Abstract
Blood dendritic cell antigen 2 (BDCA2) is exclusively expressed on plasmacytoid dendritic cells (pDCs) whose uncontrolled production of type I IFN (IFN-I) is crucial in pathogenesis of SLE and other autoimmune diseases. Although anti-BDCA2 antibody therapy reduced disease activity in SLE patients, its clinical efficacy needs further improvement. We developed a novel glucocorticoid receptor agonist and used it as a payload to conjugate with an anti-BDCA2 antibody to form an BDCA2 antibody–drug conjugate (BDCA2-ADC). The activation of BDCA2-ADC was evaluated in vitro.
Inhibitory activity of BDCA2-ADC was evaluated in peripheral blood mononuclear cells or in purified pDCs under ex vivo toll-like receptor agonistic stimulation. The global gene regulation in purified pDCs was analysed by RNA-seq. The antigen-dependent payload delivery was measured by reporter assay.
The BDCA2-ADC molecule causes total suppression of IFNα production and broader inhibition of inflammatory cytokine production compared with the parental antibody in human pDCs. Global gene expression analysis confirmed that the payload and antibody acted synergistically to regulate both type I IFN signature genes and glucocorticoid responsive genes in pDCs.
Taken together, these data suggest dual mechanisms of BDCA2-ADC on pDCs and the potential for BDCA2-ADC to be the first ADC treatment for SLE in the world and a better treatment option than anti-BDCA2 antibody for SLE patients.
BDCA2-ADC shows superior IFN-I inhibition with broader inflammatory cytokine suppression in pDCs.
BDCA2-ADC demonstrates a dual mechanism with synergistic inhibition of IFN-I and glucocorticoid responsive genes.
Introduction
SLE is an autoimmune disease characterized by the production of autoantibodies, leading to tissue damage [1]. Although the pathogenesis of SLE is heterogeneous, abnormal production of IFN-I and proinflammatory cytokines by plasmacytoid dendritic cells (pDCs) is pivotal [2]. The initial formation of immune complexes by autoantibodies with self-antigens triggers the activation of pDCs through toll-like receptors (TLRs) with the induction of IFN-I. IFN-I amplifies autoantibody production not only by inducing B cell maturation and differentiation but also by enhancing antigen presentation and co-stimulatory molecule expression by dendritic cells for adaptive immune response generation [3]. Targeting the IFN-I pathway has been validated as an effective approach for SLE therapy. Specifically, anifrolumab, an IFN-I receptor-blocking antibody, has been approved for moderate-to-severe SLE treatment [4]. Therefore, inhibition of IFN-I production or IFN signalling would block both the initiation and the development of SLE.
Blood dendritic cell antigen 2 (BDCA2) (also known as CLEC4C and CD303), a C-type lectin receptor, is uniquely expressed on pDCs [5]. Ligation of BDCA2 with a monoclonal antibody induces its internalization, suppressing IFN production in pDCs [6]. Considering the key function of pDCs in the pathogenesis of SLE, an anti-BDCA2 antibody has been developed for SLE therapy. BIIB059 (litifilimab), a humanized anti-BDCA2 antibody, shows clinical efficacy for ameliorating skin lesions in cutaneous SLE and reducing joint disease activity in SLE [7–9]. Therefore, BDCA2 is considered a potential target for SLE therapy through inhibition of the pDC–IFN axis. However, SLE is a multifactorial complex disease. Despite the important function of pDCs and IFN-I in the pathogenesis of SLE, other mechanisms exist that contribute to the disease progression. Therefore, the clinical efficacy of either anifrolumab or BIIB059 is moderate. About half of the patient population does not respond to the above-mentioned therapies [4, 9]. As a more efficacious treatment option is needed, we developed the idea of an immune modulatory antibody–drug conjugate (ADC). By conjugating an anti-BDCA2 antibody to a glucocorticoid receptor (GR) agonist payload, we reasoned that better efficacy would be achieved.
