Potent and Selective Knockdown of 2 by Antisense Oligonucleotides

Nhan Van Tran, Le Tuan Anh Nguyen, Kah Wai Lim and Anh Tuân Phan Downloaded from

ImmunoHorizons 2021, 5 (2) 70-80 doi: https://doi.org/10.4049/immunohorizons.2000108 http://www.immunohorizons.org/content/5/2/70

This information is current as of September 29, 2021. http://www.immunohorizons.org/

Supplementary http://www.immunohorizons.org/content/suppl/2021/02/03/5.2.70.DCSuppl Material emental References This article cites 69 articles, 15 of which you can access for free at: http://www.immunohorizons.org/content/5/2/70.full#ref-list-1

Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: by guest on September 29, 2021 http://www.immunohorizons.org/alerts

ImmunoHorizons is an open access journal published by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 All rights reserved. ISSN 2573-7732. RESEARCH ARTICLE

Adaptive Immunity

Potent and Selective Knockdown of Tyrosine Kinase 2 by Antisense Oligonucleotides

Nhan Van Tran,*,1 Le Tuan Anh Nguyen,*,1 Kah Wai Lim,* and Anh Tuân Phan*,† *School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore 637371, Singapore; and †NTU Institute of Structural Biology, Nanyang Technological University, Singapore 636921, Singapore Downloaded from

ABSTRACT Tyrosine kinase 2 (TYK2) is a member of the JAK family of nonreceptor tyrosine kinase, together with JAK1, JAK2, and JAK3. JAKs are http://www.immunohorizons.org/ important signaling mediators of many proinflammatory and represent compelling pharmacological targets for autoimmune and inflammatory diseases. Pan-acting small-molecule JAK inhibitors were approved for the treatment of rheumatoid arthritis and ulcerative colitis. However, their limited selectivity among JAK members have led to undesirable side effects, driving a search toward specific JAK inhibitors. Recently, TYK2 has emerged as a target of choice for the treatment of autoimmune diseases and severe COVID-19 with an optimum balance between efficacy and safety, based on observations from human genetics studies and clinical outcomes of several agents targeting pathways for which TYK2 plays an essential role. In this article, we address selective targeting of TYK2 from the genetic sequence space through development of antisense oligonucleotides (ASOs) against TYK2 mRNA. Potent ASO candidates were identified from the screening of over 200 ASOs using locked nucleic acid gapmer design. The lead ASOs exhibited potent and selective knockdown of TYK2 mRNA and protein across a panel of model human cell lines in a dose-dependent manner, showing no reduction in the mRNA and protein expression levels of other JAK paralogs. In agreement with by guest on September 29, 2021 the depletion of TYK2 proteins, several TYK2-mediated cytokine signaling pathways, including IFN-a and IL-12, were inhibited upon ASO treatment. Our results established the TYK2 ASOs as investigational tool compound and potential therapeutic agent for the treatment of autoimmune diseases and severe COVID-19. ImmunoHorizons, 2021, 5: 70–80.

INTRODUCTION phosphorylation and undergo subsequent dimerization and trans- location to the nucleus to propagate downstream signaling Tyrosine kinase 2 (TYK2) (1) is a member of the JAK family of (2). Importantly, JAK kinases are critical signal transducers for nonreceptor tyrosine kinase (2), together with JAK1, JAK2, and numerous proinflammatory cytokines (3) and thus represent JAK3. JAK kinases associate with various cytokine receptors at the compelling pharmacological targets for the treatment of autoim- cytoplasmic side, playing a crucial role as signaling mediators. mune and inflammatory diseases (4). As such, intensive efforts Upon extracellular cytokine binding to the receptors, the bound have been directed toward the development of small-molecule JAK proteins phosphorylate each other as well as the receptors, JAK inhibitors (5, 6), with the first approval for the treatment of which then serve as anchor points for transcription factors known rheumatoid arthritis by such agents being granted by the U.S. as STAT. The STAT members are in turn activated by tyrosine Foodand Drug Administration in 2012 (7). However, this andother

Received for publication December 30, 2020. Accepted for publication January 1, 2021. Address correspondence and reprint requests to: Dr. Kah Wai Lim and Prof. Anh Tuân Phan, Nanyang Technological University, 21 Nanyang Link, Singapore 637371, Singapore. E-mail addresses: [email protected] (K.W.L.) and [email protected] (A.T.P.) ORCIDs: 0000-0003-4269-9940 (L.T.A.N.); 0000-0003-3066-1494 (K.W.L.); 0000-0002-4970-3861 (A.T.P.). 1N.V.T. and L.T.A.N. contributed equally. This work was supported by the Singapore National Research Foundation Investigatorship (NRF-NRFI2017-09) and Nanyang Technological University grants to A.T.P. Abbreviations used in this article: ASO, antisense oligonucleotide; LNA, locked nucleic acid; oligo, oligonucleotide; PK, pharmacokinetic; PS, phosphorothioate; PSO, psoriasis; SLE, systemic lupus erythematosus; TYK2, tyrosine kinase 2. The online version of this article contains supplemental material. This article is distributed under the terms of the CC BY-NC-ND 4.0 Unported license. Copyright © 2021 The Authors

70 https://doi.org/10.4049/immunohorizons.2000108

ImmunoHorizons is published by The American Association of Immunologists, Inc. ImmunoHorizons POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS 71

