J mice were purchased from Jackson Laboratory (Bar Harbor, ME) and maintained by the Laboratory Animal Resource Center (LARC) at Indiana University School of Medicine, Indianapolis, IN. Mice (n = 4 per group) were injected either intravenously (IV) or intramuscularly (IM) with 1 × 10¹⁰ vg of scAAV2-EGFP or scAAV2-X-box-EGFP vectors.

Animals were kept in sterile cages until the end of the experiment. Mice were bled 3 weeks after vector administration, and the level of anti-AAV2 IgG2c capsid antibodies was measured by enzyme-linked immunosorbent assay (ELISA). All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) and performed according to the guidelines for animal care specified by the LARC at Indiana University School of Medicine, Indianapolis, IN.

Ocular injections

Intravitreal AAV vector injections were administered as previously described.³³ C57BL6/J mice were intravitreally injected in their right eyes (REs) with 3 × 10¹¹ vg/mL of scAAV2-EGFP (n = 8) or with scAAV2-X-box-EGFP (n = 8) vectors (1 μL delivered). Extreme care was taken to ensure that the retina was not perforated during the procedure so as to avoid leakage of vector into the subretinal space. This was confirmed visually for each injection using a Nikon SMZ800 dissecting microscope fitted with an Olympus C-4040 camera and F1.8 super bright zoom lens. If a perforation was suspected, that eye was excluded from the study. Next, each cohort of eight mice was subdivided into two groups (n = 4 each). In the first four mice, scAAV2-EGFP was injected intravitreally into the contralateral left eye. In the second four mice, scAAV2-mCherry was injected into the same eye (RE).

Fundoscopy

At 4 weeks postinjection, fundus images were captured using a Micron III camera (Phoenix Research Laboratories, Pleasanton, CA). Green and red fluorescent images were taken to visualize green fluorescent protein (GFP) and mCherry expression, respectively. Exposure settings remained constant between eyes. Four weeks later, fundus images were collected again in the same manner.

Statistical analysis

All results are presented as mean ± standard deviation. Statistical analyses were performed using GraphPad Prism 7 software (GraphPad Software, Inc., San Diego, CA). Differences between groups were identified by grouped-unpaired two-tailed distribution of Student's t-test. A value of p < 0.05 was considered statistically significant.

Results

AAV D-sequence shares partial homology with the X-box in the HLA-DRA promoter

Since the cellular dsD-BP protein forms a specific complex with the AAV D-sequence,²⁴ we reasoned that DNA sequences in the human genome must also exist with which these proteins might interact. Indeed, analysis of the database from the National Center for Biotechnology Information (NCBI; www.ncbi.nlm.nih.gov) by using blastn program revealed that the D-sequence-like sequences exist in the following cellular genes: (1) INK4A/ARF promoter, (2) sorting nexin 1 (SNX1), (3) LT-α regulatory region, and (4) X-box in the HLA-DRA promoter. These sequence alignments are shown in Fig. 1. Several human cellular D-sequence-like sequences were also experimentally obtained by two consecutive rounds of PCR amplification using commercially available five different human genomic libraries with APs and the AAV D-sequence primer pairs. PCR amplification products were molecularly cloned in bacterial plasmids and subjected to nucleotide sequencing as described in Materials and Methods section. Three sets of sequences obtained were homologous with SNX1, LT-α, and HLA-DRA genes. The remarkable similarity of the AAV D-sequence with the X-box sequence in the HLA-DRA promoter, to which a cellular factor, designated RFX, is known to bind,³⁴ suggested that the dsD-BP might share functional similarities with the RFX proteins. Although we obtained several D-sequence-like sequences from human genomic libraries, we only focused on HLA-DRA and LT-α gene sequences because of our interest in the regulation of genes related with immune response to AAV. Thus, all subsequent studies were focused on the X-box in the HLA-DRA gene.

Figure 1.