Glucocorticoids (GCs) are the most widely used class of anti-inflammatory and immunosuppressive agents [10]. GCs bind to the cytosolic GR and reprogram cell transcription through multiple mechanisms. The ligand-bound GR can translocate into the nucleus, where it acts as a transcription factor and binds GC response elements (GREs) to regulate gene expression [11]. Since GR is expressed by nearly all nucleated cells, GCs remain the mainstay of inflammatory and autoimmune pathology therapy. However, the high dosage or long-term systemic use of GCs is associated with potentially undesirable side effects [12, 13]. Balancing the benefits and risks of GC therapy is recommended for the development of autoimmune diseases therapy [14].
Here, we developed an immune modulatory ADC technology platform, named Duality Immune Modulating Antibody Conjugate (DIMAC). Using this technology, we further developed a BDCA2 antibody–drug conjugate (BDCA2-ADC) with a DIMAC payload, which is a novel and proprietary GC-based linker-payload, conjugated to an anti-BCDA2 antibody (BIIB059) to achieve targeted delivery of GCs to pDCs. BDCA2-ADC maintained the high affinity for BDCA2 and internalization capability of the parental antibody. It combines the suppressive effect of both the antibody and GC to pDCs. In addition to nearly completely blocking IFNα production in pDCs during in vitro TLR stimulation, BDCA2-ADC also elicited broad inhibition of proinflammatory cytokines through agonistic activation of GR transcription. Taken together, the synergistic effects of anti-BDCA2 antibody and GC payload could potentially translate to better clinical efficacy and safety for the treatment of SLE.
Methods
Preparation and characterization of the ADC
The DIMAC linker-payload (HYX0449) is composed of a conjugation moiety, a tripeptide linker and a novel GC. The ADC was prepared by reducing BIIB059, a humanized monoclonal antibody targeting BDCA2, with Tris(2-carboxyethyl)phosphine hydrochloride (TCEP-HCl) in molar excess in reduction buffer (40 mM phosphate buffer 100 mM NaCl, pH 7.4) at ambient temperature. The partially reduced BIIB059 was then conjugated with excess HYX0449 (molar ratio 6–7) without removing TCEP. The excess linker-payload was then removed with a Zebra Spin desalting column (Thermo Fisher Scientific, Waltham, MA, USA) and exchanged into 20 mM histidine buffer (pH 5.5) to give the purified conjugate with purity >95%. The drug-to-antibody ratio (DAR) was determined by hydrophobic interaction HPLC (HIC) to be 4.07. The purified ADC was filtered through a 0.2-µm sterile filter and stored at −80°C until use.
pDC isolation
pDCs were obtained from buffy coats of a single healthy donor after written informed consent was obtained under WuXi AppTec’s ethical committee review. Peripheral blood mononuclear cells (PBMCs) were isolated from donor blood by Ficoll density gradient centrifugation. The pDCs were subsequently isolated using magnetic-activated cell sorting (MACS) according to the manufacturer’s protocol (Miltenyi Biotec, Bergisch Gladbach, Germany, cat. no. 130-097-240). Briefly, PBMCs were resuspended in MACS buffer (PBS containing 1% fetal bovine serum [FBS] and 1 mM EDTA) and incubated with Non-PDC Biotin-Antibody Cocktail II (Miltenyi Biotec) for 10 min at 4°C. The cells were washed with MACS buffer, resuspended in Non-PDC MicroBead Cocktail II (Miltenyi Biotec) and incubated for 15 min at 4°C to deplete non-pDCs. The pDCs were isolated from the remaining cell population by positive selection using CD304 (BDCA4/neuropilin-1) Diamond MicroBeads (Miltenyi Biotec) following the manufacturer’s instructions. The cells were counted, and purity was assessed using flow cytometry. For this purpose, cells were washed and stained for 30 min at 4°C using BV421-labelled anti-CD303 (BD Biosciences, San Jose, CA, USA, cat. no. 566427) and phycoerythrin-labelled anti-CD123 (BioLegend, San Diego, CA, USA, cat. no. 306005) at the manufacturer’s recommend concentration. The pDCs were identified as CD303+CD123+, and the purity was >90% by flow cytometry analysis (FACS).