first-generation JAK inhibitors are pan-acting with limited just-released genome-wide association study of critically ill selectivity between different JAK paralogs. Their expansion COVID-19 patients (32) has linked high TYK2 expression to toward additional autoimmune indications has been hampered severe COVID-19, pointing to the selective targeting of TYK2 as by unfavorable safety profiles at higher therapeutic doses, most a potential treatment for these severe cases. notably through inhibition of JAK2, a critical signal transducer RNA therapeutics including antisense oligonucleotides (ASOs) in hematopoiesis (8). As such, present drug development efforts and small-interfering RNAs represent an emerging drug modal- focus on the selective targeting of individual JAK members to ity that have seen recent clinical breakthrough successes (33). achieve robust clinical efficacy while minimizing undesirable These agents operate on a genetic basis through complementary side effects (4, 5). This presents a significant challenge, owing to hybridization to their intended RNA targets (34), thus allowing the high homology among the various JAK paralogs (5). the rapid development of lead compounds against any gene target Recently, selective TYK2 targeting has emerged as an attractive of choice. In this study, we approach selective targeting of TYK2 treatment option for a range of autoimmune diseases. TYK2 plays from the genetic sequence space through the development ofASOs an essential role in the signaling pathways of multiple cytokines against TYK2 mRNA. Potent ASO candidates were identified from (9–11), including type I/III IFN, IL-12, IL-23, and IL-10. Human a screening of over 200 ASOs using locked nucleic acid (35, 36) genome-wide association studies (12–14) revealed that deactivat- (LNA) chemistry. The lead ASOs exhibited potent and selective Downloaded from ing TYK2 variants could provide protection against a broad range knockdown of TYK2 mRNA and protein across a panel of model of autoimmune conditions, including psoriasis (PSO), rheumatoid cell lines in a dose-dependent manner, showing no reduction in the arthritis, systemic lupus erythematosus (SLE), ankylosing spon- mRNA and protein expression levels of other JAK paralogs. dylitis, ulcerative colitis, Crohn’s disease, type 1 diabetes, and Consistent with the depletion of TYK2 proteins, several TYK2- multiple sclerosis. Association of TYK2 with these conditions was dependent cytokine signaling pathways, including IFN-a and http://www.immunohorizons.org/ also supported by findings from other studies (15–20). Further- IL-12, were inhibited after pretreatment with the ASOs. Our more, drug agents targeting IFN-a and IL-12/IL-23 signaling, all of results established the TYK2 ASOs as investigational tool which share TYK2 as a common signaling transducer, have shown compound and potential therapeutic agent for the treatment of clinical efficacy in various autoimmune conditions (21, 22). Among autoimmune diseases and severe COVID-19. the JAK family members, TYK2 would represent a safer target of choice, as JAK1-knockout (perinatal lethal), JAK2-knockout (embryonic lethal), and JAK3-knockout (SCID) mice displayed MATERIALS AND METHODS fatal/critical developmental defects (10, 23), whereas TYK2- knockout mice showed relatively milder deficiencies (9, 23, 24) ASO design, synthesis, and purification (susceptibility to mycobacterial, viral, and fungal infections). Unless otherwise specified, 16-nt fully phosphorothioate (37) (PS)- by guest on September 29, 2021 There are several TYK2-selective inhibitors undergoing active modified ASOs incorporating LNA (35, 36) in a 3-10-3 gapmer (38) clinical development (6, 25). Different strategies were used in the configuration were designed to be complementary to different design of the inhibitors, including active-site inhibition, simulta- segments of TYK2 mRNA (RefSeq identifier NM_003331.4). All neous inhibition of TYK2/JAK1 (26), and allosteric binding of ASOs were synthesized in-house with an ABI 394 DNA/RNA pseudokinase domain (27, 28). Early readouts from the clinical synthesizer on Glen UnySupport (Glen Research) using standard trials are encouraging (29, 30), although longer-term efficacy and phosphoramidite chemistry. LNA phosphoramidites were pur- safety of these agents are yet to be determined (31). Additional chased from either Exiqon or Sigma-Aldrich or synthesized from TYK2-targeting agents that use different modes of action will 39-hydroxyl precursors (Rasayan). Phenylacetyl disulfide (Chem- expand on the drug development repertoire and are thus highly Corporation) was used as the sulfurizing reagent. The ASOs desirable. were cleaved from solid support and deprotected with concen- With the ongoing COVID-19 pandemic around the world, trated aqueous ammonia at 55°C for 16 h and subsequently purified effective treatment options for COVID-19 are urgently needed. In using Poly-Pak II cartridges (Glen Research) according to the severe COVID-19 cases, immune overactivation leading to cytokine manufacturer’s protocol. The ASOs were desalted using Glen Pak release syndrome can be fatal. As such, pan-JAK inhibitors 2.5 desalting column (Glen Research) and dried by lyophilization. including and , both of which inhibit JAK1 The dried ASOs were resuspended with 20 mM potassium and JAK2, were brought forward as potential treatment for severe phosphate (pH 7.0) into 100 mM stock solutions and diluted with COVID-19. However, ruxolitinib failed to show clinical benefitin cell culture medium before use. All ASOs were characterized by a recently reported phase III study (https://www.novartis.com/ JEOL SpiralTOF MALDI-TOF mass spectrometer. news/media-releases/novartis-provides-update-ruxcovid-study- ruxolitinib-hospitalized-patients-covid-19). Although baricitinib Cell culture and reagents received emergency use authorization (https://www.fda.gov/ Jurkat acute T cell leukemia, MOLT-4 acute lymphoblastic news-events/press-announcements/coronavirus-covid-19-update- leukemia, K-562 chronic myelogenous leukemia, A-431 epider- fda-authorizes-drug-combination-treatment-covid-19) by the U.S. moid carcinoma, and HeLa adenocarcinoma cells were purchased Food and Drug Administration for use in combination with from American Type Culture Collection. Karpas-299 non- remdesivir, outcomes from additional clinical trials still await. A Hodgkin large cell lymphoma and HDLM-2 Hodgkin lymphoma

https://doi.org/10.4049/immunohorizons.2000108 72 POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS ImmunoHorizons cells were purchased from Sigma-Aldrich and Deutsche (reference): (forward) 59-CTG GGC TAC ACT GAG CAC C-39, Sammlung von Mikroorganismen und Zellkulturen GmbH (reverse) 59-AAG TGG TCG TTG AGG GCA ATG-39, (probe) 59- (DSMZ), respectively. NK-S1 NK/T cell lymphoma cells were TCTCCTCTGACTTCAACAGCGACACCC-39. All primers and generously provided by Prof. W.J. Chng (Cancer Science probes were purchased from Integrated DNA Technologies. Institute of Singapore). Suspension cells were typically cultured in RPMI 1640 medium (Thermo Fisher Scientific), whereas Western blot and cytokine stimulation adherent cells were cultured in DMEM (Thermo Fisher Cells were harvested and lysed in mammalian protein extraction Scientific), all supplemented with 10–20% FBS (Thermo Fisher reagent (Thermo Fisher Scientific) containing protease and Scientific). NK-S1 cell was cultured in DMEM supplemented phosphatase inhibitor mixture (Sigma-Aldrich) following the with 10% FBS and 10% equine serum (Sigma-Aldrich). All cells manufacturer’s protocol. Protein samples were quantified by BCA were grown in a 5% CO2 humidified incubator at 37°C. protein assay (Thermo Fisher Scientific). Approximately 10–30 mg of total protein was loaded and run on 10% SDS- ASO treatment PAGE with XT MOPS running buffer (Bio-Rad) and subsequently Cells were seeded out on 12- or 24-well plates (Corning) and grown transferred onto a polyvinylidene difluoride membrane (GE for 24 h prior to being treated with ASO. ASOs were diluted with Healthcare). The membrane was blocked by 5% nonfat dry milk Downloaded from fresh culture media and added directly into treatment wells (Bio-Rad) in TBST for 1 h at room temperature, followed by 5-min to attain the desired concentrations, without the use of any wash with TBST three times over. The membrane was then transfection reagents. For time-point measurements, cells were probed with primary Abs against the target proteins (TYK2: harvested at various intervals (8 and 16 h and 1, 2, 3, and 4 d) post- no. 9312; JAK1: no. 3344; JAK2: no. 3230; JAK3: no. 8827; STAT1: ASO treatment. For all other experiments, including RNA/protein no. 9172; pSTAT1-Y701: no. 7649; STAT3: no. 4904; pSTAT3- http://www.immunohorizons.org/ level measurements and cytokine stimulation assays, cells were Y705: no. 9145; pSTAT4-Y693: no. 4134; and GAPDH: no. 5174; harvested 2 or 3 d post-ASO treatment. Cell Signaling Technology) in the recommended buffers, either at room temperature for 1–2 h or at 4°C overnight. After washing for Quantitative reverse transcription PCR three times, the membrane was incubated with HRP-labeled Total RNA from each culture well was extracted by successive secondary Abs (no. 7074; Cell Signaling Technology) at 1:5000–1: addition of 1) TRIzol reagent (Thermo Fisher Scientific) for cell 2000 dilution in the blocking buffer at room temperature for 1 h. lysis, 2) chloroform for phase separation, and 3) isopropanol for The membrane was washed for three times, followed by RNA precipitation. The RNA pellet was washed with ethanol and incubation with ECL substrate (Bio-Rad) according to the allowed to air dry and subsequently resuspended in RNase-free manufacturer’s protocol. Finally, the target bands were visualized water. Quantification of the RNA was performed using Nanodrop and captured on Amersham Imager 680 (GE Healthcare) or Gel by guest on September 29, 2021 2000 (Thermo Fisher Scientific). The RNA was then reverse ChemiDoc system (Bio-Rad). In cases in which multiple JAKs had transcribed with M-MLV reverse transcriptase (Promega) follow- to be imaged, the order of Ab incubation, stripping, and reprobing ing the manufacturer’s protocol. Quantitative PCR of the cDNA were as such: TYK2 followed by JAK2 and JAK3 followed by was performed using either SYBR Green- or probe-based master JAK1. The membrane was stripped using a stripping buffer mix (Bio-Rad) on a Bio-Rad CFX96 real-time PCR detection (100 mM 2-ME, 2% SDS, 62.5 mM Tris-HCl, pH 6.7) at a system. The 22DDthreshold cycle method (39) was employed to deter- temperature of 50°C for 30 min with gentle agitation, followed by mine expression level of the target genes using GAPDH as a six-time washing of 5 min each by TBST, before proceeding to reference. reblock the membrane and reprobe with Abs. For cytokine stimulation assays, the cells lines were seeded for Primer and probe sequences 24 h, followed by ASO treatment for a further 3 d before being The primer pairs used in ASO screening and JAK selectivity subjected to stimulation by the respective cytokines as follows: measurement are as follows: TYK2:(forward)59-GGA GGA GGG IFN-a (Genscript) at 30 ng/ml for 30 min, IL-12 (Sigma-Aldrich) at TTC TAG TGG CA-39, (reverse) 59-ATG TCC CGG AAG TCA 20 ng/ml for 1 h, IL-10 (Genscript) at 100 ng/ml for 1 h, and IL-6 CAG AAG-39; JAK1:(forward)59-CTT TGC CCT GTA TGA CGA (Genscript) at 100 ng/ml for 1 h. GAA C-39,(reverse)59-ACC TCA TCC GGT AGT GGA GC-39; JAK2:(forward)59-ATC CAC CCA ACC ATG TCT TCC-39, ELISA measurement of IFN-g secretion (reverse) 59-ATT CCA TGC CGA TAG GCT CTG-39; JAK3: IFN-g secreted by NK-S1 cell upon IL-12 stimulation was (forward) 59-CCT GAT CGT GGT CCA GAG AG-39, (reverse) 59- measured by Quantikine ELISA kit (R&D Systems) following the GCA GGG ATC TTG TGA AAT GTC AT-39;andGAPDH manufacturer’s protocols. The cells were treated with 10 mM ASO-1 (reference): (forward) 59-CTG GGC TAC ACT GAG CAC C-39, for 3 d before stimulation with 20 ng/ml IL-12 (Sigma-Aldrich) for (reverse) 59-AAG TGG TCG TTG AGG GCA ATG-39.Theprimer 12 h. Following stimulation, 100 mlofcellculturewasdepositedinto pairs and probe sequences used in IC50 measurements are as individual ELISA wells and incubated at room temperature for 2 h follows: TYK2:(forward)59-CAG AGT GCC TGA AGG AGT ATA to allow IFN-g binding to precoated Abs. The wells were rinsed AG-39,(reverse)59-TCT GGC TGG AGT CAC AGT-39, (probe) 59- after removal of media, followed by 2-h incubation with 100 mlof CGT CAG CAGCTC ATA CAG GGT CAC-39 and GAPDH HRP-labeled secondary Abs. Unbound Abs were washed away