AAV D-sequence-like sequences exist in the human genome. Human genomic sequences that share sequence homology with the AAV-D-sequence are shown. AAV, adeno-associated virus.

dsD-BP also interacts with the X-box in the human HLA-DRA promoter

We next wished to determine whether dsD-BP also interacts with the X-box. This hypothesis was experimentally tested as follows: EMSAs were performed with WCE prepared from 293 cells using radiolabeled X-box- and AAV D-sequence-specific synthetic oligonucleotide probes as described in Materials and Methods section. These results are shown in Fig. 2. As can be seen, radiolabeled-free X-box (lane 1) and D-sequence (lane 6) formed the same complex with dsD-BP (lanes 2 and 7), as reported previously.²⁴ In crosscompetition experiments, where ∼200-fold molar excess of each unlabeled oligonucleotides was used as a competitor, the D-sequence competed for binding with the X-box probe (lane 4), and the X-box sequence competed with the D-sequence probe (lane 8).

Figure 2.

dsD-BP interacts with the X-box of the HLA-DRA promoter. EMSAs were performed using ³²P end-labeled X-box (CTCCGTTGCTAGGGGAAGGG; lanes 1–5) and the D-sequence (CTCCATCACTAGGGGTTCCT; lanes 6–10) oligonucleotide probes were incubated in the absence (lanes 1, 6) or the presence (lanes 2, 7) of WCE. A 200-fold molar excess of unlabeled oligonucleotides specific for the X-box (lanes 3, 8), the D-sequence (lanes 4, 9), or the LT-α upstream regulatory sequences (ATGCATCACTAGGGGTCCAT; lanes 5, 10) were used in competition experiments. The arrow indicates the specific DNA:protein complex with the X-box- or the AAV D-sequence-specific double-stranded oligonucleotides. dsD-BP, double-stranded D-sequence-binding protein; EMSA, electrophoretic mobility-shift assay; HLA-DRA, human leukocyte antigen DRA; LT-α, lymphotoxin-α; WCE, whole-cell extract.

Competition with the homologous sequences to each of the probes was also observed (lanes 3 and 9). The LT-α sequence competed for binding with the X-box probe (lane 5) and the D-sequence probe (lane 10). Based on these results, we concluded that the dsD-BP is a putative RFX transcription factor. Indeed, RFX1 and RFX3 transcription factors were recently shown to interact with the AAV D-sequence.²⁶

AAV D-sequence downregulates transcriptional activity of the HLA-DRA promoter through inhibition of RFX binding

Beyond physical binding of dsD-BP to the X-box, we tested our hypothesis that the D-sequence competes with the X-box for binding with the RFX transcription factor, which is essential for expression from the HLA-DRA promoter.^35

–39 The regulatory structure of the human HLA-DRA promoter is depicted schematically in Fig. 3A. In brief, binding of RFX factor to the X-box allows the binding of cAMP-response element binding protein (CREB) and NF-Y to X2 and Y boxes, respectively, which allows the binding of a master regulator, CIITA, which regulates HLA-DRA gene expression.⁴⁰ An HLA-DRAp-FLuc reporter plasmid, constructed as described in Materials and Methods section, is depicted schematically in Fig. 3B.

Figure 3.

(A) Schematic representation of the HLA-DRA promoter and binding sites for the cognate transcription factors. (B) Schematic structure of a recombinant plasmid containing the FLuc reporter gene under the control of the HLA-DRA promoter. The regulatory structure of the human HLA-DRA promoter is depicted schematically (A). Binding of RFX factor to the X-box allows the binding of CREB and NF-Y to X2 and Y boxes, respectively, which allows the binding of a master regulator, CIITA, which regulates HLA-DRA gene expression. An HLA-DRAp-FLuc reporter plasmid was constructed as described in Materials and Methods section. CIITA, class II major histocompatibility complex trans-activator; CREB, cAMP-response element binding protein; FLuc, firefly luciferase; HLA-DRAp-FLuc, HLA-DRA promoter-driven firefly luciferase; RFX, regulatory factor binding to the X-box.