PBMC and pDC stimulation
Purified PBMCs or pDCs were diluted in fresh culture medium to reach a cell concentration of 2 × 106 per ml. Then, 100 μl of the cell suspension was transferred to a 96-well plate, and the cells were treated with the indicated concentration of antibody or ADC overnight. The cells were stimulated with 2.5 µM CpG-A (InvivoGen, San Diego, CA, USA, cat. no. tlrl-2216-5) or 5 μM R848 (InvivoGen, tlrl-r848) for the indicated duration. The supernatant was collected after stimulation, and the cytokine concentrations were measured by ELISA or Luminex assessment.
ELISA analysis of cytokine concentration
IFNα, TNFα, IL-6 and IL-8 concentrations were measured using ELISA (Thermo Fisher Scientific) according to the manufacturer’s protocol.
Luminex analysis of cytokine concentrations
IFNα, TNFα, MIP-1α, IL-6 and IP-10 concentrations were measured using a Luminex ProcartaPlex system (Thermo Fisher Scientific) according to the manufacturer’s instructions. Briefly, 50 μl of microparticle cocktail was mixed with 50 μl of supernatant sample or standard in a plate well. The plate was sealed with foil and incubated for 2 h at room temperature on a horizontal orbital microplate shaker. The microparticles were washed three times and incubated with 25 μl of biotin-antibody cocktail for 30 min at room temperature. Next, the microparticles were washed three times and stained with streptavidin–phycoerythrin. Finally, the beads were washed three times and resuspended by adding 120 μl of sample reading buffer to each well. The fluorescence was read using a Bio-Plex 200 analyser (Bio-Rad Laboratories, Hercules, CA, USA), and the data were analysed with Bio-Plex Manager software version 6.2. The median fluorescence intensity was used to calculate cytokine/chemokine concentrations with a five-parameter logistic method.
RNA isolation and library preparation for transcriptome sequencing
RNA was isolated from the treated pDCs using TRIzol (Thermo Fisher Scientific, cat. no. 15596026). The total amount and integrity of the RNA were assessed using the RNA Nano 6000 Assay Kit and a Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA). Total RNA was used as input material for RNA sample preparation. Briefly, mRNA was purified from total RNA by using poly(T) oligo-attached magnetic beads. Fragmentation was carried out using divalent cations under elevated temperature in First-Strand Synthesis Reaction Buffer (5×). First-strand cDNA was synthesized using a random hexamer primer and M-MuLV Reverse Transcriptase, and then RNaseH was used to degrade the RNA. Second-strand cDNA synthesis was subsequently performed using random primer, DNA polymerase I and dNTPs. The remaining overhangs were converted into blunt ends via exonuclease/polymerase activity. After adenylation of the 3′ ends of the DNA fragments, adaptors with hairpin loop structures were ligated to prepare for hybridization. To preferentially select cDNA fragments 370∼420 bp in length, the library fragments were purified with an AMPure XP system (Beckman Coulter, Beverly, MA, USA). Then, PCR amplification was performed, the PCR products were purified with AMPure XP beads, and the library was finally obtained. The quality of the library was initially assessed with a Qubit 2.0 fluorometer (Thermo Fisher Scientific), followed by insert size detection with an Agilent 2100 Bioanalyzer.