https://doi.org/10.4049/immunohorizons.2000108 ImmunoHorizons POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS 73 before HRP substrate was added to the wells for color development RESULTS over 30 min. The reaction was quenched with 0.2 M H2SO4,and OD450 for individual wells was recorded by mQuant microplate Potent knockdown of TYK2 mRNA and protein by TYK2- reader (BioTek). specific ASOs To identify potent TYK2 ASO candidates, we performed screening Cytotoxicity measurements of ASOs on over 200 TYK2-specificASOsfortheirpotentialtoinduce Cytotoxicity of ASOs was measured based on their activation of RNase H-mediated mRNA cleavage in Jurkat leukemia cells, a cellular caspase 3/7. Ten thousand HeLa cells were seeded in 100 T-lymphocyte model cell line. Unless otherwise specified, the ASO ml of DMEM into 96-well plates for 24 h before being transfected configuration comprises 16-nt fully PS (37)-modified gapmer (38) with 100 nM or 1 mMASOsusingLipofectamineRNAiMAX with a middle 10-nt DNA stretch flanked on both ends by 3-nt LNA (Thermo Fisher Scientific) for an additional 24 h. Subsequently, (35, 36) segments (i.e., a wing-gap-wing composition of 3-10-3). 100 ml of Caspase Glo 3/7 Assay (Promega) was added to each well, The high potency of LNA gapmers enable cellular mRNA and the plates were kept in the dark for 30 min. The resulting knockdown to be achieved in vitro simply by free incubation luminescence was recorded by a TECAN Infinite M200 plate with ASOs [otherwise termed as free uptake (40, 41) or gymnosis reader and background subtracted according to the manufac- (42) in the literature] without the use of any delivery agent. Five Downloaded from turer’s protocol. Relative caspase activation was calculated as the selected ASOs (ASO-1–5) demonstrated clear dose-response ratio of luminescence reading between the treated samples and the knockdown of TYK2 mRNA (Fig. 1A) with half maximal IC50 untreated control, of which a high value indicates high cytotoxicity values in the range of ;90–240 nM following 72 h of ASO of the ASO. treatment. Potency of a selected ASO, ASO-1, was further http://www.immunohorizons.org/ by guest on September 29, 2021

FIGURE 1. ASO knockdown of TYK2 mRNA and protein. (A) TYK2 mRNA levels, normalized against GAPDH, of Jurkat cell line after incubation with varying doses of ASOs for 72 h. n = 3, from three independent experiments. IC50 values of the ASOs are shown in the legend. (B) TYK2 mRNA levels, normalized against GAPDH, of different cell lines (Jurkat, K-562, Karpas-299, MOLT-4, A-431, and HDLM-2) after incubation with varying doses of ASO-1 for 72 h. n = 3 for Jurkat, K-562, Karpas-

299, and MOLT-4; n = 2 for A-431 and HDLM-2, from at least two independent experiments. The IC50 values across the cell lines are shown in the legend. (C) TYK2 mRNA levels, normalized against GAPDH, of different cell lines (Jurkat, Karpas-299, and MOLT-4) after incubation with 0.5 mM ASO-1 or the control oligonucleotide (oligo) ASO-ctrl across varying time points from 8 h to 4 d. n = 2, from two independent experiments. Karpas- 299 showed a rapid depletion of TYK2 mRNA upon ASO-1 treatment, whereas MOLT-4 showed a slower rate of depletion. (D) Western blot showing the dose-dependent knockdown of TYK2 protein in Jurkat cell line after incubation with varying doses of ASO-1 for 72 h. n = 3, from three independent experiments. TYK2 protein levels were not affected by the control oligo ASO-ctrl.

https://doi.org/10.4049/immunohorizons.2000108 74 POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS ImmunoHorizons

validating the on-target effect of the TYK2-specificASOs.The dose-dependent experiments were repeated on various suspen- sion cell lines and showed the near-complete depletion of TYK2 across these cell lines upon ASO-1 treatment, even at the lower dosage of 100 nM (Fig. 2). Time-course measurement of protein levels following incubation of Jurkat cell with 1 mMofASO-1 (Supplemental Fig. 1A) showed protein knockdown as soon as 24 h posttreatment, with progressively more knockdown observed at 48 and 72 h posttreatment.