HEK293 cells were either mock-transfected, or transfected with either the vector alone, or with HLA-DRAp-FLuc plasmid with or without various amounts ranging from 10 to 200 ng of AAV D-sequence- or X-box-specific synthetic double-stranded oligonucleotides. Two additional nonspecific synthetic double-stranded oligonucleotides were also used as appropriate controls. FLuc activity was determined 48 h post-transfection. In preliminary experiments, both D-sequence and X-box oligonucleotides were found to inhibit the reporter gene expression in a dose-dependent manner (data not shown). In subsequent experiments, 200 ng of each oligonucleotide was used. These results are shown in Fig. 4. It is interesting to note that both AAV D-sequence and X-box sequence inhibited the reporter gene expression in 293 cells by ∼93% and 96%, respectively (Fig. 4A). No inhibition was detected when two different nonspecific synthetic oligonucleotide sequences (data not shown), or when single-stranded X-box-specific oligonucleotides were used (Fig. 4B).

Figure 4.

Effect of AAV D-sequence and X-box sequence on HLA-DRA promoter-driven FLuc reporter gene expression in 293 cells. (A) HEK293 cells were either mock-transfected, or transfected with the vector plasmid or with the HLA-DRAp-Fluc FLuc plasmid with or without cotransfection with double-stranded synthetic oligonucleotides specific for AAV D-sequence or the X-box sequence FLuc activity was determined 48 h post-transfection. (B) Synthetic single-stranded oligonucleotides specific for the X-box sequences were used as appropriate controls, p = 0.001. (C) Expression from the CMV promoter was not affected by either AAV D-sequence, the X-box sequence, or AAV DNA, p = n.s. CMV, cytomegalovirus.

Furthermore, the observed inhibition of expression from the HLA-DRA promoter was specific since neither the D-sequence nor X-box sequence-specific oligonucleotides had any significant effect on expression of the LacZ reporter gene from the CMV promoter (Fig. 4C). Strong inhibition of expression from the HLA-DRA promoter was also observed with AAV DNA in 293 cells (Fig. 5A) as well as in HeLa cells (Fig. 5B). These results further suggest that AAV D-sequence-mediated downregulation of the MHC-II promoter is mediated by recruitment of a putative RFX transcription factor(s) away from the X-box.

Figure 5.

Effect of AAV DNA on HLA-DRA promoter-driven FLuc reporter gene expression in 293 and HeLa cells, and effect of IFN-γ on expression from HLA-DRA promoter and its suppression by D-sequence or the X-box sequence in HeLa cells. (A) HEK293 and (B) HeLa cells were transfected with either 200 ng of D-sequence or 1 μg of AAV DNA. FLuc expression was determined as described in the legend to Figure 4. (C) HeLa cells were treated with IFN-γ followed by transfection with the vector plasmid or with the HLA-DRAp-FLuc plasmid with or without cotransfection with D-sequence oligonucleotides. IFN-γ treatment increased the extent of expression from the HLA-DRA promoter. The rest of the steps were the same as described above, p = 0.001. IFN-γ, interferon-γ

IFN-γ-induced MHC-II promoter activation is inhibited by AAV D-sequence

Expression from the endogenous MHC-II promoter in 293 or HeLa cells does not normally occur, but in HeLa cells it can be induced after treatment with IFN-γ⁴¹ We next examined the effect of AAV D-sequence on expression from the HLA-DRA promoter in HeLa cells, with and without pretreatment with 100 U/mL of IFN-γ. These results are shown in Fig. 5C. As is evident, although IFN-γ-treatment increased the extent of expression from the HLA-DRA promoter, presumably because of activation of the CIITA master regulator,¹⁹ HLA-DRA promoter-mediated expression of the FLuc reporter gene was also significantly inhibited by D-sequence oligonucleotides. These results further corroborate that AAV D-sequence-mediated downregulation of the HLA-DRA promoter is mediated by recruitment of a putative RFX transcription factor(s) away from the X-box.

Since all the data obtained thus far were derived from DNA transfections, we wished to evaluate the effect of AAV infection on the MHC-II gene expression. IFN-γ-treated HeLa cells were either mock-infected, or infected with either the wild-type AAV2 or the recombinant AAV2- LacZ vectors, with and without pretreatment of cells with tyrphostin 1, known to augment AAV second-strand DNA synthesis,⁴² and analyzed for MHC-II expression by fluorescence-activated cell sorting using mouse antihuman MHC-II (DR and DP) monoclonal antibody isotype IgG3 (Chemicon International, Inc., Temecula, CA). In preliminary experiments, surface expression of MHC-II proteins in IFN-γ-treated HeLa cells was also downregulated after infection by both wild-type and recombinant AAV (data not shown).