Sequencing and data analysis
Different libraries were pooled according to the effective concentration of the library and the desired quantity of reads and then sequenced with an Illumina (San Diego, CA, USA) NovaSeq 6000. Clean data (clean reads) were obtained by removing reads containing the adapter sequence, reads containing N bases and low-quality reads from the raw data. Reference genome and gene model annotation files were downloaded from the genome website directly (ftp://ftp.ensembl.org/pub/release-94/fasta/homo_sapiens/dna/Homo_sapiens.GRCh38.dna.toplevel.fa.gz). The index of the reference genome was built using HISAT2 (v2.0.5), and paired-end clean reads were aligned to the reference genome using HISAT2 (v2.0.5). FeatureCounts v1.5.0-p3 was used to count the reads mapped to each gene. Then, the fragments per kilobase per million mapped reads (FPKM) of each gene was calculated based on the length of the gene and the number of reads mapped to this gene. Prior to differential gene expression analysis, for each sequenced library, the read counts were adjusted with the edgeR package through one scaling normalized factor. Differential expression analysis between two conditions was performed using the edgeR R package (3.24.3). The P-values were adjusted using the Benjamini and Hochberg method. A Padj < 0.05 and a |log2(fold-change)| > 1 were set as the criteria for significantly differential expression. We used the clusterProfiler R package (3.8.1) to test the statistical enrichment of differentially expressed genes in Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways.
Antibody staining and flow cytometry analysis
Cells were washed once with FACS buffer (PBS + 1% FBS) and blocked with Human TruStain FcX (BioLegend, 422302) at 4°C for 10 min. The cells were washed and incubated with indicated antibodies against surface proteins in FACS buffer (PBS + 1% FBS) at 4°C for 30 min. Afterward, the cells were washed with FACS buffer and resuspended in PBS containing 1% paraformaldehyde. All cells were kept at 4°C until acquisition by FACS. The flow cytometry data were analysed using FlowJo X (BD Biosciences, San Jose, CA, USA).
HEK293 GRE luciferase reporter assay
A HEK-293 glucocorticoid response element (GRE) luciferase reporter cell line was purchased from Panomics (Fremont, CA, USA) (cat. no. RC0018). Cells were transfected with hBDCA2 expression plasmid or empty vector via Lipofectamine 2000 (Thermo Fisher Scientific). The transfected cells were seeded at 5 × 104 cells/well in a 96-well cell assay plate in complete medium (DMEM with 10% FBS, 1% penicillin–streptomycin and 100 μg/ml hygromycin B) at 37°C and 5% CO2 overnight. On the following day, the cells were incubated with various concentrations of BIIB059, BDCA2-ADC or human IgG1-ADC (Iso-ADC) for 24 h. Luciferase GRE reporter activity was measured according to the manufacturer’s instructions for the Steadylite plus system (PerkinElmer, Waltham, MA, USA) and being read on an Envision plate reader (PerkinElmer).
Statistical analysis
Student’s t-test was used to compare two sets of data. A value of P < 0.05 was considered to be statistically significant.
Results
BDCA2-ADCs have similar binding affinity to BDCA2 and internalization to the same extent as parental antibody
Anti-BDCA2 antibody BIIB059 has demonstrated preliminary clinical efficacy in SLE patients. However, the results of clinical trials suggest the efficacy of BIIB059 is moderate in SLE patients. We aim to generate a BDCA2 specific ADC that is able to produce greater clinical efficacy by combining the effects of a biologic and a small-molecule drug. As the first step, we designed a DIMAC payload that has high potency and specificity to GR without activation of the mineralocorticoid receptor (Supplementary Fig. S1, available at Rheumatology online). The free DIMAC payload has a pharmacokinetic profile with rapid clearance, with a half-life of 1.37 h in rats (Supplementary Fig. S2 and Table S1, available at Rheumatology online).