Selective TYK2 knockdown by ASO-1 with no perturbation on other JAK paralogs To be an effective therapeutic option, the ASOs should be specific for TYK2 while selective against all other JAK family members. Toward that end, JAK1, JAK2, and JAK3 protein levels in various Downloaded from cell lines were not reduced by treatment with ASO-1,evenat dosages whereby the TYK2 protein was completely depleted (Fig. 2, Supplemental Fig. 1A). The TYK2 specificity of ASO-1 was also validated in the context of mRNA, for which only TYK2 mRNA knockdown was observed, whereas JAK1, JAK2, and JAK3 mRNA http://www.immunohorizons.org/ levels remained largely unchanged when 1 mM ASO-1 was incubated across different cell lines (Fig. 3). High TYK2 selectivity was similarly observed for ASO2–5, as JAK1, JAK2, and JAK3 mRNA levels in Jurkat cell remained largely constant following ASO treatment (Supplemental Fig. 1B). FIGURE 2. Selective knockdown of TYK2 protein by ASO-1. (A) Western blot showing the protein levels of all four JAKs in Jurkat and ASO-1 inhibition of TYK2-mediated cytokine Karpas-299 cell lines after incubation with varying doses of ASO-1 for signaling pathways 72 h. (B) Western blot showing the protein levels of JAK1, JAK3, and Next,weevaluatedtheeffects of TYK2 knockdown on various TYK2 in MOLT-4 and K-562 cell lines after incubation with varying TYK2-associated cytokine signaling pathways (10, 11) (i.e., IFN-a, doses of ASO-1 for 72 h. Experiment was performed on four in- by guest on September 29, 2021 IL-12, IL-10, and IL-6) across different cell lines in vitro. For each dependent cell lines (n = 4). Potential slight increase of JAK1 and JAK2 signaling pathway, the cell lines were pretreated with ASO-1 for was observed in Jurkat, Karpas-299, and K-562 upon TYK2 knockdown. 3 d prior to being subjected to cytokine stimulation. Phosphory- This could have resulted from technical issues from stripping and lation status of the relevant STAT members was then quantified by reprobing (see Materials and Methods) or actual upregulation due to possible compensatory mechanisms following TYK2 depletion. measured across different cell lines, including suspension (Karpas-299 non-Hodgkin lymphoma, MOLT-4 acute lympho- blastic leukemia, K-562 chronic myelogenous leukemia, and HDLM-2 Hodgkin lymphoma) and adherent (A-431 epidermoid carcinoma) cells, giving rise to IC50 values ranging from ;13 to 155 nM (Fig. 1B). Time-course measurement of RNA levels following ASO-1 treatment was performed across different cell lines (Fig. 1C); maximal knockdown was generally observed after 24–48 h and maintained thereafter for at least 4 d posttreatment with no replacement of media and ASO. In the same experiment, a control oligonucleotide (oligo) comprising the same LNA gapmer config- uration (ASO-ctrl) did not lead to reduction in TYK2 mRNA level, thereby confirming the on-target effect of the TYK2-specificASOs. TYK2 protein levels in Jurkat cell were measured after FIGURE 3. Selective knockdown of TYK2 mRNA by ASO-1. treatment with varying concentrations of ASO-1 for 72 h. Near- TYK2 mRNA levels, normalized against GAPDH, of different cell lines complete depletion of the TYK2 protein was observed at an ASO (Jurkat, K-562, Karpas-299, MOLT-4, A-431, and HDLM-2) after in- concentration of 1 mM (Fig. 1D). At the same dosage, the control cubation with 1 mM ASO-1 for 72 h. n = 2, from two independent oligo ASO-ctrl did not affect the level of TYK2 protein, further experiments. mRNA levels of JAK1/2/3 were largely not affected.

https://doi.org/10.4049/immunohorizons.2000108 ImmunoHorizons POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS 75

findings from human TYK2 deficiency (9) and TYK2-knockout mice (47, 48), albeit in the mice knockouts, TYK2 protein was found to play a more-limited role in type I IFN. IL-12 signaling occurs through its binding to the IL-12R, a heterodimeric complex consisting of IL-12Rb1 and IL-12Rb2(49). Upon IL-12 binding to the receptor, the bound TYK2 and JAK2 are activated and phosphorylate STAT1/3/4 (50, 51), thereby leading to downstream gene regulation events, among which include IFN-ɣ production. After ASO treatment for 3 d, NK-S1 cells were stimulated with 20 ng/ml IL-12 for 1 h. Western blot of the cell lysate showed that phosphorylated STAT4 levels following IL-12 stimulation were reduced in cells that were pretreated with ASO-1, as compared with the controls (Fig. 5). Effects of TYK2 knockdown were also evident from IFN-ɣ quantification through ELISA measurement, wherein IFN-ɣ secretion (stimulated over a duration Downloaded from of 12 h with 20 ng/ml IL-12) was drastically reduced in NK-S1 cells pretreated with ASO-1,ascomparedwithnontreated controls and cells pretreated with ASO-ctrl (Supplemental Fig. 2). Our results pointed to the essential role of TYK2 protein in mediating IL-12 signal transduction (9, 47, 48, 52), as has been reported previously http://www.immunohorizons.org/ fromstudiesinhumans(9,52)andmice(47,48). IL-10 signaling occurs through a heterotetrameric receptor complex comprising two each of IL-10R1 and IL-10R2 (53), which is mediated by TYK2 and JAK1, subsequently leading to STAT3 phosphorylation (54). Jurkat and K-562 cells were stimulatedwith100ng/mlIL-10for1hfollowingpre- treatment with ASO-1 for 3 d. Western blot of the cell lysate showed that STAT3 phosphorylation was inhibited in a dose- dependent manner (Fig. 6), in agreement with TYK2 protein depletion, whereas normal STAT3 protein levels were not by guest on September 29, 2021 affected. Our results suggested at least a similar level of FIGURE 4. Dose-dependent inhibition of TYK2-mediated IFN-a involvement from TYK2 in IL-10 signaling as compared with signaling by ASO-1. JAK1, which contrast with results from previous studies (A) Western blot showing the protein levels of TYK2, STAT1/3, and reporting a dominant role of JAK1 over TYK2 in IL-10 signal phosphorylated STAT1/3 in Jurkat cell after IFN-a stimulation for transduction (48, 55, 56). 30 min following incubation with varying doses of ASO-1 for 72 h. (B) IL-6 signaling is mediated through its receptor complex (57), Western blot showing the protein levels of STAT1 and phosphorylated which is composed of gp130 and IL-6Ra.UponIL-6binding,the STAT1 in Karpas-299 cell, after IFN-a stimulation for 30 min following associated JAK1, JAK2, and TYK2 are activated, thereby leading incubation with varying doses of ASO-1 for 72 h. Experiment was per- to phosphorylation of STAT3 protein (9, 10). In this study, we formed on two independent cell lines (n = 2).

Western blot of the cell lysate to determine the potential inhibition of pathway stimulation by ASO-1. TypeIIFNs(IFN-a/IFN-b) signal through binding to the IFN- a/b receptor (IFNAR), which is mediated by TYK2 and JAK1 (43), and eventually lead to the activation of several STAT members, including STAT1/3 (44–46). Following ASO-1 treatment for 3 d, Jurkat and Karpas-299 cells were stimulated with 30 ng/ml IFN-a for 30 min. Western blot of cell lysate showed that STAT1 and STAT3 phosphorylation within both cell lines was inhibited by FIGURE 5. Inhibition of TYK2-mediated IL-12 signaling by ASO-1. ASO-1 in a dose-dependent manner (Fig. 4), whereas the total Western blot showing the protein levels of phosphorylated STAT4 in amounts of nonphosphorylated STAT1 and STAT3 remained NK-S1 cell, with or without IL-12 stimulation for 1 h following in- unchanged. Our results indicated the dependence of type I IFN cubation with 10 mM ASO-1 or the control oligo ASO-ctrl for 72 h. n =2, signaling on TYK2 protein. These are consistent with previous from two independent experiments.

https://doi.org/10.4049/immunohorizons.2000108 76 POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS ImmunoHorizons

Fig. 4A) while still retaining their TYK2 mRNA knockdown activity (Supplemental Fig. 4B).