Based on all the available data, we propose a model, shown in Fig. 6, in which expression from the MHC-II promoter does not occur in the absence of RFX transcription factor binding to the X-box (Fig. 6A). After RFX transcription factor binding to the X-box, and additional transcription factor binding to X2 and Y boxes, expression from the MHC-II promoter ensues (Fig. 6B). In the presence of AAV genomes, either as a panhandle structure in a HP configuration after viral second-strand DNA synthesis, or in a duplex form after complementary DNA stand annealing, the double-stranded D-sequence effectively competes for the RFX transcription factor binding, thereby rendering the MHC-II promoter inactive leading to downregulation of the MHC-II genes (Fig. 6C). These studies suggest that the D-sequence-mediated downregulation of the MHC-II genes might be exploited toward the development of recombinant AAV vectors capable of dampening the host humoral immune response, which was tested experimentally as follows.

Figure 6.

A simplified model for AAV D-sequence-mediated downregulation of MHC-II gene expression. In the absence of RFX transcription factor binding to the X-box, expression from the MHC-II promoter does not occur (A). After RFX transcription factor binding to the X-box, transcription factors CREB and NF-Y bind to X2 and Y boxes, respectively, followed by CIITA master regulator-mediated expression from the MHC-II promoter (B). After infection with AAV, the double-stranded D-sequence effectively competes for the RFX transcription factor binding, leading to downregulation of the MHC-II gene expression (C). See the text for details. MHC-II, major histocompatibility complex class II.

Development of X-box-containing AAV vectors and evaluation of their biological activity

Since the D-sequence in the AAV-ITR, which shares partial sequence homology with the X-box in the HLA-DRA promoter, can functionally compete for the binding of RFX transcription factor, we next generated a novel AAV vector in which one D-sequence in the ITR was replaced with an authentic X-box sequence in a scAAV2 vector (Supplementary Fig. S1). The following two sets of experiments were carried out. In the first set, scAAV2-EGFP or scAAV2-X-box-EGFP vectors were administered, either IV or IM, in C57BL6/J mice (n = 4 in each group). Three weeks postvector administration, the levels of anti-AAV antibody production were determined. These data are shown in Supplementary Fig. S2. No significant differences were observed in the levels of anti-AAV2 IgG2c antibodies under either experimental condition.

In the second set of experiments, attempts were made to determine whether repeat intravitreal administration of AAVs would be facilitated by initial injection with an X-box-containing vector. scAAV2-X-box-EGFP vectors in one eye, transgene expression from a second scAAV-mCherry vector in the contralateral eye could be observed. To this end, two groups of C57BL6/J mice (n = 8 in each group) received intravitreal injections in their RE with scAAV2-EGFP or scAAV2-X-box-EGFP vectors. Four weeks later, fundus images were acquired to document the extent of GFP expression. Next, both groups of mice were further subdivided into two groups (n = 4 each).

In the first group of four mice, scAAV2-EGFP vector was intravitreally injected into the contralateral (left) eye. In the second group of four mice, scAAV2-mCherry vector was intravitreally injected into the same (right) eye. This was done to ask whether initial administration of scAAV2-X-box-EGFP would facilitate transgene expression by a second intravitreally delivered AAV in either the contralateral or same eye, respectively. Four weeks later (and 8 weeks postinitial injections), fundus images were captured as described previously in this section. These images are shown in Supplementary Fig. S3.

As can be seen, the majority of eyes that received the second intravitreal injection failed to express either EGFP (contralateral eye) or mCherry (same eye), suggesting that either the scAAV2-X-box-EGFP vectors were unable to dampen the humoral immune response, or the subsequent scAAV2-mCherry and scAAV2-GFP vectors failed to overcome antibody-mediated neutralization.