Then we generated the BDCA2-ADC by ligation of the DIMAC payload to BIIB059 using a cleavable linker containing maleimide tetrapeptide (Supplementary Fig. S3, available at Rheumatology online). The BDCA2-ADC showed good stability; it remained stable in human plasma up to 21 days with only a small change in concentration and DAR value (Supplementary Fig. S4, available at Rheumatology online). We examined the binding of BDCA2-ADC to cell-surface BDCA2 on engineered CHOK1 cells by flow cytometry. The binding affinities of BDCA2-ADC and BIIB059 for human or cynomolgus monkey BDCA2 did not differ significantly (Fig. 1A). BDCA2 was rapidly internalized upon BIIB059 binding in a time-dependent manner [6]. BDCA2-ADC was also quickly internalized upon BDCA2 binding (Fig. 1B). These results show that the ADC has binding and internalization characteristics similar to those of the parental antibody.
BDCA2-ADC binds to and is internalized by BDCA2-expressing cells as efficiently as the parental antibody. (A) Human BDCA2-expressing CHOK1 cells or cynomolgus monkey BDCA2-expressing 293T cells were treated with BDCA2-ADC or BIIB059 at different concentrations. Cell-surface binding of the BDCA2-ADC or BIIB059 was measured by FACS. (B) Human BDCA2-expressing CHOK1 cells were stained with BDCA2-ADC or BIIB059 on ice (binding) and incubated at 37°C for 1, 2 and 4 h (internalization). BDCA2 internalization was determined based on the proportion of BDCA2-ADC or BIIB059 that stayed bound to the cell surface. ADC: antibody–drug conjugate; BDCA2: blood dendritic cell antigen 2; FACS: flow cytometry analysis; hIgG1: human immunoglobulin gamma-1; MFI: mean fluorescence intensity
BDCA2-ADCs inhibited cytokine production in PBMCs upon TLR stimulation to a greater extent than the parental antibody
Uncontrolled production of IFN-I and proinflammatory cytokines is triggered by TLR activation in pDCs during SLE pathogenesis. The ligation of BDCA2 and BIIB059 can inhibit TLR7/8 and TLR9 agonist-induced IFNα production by human PBMCs [6]. To investigate whether BDCA2-ADC can also suppress cytokine production, PBMCs from healthy donors (n = 3) were treated with BDCA2-ADC or BIIB059 and then stimulated with R848, which activates TLR7/TLR8 intracellularly. IFNα secretion was significantly inhibited by BDCA2-ADC treatment but not BIIB059 treatment (Fig. 2). The secretion of the proinflammatory cytokines TNFα, IL-6 and IL-8 was also blocked by BDCA2-ADC but not BIIB059 treatment (Fig. 2).
BDCA2-ADC suppresses IFNα and proinflammatory cytokine secretion from TLR-activated PBMCs. PBMCs from three healthy donors were pretreated with 300 nM BIIB059 or BDCA2-ADC and then stimulated with 2.5 μM CpG-A or 5 μM R848 for 24 h. The secretion of IFNα, TNFα, IL-6 and IL-8 was measured by ELISA. ADC: antibody–drug conjugate; BDCA2: blood dendritic cell antigen 2; PBMC: peripheral blood mononuclear cell; TLR: toll-like receptor
Autoantibody–DNA complexes stimulate pDCs to produce cytokine through TLR9 [15]. The suppression of IFNα and proinflammatory cytokine secretion was also tested under TLR9 ligand stimulation. CpG-A, which is a TLR9 agonist, induced IFNα secretion in PBMCs, while this cytokine secretion was decreased dramatically by both BIIB059 and BDCA2-ADC. BDCA2-ADC showed stronger IFNα and proinflammatory cytokine suppression than BIIB059 (Fig. 2 and Supplementary Fig. S5, available at Rheumatology online). Taken together, these findings indicate that BDCA2-ADC achieved superior IFNα suppression than BIIB059 with broad proinflammatory cytokine inhibition under TLR stimulation.