DISCUSSION

JAK proteins, acting as critical signaling mediators of many proinflammatory cytokines, represent attractive drug targets for the treatment of autoimmune and inflammatory diseases. Mount- ing evidence from human genetics studies and clinical findings of several drug agents against TYK2-dependent cytokine signal- ing pathways (e.g., IFN-a and IL-12/IL-23), combined with the less-severe phenotype of TYK2-knockouts as compared with the other JAK paralogs, indicate TYK2 as an excellent target for the treatment of autoimmune diseases with an optimal balance Downloaded from between efficacy and safety. Additionally, high TYK2 expression has been linked to severe COVID-19 from a recent genome-wide association study of critically ill COVID-19 patients (32), pointing to its selective targeting as a potential treatment for these severe cases. Among small-molecule inhibitors of TYK2 that have been http://www.immunohorizons.org/ advanced to clinical testing, nearly all were designed to block or FIGURE 6. Dose-dependent inhibition of TYK2-mediated IL-10 compete with ATP binding to the catalytic domain, also known as signaling by ASO-1. the Janus Homology 1 (JH1) domain, where selectivity is especially (A) Western blot showing the protein levels of STAT3 and phosphory- challenging to achieve, given the high homology among the JAK lated STAT3 in Jurkat cell line, after IL-10 stimulation for 1 h following paralogs. A small-molecule inhibitor using an alternative approach incubation with varying doses of ASO-1 for 72 h. (B) Western blot of allosteric inhibition through selective binding to the pseudoki- showing the protein levels of STAT3 and phosphorylated STAT3 in nase (JH2) domain has shown a remarkable selectivity for TYK2 K-562 cell line, after IL-10 stimulation for 1 h following incubation with against all other JAK paralogs, and early clinical results of the varying doses of ASO-1 for 72 h. Experiment was performed on two inhibitor against PSO (29) look promising. Despite its high independent cell lines (n = 2). selectivity for TYK2, the inhibitor still binds to JAK1 (JH2 by guest on September 29, 2021 domain) and BMPR2 kinase with high affinity (27). Longer-term efficacy and safety of this and other TYK2 inhibitors remain to be measured the levels of phosphorylated STAT3 in A-431 cells after determined, and additional TYK2-targeting agents employing IL-6 (100 ng/ml) stimulation for 1 h following pretreatment with other modes of action will build on the drug development toolbox ASO-1 for 3 d. In this case, TYK2 depletion with ASO-1 did not lead toaddresstherangeofautoimmunediseases.Withtheongoing to any visible changes on both phosphorylated and normal STAT3 pandemic around the world, effective treatment options for severe (Supplemental Fig. 3), even at the highest dose tested (2 mM), COVID-19 are urgently needed. suggesting minimal impact of TYK2 knockdown on IL-6 signaling In this study, we have identified potent ASO compounds through STAT3. Our results were consistent with multiple reports against TYK2 mRNA. In contrast to small-molecule inhibitors that (48, 52, 55, 56, 58) showing a dominant role of JAK1 over JAK2 and act through direct binding to the protein target, ASOs operate on TYK2 in mediating IL-6 signal transduction. the basis of hybridization to the RNA target through complemen- We further measured the cytotoxicity of the ASOs through tary base pairing and hence achieve selectivity from a sequence their activation of cellular caspase 3/7 (59, 60) (Supplemental Fig. standpoint. In the case of TYK2 ASO gapmers, RNase H are 4A), which was shown to be a good predictor of their in vivo recruited upon ASO hybridization to the sequence target, thereby toxicity. ASO-1 exhibited an excellent safety profile, eliciting a low- resulting in specific cleavage and subsequent degradation of TYK2 caspase activation close to basal level in HeLa adenocarcinoma mRNA and eventually depletion of TYK2 protein. Fully PS- cells even at a high dose of 1 mM with transfection. In contrast, modified LNA gapmers were used in the design of the TYK2 ASOs,

ASO-3 and ASO-4 led to high relative caspase activation levels (up which exhibit IC50 values matching those of recent ASO to ;6–10 times that of untreated control) at the same or lower dose candidates with a similar chemical configuration that have been (100 nM) with transfection. It has been reported that cytotoxicity advanced to clinical testing. Such fully chemically modified of PS-ASO gapmers could be reduced by chemical modifications ASOs are resistant against endo- and exonucleases and remain (60), specifically through incorporation of 2’-O-methyl (2’-OMe) stable in cells and in vivo for days, even weeks (34, 61). We have at gap position 2 (i.e., position 5 of 3-10-3 gapmers). Accordingly, also shown that ASO-mediated TYK2 mRNA knockdown can be corresponding modifications to ASO-1, ASO-3,andASO-4 reduced sustained for at least 4 d posttreatment. Sequence alignments of caspase activation in HeLa cells back to basal levels (Supplemental ASO-1–5 against the transcript database found

https://doi.org/10.4049/immunohorizons.2000108 ImmunoHorizons POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS 77 no off-target binding sites with significant overlap for the TYK2 in IFN-a signaling, supporting further development of the majority of these sequences, hence supporting their specificity TYK2 ASOs toward treatment/application on autoimmune for TYK2 mRNA. Experimentally, we have demonstrated their diseases such as SLE in which elevated IFN-a signaling was selectivity against the other JAK paralogs, as the mRNA and identified as a major driver of pathology (65). On a similar note, protein levels of JAK1/2/3 were not reduced by TYK2 ASO we showed that ASO-mediated TYK2 knockdown led to treatment. reduction of IL-12 signaling as measured by pSTAT4 levels. Thankstotheirlowmolecularmassaswellaschemicaldesign, This observation was consistent with TYK2 being an important most small-molecule therapeutics typically exhibit excellent signaling mediator of the IL-12/IL-23 pathway, for which an cellular uptake profile across all cell types. ASOs, in contrast, mAb against the common p40 subunit of IL-12/IL-23 was display a drastically distinct cellular uptake and pharmacoki- approved for the treatment of PSO, psoriatic arthritis, Crohn’s netics (PK) profile (61). The differential knockdown efficacies disease, and ulcerative colitis (66). We have also shown that IL- of ASO-1 in various cell lines (IC50 values ranging from ;13 to 10 signaling was inhibited upon TYK2 depletion through ASO 155 nM) were consistent with previous observation of variable treatment, as measured by pSTAT3 levels. Unlike IFN-a and ASO uptake efficiency across different cell lines (62), even IL-12/IL-23, IL-10 is generally regarded as an anti-inflammatory among those originating from the same tissue type. Generally, cytokine (3), although its overproduction has been raised as a Downloaded from this would be construed as a disadvantage, as one would predisposing factor for SLE (67). In this case, our results showed normally hope for maximal cellular uptake of a drug compound. that TYK2 exerts an essential role as signaling mediator for However, there could be circumstances under which selective IL-10, which contrasts with previous reports on the dominant uptake by cellular subsets is desired or cellular uptake into certain role of JAK1 over TYK2 in IL-10 signal transduction (48, 55, 56). cell types is to be avoided. In these cases, selective distribution and In contrast, TYK2 depletion showed no effects on IL-6 signaling http://www.immunohorizons.org/ uptake of ASO into different cellular subtypes could work to our through pSTAT3, suggesting a limited role for TYK2 in signal advantage. transduction of the IL-6 pathway. This was consistent with To achieve targeted delivery of ASOs in vivo, intensive multiple reports of the dominant role of JAK1 in mediating IL-6 efforts have been driven toward direct chemical conjugation of signal transduction (48, 52, 55, 56, 58). We further measured the ASO with cell type– and organ-selective ligands (63). Tremen- cytotoxicity of the TYK2 ASOs through cellular caspase 3/7 dous success has been observed for their targeted delivery to activation (59, 60), which was shown to be a good predictor of their liver hepatocytes upon conjugation with multiple copies of the in vivo toxicity. ASO-1 exhibited an excellent safety profile, whereas simple carbohydrate N-acetylgalactosamine (64) (GalNAc). cytotoxicity of the other ASOs could be largely eliminated through Conjugation of TYK2 ASO with ligands that are specificfor judicious modifications (60). immune cell/tissue subtypes of interest thus holds great promise The high knockdown efficacy and specificity of the TYK2 ASOs, by guest on September 29, 2021 for its targeted delivery to address particular autoimmune coupled with their ease of application, make them excellent tool diseases. In principle, conjugation can be applied onto compoundsforfurtherinvestigationsonfunctionalrolesofTYK2in small molecules for targeted delivery as well but with certain immunological settings/autoimmune diseases. Aside from autoim- caveats. First, structure of a small molecule has already been mune diseases, TYK2 was also implicated in the pathology of various chemically optimized, hence complicating the choices for cancers, including T cell acute lymphoblastic leukemia (68), conjugation chemistry and site of conjugation, and this has to be anaplastic large cell lymphoma (69), and nerve sheath tumors (70). considered on a case-by-case basis. Compare this against ASO, The TYK2 ASOs presented in this work could be applied to further which is modular by nature and thus would allow the understand the involvement of TYK2 in the disease pathology of incorporation of diverse chemical functionalities more readily these cancers and serve as a basis for therapeutic development. across different ASO sequences. Second, as small molecules are In conclusion, we have identified a series of potent TYK2 ASOs optimizedforPKandpharmacodynamics,theymightnotbe that are selective against all other JAK paralogs. The ASOs induce compatible with the chemical modification and conjugation RNase H-mediated RNA degradation upon their binding of TYK2 method requirements of the ligand. Furthermore, conjugation mRNA, thereby achieving selectivity from the sequence space. with ligand could tremendously alter the PK/pharmacodynamics This is completely different from small-molecule JAK inhibitors profile of the original small molecule. that act by direct binding to the catalytic/allosteric site of TYK2 Beyond TYK2 mRNA and protein level measurements, we protein. ASO-mediated knockdown of TYK2 was examined both at further examined the functional effects of ASO-mediated TYK2 the mRNA and protein levels as well as functional measurement of knockdown on various TYK2-associated cytokine signaling various cytokine signaling pathways that are associated with pathways (i.e., IFN-a, IL-12, IL-10, and IL-6). In agreement with TYK2. In particular, we have shown that ASO-mediated depletion TYK2 protein depletion by ASO-mediated mRNA degradation, of TYK2 protein led to reduced activation of IFN-a and IL-12 we observed dose-dependent reduction of IFN-a signaling with pathways, for which aberrant signaling has been implicated in TYK2 ASO treatment, as measured by pSTAT1/3 levels. Abs the pathogenesis of various autoimmune diseases. These results against IFN-a and IFNAR have been evaluated for the treatment established the TYK2 ASOs as investigational tool compound and of SLE, with the latter showing clinical efficacy in a phase III potential therapeutic agent for the treatment of autoimmune trial (21). Our results were consistent with the essential role of diseases and severe COVID-19.