Discussion

Since a significant fraction of the patient population with pre-existing antibodies to AAV capsids is currently ineligible to be enrolled in gene therapy trials with AAV vectors, and repeat dosing with AAV vectors is presently not possible, it is very clear that novel strategies are needed to overcome this major barrier of vector neutralization. Indeed, efforts by a number of investigators are afoot to this end, which include the following: (1) use of alternative AAV serotypes,⁴³ (2) transient depletion of B lymphocytes,⁴⁴ (3) structure-guided design of AAV vectors devoid of known antigenic epitopes,⁴⁵ and (4) site-directed mutagenesis of AAV capsids to develop AAV vectors capable of evading pre-existing antibodies.⁴⁶

In our present studies, we were able to document that AAV D-sequence-like sequences do exist in the human genome, which, beyond computer-based homology searches, could be isolated after PCR-based amplifications. These sequences were found in the following genes: INK4A/ARF, sorting nexin (SNX1), and LT-α. INK4A/ARF gene-encoded proteins are involved in pRB and p53 tumor suppressor pathways.⁴⁷ Whether the well-known antitumor property of AAV^{48

–55} involves the D-sequence remains to be explored. Similarly, further studies are warranted to evaluate the role of SNX1 gene, known to encode a diverse group of cellular trafficking proteins and their propensity to form protein–protein complexes,⁵⁶ in the life cycle of AAV. LT-α gene encodes a protein, also known as tumor necrosis factor-β, that possesses antiproliferative activity,⁵⁷ Again, what role, if any, LT-α plays in AAV biology will require further studies.

One of these sequences was of significant interest since a cellular protein, which we had termed double-stranded D-sequence-binding protein,²⁴ could bind specifically to the X-box in the HLA-DRA promoter, and we speculated that the dsD-BP was a putative RFX transcription factor.⁵⁸ Indeed, D-sequences from AAV1 and AAV2 viral genomes were recently shown to interact with RFX1 and RFX3 transcription factors.²⁶ D-sequence in the AAV-ITR, which shares a partial sequence homology with the X-box in the HLA-DRA promoter, can functionally compete for the binding of RFX transcription factor. This observation led to pursuit of a novel AAV vector in which one D-sequence in the ITR was replaced with an authentic X-box sequence in a scAAV2 vector.

However, the extent of anti-AAV antibody production from both vectors after intravenous or intramuscular administration in a murine model was not significantly different. Similarly, intravitreal administration of X-box-containing scAAV2 vectors in one eye was insufficient to permit transduction by conventional AAV2 vectors in the contralateral or same eye in mice. This apparent paradox between the in vitro and the in vivo data could possibly be explained by the following: (1) AAV2 serotype vectors, known not to transduce murine APCs, were used; (2) in contrast to the use of a large excess of D-sequence oligonucleotides in transfection experiments to achieve near total suppression of expression from the HLA-DRA promoter in vitro, AAV vectors containing only one X-box sequence were used for intravenous and intramuscular injections in vivo; (3) scAAV2 and scAAV2-X-box vectors were used at only a 1:1 ratio for intravitreal injections; (4) the number of resident APCs in retina is generally low given the immune privileged status of this organ; and (5) D-sequence, and by inference, human X-box sequence, which interacts with human RFX1 and RFX3 transcription factors, was used in mouse models, whereas mouse X-box sequence is known to bind with mouse RFX5 transcription factor.⁵⁹

Thus, it is possible that the murine model does not necessarily recapitulate the transcription control mechanism that is operational in human MHC-II gene expression. These studies further underscore the need to exercise caution in relying on and drawing conclusions from rodent models, as well as the need to pursue further studies using the X-box sequence from nonhuman primates that may more closely mimic the human system. Ultimately, however, the validity of such an approach to dampen the host humoral response to AAV vectors will only be corroborated from phase I clinical trials in humans.

Footnotes

Acknowledgments

We thank Dr. Kenneth I. Berns for a critical review of this article. We also thank our laboratory colleagues, both past and current, for scientific discussions and helpful suggestions. We also thank Dr. Roland W. Herzog for his support and general scientific counsel.

Author Disclosure

A.S. is a cofounder of, and holds equity in, Lacerta Therapeutics and Nirvana Therapeutics, and is an inventor on several issued patents on recombinant AAV vectors that have been licensed to various gene therapy companies. All other authors declare no conflict of interests.

Funding Information

This research was supported in part by Public Health Service grants R01 HL-097088, R41 AI-122735, and R21 EB-015684 from the National Institutes of Health; a grant from the Children's Miracle Network; and supported by the Kitzman Foundation (to A.S.).