BDCA2-ADC inhibits IFNα and proinflammatory cytokine production by pDCs
pDCs are the major cell population in the blood that secretes IFNα in response to TLR9 stimulation [16, 17]. To determine whether the changes in the PBMCs were due to BDCA2-ADC-mediated suppression of pDCs, purified pDCs were incubated with BDCA2-ADC or BIIB059 and then stimulated with CpG-A (Fig. 3A). PDC-surface BDCA2 was internalized upon BIIB059 or BDCA2-ADC treatment, and BDCA2 internalization persisted during CpG-A treatment (Fig. 3B). CpG-A-induced IFNα secretion in pDCs was inhibited by BIIB059 and nearly completely blocked by BDCA2-ADC (Fig. 3C). Suppression was also observed at the transcriptional level. Specifically, IFNα mRNA levels were reduced by BIIB059 treatment and reduced to an undetectable level by BDCA2-ADC treatment (Fig. 3C). In addition to IFNα, both the transcription and the secretion of TNFα, MIP-1α, IL-6 and IP-10 were inhibited by BDCA2-ADC but not by BIIB059 (Fig. 3C). In conclusion, BDCA2-ADC inhibited CpG-A-induced IFNα production in pDCs to a much greater extent than the parental antibody and broadly inhibited proinflammatory cytokine secretion.
BDCA2-ADC induced BDCA2 internalization and strongly inhibited cytokine secretion by purified pDCs. (A) Scheme of the experimental process. (B) pDCs from healthy donor were stained with BDCA2 and CD123 antibodies at the indicated time points after BDCA2-ADC or BIIB059 treatment. Cell-surface BDCA2 levels were measured by FACS. (C) IFNα, TNFα, MIP-1α, IL-6 and IP-10 secretion was measured by Luminex assay at the end of the experiment. mRNA expression was determined by RNA-seq and calculated as the FPKM. ADC: antibody–drug conjugate; BDCA2: blood dendritic cell antigen 2; FACS: fluorescence-activated cell sorting; FPKM: fragments per kilobase per million mapped reads; pDC: plasmacytoid dendritic cell
BDCA2-ADC downregulates IFN response genes
Expression of IFN response gene (IRGs), also termed IFN signature genes, is associated with SLE disease activity. IRG levels were elevated in SLE patients and downregulated by clinical BIIB059 treatment [3, 8]. We assessed whether BDCA2-ADC also modulates IFNα downstream pathway signalling. IFN.9 IRGs and MEDI.21 IRGs were IFN response genes selected for analysis on the basis of previous BIIB059 clinical study [8]. Their expression was measured in BIIB059- or BDCA2-ADC-pretreated pDCs under CpG-A stimulation using RNA-seq. In the current study, the expression of both IFN.9 IRGs and MEDI.21 IRGs was downregulated in the BDCA2-ADC-treated group compared with the BIIB059-treated or IgG control groups (Fig. 4A, B).
BDCA2-ADC suppressed IFN response genes (IRGs) and IFN expression regulatory genes (IERGs) in purified pDCs stimulated by CpG-A in vitro. RNA-seq data from in vitro BIIB059 or BDCA2-ADC treated pDCs were used for the analysis. (A) IFN.9 IRG expression was calculated as the FPKM. (B, C) Cluster for MEDI.21 IRG (B) and IERG expression analysis (C). ADC: antibody–drug conjugate; BDCA2: blood dendritic cell antigen 2; FPKM: fragments per kilobase per million mapped reads; pDC: plasmacytoid dendritic cell
Single-cell sequencing revealed that the expression of IFN expression regulatory genes (IERGs) was induced in pDCs following CpG-A stimulation in vitro [18]. In the present study, the induction of IERGs was blocked by BDCA2-ADC treatment and blocked to a lesser extent by BIIB059 treatment (Fig. 4C). These data confirm that BDCA2-ADC inhibits type I IFN signalling not only by blocking IFN induction but also by targeting IFN response genes.