https://doi.org/10.4049/immunohorizons.2000108 78 POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS ImmunoHorizons

DISCLOSURES 15. Marroqui, L., R. S. Dos Santos, T. Fløyel, F. A. Grieco, I. Santin, A. Op de Beeck, L. Marselli, P. Marchetti, F. Pociot, and D. L. Eizirik. 2015. ApatentonTYK2 ASOs has been filed by Nanyang Technological TYK2, a candidate gene for type 1 diabetes, modulates apoptosis and the innate immune response in human pancreatic b-cells. Diabetes 64: University with the authors as inventors. 3808–3817. 16. Inshaw, J. R. J., A. J. Cutler, O. S. Burren, M. I. Stefana, and J. A. Todd. 2018. Approaches and advances in the genetic causes of auto- ACKNOWLEDGMENTS immune disease and their implications. [Published erratum appears in 2019 Nat. Immunol. 20: 375; Published erratum appears in 2020 Nat. We thank Wilson Ong Wei Shiuan and Qi Shi for synthesis and purification Immunol. 21: 695.] Nat. Immunol. 19: 674–684. of ASOs and Fernaldo Richtia Winnerdy for help in plotting IC50 and time- 17. Ishizaki, M., R. Muromoto, T. Akimoto, Y. Ohshiro, M. Takahashi, course measurements of the ASOs. Y. Sekine, H. Maeda, K. Shimoda, K. Oritani, and T. Matsuda. 2011. Tyk2 deficiency protects joints against destruction in anti-type II col- lagen antibody-induced arthritis in mice. Int. Immunol. 23: 575–582. REFERENCES 18. Diogo, D., L. Bastarache, K. P. Liao, R. R. Graham, R. S. Fulton, J. D. Greenberg, S. Eyre, J. Bowes, J. Cui, A. Lee, et al. 2015. TYK2 protein-

1. Firmbach-Kraft, I., M. Byers, T. Shows, R. Dalla-Favera, and J. J. coding variants protect against rheumatoid arthritis and autoimmu- Downloaded from ff Krolewski. 1990. tyk2, prototype of a novel class of non-receptor ty- nity, with no evidence of major pleiotropic e ects on non-autoimmune rosine kinase genes. Oncogene 5: 1329–1336. complex traits. PLoS One 10: e0122271. 2. O’Shea, J. J., D. M. Schwartz, A. V. Villarino, M. Gadina, I. B. 19. Sigurdsson, S., G. Nordmark, H. H. Goring,¨ K. Lindroos, A. C. Wiman, McInnes, and A. Laurence. 2015. The JAK-STAT pathway: impact on G. Sturfelt, A. Jonsen,¨ S. Rantapa¨a-Dahlqvist,¨ B. Moller,¨ J. Kere, et al. human disease and therapeutic intervention. Annu. Rev. Med. 66: 2005. Polymorphisms in the tyrosine kinase 2 and regu- 311–328. latory factor 5 genes are associated with systemic lupus eryth-

ematosus. Am. J. Hum. Genet. 76: 528–537. http://www.immunohorizons.org/ 3. Schwartz, D. M., M. Bonelli, M. Gadina, and J. J. O’Shea. 2016. Type 20. Cunninghame Graham, D. S., D. L. Morris, T. R. Bhangale, L. A. I/II cytokines, JAKs, and new strategies for treating autoimmune Criswell, A. C. Syvanen,¨ L. Ronnblom,¨ T. W. Behrens, R. R. Graham, diseases. Nat. Rev. Rheumatol. 12: 25–36. and T. J. Vyse. 2011. Association of NCF2, IKZF1, IRF8, IFIH1, and 4. Schwartz, D. M., Y. Kanno, A. Villarino, M. Ward, M. Gadina, and J. J. TYK2 with systemic lupus erythematosus. PLoS Genet. 7: e1002341. O’Shea. 2017. JAK inhibition as a therapeutic strategy for immune and 21. Morand, E. F., R. Furie, Y. Tanaka, I. N. Bruce, A. D. Askanase, inflammatory diseases. [Published erratum appears in 2018 Nat. Rev. C. Richez, S. C. Bae, P. Z. Brohawn, L. Pineda, A. Berglind, and Drug Discov. 17: 78.] Nat. Rev. Drug Discov. 16: 843–862. R. Tummala, TULIP-2 Trial Investigators. 2020. Trial of 5. Clark, J. D., M. E. Flanagan, and J. B. Telliez. 2014. Discovery and in active systemic lupus erythematosus. N. Engl. J. Med. 382: 211–221. development of (JAK) inhibitors for inflammatory dis- 22. Savage, L. J., M. Wittmann, D. McGonagle, and P. S. Helliwell. 2015. eases. J. Med. Chem. 57: 5023–5038. Ustekinumab in the treatment of psoriasis and psoriatic arthritis. 6. Bryan, M. C., and N. S. Rajapaksa. 2018. Kinase inhibitors for the Rheumatol. Ther. 2: 1–16. treatment of immunological disorders: recent advances. J. Med. Chem. 23. Kiu, H., and S. E. Nicholson. 2012. Biology and significance of the by guest on September 29, 2021 61: 9030–9058. JAK/STAT signalling pathways. Growth Factors 30: 88–106. 7. Traynor, K. 2012. FDA approves for rheumatoid arthritis. 24. Watford, W. T., and J. J. O’Shea. 2006. Human tyk2 kinase deficiency: Am. J. Health Syst. Pharm. 69: 2120. another primary immunodeficiency syndrome. Immunity 25: 695–697. ff 8. Neubauer, H., A. Cumano, M. Muller,¨ H. Wu, U. Hu stadt, and 25. He, X., X. Chen, H. Zhang, T. Xie, and X. Y. Ye. 2019. Selective Tyk2 ff fi fi K. Pfe er. 1998. Jak2 de ciency de nes an essential developmental inhibitors as potential therapeutic agents: a patent review (2015- fi – checkpoint in de nitive hematopoiesis. Cell 93: 397 409. 2018). Expert Opin. Ther. Pat. 29: 137–149. 9. Minegishi, Y., M. Saito, T. Morio, K. Watanabe, K. Agematsu, 26. Fensome, A., C. M. Ambler, E. Arnold, M. E. Banker, M. F. Brown, S. Tsuchiya, H. Takada, T. Hara, N. Kawamura, T. Ariga, et al. 2006. J. Chrencik, J. D. Clark, M. E. Dowty, I. V. Efremov, A. Flick, et al. fi Human tyrosine kinase 2 de ciency reveals its requisite roles in 2018. Dual inhibition of TYK2 and JAK1 for the treatment of autoimmune multiple cytokine signals involved in innate and acquired immunity. diseases: discovery of (( S)-2,2-Difluorocyclopropyl)((1 R,5 S)-3-(2-((1- – Immunity 25: 745 755. methyl-1 H-pyrazol-4-yl)amino)pyrimidin-4-yl)-3,8-diazabicyclo[3.2.1] 10. Ghoreschi, K., A. Laurence, and J. J. O’Shea. 2009. Janus kinases in octan-8-yl)methanone (PF-06700841). J. Med. Chem. 61: 8597–8612. – immune cell signaling. Immunol. Rev. 228: 273 287. 27. Wrobleski, S. T., R. Moslin, S. Lin, Y. Zhang, S. Spergel, J. Kempson, 11. Strobl, B., D. Stoiber, V. Sexl, and M. Mueller. 2011. Tyrosine kinase 2 J. S. Tokarski, J. Strnad, A. Zupa-Fernandez, L. Cheng, et al. 2019. (TYK2) in cytokine signalling and host immunity. Front. Biosci. 16: Highly selective inhibition of tyrosine kinase 2 (TYK2) for the 3214–3232. treatment of autoimmune diseases: discovery of the allosteric in- 12. Ban, M., A. Goris, A. R. Lorentzen, A. Baker, T. Mihalova, G. Ingram, hibitor BMS-986165. J. Med. Chem. 62: 8973–8995. D. R. Booth, R. N. Heard, G. J. Stewart, E. Bogaert, et al; Wellcome 28. Burke, J. R., L. Cheng, K. M. Gillooly, J. Strnad, A. Zupa-Fernandez, Trust Case-Control Consortium (WTCCC). 2009. Replication analysis I. M. Catlett, Y. Zhang, E. M. Heimrich, K. W. McIntyre, M. D. identifies TYK2 as a multiple sclerosis susceptibility factor. Eur. J. Cunningham, et al. 2019. Autoimmune pathways in mice and humans Hum. Genet. 17: 1309–1313. are blocked by pharmacological stabilization of the TYK2 pseudoki- 13. Tao, J. H., Y. F. Zou, X. L. Feng, J. Li, F. Wang, F. M. Pan, and D. Q. nase domain. Sci. Transl. Med. 11: eaaw1736. Ye. 2011. Meta-analysis of TYK2 gene polymorphisms association 29. Papp, K., K. Gordon, D. Thaçi, A. Morita, M. Gooderham, P. Foley, with susceptibility to autoimmune and inflammatory diseases. Mol. I. G. Girgis, S. Kundu, and S. Banerjee. 2018. Phase 2 trial of selective Biol. Rep. 38: 4663–4672. tyrosine kinase 2 inhibition in psoriasis. N. Engl. J. Med. 379: 14. Dendrou, C. A., A. Cortes, L. Shipman, H. G. Evans, K. E. Attfield, 1313–1321. L. Jostins, T. Barber, G. Kaur, S. B. Kuttikkatte, O. A. Leach, et al. 2016. 30. Banfield, C., M. Scaramozza, W. Zhang, E. Kieras, K. M. Page, Resolving TYK2 genotype-to-phenotype differences in autoim- A. Fensome, M. Vincent, M. E. Dowty, K. Goteti, P. J. Winkle, and munity. Sci. Transl. Med. 8: 363ra149. E. Peeva. 2018. The safety, tolerability, pharmacokinetics, and