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

References

Bainbridge

, Smith

, Barker

, et al. Effect of gene therapy on visual function in Leber's congenital amaurosis. N Engl J Med, 2008; 358:2231–2239.

Maguire

, Simonelli

, Pierce

, et al. Safety and efficacy of gene transfer for Leber's congenital amaurosis. N Engl J Med, 2008; 358:2240–2248.

Hauswirth

, Aleman

, Kaushal

, et al. Treatment of Leber congenital amaurosis due to RPE65 mutations by ocular subretinal injection of adeno-associated virus gene vector: short-term results of a phase I trial. Hum Gene Ther, 2008; 19:979–990.

Cideciyan

, Aleman

, Boye

, et al. Human gene therapy for RPE65 isomerase deficiency activates the retinoid cycle of vision but with slow rod kinetics. Proc Natl Acad Sci U S A, 2008; 105:15112–15117.

Gaudet

, Methot

, Dery

, et al. Efficacy and long-term safety of alipogene tiparvovec (AAV1- LPLS447X) gene therapy for lipoprotein lipase deficiency: an open-label trial. Gene Ther, 2013; 20:361–369.

Nathwani

, Rosales

, McIntosh

, et al. Long-term safety and efficacy following systemic administration of a self-complementary AAV vector encoding human FIX pseudotyped with serotype 5 and 8 capsid proteins. Mol Ther, 2011; 19:876–885.

Nathwani

, Reiss

, Tuddenham

, et al. Long-term safety and efficacy of factor IX gene therapy in hemophilia B. N Engl J Med, 2014; 371:1994–2004.

Miesbach

, Meijer

, Coppens

, et al. Gene therapy with adeno-associated virus vector 5-human factor IX in adults with hemophilia B. Blood, 2018; 131:1022–1031.

George

, Sullivan

, Giermasz

, et al. Hemophilia B gene therapy with a high-specific- activity factor IX variant. N Engl J Med, 2017; 377:2215–2227.

10.

Rangarajan

, Walsh

, Lester

, et al. AAV5-factor VIII gene transfer in severe hemophilia A. N Engl J Med, 2017; 377:2519–2530.

11.

Pasi

, Rangarajan

, Mitchell

, et al. Multiyear follow-up of AAV5-hFVIII-SQ gene therapy for hemophilia A. N Engl J Med, 2020; 382:29–40.

12.

Hwu

, Muramatsu

, Tseng

, et al. Gene therapy for aromatic L-amino acid decarboxylase deficiency. Sci Transl Med, 2012; 4:134ra161.

13.

MacLaren

, Groppe

, Barnard

, et al. Retinal gene therapy in patients with choroideremia: initial findings from a phase 1/2 clinical trial. Lancet, 2014; 383:1129–1137.

14.

Guy

, Feuer

, Davis

, et al. Gene therapy for leber hereditary optic neuropathy: low- and medium-dose visual results. Ophthalmology, 2017; 124:1621–1634.

15.

Mendell

, Al-Zaidy

, Shell

, et al. Single-dose gene-replacement therapy for spinal muscular atrophy. N Engl J Med, 2017; 377:1713–1722.

16.

Fitzpatrick

, Leborgne

, Barbon

, et al. Influence of pre-existing anti-capsid neutralizing and binding antibodies on AAV vector transduction. Mol Ther Methods Clin Dev, 2018; 9:119–129.

17.

Cooper

. The early history of B cells. Nat Rev Immunol, 2015; 15:191–197.

18.

Roche

, Furuta

. The ins and outs of MHC class II-mediated antigen processing and presentation. Nat Rev Immunol, 2015; 15:203–216.

19.

Nakamura

. CIITA: a master regulator of adaptive immunity shows its innate side in the bone. J Bone Miner Res, 2014; 29:287–289.

20.

Srivastava

, Lusby

, Berns

. Nucleotide sequence and organization of the adeno-associated virus 2 genome. J Virol, 1983; 45:555–564.

21.

Srivastava

. Replication of the adeno-associated virus DNA termini in vitro. Intervirology, 1987; 27:138–147.

22.