ADC treatment-induced transcriptional pathway analysis
To further explore the signalling pathways regulated by BDCA2-ADC treatment, we compared the gene expression profiles of CpG-A-stimulated pDCs under BIIB059 or BDCA2-ADC treatments. Assessment of gene enrichment and functional annotation using KEGG pathway analysis showed that autoimmune thyroid disease, cytokine–cytokine receptor interaction pathway and Janus kinase (JAK)–signal transducer and activator of transcription (STAT) signalling pathway genes were significantly downregulated by BIIB059 treatment. Even the most significantly downregulated genes under BDCA2-ADC or BIIB059 treatment were enriched in the cytokine–cytokine receptor interaction pathway, and BDCA2-ADC suppressed cytokine–cytokine receptor interactions, autoimmune thyroid diseases and JAK–STAT signalling pathway genes to a substantially greater degree than BIIB059 treatment (Fig. 5). This result is consistent with the downregulation of proinflammatory cytokine secretion observed following CpG-A-treated pDCs.
KEGG enrichment analysis of differentially expressed genes from RNA-seq data from in vitro BIIB059 or BDCA2-ADC treated pDCs. ADC: antibody–drug conjugate; BDCA2: blood dendritic cell antigen 2; JAK-STAT: Janus kinase–signal transducer and activator of transcription; KEGG: Kyoto Encyclopedia of Genes and Genomes; pDC: plasmacytoid dendritic cell; PI3K: phosphoinositide 3-kinase
The GC payload contributes to the immunosuppressive effect of the ADC
Since the addition of GC to the antibody increased suppression of proinflammatory cytokine secretion compared with that with the antibody alone, we hypothesized that the GC payload regulates genes contributing to the negative regulation. We thus evaluated the expression of GC-regulated genes under BDCA2-ADC or BIIB059 treatment. The anti-inflammatory mechanism of GCs is primarily mediated by the inhibition of nuclear factor κB (NF-κB) in immune cells [19]. However, there was no difference in the expression of NF-κB inhibitor family members between the groups (Fig. 6A). In addition to inhibiting NF-κB, GC inhibits inflammation through transcriptional regulation of transcription factors in multiple pathways. GC-induced leucine zipper protein (GILZ), a GC-responsive molecule, was shown to be upregulated under GC stimulation [20]. GILZ mRNA expression was induced under BDCA2-ADC treatment (Fig. 6B). In addition to GILZ, IL-1 receptor associated kinase 3 (IRAK3) and dual specificity protein phosphatase 1 (DUSP1) are negative regulators induced by GC that engage in crosstalk with TLR signalling [21, 22]. Expression of these proteins was also upregulated only under BDCA2-ADC treatment (Fig. 6B). These findings indicate that the addition of the GC payload in BDCA2-ADC induces the transcription of negative regulators to inhibit pDC function.
The GC payload contributes to the immunosuppressive effect of BDCA2-ADC. (A, B) The expression of NF-κB pathway members (A) and select GC-regulated genes (B) in in vitro-treated pDCs was analysed by RNA-seq. (C) 293T cells were transfected with a GRE-containing luciferase reporter with or without the BDCA2 plasmid. The cells were stimulated with either BIIB059 or BDCA2-ADC. GR activation was measured based on the results of a luciferase assay after 24 h. ADC: antibody–drug conjugate; BDCA2: blood dendritic cell antigen 2; DUSP1: dual specificity protein phosphatase 1; GC: glucocorticoid; GILZ: GC-induced leucine zipper protein; GRE: GC response element; FPKM: fragments per kilobase per million mapped reads; IRAK3: IL-1 receptor associated kinase 3; ISO ADC: human IgG1 Isotype-ADC; NF-κB: nuclear factor κB; NFKBIA–NFKBIE: NF-κB inhibitor α–ε; pDC: plasmacytoid dendritic cell
To further confirm that the BDCA2 targeting-delivered GC payload could induce GR-mediated transcription, we tested BDCA2-ADC-induced GR activation in a HEK293 GRE luciferase reporter system. Ligand-bound GR induced GRE transcriptional activation with luciferase expression in reporter cells. BDCA2-ADC or BIIB059 treatment induced BDCA2 internalization, while GR activation was observed only after BDCA2-ADC treatment, not after BIIB059 treatment (Fig. 6C;Supplementary Table S2, available at Rheumatology online). Therefore, the BDCA2-ADC can enter cells in a BDCA2-dependent manner and activate GR through GC payload release. Taken together, BDCA2-ADC applied a dual suppressive mechanism with both antibody and GC payload to achieve stronger inhibition to pDC function than antibody alone.