https://doi.org/10.4049/immunohorizons.2000108 ImmunoHorizons POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS 79

pharmacodynamics of a TYK2/JAK1 inhibitor (PF-06700841) in Partial impairment of cytokine responses in Tyk2-deficient mice. healthy subjects and patients with plaque psoriasis. J. Clin. Pharma- Immunity 13: 549–560. col. 58: 434–447. 48. Shimoda, K., K. Kato, K. Aoki, T. Matsuda, A. Miyamoto, M. Shibamori, 31. Winthrop, K. L. 2017. The emerging safety profile of JAK inhibitors M. Yamashita, A. Numata, K. Takase, S. Kobayashi, et al. 2000. Tyk2 in rheumatic disease. [Published erratum appears in 2017 Nat. Rev. plays a restricted role in IFN a signaling, although it is required for IL- Rheumatol. 13: 320.] Nat. Rev. Rheumatol. 13: 234–243. 12-mediated T cell function. Immunity 13: 561–571. 32. Pairo-Castineira, E., S. Clohisey, L. Klaric, A. D. Bretherick, K. Rawlik, 49. Collison, L. W., and D. A. Vignali. 2008. Interleukin-35: odd one out or D. Pasko, S. Walker, N. Parkinson, M. H. Fourman, C. D. Russell, et al. part of the family? Immunol. Rev. 226: 248–262. 2020. Genetic mechanisms of critical illness in Covid-19. Nature. DOI: 50. Bacon, C. M., E. F. Petricoin III., J. R. Ortaldo, R. C. Rees, A. C. https://doi.org/10.1038/s41586-020-03065-y. Larner, J. A. Johnston, and J. J. O’Shea. 1995. induces 33. Levin, A. A. 2019. Treating disease at the RNA level with oligonu- tyrosine phosphorylation and activation of STAT4 in human lym- cleotides. N. Engl. J. Med. 380: 57–70. phocytes. Proc. Natl. Acad. Sci. USA 92: 7307–7311. 34. Bennett, C. F., B. F. Baker, N. Pham, E. Swayze, and R. S. Geary. 2017. 51. Jacobson, N. G., S. J. Szabo, R. M. Weber-Nordt, Z. Zhong, R. D. Pharmacology of antisense drugs. Annu. Rev. Pharmacol. Toxicol. 57: Schreiber, J. E. Darnell Jr., and K. M. Murphy. 1995. Interleukin 12 81–105. signaling in T helper type 1 (Th1) cells involves tyrosine phosphory- 35. Obika, S., D. Nanbu, Y. Hari, K. I. Morio, Y. In, T. Ishida, and lation of signal transducer and activator of transcription (Stat)3 and – T. Imanishi. 1997. Synthesis of 29-O,49-C-methyleneuridine and -cytidine. Stat4. J. Exp. Med. 181: 1755 1762. Downloaded from

Novel bicyclic nucleosides having a fixed C39 -endo sugar puckering. 52. Kreins, A. Y., M. J. Ciancanelli, S. Okada, X. F. Kong, N. Ramı´rez- Tetrahedron Lett. 38: 8735–8738. Alejo, S. S. Kilic, J. El Baghdadi, S. Nonoyama, S. A. Mahdaviani, 36. Koshkin, A. A., S. K. Singh, P. Nielsen, V. K. Rajwanshi, R. Kumar, F. Ailal, et al. 2015. Human TYK2 deficiency: mycobacterial and viral M. Meldgaard, C. E. Olsen, and J. Wengel. 1998. LNA (Locked Nucleic infections without hyper-IgE syndrome. J. Exp. Med. 212: 1641–1662. Acids): synthesis of the adenine, cytosine, guanine, 5-methylcytosine, 53. Pestka, S., C. D. Krause, D. Sarkar, M. R. Walter, Y. Shi, and P. B. thymine and uracil bicyclonucleoside monomers, oligomerisation, Fisher. 2004. Interleukin-10 and related cytokines and receptors.

and unprecedented nucleic acid recognition. Tetrahedron Lett. 54: Annu. Rev. Immunol. 22: 929–979. http://www.immunohorizons.org/ 3607–3630. 54. Murray, P. J. 2006. Understanding and exploiting the endogenous 37. Eckstein, F. 2014. Phosphorothioates, essential components of ther- interleukin-10/STAT3-mediated anti-inflammatory response. Curr. apeutic oligonucleotides. Nucleic Acid Ther. 24: 374–387. Opin. Pharmacol. 6: 379–386. 38. Monia, B. P., E. A. Lesnik, C. Gonzalez, W. F. Lima, D. McGee, C. J. 55. Rodig, S. J., M. A. Meraz, J. M. White, P. A. Lampe, J. K. Riley, C. D. Guinosso, A. M. Kawasaki, P. D. Cook, and S. M. Freier. 1993. Eval- Arthur, K. L. King, K. C. Sheehan, L. Yin, D. Pennica, et al. 1998. uation of 29-modified oligonucleotides containing 29-deoxy gaps Disruption of the Jak1 gene demonstrates obligatory and non- as antisense inhibitors of . J. Biol. Chem. 268: redundant roles of the Jaks in cytokine-induced biologic responses. 14514–14522. Cell 93: 373–383. 39. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene 56. Sohn, S. J., K. Barrett, A. Van Abbema, C. Chang, P. B. Kohli, expression data using real-time quantitative PCR and the 2(-DDC(T)) H. Kanda, J. Smith, Y. Lai, A. Zhou, B. Zhang, et al. 2013. A restricted Method. Methods 25: 402–408. role for TYK2 catalytic activity in human cytokine responses revealed