Lusby

, Fife

, Berns

. Nucleotide sequence of the inverted terminal repetition in adeno- associated virus DNA. J Virol, 1980; 34:402–409.

23.

Wang

, Ponnazhagan

, Srivastava

. Rescue and replication signals of the adeno-associated virus 2 genome. J Mol Biol, 1995; 250:573–580.

24.

Wang

, Ponnazhagan

, Srivastava

. Rescue and replication of adeno-associated virus type 2 as well as vector DNA sequences from recombinant plasmids containing deletions in the viral inverted terminal repeats: selective encapsidation of viral genomes in progeny virions. J Virol, 1996; 70:1668-1677.

25.

Wang

, Qing

, Ponnazhagan

, et al. Adeno-associated virus type 2 DNA replication in vivo: mutation analyses of the D sequence in viral inverted terminal repeats. J Virol, 1997; 71:3077–3082.

26.

Julien

, Chassagne

, Peccate

, et al. RFX1 and RFX3 transcription factors interact with the D sequence of adeno-associated virus inverted terminal repeat and regulate AAV transduction. Sci Rep, 2018; 8:210.

27.

Qing

, Wang

, Kube

, et al. Role of tyrosine phosphorylation of a cellular protein in adeno- associated virus 2-mediated transgene expression. Proc Natl Acad Sci U S A, 1997; 94:10879–10884.

28.

Qing

, Khuntirat

, Mah

, et al. Adeno-associated virus type 2-mediated gene transfer: correlation of tyrosine phosphorylation of the cellular single-stranded D sequence-binding protein with transgene expression in human cells in vitro and murine tissues in vivo. J Virol, 1998; 72:1593–1599.

29.

Ponnazhagan

, Wang

, Woody

, et al. Differential expression in human cells from the p6 promoter of human parvovirus B19 following plasmid transfection and recombinant adeno-associated virus 2 (AAV) infection: human megakaryocytic leukaemia cells are non-permissive for AAV infection. J Gen Virol, 1996; 77(Pt 6):1111–1122.

30.

Ponnazhagan

, Mukherjee

, Wang

, et al. Adeno-associated virus type 2-mediated transduction in primary human bone marrow-derived CD34+ hematopoietic progenitor cells: donor variation and correlation of transgene expression with cellular differentiation. J Virol, 1997; 71:8262–8267.

31.

Ling

, Wang

, Lu

, et al. Enhanced transgene expression from recombinant single-stranded D-sequence-substituted adeno-associated virus vectors in human cell lines in vitro and in murine hepatocytes in vivo. J Virol, 2015; 89:952–961.

32.

Kube

, Srivastava

. Quantitative DNA slot blot analysis: inhibition of DNA binding to membranes by magnesium ions. Nucleic Acids Res, 1997; 25:3375–3376.

33.

Kay

, Ryals

, Aslanidi

, et al. Targeting photoreceptors via intravitreal delivery using novel, capsid-mutated AAV vectors. PLoS One, 2013; 8:e62097.

34.

Reith

, Kobr

, Emery

, et al. Cooperative binding between factors RFX and X2bp to the X and X2 boxes of MHC class II promoters. J Biol Chem, 1994; 269:20020–20025.

35.

Steimle

, Durand

, Barras

, et al. A novel DNA-binding regulatory factor is mutated in primary MHC class II deficiency (bare lymphocyte syndrome). Genes Dev, 1995; 9:1021–1032.

36.

Masternak

, Barras

, Zufferey

, et al. A gene encoding a novel RFX-associated transactivator is mutated in the majority of MHC class II deficiency patients. Nat Genet, 1998; 20:273–277.

37.

Nagarajan

, Louis-Plence

, DeSandro

, et al. RFX-B is the gene responsible for the most common cause of the bare lymphocyte syndrome, an MHC class II immunodeficiency. Immunity, 1999; 10:153–162.

38.

Moreno

, Beresford

, Louis-Plence

, et al. CREB regulates MHC class II expression in a CIITA-dependent manner. Immunity, 1999; 10:143–151.

39.

Lee

, Lee

, Kim

, Kwon

. Direct role of NF-kappaB activation in Toll-like receptor- triggered HLA-DRA expression. Eur J Immunol, 2006; 36:1254–1266.

40.