Discussion
Even though BIIB059 treatment reduced SLE patient disease severity in clinical trials, more effective therapy would further benefit patients. The aim with BDCA2-ADC is to create the first ADC treatment for SLE patients in the world by generating a pDC specific ADC that is able to produce greater clinical efficacy through synergistic effects of the parental antibody and GC. Compared with BIIB059, BDCA2-ADC showed superior suppression of pDC functions, especially IFNα and proinflammatory cytokine secretion in vitro. BDCA2-ADC further reduced TLR9 agonist-induced IRG and IERG upregulation in purified pDCs. In summary, the BDCA2-ADC shows immune-suppressive activity superior to that of the antibody alone as it combines suppression of the IFN-I expression pathway, IFN regulation pathway and pathways beyond IFN regulation.
The BDCA2-ADC’s regulatory function was confirmed in healthy donor samples, not in samples from an SLE patient, in the current study due to resource limitation. We speculate it would work for samples from the SLE condition. BIIB059-mediated BDCA2 internalization inhibits IFNα production of PBMCs from both healthy people and SLE patients [6]. Circulating pDCs have a similar BDCA2 expression level in both SLE patients and the healthy population, and a decrease in cell number and secretory and antigen-presenting function was observed in the disease condition [23, 24]. BDCA2-ADC shares the antibody-mediated suppressive function with the parental antibody to inhibit BDCA2 positive cells using the same pathway with additional GC payload release. We expect BDCA2-ADC will have a regulatory function in pDCs from SLE patients also, and we would like to test this in a future study.
A similar strategy with anti-TNFα-steroid conjugation has shown promising efficacy in RA therapy [25, 26]. Specifically, the internalization of transmembrane TNFα was shown to transport the TNFα–anti-TNF-steroid complex into early endosomes and lysosomes. In the current study, BDCA2-ADC achieves targeted delivery of GC intracellularly utilizing the high internalization capability of liganded BDCA2. In addition to GC inhibition, BIIB059 could suppress pDC function as a naked antibody. Therefore, the synergistical inhibition of antibody and GC could further downregulate the overactive pDC in SLE patients.
The use of BDCA2-ADC can avoid the side effects of systemic GC exposure in three ways. First, the ADC remains stable in human plasma without releasing the conjugated GC payload into the circulation. Second, BDCA2-ADC binds and enters BDCA2 positive cells exclusively and BDCA2 was dominantly expressed on pDCs (Supplementary Fig. S6, available at Rheumatology online). Finally, the selected GC payload has a rapid clearance pharmacokinetic profile: it is excreted quickly, which reduces its off-target toxicity effects. Therefore, the use of BDCA2-ADC has the potential to reduce side effects and enhance therapeutic efficacy.
BDCA2-ADC, which contains a conjugated GC, is the first immunosuppressive ADC in the world that specifically targets pDCs. In conclusion, our data show that the ADC has a better pDC-suppressive function than the unconjugated antibody, and the targeted delivery of the GC payload to pDCs further reduces systemic GC exposure. These characteristics may contribute to improved efficacy and safety in clinical applications.
Supplementary material
Supplementary material is available at Rheumatology online.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Funding
This study was fully supported by Duality Biologics, Ltd.
Disclosure statement: All authors are or were employees of Duality Biologics, Ltd, which provided support in the form of salaries for authors and funding for research materials, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Acknowledgements
We thank Dr Xiong Li for his valuable suggestions on study design and data interpretation. The design, study conduct and financial support for this research were provided by Duality Biologics, Ltd.
Comments