40.Hong,D.,R.Kurzrock,Y.Kim,R.Woessner,A.Younes,J.Nemunaitis, by novel TYK2-selective inhibitors. J. Immunol. 191: 2205–2216. by guest on September 29, 2021 N.Fowler,T.Zhou,J.Schmidt,M.Jo,etal.2015.AZD9150,anext- 57. Stahl, N., T. G. Boulton, T. Farruggella, N. Y. Ip, S. Davis, B. A. generation antisense oligonucleotide inhibitor of STAT3 with early evi- Witthuhn, F. W. Quelle, O. Silvennoinen, G. Barbieri, S. Pellegrini, dence of clinical activity in lymphoma and lung cancer. Sci. Transl. Med. et al. 1994. Association and activation of Jak-Tyk kinases by CNTF- 7: 314ra185. LIF-OSM-IL-6 beta receptor components. Science 263: 92–95. 41. Ross, S. J., A. S. Revenko, L. L. Hanson, R. Ellston, A. Staniszewska, 58. Parganas, E., D. Wang, D. Stravopodis, D. J. Topham, J. C. Marine, N. Whalley, S. K. Pandey, M. Revill, C. Rooney, L. K. Buckett, et al. S. Teglund, E. F. Vanin, S. Bodner, O. R. Colamonici, J. M. van 2017. Targeting KRAS-dependent tumors with AZD4785, a high- Deursen, et al. 1998. Jak2 is essential for signaling through a variety of affinity therapeutic antisense oligonucleotide inhibitor of KRAS. Sci. cytokine receptors. Cell 93: 385–395. Transl. Med. 9: eaal5253. 59. Dieckmann, A., P. H. Hagedorn, Y. Burki, C. Brugmann,¨ M. Berrera, 42. Stein, C. A., J. B. Hansen, J. Lai, S. Wu, A. Voskresenskiy, A. Høg, M. Ebeling, T. Singer, and F. Schuler. 2018. A sensitive in vitro ap- J. Worm, M. Hedtjarn,¨ N. Souleimanian, P. Miller, et al. 2010. Effi- proach to assess the hybridization-dependent toxic potential of high cient gene silencing by delivery of locked nucleic acid antisense oli- affinity gapmer oligonucleotides. Mol. Ther. Nucleic Acids 10: 45–54. gonucleotides, unassisted by transfection reagents. Nucleic Acids Res. 60. Shen, W., C. L. De Hoyos, M. T. Migawa, T. A. Vickers, H. Sun, 38: e3. A. Low, T. A. Bell III., M. Rahdar, S. Mukhopadhyay, C. E. Hart, et al. 43. Silvennoinen, O., J. N. Ihle, J. Schlessinger, and D. E. Levy. 1993. 2019. Chemical modification of PS-ASO therapeutics reduces cellular Interferon-induced nuclear signalling by Jak protein tyrosine kinases. protein-binding and improves the therapeutic index. Nat. Biotechnol. Nature 366: 583–585. 37: 640–650. 44. Velichko, S., T. C. Wagner, J. Turkson, R. Jove, and E. Croze. 2002. 61. Geary, R. S., D. Norris, R. Yu, and C. F. Bennett. 2015. Pharmacoki- STAT3 activation by type I is dependent on specificty- netics, biodistribution and cell uptake of antisense oligonucleotides. rosines located in the cytoplasmic domain of interferon receptor chain Adv. Drug Deliv. Rev. 87: 46–51. 2c. Activation of multiple STATS proceeds through the redundant us- 62. Linnane, E., P. Davey, P. Zhang, S. Puri, M. Edbrooke, E. Chiarparin, age of two tyrosine residues. J. Biol. Chem. 277: 35635–35641. A. S. Revenko, A. R. Macleod, J. C. Norman, and S. J. Ross. 2019. 45. Platanias, L. C. 2005. Mechanisms of type-I- and type-II-interferon- Differential uptake, kinetics and mechanisms of intracellular traf- mediated signalling. Nat. Rev. Immunol. 5: 375–386. ficking of next-generation antisense oligonucleotides across human 46. van Boxel-Dezaire, A. H., M. R. Rani, and G. R. Stark. 2006. Complex cancer cell lines. Nucleic Acids Res. 47: 4375–4392. modulation of cell type-specific signaling in response to type I interferons. 63. Dowdy, S. F. 2017. Overcoming cellular barriers for RNA therapeutics. Immunity 25: 361–372. Nat. Biotechnol. 35: 222–229. 47. Karaghiosoff, M., H. Neubauer, C. Lassnig, P. Kovarik, H. Schindler, 64. Prakash, T. P., M. J. Graham, J. Yu, R. Carty, A. Low, A. Chappell, H. Pircher, B. McCoy, C. Bogdan, T. Decker, G. Brem, et al. 2000. K. Schmidt, C. Zhao, M. Aghajan, H. F. Murray, et al. 2014. Targeted

https://doi.org/10.4049/immunohorizons.2000108 80 POTENT AND SELECTIVE KNOCKDOWN OF TYK2 BY ANTISENSE OLIGOS ImmunoHorizons

delivery of antisense oligonucleotides to hepatocytes using trianten- 68. Sanda, T., J. W. Tyner, A. Gutierrez, V. N. Ngo, J. Glover, B. H. Chang, nary N-acetyl galactosamine improves potency 10-fold in mice. A. Yost, W. Ma, A. G. Fleischman, W. Zhou, et al. 2013. TYK2-STAT1- Nucleic Acids Res. 42: 8796–8807. BCL2 pathway dependence in T-cell acute lymphoblastic leukemia. 65. Crow, M. K. 2014. Type I interferon in the pathogenesis of lupus. J. Cancer Discov. 3: 564–577. Immunol. 192: 5459–5468. 69. Prutsch, N., E. Gurnhofer, T. Suske, H. C. Liang, M. Schlederer, 66. Sands, B. E., W. J. Sandborn, R. Panaccione, C. D. O’Brien, H. Zhang, S. Roos, L. C. Wu, I. Simonitsch-Klupp, A. Alvarez-Hernandez, J. Johanns, O. J. Adedokun, K. Li, L. Peyrin-Biroulet, G. Van Assche, C. Kornauth, et al. 2019. Dependency on the TYK2/STAT1/MCL1 axis et al; UNIFI Study Group. 2019. Ustekinumab as induction and main- in anaplastic large cell lymphoma. [Published erratum appears in tenance therapy for ulcerative colitis. N. Engl. J. Med. 381: 1201–1214. 2020 Leukemia 34: 3105.] Leukemia 33: 696–709. 67.Beebe,A.M.,D.J.Cua,andR.deWaalMalefyt.2002.Theroleof 70. Qin, W., A. Godec, X. Zhang, C. Zhu, J. Shao, Y. Tao, X. Bu, and A. C. interleukin-10 in autoimmune disease: systemic lupus erythematosus (SLE) Hirbe. 2019. TYK2 promotes malignant peripheral nerve sheath tumor and multiple sclerosis (MS). Cytokine Growth Factor Rev. 13: 403–412. progression through inhibition of cell death. Cancer Med. 8: 5232–5241. Downloaded from http://www.immunohorizons.org/ by guest on September 29, 2021

https://doi.org/10.4049/immunohorizons.2000108