Gauvreau

, Côté

, Bourgeois-Daigneault

, et al. Sorting of MHC class II molecules into exosomes through a ubiquitin-independent pathway. Traffic, 2009; 10:1518–1527.

41.

Abendroth

, Slobedman

, Lee

, et al. Modulation of major histocompatibility class II protein expression by varicella-zoster virus. J Virol, 2000; 74:1900–1907.

42.

Mah

, Qing

, Khuntirat

, et al. Adeno-associated virus type 2-mediated gene transfer: role of epidermal growth factor receptor protein tyrosine kinase in transgene expression. J Virol, 1998; 72:9835–9843.

43.

Davidoff

, Gray

, Ng

, et al. Comparison of the ability of adeno-associated viral vectors pseudotyped with serotype 2, 5, and 8 capsid proteins to mediate efficient transduction of the liver in murine and nonhuman primate models. Mol Ther, 2005; 11:875–888.

44.

Sack

, Merchant

, Markusic

, et al. Transient B cell depletion or improved transgene expression by codon optimization promote tolerance to factor VIII in gene therapy. PLoS One, 2012; 7:e37671.

45.

Tse

, Klinc

, Madigan

, et al. Structure-guided evolution of antigenically distinct adeno- associated virus variants for immune evasion. Proc Natl Acad Sci U S A, 2017; 114:E4812–E4821.

46.

van Lieshout

, Domm

, Rindler

, et al. A novel triple-mutant AAV6 capsid induces rapid and potent transgene expression in the muscle and respiratory tract of mice. Mol Ther Methods Clin Dev, 2018; 9:323–329.

47.

Sharpless

. INK4a/ARF: a multifunctional tumor suppressor locus. Mutat Res, 2005; 576:22–38.

48.

de la Maza

, Carter

. Inhibition of adenovirus oncogenicity in hamsters by adeno-associated virus DNA. J Natl Cancer Inst, 1981; 67:1323–1326.

49.

Hermonat

. Inhibition of H-ras expression by the adeno-associated virus Rep78 transformation suppressor gene product. Cancer Res, 1991; 51:3373–3377.

50.

Hermonat

. Inhibition of bovine papillomavirus plasmid DNA replication by adeno-associated virus. Virology, 1992; 189:329–333.

51.

Hermonat

. Down-regulation of the human c-fos and c-myc proto-oncogene promoters by adeno- associated virus Rep78. Cancer Lett, 1994; 81:129–136.

52.

Hermonat

. Adeno-associated virus inhibits human papillomavirus type 16: a viral interaction implicated in cervical cancer. Cancer Res, 1994; 54:2278–2281.

53.

Mayor

, Houlditch

, Mumford

. Influence of adeno-associated satellite virus on adenovirus-induced tumours in hamsters. Nat New Biol, 1973; 241:44–46.

54.

Ostrove

, Duckworth

, Berns

. Inhibition of adenovirus-transformed cell oncogenicity by adeno-associated virus. Virology, 1981; 113:521–533.

55.

Raj

, Ogston

, Beard

. Virus-mediated killing of cells that lack p53 activity. Nature, 2001; 412:914–917.

56.

Cullen

. Endosomal sorting and signalling: an emerging role for sorting nexins. Nat Rev Mol Cell Biol, 2008; 9:574–582.

57.

Nedwin

, Naylor

, Sakaguchi

, et al. Human lymphotoxin and tumor necrosis factor genes: structure, homology and chromosomal localization. Nucleic Acids Res, 1985; 13:6361–6373.

58.

Kwon

H-J

, Weigel

, Hansen

, et al. AAV-mediated gene transfer: effect on MHC Class II gene expression. Mol Ther, 2000; 1: S190.

59.

Villard

, Peretti

, Masternak

, et al. A functionally essential domain of RFX5 mediates activation of major histocompatibility complex class II promoters by promoting cooperative binding between RFX and NF-Y. Mol Cell Biol, 2000; 20:3364–3376.

60.

Wang

, Ma

, Li

, et al. Rapid and highly efficient transduction by double-stranded adeno-associated virus vectors in vitro and in vivo . Gene Ther, 2003; 10:2105–2111.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.04 MB

0.15 MB